1 DO NOT CITE OR QUOTE EPA/635/R-09/01 ID
www.epa.gov/iris
flL XH H H
Otrn
TOXICOLOGICAL REVIEW
OF
T richloroethylene
(CAS No. 79-01-6)
In Support of Summary Information on the
Integrated Risk Information System (IRIS)
June 2011
NOTICE
This document is a Final Agency/Interagency Review draft. This information is distributed
solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by U.S. EPA. It does not represent and should
not be construed to represent any Agency determination or policy. It is being circulated for
review of its technical accuracy and science policy implications.
U.S. Environmental Protection Agency
Washington, DC
-------
DISCLAIMER
This document is a preliminary draft for review purposes only. This information is
distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and
should not be construed to represent any Agency determination or policy. Mention of trade
names or commercial products does not constitute endorsement or recommendation for use.
This document is a draft for review purposes only and does not constitute Agency policy.
ii DRAFT—DO NOT CITE OR QUOTE
-------
GUIDE TO READERS OF THIS DOCUMENT
Due to the length of the TCE toxicological review, it is recommended that
Chapters 1 and 6 be read prior to Chapters 2-5.
Chapter 1 is the standard introduction to an IRIS Toxicological Review, describing the
purpose of the assessment and the guidelines used in its development.
Chapter 2 is an exposure characterization that summarizes information about TCE
sources, releases, media levels and exposure pathways for the general population (occupational
exposure is also discussed to a lesser extent).
Chapter 3 describes the toxicokinetics and physiologically based pharmacokinetic
(PBPK) modeling of TCE and metabolites (PBPK modeling details are in Appendix A).
Chapter 4 is the hazard characterization of TCE. Section 4.1 summarizes the evaluation
of epidemiologic studies of cancer and TCE (qualitative details in Appendix B; meta-analyses in
Appendix C). Each of the Sections 4.2-4.9 provides self-contained summary and syntheses of
the epidemiologic and laboratory studies on TCE and metabolites, organized by tissue/type of
effects, in the following order: genetic toxicity, central nervous system (CNS), kidney, liver,
immune system, respiratory tract, reproduction and development, and other cancers. Additional
details are provided in Appendix D for CNS effects and Appendix E for liver effects.
Section 4.10 summarizes the available data on susceptible lifestages and populations.
Section 4.11 describes the overall hazard characterization, including the weight of evidence for
noncancer effects and for carcinogenicity.
Chapter 5 is the dose-response assessment of TCE. Section 5.1 describes the dose-
response analyses for noncancer effects, and Section 5.2 describes the dose-response analyses for
cancer. Additional computational details are described in Appendix F for noncancer dose-
response analyses, Appendix G for cancer dose-response analyses based on rodent bioassays, and
Appendix H for cancer dose-response analyses based on human epidemiologic data.
Chapter 6 is the summary of the major conclusions in the characterization of TCE hazard
and dose response.
This document is a draft for review purposes only and does not constitute Agency policy.
iii DRAFT—DO NOT CITE OR QUOTE
-------
CONTENTS of TOXICOLOGICAL REVIEW for TRICHLOROETHYLENE
(CAS No. 79-01-6)
TOXICOLOGICAL REVIEW OF Trichloroethylene i
DISCLAIMER ii
GUIDE TO READERS OF THIS DOCUMENT iii
CONTENTS of TOXICOLOGICAL REVIEW for TRICHLOROETHYLENE iv
LIST OF TABLES xv
LIST OF FIGURES XXV
LIST OF ABBREVIATIONS AND ACRONYMS xxxii
FOREWORD xxxviii
AUTHORS, CONTRIBUTORS. AND REVIEWERS xxxix
EXECUTIVE SUMMARY xliv
1. INTRODUCTION 1
2. EXPOSURE CHARACTERIZATION 1
1 2.1. ENVIRONMENTAL SOURCES 2
2 2.2. ENVIRONMENTAL FATE 7
2.2.1. Fate in Terrestrial Environments 7
2.2.2. Fate in the Atmosphere 7
2.2.3. Fate in Aquatic Environments 8
3 2.3. EXPOSURE CONCENTRATIONS 8
2.3.1. Outdoor Air—Measured Level s 8
2.3.2. Outdoor Air—Modeled Levels 9
2.3.3. Indoor Air 12
This document is a draft for review purposes only and does not constitute Agency policy.
iv DRAFT—DO NOT CITE OR QUOTE
-------
2.3.4. Water 15
2.3.5. Other Media 17
2.3.6. Biological Monitoring 19
1 2.4. EXPOSURE PATHWAYS AND LEVELS 19
2.4.1. General Population 19
2 2.4.1.1.1. Inhalation 20
3 2.4.1.1.2. Ingestion 21
4 2.4.1.1.3. Dermal 22
5 2.4.1.1.4. Exposure to Trichloroethylene (TCE) Related Compounds 23
2.4.2. Potentially Highly Exposed Populations 24
6 2.4.2.1.1. Occupational Exposure 24
7 2.4.2.1.2. Consumer Exposure 26
2.4.3. Exposure Standards 26
8 2.5. EXPOSURE SUMMARY 27
3. TOXICOKINETICS 1
9 3.1. ABSORPTION 2
3.1.1. Oral 2
3.1.2. Inhalation 4
3.1.3. Dermal 11
10 3.2. DISTRIBUTION AND BODY BURDEN 13
11 3.3. METABOLISM 24
3.3.1. Introducti on 24
3.3.2. Extent of Metabolism 25
3.3.3. Pathways of Metabolism 28
12 3.3.3.1.1. Cytochrome P450-Dependent Oxidation 28
13 3.3.3.1.3. Formation of chloral hydrate (CH), trichloroethanol (TCOH)
14 and trichloroacetic acid (TCA) 33
15 3.3.3.1.8. Glutathione (GSH) Conjugation Pathway 46
16 3.3.3.1.17. Relative Roles of the Cytochrome P450 (CYP) and
17 Glutathione (GSH) Pathways 64
18 3.4. TRICHLOROETHYLENE (TCE) EXCRETION 68
3.4.1. Exhaled Air 68
3.4.2. Urine 71
3.4.3. Feces 73
19 3.5. PHYSIOLOGICALLY BASED PHARMACOKINETIC (PBPK)
20 MODELING OF TRICHLOROETHYLENE (TCE) AND ITS
21 METABOLITES 74
3.5.1. Introducti on 74
3.5.2. Previous Physiologically Based Pharmacokinetic (PBPK) Modeling of
Trichloroethylene (TCE) for Risk Assessment Application 74
This document is a draft for review purposes only and does not constitute Agency policy.
v DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
3.5.3. Development and Evaluation of an Interim "Harmonized"
Trichloroethylene (TCE) Physiologically Based Pharmacokinetic
(PBPK) Model 76
3.5.4. Physiologically Based Pharmacokinetic (PBPK) Model for
Trichloroethylene (TCE) and Metabolites Used for This Assessment 77
3.5.4.1.1. Introduction 77
3.5.4.1.2. Updated Physiologically Based Pharmacokinetic (PBPK)
Model Structure 80
3.5.4.1.3. Specification of Baseline Physiologically Based
Pharmacokinetic (PBPK) Model Parameter 84
3.5.4.1.4. Dose-Metric Predictions 85
3.5.5. Bayesian Estimation of Physiologically Based Pharmacokinetic
(PBPK) Model Parameters, and Their Uncertainty and Variability 86
3.5.5.1.1. Updated Pharmacokinetic Database 86
3.5.5.1.2. Updated Hierarchical Population Statistical Model and Prior
Distributions 94
3.5.5.1.3. Use of Interspecies Scaling to Update Prior Distributions in
the Absence of Other Data 96
3.5.5.1.4. Implementation 100
3.5.6. Evaluation of Updated Physiologically Based Pharmacokinetic
(PBPK) Model 100
3.5.6.1.1. Convergence 100
3.5.6.1.2. Evaluation of Posterior Parameter Distributions 103
3.5.6.1.3. Comparison of Model Predictions With Data 126
3.5.6.1.7. Sensitivity Analysis With Respect to Calibration Data 150
3.5.6.1.8. Summary Evaluation of Updated Physiologically Based
Pharmacokinetic (PBPK) Model 157
3.5.7. Physiologically Based Pharmacokinetic (PBPK) Model Dose-Metric
Predictions 157
3.5.7.1.1. Characterization of Uncertainty and Variability 157
3.5.7.1.2. Local Sensitivity Analysis With Respect to Dose-Metric
Predictions 158
3.5.7.1.3. Implications for the Population Pharmacokinetics of
Trichloroethylene (TCE) 174
3.5.7.1.6. Key Limitations and Potential Implications of Violating Key
Assumptions 189
3.5.7.1.7. Overall Evaluation of Physiologically Based
Pharmacokinetic (PBPK) Model-Based Internal Dose
Predictions 190
4. HAZARD CHARACTERIZATION 1
4.1. EPIDEMIOLOGIC STUDIES ON CANCER AND
TRICHLOROETHYLENE (TCE)—METHODOLOGICAL OVERVIEW 1
4.2. GENETIC TOXICITY 33
4.2.1. Trichloroethylene (TCE) 34
This document is a draft for review purposes only and does not constitute Agency policy.
vi DRAFT—DO NOT CITE OR QUOTE
-------
1 4.2.1.1.1. DNA Binding Studies 34
2 4.2.1.1.2. Bacterial Systems—Gene Mutations 36
3 4.2.1.1.3. Fungal and Yeast Systems—Gene Mutations, Conversions
4 and Recombination 40
5 4.2.1.1.4. Mammalian Systems Including Human Studies 44
6 4.2.1.1.14. Summary 57
4.2.2. Trichloroacetic Acid (TCA) 58
7 4.2.2.1.1. Bacterial Systems—Gene Mutations 59
8 4.2.2.1.2. Mammalian Systems 60
9 4.2.2.1.7. Summary 67
4.2.3. Dichloroacetic Acid (DCA) 67
10 4.2.3.1.1. Bacterial and Fungal Systems—Gene Mutations 67
11 4.2.3.1.2. Mammalian Systems 72
12 4.2.3.1.6. Summary 74
4.2.4. Chloral Hydrate 74
13 4.2.4.1.1. DNA Binding Studies 75
14 4.2.4.1.2. Bacterial and Fungal Systems—Gene Mutations 82
15 4.2.4.1.3. Mammalian Systems 83
16 4.2.4.1.9. Summary 86
4.2.5. Dichlorovinyl Cysteine (DCVC) and S-Dichlorovinyl Glutathione
(DCVG) 87
4.2.6. Trichloroethanol (TCOH) 93
4.2.7. Synthesis and Overall Summary 94
17 4.3. CENTRAL NERVOUS SYSTEM (CNS) TOXICITY 98
4.3.1. Alterations in Nerve Conduction 99
18 4.3.1.1.1. Trigeminal Nerve Function: Human Studies 99
19 4.3.1.1.2. Nerve Conduction Velocity—Human Studies 106
20 4.3.1.1.3. Trigeminal Nerve Function: Laboratory Animal Studies 106
21 4.3.1.1.4. Discussion and Conclusions: Trichloroethylene (TCE)-
22 Induced Trigeminal Nerve Impairment 109
4.3.2. Auditory Effects 110
23 4.3.2.1.1. Auditory Function: Human Studies 110
24 4.3.2.1.2. Auditory Function: Laboratory Animal Studies 114
25 4.3.2.1.3. Summary and Conclusion of Auditory Effects 119
4.3.3. Vestibular Function 121
26 4.3.3.1.1. Vestibular Function: Human Studies 121
27 4.3.3.1.2. Vestibular Function: Laboratory Animal Data 121
28 4.3.3.1.3. Summary and Conclusions for the Vestibular Function
29 Studies 123
4.3.4. Visual Effects 124
30 4.3.4.1.1. Visual Effects: Human Studies 124
31 4.3.4.1.2. Visual Effects: Laboratory Animal Data 127
32 4.3.4.1.3. Summary and Conclusion of Visual Effects 129
4.3.5. Cognitive Function 131
33 4.3.5.1.1. Cognitive Effects: Human Studies 131
34 4.3.5.1.2. Cognitive Effects: Laboratory Animal Studies 132
This document is a draft for review purposes only and does not constitute Agency policy.
vii DRAFT—DO NOT CITE OR QUOTE
-------
1 4.3.5.1.3. Summary and Conclusions of Cognitive Function Studies 133
4.3.6. Psychomotor Effects 138
2 4.3.6.1.1. Psychomotor Effects: Human Studies 139
3 4.3.6.1.4. Psychomotor Effects: Laboratory Animal Data 143
4 4.3.6.1.8. Summary and Conclusions for Psychomotor Effects 149
4.3.7. Mood Effects and Sleep Disorders 150
5 4.3.7.1.1. Effects on Mood: Human Studies 150
6 4.3.7.1.2. Effects on Mood: Laboratory Animal Findings 150
7 4.3.7.1.3. Sleep Disturbances 151
4.3.8. Developmental Neurotoxicity 152
8 4.3.8.1.1. Human Studies 152
9 4.3.8.1.2. Animal Studies 153
10 4.3.8.1.3. Summary and Conclusions for the Developmental
11 Neurotoxicity Studies 158
4.3.9. Mechani sti c Studi e s of Tri chl oroethyl ene (TCE) Neurotoxi city 159
12 4.3.9.1.1. Dopamine Neuron Disruption 159
13 4.3.9.1.5. Neurochemical and Molecular Changes 161
4.3.10. Potential Mechanisms for Tri chl oroethyl ene (TCE)-Mediated
Neurotoxicity 164
4.3.11. Overall Summary and Conclusions—Weight of Evidence 167
14 4.4. KIDNEY TOXICITY AM) CANCER 172
4.4.1. Human Studies of Kidney 172
15 4.4.1.1.1. Nonspecific Markers of Nephrotoxicity 172
16 4.4.1.1.2. End-Stage Renal Disease 179
4.4.2. Human Studies of Kidney Cancer 179
17 4.4.2.1.1. Studies of Job Titles and Occupations with Historical
18 Tri chl oroethyl ene (TCE) Usage 188
19 4.4.2.1.2. Cohort and Case-Controls Studies of Trichloroethylene
20 (TCE) Exposure 192
21 4.4.2.1.4. Examination of Possible Confounding Factors 197
22 4.4.2.1.5. Susceptible Populations—Kidney Cancer and
23 Trichloroethylene (TCE) Exposure 202
24 4.4.2.1.6. Meta-Analysis for Kidney Cancer 205
4.4.3. Human Studies of Somatic Mutation of von Hippel-Lindau (VHL)
Gene 211
4.4.4. Kidney Noncancer Toxicity in Laboratory Animals 217
4.4.5. Kidney Cancer in Laboratory Animals 228
25 4.4.5.1.1. Inhalation Studies of Trichloroethylene (TCE) 228
26 4.4.5.1.2. Gavage and Drinking Water Studies of Trichloroethylene
27 (TCE) 231
28 4.4.5.1.3. Conclusions: Kidney Cancer in Laboratory Animals 233
4.4.6. Role of Metabolism in Trichloroethylene (TCE) Kidney Toxicity 233
29 4.4.6.1.1. In Vivo Studies of the Kidney Toxicity of Trichloroethylene
30 (TCE) Metabolites 234
31 4.4.6.1.4. In Vitro Studies of Kidney Toxicity of Trichloroethylene
32 (TCE) and Metabolites 243
This document is a draft for review purposes only and does not constitute Agency policy.
viii DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
4.4.6.1.5. Conclusions as to the Active Agents of Trichloroethylene
(TCE)-Induced Nephrotoxicity 245
4.4.7. Mode(s) of Action for Kidney Carcinogenicity 246
4.4.7.1.1. Hypothesized Mode of Action: Mutagenicity 246
4.4.7.1.3. Hypothesized Mode of Action: Cytotoxicity and
Regenerative Proliferation 253
4.4.7.1.5. Additional Hypothesized Modes of Action with Limited
Evidence or Inadequate Experimental Support 256
4.4.7.1.9. Conclusions About the Hypothesized Modes of Action 258
4.4.8. Summary: Trichloroethylene (TCE) Kidney Toxicity, Carcinogenicity,
and Mode-of-Action 261
4.5. LIVER TOXICITY AND CANCER 264
4.5.1. Liver Noncancer Toxicity in Humans 264
4.5.2. Liver Cancer in Humans 269
4.5.3. Experimental Studies of Trichloroethylene (TCE) in Rodents—
Introduction 287
4.5.4. Trichloroethylene (TCE)-Induced Liver Noncancer Effects 290
4.5.4.1.1. Liver Weight 290
4.5.4.1.2. Cytotoxicity and Histopathology 301
4.5.4.1.3. Measures ofDNA Synthesis, Cellular Proliferation, and
Apoptosis 308
4.5.4.1.4. Peroxisomal Proliferation and Related Effects 311
4.5.4.1.5. Oxidative Stress 313
4.5.4.1.6. Bile Production 315
4.5.4.1.7. Summary: Trichloroethylene (TCE)-Induced Noncancer
Effects in Laboratory Animals 316
4.5.5. Trichloroethylene (TCE)-Induced Liver Cancer in Laboratory Animals 317
4.5.5.1.1. Negative or Inconclusive Studies of Mice and Rats 317
4.5.5.1.2. Positive Trichloroethylene (TCE) Studies of Mice 327
4.5.5.1.3. Summary: Trichloroethylene (TCE)-Induced Cancer in
Laboratory Animals 329
4.5.6. Role of Metabolism in Liver Toxicity and Cancer 329
4.5.6.1.1. Pharmacokinetics of Chloral Hydrate (CH), Trichloroacetic
Acid (TCA), and Dichloroacetic Acid (DCA) From
Trichloroethylene (TCE) Exposure 330
4.5.6.1.2. Comparisons Between Trichloroethylene (TCE) and
Trichloroacetic Acid (TCA), Dichloroacetic Acid (DCA),
and Chloral Hydrate (CH) Noncancer Effects 331
4.5.6.1.10. Comparisons of Trichloroethylene (TCE)-Induced
Carcinogenic Responses With Trichloroacetic Acid (TCA),
Dichloroacetic Acid (DCA), and Chloral Hydrate (CH)
Studies 352
4.5.6.1.15. Conclusions Regarding the Role of Trichloroacetic Acid
(TCA), Dichloroacetic Acid (DCA), and Chloral Hydrate
(CH) in Trichloroethylene (TCE)-Induced Effects in the
Liver 380
This document is a draft for review purposes only and does not constitute Agency policy.
ix DRAFT—DO NOT CITE OR QUOTE
-------
4.5.7. Mode of Action (MOA) for Trichloroethylene (TCE) Liver
Carcinogenicity 380
1 4.5.7.1.1. Mutagenicity 381
2 4.5.7.1.3. Peroxisome Proliferator Activated Receptor Alpha (PPARa)
3 Receptor Activation 3 82
4 4.5.7.1.6. Additional Proposed Hypotheses and Key Events with
5 Limited Evidence or Inadequate Experimental Support 392
6 4.5.7.1.15. Mode of Action (MOA) Conclusions 402
7 4.6. IMMUNOTOXICITY AND CANCERS OF THE IMMUNE SYSTEM 412
4.6.1. Human Studies 399
8 4.6.1.1.1. Noncancer Immune-Related Effects 399
9 4.6.1.1.6. Cancers of the Immune System, Including Childhood
10 Leukemia 413
4.6.2. Animal Studies 453
11 4.6.2.1.1. Immunosuppression 453
12 4.6.2.1.5. Hypersensitivity 463
13 4.6.2.1.6. Autoimmunity 467
14 4.6.2.1.7. Cancers of the Immune System 481
4.6.3. Summary 484
15 4.6.3.1.1. Noncancer Effects 484
16 4.6.3.1.2. Cancer 486
17 4.7. RESPIRATORY TRACT TOXICITY AND CANCER 488
4.7.1. Epidemiologic Evidence 488
18 4.7.1.1.1. Chronic Effects: Inhalation 488
19 4.7.1.1.2. Cancer 489
4.7.2. Laboratory Animal Studies 501
20 4.7.2.1.1. Respiratory Tract Animal Toxicity 501
21 4.7.2.1.3. Respiratory Tract Cancer 510
4.7.3. Role of Metabolism in Pulmonary Toxicity 515
4.7.4. Mode of Action for Pulmonary Carcinogenicity 520
22 4.7.4.1.1. Mutagenicity via Oxidative Metabolism 520
23 4.7.4.1.3. Cytotoxicity Leading to Increased Cell Proliferation 522
24 4.7.4.1.5. Additional Hypothesized Modes of Action with Limited
25 Evidence or Inadequate Experimental Support 524
26 4.7.4.1.7. Conclusions About the Hypothesized Modes of Action 525
4.7.5. Summary and Conclusions 527
27 4.8. REPRODUCTIVE AND DEVELOPMENTAL TOXICITY 529
4.8.1. Reproductive Toxicity 529
28 4.8.1.1.1. Human Reproductive Outcome Data 530
29 4.8.1.1.6. Animal Reproductive Toxicity Studies 538
30 4.8.1.1.9. Discussion/Synthesis of Noncancer Reproductive Toxicity
31 Findings 553
4.8.2. Cancers of the Reproductive System 561
32 4.8.2.1.1. Human Data 561
This document is a draft for review purposes only and does not constitute Agency policy.
x DRAFT—DO NOT CITE OR QUOTE
-------
1 4.8.2.1.5. Animal Studies 578
2 4.8.2.1.6. Mode of Action for Testicular Tumors 581
4.8.3. Developmental Toxicity 582
3 4.8.3.1.1. Human Developmental Data 582
4 4.8.3.1.5. Animal Developmental Toxicology Studies 608
5 4.8.3.1.9. Discussion/Synthesis of Developmental Data 635
6 4.9. OTHER SITE-SPECIFIC CANCERS 653
4.9.1. Esophageal Cancer 653
4.9.2. Bladder Cancer 664
4.9.3. Central Nervous System and Brain Cancers 671
7 4.10. SUSCEPTIBLE LIFESTAGES AND POPULATIONS 678
4.10.1. Lifestages 678
8 4.10.1.1.1. Early Lifestages 679
9 4.10.1.1.5. Later Lifestages 690
4.10.2. Other Susceptibility Factors 691
10 4.10.2.1.1. Gender 691
11 4.10.2.1.4. Genetic Variability 698
12 4.10.2.1.8. Race/Ethnicity 699
13 4.10.2.1.9. Preexisting Health Status 700
14 4.10.2.1.13. Lifestyle Factors and Nutrition Status 701
15 4.10.2.1.19. Mixtures 704
4.10.3. Uncertainty of Database and Research Needs for Susceptible
Populations 705
16 4.11. HAZARD CHARACTERIZATION 706
4.11.1. Characterization of Noncancer Effects 706
17 4.11.1.1.1. Neurotoxicity 706
18 4.11.1.1.2. Kidney Toxicity 711
19 4.11.1.1.3. Liver Toxicity 712
20 4.11.1.1.4. Immunotoxicity 714
21 4.11.1.1.5. Respiratory Tract Toxicity 715
22 4.11.1.1.6. Reproductive Toxicity 716
23 4.11.1.1.7. Developmental Toxicity 717
4.11.2. Characterization of Carcinogenicity 721
24 4.11.2.1.1. Summary Evaluation of Epidemiologic Evidence of
25 Trichloroethylene (TCE) and Cancer 722
26 4.11.2.1.12. Summary of Evidence for Trichloroethylene (TCE)
27 Carcinogenicity in Rodents 731
28 4.11.2.1.13. Summary of Additional Evidence on Biological
29 Plausibility 734
4.11.3. Characterization ofFactors Impacting Susceptibility 739
5. DOSE-RESPONSE ASSESSMENT 1
30 5.1. DOSE-RESPONSE ANALYSES FOR NONCANCER ENDPOINTS 1
This document is a draft for review purposes only and does not constitute Agency policy.
xi DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
5.1.1. Modeling Approaches and Uncertainty Factors for Developing
Candidate Reference Values Based on Applied Dose 3
5.1.2. Candidate Critical Effects by Effect Domain 7
5.1.2.1.1. Candidate Critical Neurological Effects on the Basis of
Applied Dose 8
5.1.2.1.2. Candidate Critical Kidney Effects on the Basis of Applied
Dose 16
5.1.2.1.3. Candidate Critical Liver Effects on the Basis of Applied
Dose 22
5.1.2.1.4. Candidate Critical Body Weight Effects on the Basis of
Applied Dose 23
5.1.2.1.5. Candidate Critical Immunological Effects on the Basis of
Applied Dose 24
5.1.2.1.6. Candidate Critical Respiratory Tract Effects on the Basis of
Applied Dose 29
5.1.2.1.7. Candidate Critical Reproductive Effects on the Basis of
Applied Dose 30
5.1.2.1.10. Candidate Critical Developmental Effects on the Basis of
Applied Dose 42
5.1.2.1.11. Summary of cRfCs, cRfDs, and Candidate Critical Effects 54
5.1.3. Application of Physiologically Based Pharmacokinetic (PBPK) Model
to Inter- and Intraspecies Extrapolation for Candidate Critical Effects 54
5.1.3.1.1. Selection of Dose-metrics for Different Endpoints 58
5.1.3.1.7. Methods for Inter- and Intraspecies Extrapolation Using
Internal Doses 66
5.1.3.1.8. Results and Discussion of p-RfCs and p-RfDs for Candidate
Critical Effects 85
5.1.4. Uncertainties in cRfCs and cRfDs 86
5.1.4.1.1. Qualitative Uncertainties 86
5.1.4.1.2. Quantitative Uncertainty Analysis of Physiologically Based
Pharmacokinetic (PBPK) Model-Based Dose-metrics for
Lowest-Ob served-Adverse-Effect Level (LOAEL) or
No-Observed-Adverse-Effect Level (NOAEL)-Based Points
of Departure (PODs) 89
5.1.5. Summary of Noncancer Reference Values 101
5.1.5.1.1. Preferred Candidate Reference Values (cRfCs, cRfD,
p-cRfCs and p-cRfDs) for Candidate Critical Effects 101
5.1.5.1.2. Reference Concentration 108
5.1.5.1.3. Reference Dose 111
5.2. DOSE-RESPONSE ANALYSIS FOR CANCER ENDPOINTS 115
5.2.1. Dose-Response Analyses: Rodent Bioassays 115
5.2.1.1.1. Rodent Dose-Response Analyses: Studies and Modeling
Approaches 116
5.2.1.1.2. Rodent Dose-Response Analyses: Dosimetry 125
5.2.1.1.5. Rodent Dose-Response Analyses: Results 134
This document is a draft for review purposes only and does not constitute Agency policy.
xii DRAFT—DO NOT CITE OR QUOTE
-------
1 5.2.1.1.6. Uncertainties in Dose-Response Analyses of Rodent
2 Bioassays 146
5.2.2. Dose-Response Analyses: Human Epidemiologic Data 154
3 5.2.2.1.1. Inhalation Unit Risk Estimate for Renal Cell Carcinoma
4 Derived from Charbotel et al. (2006) Data 158
5 5.2.2.1.6. Adjustment of the Inhalation Unit Risk Estimate for Multiple
6 Sites 167
7 5.2.2.1.7. Route-to-Route Extrapolation Using Physiologically Based
8 Pharmacokinetic (PBPK) Model 171
5.2.3. Summary of Unit Risk Estimates 177
9 5.2.3.1.1. Inhalation Unit Risk Estimate 177
10 5.2.3.1.2. Oral Slope Factor Estimate 178
11 5.2.3.1.3. Application of Age-Dependent Adjustment Factors 179
12 5.3. KEY RESEARCH NEEDS FOR TCE DOSE-RESPONSE ANALYSES 189
6. MAJOR CONCLUSIONS IN THE CHARACTERIZATION OF HAZARD AND
DOSE RESPONSE 1
13 6.1. HUMAN HAZARD POTENTIAL 1
6.1.1. Exposure (see Section 2) 1
6.1.2. Toxicokinetics and Physiologically Based Pharmacokinetic Modeling
(see Section 3 and Appendix A) 2
6.1.3. Noncancer Toxicity 4
14 6.1.3.1.1. Neurological Effects (see Sections 4.3 and 4.11.1.1 and
15 Appendix D) 4
16 6.1.3.1.2. Kidney Effects (see Sections 4.4.1, 4.4.4, 4.4.6, and
17 4.11.1.2) 5
18 6.1.3.1.3. Liver Effects (see Sections 4.5.1, 4.5.3, 4.5.4, 4.5.6, and
19 4.11.1.3, and Appendix E) 6
20 6.1.3.1.4. Immunological Effects (see Sections 4.6.1.1, 4.6.2, and
21 4.11.1.4) 7
22 6.1.3.1.5. Respiratory Tract Effects (see Sections 4.7.1.1, 4.7.2.1,
23 4.7.3, and 4.11.1.5) 8
24 6.1.3.1.6. Reproductive Effects (see Sections 4.8.1 and 4.11.1.6) 9
25 6.1.3.1.7. Developmental Effects (see Sections 4.8.3 and 4.11.1.7) 10
6.1.4. Carcinogenicity (see Sections 4.1, 4.2, 4.4.2, 4.4.5, 4.4.7, 4.5.2, 4.5.5,
4.5.6, 4.5.7, 4.6.1.2, 4.6.2.4, 4.7.1.2, 4.7.2.2, 4.7.4, 4.8.2, 4.9, and
4.11.2, and Appendices B and C) 12
6.1.5. Susceptibility (see Sections 4.10 and 4.11.3) 17
26 6.2. DOSE-RESPONSE ASSESSMENT 19
6.2.1. Noncancer Effects (see Section 5.1) 19
27 6.2.1.1.1. Background and Methods 19
28 6.2.1.1.2. Uncertainties and Application of Uncertainty Factors (UFs)
29 (see Section 5.1.1 and 5.1.4) 20
This document is a draft for review purposes only and does not constitute Agency policy.
xiii DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
6.2.1.1.3. Candidate Critical Effects and Reference Values (see
Sections 5.1.2 and 5.1.3) 24
6.2.1.1.11. Noncancer Reference Values (see Section 5.1.5) 30
6.2.2. Cancer (see Section 5.2) 33
6.2.2.1.1. Background and Methods (rodent: see Section 5.2.1.1;
human: see Section 5.2.2.1) 33
6.2.2.1.2. Inhalation Unit Risk Estimate (rodent: see Section 5.2.1.3;
human: see Section 5.2.2.1 and 5.2.2.2) 34
6.2.2.1.3. Oral Slope Factor Estimate (rodent: see Section 5.2.1.3;
human: see Section 5.2.2.3) 36
6.2.2.1.4. Uncertainties in Cancer Dose-Response Assessment 38
6.2.2.1.7. Application of Age-Dependent Adjustment Factors (see
Section 5.2.3.3) 43
6.3. OVERALL CHARACTERIZATION OF TCE HAZARD AND DOSE
RESPONSE 44
7. References 1
APPENDIX A: PBPK MODELING OF TCE AND METABOLITES-DETAILED
METHODS AND RESULTS A-l
APPENDIX B: SYSTEMATIC REVIEW OF EPIDEMIOLOGIC STUDIES ON
CANCER AND TRICHLOROETHYLENE (TCE) EXPOSURE B-l
APPENDIX C: META-ANALYSIS OF CANCER RESULTS FROM
EPIDEMIOLOGICAL STUDIES C-l
APPENDIX D: NEUROLOGICAL EFFECTS OF TRICHLOROETHYLENE D-l
APPENDIX E: ANALYSIS OF LIVER AND COEXPOSURE ISSUES FOR
THE TCE TOXICOLOGICAL REVIEW E-l
APPENDIX F: TCE NONCANCER DOSE-RESPONSE ANALYSES F-l
APPENDIX G: TCE CANCER DOSE-RESPONSE ANALYSES WITH RODENT
CANCER BIO AS SAY DATA G-l
APPENDIX H: LIFETABLE ANALYSIS AND WEIGHTED LINEAR REGRESSION
BASED ON RESULTS FROM CHARBOTEL ET AL II-l
APPENDIX I: EPA RESPONSE TO MAJOR PEER REVIEW AND PUBLIC
COMMENTS 1-1
This document is a draft for review purposes only and does not constitute Agency policy.
xiv DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
LIST OF TABLES
Table 2-1. TCE metabolites and related parent compounds51 1
Table 2-2. Chemical properties of TCE 2
Table 2-3. Properties and uses of TCE related compounds 3
Table 2-4. TRI releases of TCE (pounds/year) 5
Table 2-5. Concentrations of trichloroethylene in ambient air 10
Table 2-6. TCE ambient air monitoring data ((J,g/m ) 11
Table 2-7. Mean TCE air levels across monitors by land setting and use (1985-1998) 11
Table 2-8. Concentrations of trichloroethylene in water based on pre-1990 studies 15
Table 2-9. Levels in food 18
Table 2-10. TCE levels in whole blood by population percentile 19
"3
Table 2-11. Modeled 1999 annual exposure concentrations ((J,g/m ) for trichloroethylene 20
Table 2-12. Preliminary estimates of TCE intake from food ingestion 22
Table 2-13. Preliminary intake estimates of TCE and TCE-related chemicals 23
Table 2-14. Years of solvent use in industrial degreasing and cleaning operations 25
Table 2-15. TCE standards 27
Table 3-1. Blood:air PC values for humans 6
Table 3-2. Blood:air PC values for rats and mice 7
Table 3-3. Air and blood concentrations during exposure to TCE in humans (Astrand and
Ovrum, 1976) 9
Table 3-4. Retention of inhaled TCE vapor in humans (Jakubowski and Wieczorek, 1988) 10
Table 3-5. Uptake of TCE in human volunteers following 4 hour exposure to 70 ppm (Monster
etal., 1979) 10
Table 3-6. Concentrations of TCE in maternal and fetal blood at birth 15
Table 3-7. Distribution of TCE to rat tissues" following inhalation exposure (Savolainen et al.,
1977) 17
Table 3-8. Tissue:blood partition coefficient values for TCE 18
Table 3-9. Age-dependence of tissue:air partition coefficients in rats 21
Table 3-10. Predicted maximal concentrations of TCE in rat blood following a 6-hour inhalation
exposure (Rodriguez et al., 2007) 21
Table 3-11. Tissue distribution of TCE metabolites following inhalation exposure 22
Table 3-12. Binding of 14C from [14C]TCE in rat liver and kidney at 72 hours after oral
administration of 200 mg/kg [14C]TCE (Dekant et al., 1986b) 24
Table 3-13. In vitro TCE oxidative metabolism in hepatocytes and microsomal fractions 30
This document is a draft for review purposes only and does not constitute Agency policy.
XV DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Table 3-14. In vitro kinetics of trichloroethanol and trichloroacetic acid formation from chloral
hydrate in rat, mouse, and human liver homogenates 36
Table 3-15. In vitro kinetics of DCA metabolism in hepatic cytosol of mice, rats, and humans 38
Table 3-16. TCOH and TCA formed from CH in vitro in lysed whole blood of rats and mice or
fractionated blood of humans (nmoles formed in 400 [j,L samples over 30 minutes) 40
Table 3-17. Reported TCA plasma binding parameters 41
Table 3-18. Partition coefficients for TCE oxidative metabolites 41
Table 3-19. Urinary excretion of trichloroacetic acid by various species exposed to
trichloroethylene [based on data reviewed in (Fisher et al., 1991)] 44
Table 3-20. P450 isoform kinetics for metabolism of TCE to CH in human, rat, and mouse
recombinant P450s 45
Table 3-21. P450 isoform activities in human liver microsomes exhibiting different affinities for
TCE 45
Table 3-22. Comparison of peak blood concentrations in humans exposed to 100 ppm (537
mg/m3) TCE for 4 hours (Fisher et al., 1998; Lash et al., 1999a) 50
Table 3-23. GSH conjugation of TCE (at 1-2 mM) in liver and kidney cellular fractions in
humans, male F344 rats, and male B6C3F1 mice from Lash laboratory 51
Table 3-24. Kinetics of TCE metabolism via GSH conjugation in male F344 rat kidney and
human liver and kidney cellular and subcellular fractions from Lash laboratory 52
Table 3-25. GSH conjugation of TCE (at 1.4-4 mM) in liver and kidney cellular fractions in
humans, male F344 rats, and male B6C3F1 mice from Green and Dekant laboratories 53
Table 3-26. GGT activity in liver and kidney subcellular fractions of mice, rats, and humans .. 61
Table 3-27. Multispecies comparison of whole-organ activity levels of GGT and dipeptidase.. 62
Table 3-28. Comparison of hepatic in vitro oxidation and conjugation of TCE 67
Table 3-29. Estimates of DCVG in blood relative to inhaled TCE dose in humans exposed to 50
and 100 ppm (269 and 537 mg/m3) (Fisher et al., 1998; Lash et al., 1999b) 68
Table 3-30. Concentrations of TCE in expired breath from inhalation-exposed humans (Astrand
and Ovrum, 1976) 71
Table 3-31. Conclusions from evaluation of Hack et al. (2006), and implications for PBPK
model development 78
Table 3-32. Discussion of changes to the Hack et al. (2006) PBPK model implemented for this
assessment 83
Table 3-33. PBPK model-based dose-metrics 87
Table 3-34. Rodent studies with pharmacokinetic data considered for analysis 88
Table 3-35. Human studies with pharmacokinetic data considered for analysis 92
This document is a draft for review purposes only and does not constitute Agency policy.
xvi DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Table 3-36. Parameters for which scaling from mouse to rat, or from mouse and rat to human,
was used to update the prior distributions 99
Table 3-37. Prior and posterior uncertainty and variability in mouse PBPK model parametersl06
Table 3-38. Prior and posterior uncertainty and variability in rat PBPK model parameters Ill
Table 3-39. Prior and posterior uncertainty and variability in human PBPK model parameters
116
Table 3-40. Confidence interval (CI) widths (ratio of 97.5% to 2.5% estimates) and fold-shift in
median estimate for the PBPK model population median parameters, sorted in order of
decreasing CI width. Shifts in the median estimate greater than threefold are in bold to denote
larger shifts between the prior and posterior distributions 121
Table 3-41. Estimates of the residual-error 128
Table 3-42. Summary comparison of updated PBPK model predictions and in vivo data in mice
132
Table 3-43. Summary comparison of updated PBPK model predictions and in vivo data used for
"calibration" in rats 138
Table 3-44. Summary comparison of updated PBPK model predictions and in vivo data used for
"out-of-sample" evaluation in rats 141
Table 3-45. Summary comparison of updated PBPK model predictions and in vivo data used for
"calibration" in humans 147
Table 3-46. Summary comparison of updated PBPK model predictions and in vivo data used for
"out-of-sample" evaluation in humans 149
Table 3-47. Summary of scaling parameters ordered by fraction of calibration data of moderate
or high sensitivity 155
Table 3-48. Posterior predictions for representative internal doses: mouse 169
Table 3-49. Posterior predictions for representative internal doses: rat 170
Table 3-50. Posterior predictions for representative internal doses: human 171
Table 3-51. Degree of variance in dose-metric predictions due to incomplete convergence
(columns 2-4), combined uncertainty and population variability (columns 5-7), uncertainty in
particular human population percentiles (columns 8-10), model fits to in vivo data (column 11).
The GSD is the geometric standard deviation, which is a "fold-change" from the central
tendency 192
Table 4-1. Description of epidemiologic cohort and proportionate mortality ratio (PMR) studies
assessing cancer and TCE exposure 2
Table 4-2. Case-control epidemiologic studies examining cancer and TCE exposure 9
Table 4-3. Geographic-based studies assessing cancer and TCE exposure 22
This document is a draft for review purposes only and does not constitute Agency policy.
xvii DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Table 4-4. Standards of epidemiologic study design and analysis use for identifying cancer
hazard and TCE exposure 25
Table 4-5. Summary of criteria for meta-analysis study selection 29
Table 4-6. TCE genotoxicity: bacterial assays 38
Table 4-7. TCE genotoxicity: fungal and yeast systems 42
Table 4-8. TCE genotoxicity: mammalian systems—gene mutations and chromosome
aberrations 46
Table 4-9. TCE genotoxicity: mammalian systems—micronucleus, sister chromatic exchanges
50
Table 4-10. TCE genotoxicity: mammalian systems—unscheduled DNA synthesis, DNA strand
breaks/protein crosslinks, cell transformation 55
Table 4-11. Genotoxicity of trichloroacetic acid—bacterial systems 59
Table 4-12. TCA Genotoxicity—mammalian systems (both in vitro and in vivo) 62
Table 4-13. Genotoxicity of dichloroacetic acid (bacterial systems) 69
Table 4-14. Genotoxicity of dichloroacetic acid—mammalian systems 70
Table 4-15. Chloral hydrate genotoxicity: bacterial, yeast and fungal systems 76
Table 4-16.. Chloral hydrate genotoxicity: mammalian systems—all genetic endpoints, in vitro
78
Table 4-17.. Chloral hydrate genotoxicity: mammalian systems—all genetic damage, in vivo . 80
Table 4-18. TCE GSH conjugation metabolites genotoxicity 89
Table 4-19. Genotoxicity of trichloroethanol 93
Table 4-20. Summary of human trigeminal nerve and nerve conduction velocity studies 100
Table 4-21. Summary of animal trigeminal nerve studies 108
Table 4-22. Summary of human auditory function studies 112
Table 4-23. Summary of animal auditory function studies 117
Table 4-26. Summary of animal visual system studies 129
Table 4-27. Summary of human cognition effect studies 134
Table 4-28. Summary of animal cognition effect studies 137
Table 4-29. Summary of human choice reaction time studies 141
Table 4-30. Summary of animal psychomotor function and reaction time studies 145
Table 4-31. Summary of animal locomotor activity studies 147
Table 4-33. Summary of human developmental neurotoxicity associated with TCE exposures
152
Table 4-34. Summary of mammalian in vivo developmental neurotoxicity studies—oral
exposures 154
Table 4-35. Summary of animal dopamine neuronal studies 160
This document is a draft for review purposes only and does not constitute Agency policy.
xviii DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Table 4-36. Summary of neurophysiological, neurochemical, and neuropathological effects with
TCE exposure 161
Table 4-37. Summary of in vitro ion channel effects with TCE exposure 165
Table 4-38. Summary of human kidney toxicity studies 174
Table 4-39. Summary of human studies on TCE exposure and kidney cancer 181
Table 4-40. Summary of case-control studies on kidney cancer and occupation or job title .... 189
Table 4-41. Summary of lung and kidney cancer risks in active smokers (from IARC, 2004b)
200
Table 4-42. Summary of human studies on somatic mutations of the VHL genea 214
Table 4-43. Inhalation studies of kidney noncancer toxicity in laboratory animals 219
Table 4-44. Oral and i.p. studies of kidney noncancer toxicity in laboratory animals 220
Table 4-46. Summary of renal toxicity and tumor findings in gavage studies of trichloroethylene
by NCI (1976) 224
Table 4-47. Summary of renal toxicity findings in gavage studies of trichloroethylene by
Maltoni et al. (1988) 225
Table 4-48. Summary of renal toxicity and tumor incidence in gavage studies of
trichloroethylene by NTP (1988) 226
Table 4-49. Summary of renal toxicity and tumor findings in inhalation studies of
trichloroethylene by Maltoni et al. (1988)a 227
Table 4-50. Summary of renal tumor findings in inhalation studies of trichloroethylene by
Henschler et al. (1980)a and Fukuda et al. (1983)b 230
Table 4-51. Summary of renal tumor findings in gavage studies of trichloroethylene by
Henschler et al. (1984)a and Van Duuren et al. (1979)b 232
Table 4-52. Laboratory animal studies of kidney noncancer toxicity of TCE metabolites 236
Table 4-53. Summary of histological changes in renal proximal tubular cells induced by chronic
exposure to TCE, DCVC, and TCOH 238
Table 4-54. Summary of major mode of action conclusions for TCE kidney carcinogenesis .. 248
Table 4-55. Summary of human liver toxicity studies 265
Table 4-56. Selected results from epidemiologic studies of TCE exposure and cirrhosis 271
Table 4-57. Selected results from epidemiologic studies of TCE exposure and liver cancer ... 274
Table 4-58. Oral studies of TCE-induced liver effects in mice and rats 292
Table 4-59. Inhalation and i.p. studies of TCE-induced liver effects in mice and rats 295
Table 4-60. Summary of liver tumor findings in gavage studies of trichloroethylene by NTP
(1990 )a 319
Table 4-61. Summary of liver tumor findings in gavage studies of trichloroethylene by NCI
(1976) 320
This document is a draft for review purposes only and does not constitute Agency policy.
xix DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Table 4-62. Summary of liver tumor incidence in gavage studies of trichloroethylene by NTP
(1988) 321
Table 4-63. Summary of liver tumor findings in inhalation studies of trichloroethylene by
Maltoni et al. (1988)a 323
Table 4-64. Summary of liver tumor findings in inhalation studies of trichloroethylene by
Henschler et al. (1980)a and Fukuda et al. (1983) 324
Table 4-65. Summary of liver tumor findings in gavage studies of trichloroethylene by
Henschler et al. (1984)a 325
Table 4-66. Potency indicators for mouse hepatocarcinogenicity and in vitro transactivation of
mouse PPARa for four PPARa agonists 390
Table 4-67. Potency indicators for rat hepatocarcinogenicity and common short-term markers of
PPARa activation for four PPARa agonists 391
Table 4-68. Summary of mode of action conclusions for TCE-induced liver carcinogenesis .. 403
Table 4-69. Studies of immune parameters (IgE antibodies and cytokines) and trichloroethylene
in humans 401
Table 4-70. Case-control studies of autoimmune diseases with measures of trichloroethylene
exposure 411
Table 4-71. Incidence cohort studies of TCE exposure and lymphopoietic and hematopoietic
cancer risk 418
Table 4-72. Mortality cohort and PMR studies of TCE exposure and lymphopoietic and
hematopoietic cancer risk 422
Table 4-73. Case-control studies of TCE exposure and lymphopoietic cancer, leukemia or
multiple myeloma 433
Table 4-74. Geographic-based studies of TCE and non-Hodgkin lymphoma or leukemia in
adults 438
Table 4-75. Selected results from epidemiologic studies of TCE exposure and childhood
leukemia 443
Table 4-76. Summary of TCE immunosuppression studies 455
Table 4-77. Summary of TCE hypersensitivity studies 465
Table 4-78. Summary of autoimmune-related studies of TCE and metabolites in mice and rats
(by sex, strain, and route of exposure)51 469
Table 4-79. Malignant lymphomas incidence in mice exposed to TCE in gavage and inhalation
exposure studies 482
Table 4-80. Leukemia incidence in rats exposed to TCE in gavage and inhalation exposure
studies 483
Table 4-81. Selected results from epidemiologic studies of TCE exposure and lung cancer.... 491
This document is a draft for review purposes only and does not constitute Agency policy.
xx DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Table 4-82. Selected results from epidemiologic studies of TCE exposure and laryngeal cancer
498
Table 4-83. Animal toxicity studies of trichloroethylene 503
Table 4-84. Animal carcinogenicity studies of trichloroethylene 512
Table 4-85. Human reproductive effects 531
Table 4-86. Summary of mammalian in vivo reproductive toxicity studies—inhalation exposures
539
Table 4-87. Summary of mammalian in vivo reproductive toxicity studies—oral exposures .. 541
Table 4-88. Summary of adverse female reproductive outcomes associated with TCE exposures
553
Table 4-89. Summary of adverse male reproductive outcomes associated with TCE exposures
555
Table 4-90. Summary of human studies on TCE exposure and prostate cancer 562
Table 4-91. Summary of human studies on TCE exposure and breast cancer 567
Table 4-92. Summary of human studies on TCE exposure and cervical cancer 571
Table 4-93. Histopathology findings in reproductive organs 580
Table 4-94. Testicular tumors in male rats exposed to TCE, adjusted for reduced survivaf.... 581
Table 4-95. Developmental studies in humans 584
Table 4-96. Summary of mammalian in vivo developmental toxicity studies—inhalation
exposures 608
Table 4-97. Ocular defects observed (Narotsky et al., 1995) 610
Table 4-98. Summary of mammalian in vivo developmental toxicity studies—oral exposures 612
Table 4-99. Types of congenital cardiac defects observed in TCE-exposed fetuses (Dawson et
al., 1993, Table 3) 621
Table 4-100. Types of heart malformations per 100 fetuses (Johnson et al., 2003, Table 2, p.
290) 622
Table 4-101. Congenital cardiac malformations (Johnson et al., 1998a, Table 2, p. 997) 625
Table 4-102. Summary of adverse fetal and early neonatal outcomes associated with TCE
exposures 636
Table 4-103. Summary of studies that identified cardiac malformations associated with TCE
exposures 637
Table 4-104. Events in cardiac valve formation in mammals and birds3 641
Table 4-105. Summary of other structural developmental outcomes associated with TCE
exposures 645
Table 4-106. Summary of developmental neurotoxicity associated with TCE exposures 647
Table 4-107. Summary of developmental immunotoxicity associated with TCE exposures.... 650
This document is a draft for review purposes only and does not constitute Agency policy.
xxi DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Table 4-108. Summary of childhood cancers associated with TCE exposures 652
Table 4-109. Selected observations from case-control studies of TCE exposure and esophageal
cancer 655
Table 4-110. Summary of human studies on TCE exposure and esophageal cancer 658
Table 4-111. Summary of human studies on TCE exposure and bladder cancer 666
Table 4-112. Summary of human studies on TCE exposure and brain cancer 673
Table 4-113. Estimated lifestage-specific daily doses for TCE in watera 682
Table. 5-1 Summary of studies of neurological effects suitable for dose-response assessment.... 9
Table 5-2. Neurological effects in studies suitable for dose-response assessment, and
corresponding cRfCs and cRfDs 11
Table 5-3. Summary of studies of kidney, liver, and body weight effects suitable for
dose-response assessment 17
Table 5-4. Kidney, liver, and body weight effects in studies suitable for dose-response
assessment, and corresponding cRfCs and cRfDs 19
Table 5-5. Summary of studies of immunological effects suitable for dose-response assessment
26
Table 5-6. Immunological effects in studies suitable for dose-response assessment, and
corresponding cRfCs and cRfDs 28
Table 5-7. Summary of studies of reproductive effects suitable for dose-response assessment. 32
Table 5-8. Reproductive effects in studies suitable for dose-response assessment, and
corresponding cRfCs and cRfDs 37
Table 5-9. Summary of studies of developmental effects suitable for dose-response assessment
45
Table 5-10. Developmental effects in studies suitable for dose-response assessment, and
corresponding cRfCs and cRfDs 49
Table 5-11. Ranges of cRfCs based on applied dose for various noncancer effects associated
with inhalation TCE exposure 56
Table 5-12. Ranges of cRfDs based on applied dose for various noncancer effects associated
with oral TCE exposure 57
Table 5-13. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on
PBPK modeled internal dose-metrics) for candidate critical neurological effects 71
Table 5-14. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on
PBPK modeled internal dose-metrics) for candidate critical kidney effects 73
Table 5-15. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on
PBPK modeled internal dose-metrics) for candidate critical liver effects 76
This document is a draft for review purposes only and does not constitute Agency policy.
xxii DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Table 5-16. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on
PBPK modeled internal dose-metrics) for candidate critical immunological effects 77
Table 5-17. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on
PBPK modeled internal dose-metrics) for candidate critical reproductive effects 79
Table 5-18. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on
PBPK modeled internal dose-metrics) for candidate critical developmental effects 82
Table 5-19. Comparison of "sensitive individual" HECs or HEDs for neurological effects based
on PBPK modeled internal dose-metrics at different levels of confidence and sensitivity, at the
NOAEL or I .OA HI. 93
Table 5-20. Comparison of "sensitive individual" HECs or HEDs for kidney and liver effects
based on PBPK modeled internal dose-metrics at different levels of confidence and sensitivity, at
the NOAEL or LOAEL 94
Table 5-21. Comparison of "sensitive individual" HECs or HEDs for immunological effects
based on PBPK modeled internal dose-metrics at different levels of confidence and sensitivity, at
the NOAEL or LOAEL 96
Table 5-22. Comparison of "sensitive individual" HECs or HEDs for reproductive effects based
on PBPK modeled internal dose-metrics at different levels of confidence and sensitivity, at the
NOAEL or LOAEL 97
Table 5-23. Comparison of "sensitive individual" HECs or HEDs for developmental effects
based on PBPK modeled internal dose-metrics at different levels of confidence and sensitivity, at
the NOAEL or LOAEL 99
Table 5-24. Lowest p-cRfCs or cRfCs for different effect domains 103
Table 5-25. Lowest p-cRfDs or cRfDs for different effect domains 105
Table 5-26. Lowest p-cRfCs for candidate critical effects for different types of effect based on
primary dose-metric 107
Table 5-27. Lowest p-cRfDs for candidate critical effects for different types of effect based on
primary dose-metric 107
Table 5-28. Summary of critical studies, effects, PODs, and UFs used to derive the RfC 109
Table 5-29. Summary of supporting studies, effects, PODs, and UFs for the RfC 109
Table 5-30. Summary of critical studies, effects, PODs, and UFs used to derive the RfD 112
Table 5-31. Summary of supporting studies, effects, PODs, and UFs for the RfD 114
Table 5-32. Inhalation bioassays 116
Table 5-33. Oral bioassays 118
Table 5-34. Specific dose-response analyses performed and dose-metrics used 122
Table 5-35. Mean PBPK model predictions for weekly internal dose in humans exposed
continuously to low levels of TCE via inhalation (ppm) or orally (mg/kg/day) 133
This document is a draft for review purposes only and does not constitute Agency policy.
xxiii DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Table 5-36. Summary of PODs and unit risk estimates for each sex/species/bioassay/tumor type
(inhalation) 135
Table 5-37. Summary of PODs and slope factor estimates for each sex/species/bioassay/tumor
type (oral) 137
Table 5-38. Comparison of survival-adjusted results for 3 oral male rat data setsa 141
Table 5-39. Inhalation: most sensitive bioassay for each sex/species combination51 145
Table 5-40. Oral: most sensitive bioassay for each sex/species combination3 145
Table 5-41. Summary of PBPK model-based uncertainty analysis of unit risk estimates for each
sex/species/bioassay/tumor type (inhalation) 155
Table 5-42. Summary of PBPK model-based uncertainty analysis of slope factor estimates for
each sex/species/bioassay/tumor type (oral) 156
Table 5-43. Results from Charbotel et al. (2006) on relationship between TCE exposure and
RCC 160
Table 5-44. Extra risk estimates for RCC incidence from various levels of lifetime exposure to
TCE, using linear cumulative exposure model 162
Table 5-45. ECoi, LECoi, and unit risk estimates for RCC incidence, using linear cumulative
exposure model 163
Table 5-46. Relative contributions to extra risk for cancer incidence from TCE exposure for
multiple tumor types 171
Table 5-47. Route-to-route extrapolation of site-specific inhalation unit risks to oral slope
factors 176
Table 5-48. Sample calculation for total lifetime cancer risk based on the kidney unit risk
estimate, potential risk for NHL and liver cancer, and potential increased early-life susceptibility,
"3
assuming a constant lifetime exposure to 1 (J,g/m of TCE in air 183
Table 5-49. Sample calculation for total lifetime cancer risk based on the kidney cancer slope
factor estimate, potential risk for NHL and liver cancer, and potential increased early-life
susceptibility, assuming a constant lifetime exposure to 1 [j,g/L of TCE in drinking water 187
LIST OF FIGURES
Figure 2-1. Molecular structure of TCE 2
Figure 2-2. Source contribution to TCE emissions 6
Figure 2-3. Annual emissions of TCE 6
Figure 2-4. Modeled ambient air concentrations of TCE 13
Figure 3-1. Gas uptake data from closed-chamber exposure of rats to TCE. Symbols represent
measured chamber concentrations. Source: Simmons et al. (2002) 12
This document is a draft for review purposes only and does not constitute Agency policy.
xxiv DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Figure 3-2. Disposition of [14C]TCE administered by oral gavage in mice (Dekant et al., 1984;
Dekant et al., 1986b; Green and Prout, 1985; Prout et al., 1985) 26
Figure 3-3. Disposition of [14C]TCE administered by oral gavage in rats (Dekant et al., 1984;
Dekant et al., 1986b; Green and Prout, 1985; Prout et al., 1985) 27
Figure 3-4. Scheme for the oxidative metabolism of TCE 29
Figure 3-5. Scheme for GSH-dependent metabolism of TCE 48
Figure 3-6. Interorgan TCE transport and metabolism via the GSH pathway. See Figure 3-5 for
enzymes involved in metabolic steps. Source: Lash et al.(2000a; 2000b); NRC (2006) 58
Figure 3-7. Overall structure of PBPK model for TCE and metabolites used in this assessment.
Boxes with underlined labels are additions or modifications of the Hack et al. (2006) model,
which are discussed in Table 3-32 81
Figure 3-8. Schematic of how posterior predictions were generated for comparison with
experimental data. Two sets of posterior predictions were generated: population predictions
(diagonal hashing) and subject-specific predictions (vertical hashing). (Same as Figure A-2 in
Appendix A) 127
Figure 3-9. Comparison of mouse data and PBPK model predictions from a random posterior
sample. Each panel shows results for a different measurement. The solid line represents
prediction = data, and the grey dotted lines show prediction = data x GSDerr and data ^ GSDerr,
where GSDerr is the median estimate of the residual-error GSD shown in Table 3-41 130
Figure 3-10. Comparison of rat data and PBPK model predictions from a random posterior
sample. Each panel shows results for a different measurement. The solid line represents
prediction = data, and the grey lines show prediction = data x GSDerr and data ^ GSDerr, where
GSDerr is the lowest (dotted) and highest (dashed) median estimate of the residual-error GSD
shown in Table 3-41 134
Figure 3-11. Comparison of urinary excretion data for NAcDCVC and predictions from the
Hack et al. (2006) and the updated PBPK models. Data are from Bernauer et al. (1996) for (A
and B) rats or (C and D) humans exposed for 6 hour to 40 (o), 80 (A), or 160 (+) ppm in air
(thick horizontal line denotes the exposure period). Predictions from Hack et al. (2006) and the
corresponding data (A and C) are only for the 1,2 isomer, whereas those from the updated model
(B and D) are for both isomers combined. Parameter values used for each prediction are a
random sample from the subject-specific parameters from the rat and human MCMC chains (the
last iteration of the first chain was used in each case). Note that in the Hack et al. (2006) model,
each dose group had different model parameters, whereas in the updated model, all dose groups
are required to have the same model parameters. See files linked to Appendix A for comparisons
with the full distribution of predictions 143
This document is a draft for review purposes only and does not constitute Agency policy.
XXV DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Figure 3-12. Comparison of human data and PBPK model predictions from a random posterior
sample. Each panel shows results for a different measurement. The solid line represents
prediction = data, and the grey lines show prediction = data x GSDerr and data ^ GSDerr, where
GSDerr is the lowest (dotted) and highest (dashed) median estimate of the residual-error GSD
shown in Table 3-41 145
Figure 3-13. Comparison of DCVG concentrations in human blood and predictions from the
updated model. Data are mean concentrations for males (A) and females (o) reported in Lash
et al. (1999a) for humans exposed for 4 hours to 100 ppm TCE in air (thick horizontal line
denotes the exposure period). Data for oxidative metabolites from the same individuals were
reported in Fisher et al. (1998) but could not be matched with the individual DCVG data (Lash
2007, personal communication). The vertical error bars are standard errors of the mean as
reported in Lash et al. (1999a) (n = 8, so standard deviation is 80.5-fold larger). Lines are PBPK
model predictions for individual male (solid) and female (dashed) subjects. Parameter values
used for each prediction are a random sample from the individual-specific parameters from the
human MCMC chains (the last iteration of the 1st chain was used). See files linked to Appendix
A for comparisons with the full distribution of predictions 151
Figure 3-14. Sensitivity analysis results: Number of mouse calibration data points with SC in
various categories for each scaling parameter 152
Figure 3-15. Sensitivity analysis results: Number of rat calibration data points with SC in
various categories for each scaling parameter 153
Figure 3-16. Sensitivity analysis results: Number of human calibration data points with SC in
various categories for each scaling parameter 154
Figure 3-17. PBPK model predictions for the fraction of intake that is metabolized under
continuous inhalation (A) and oral (B) exposure conditions in mice (white), rats (diagonal
hashing), and humans (horizontal hashing). Bars and thin error bars represent the median
estimate and 95% CI for a random subject, and reflect combined uncertainty and variability.
Circles and thick error bars represent the median estimate and 95% CI for the population mean,
and reflect uncertainty only 160
Figure 3-18. PBPK model predictions for the fraction of intake that is metabolized by oxidation
(in the liver and lung) under continuous inhalation (A) and oral (B) exposure conditions in mice
(white), rats (diagonal hashing), and humans (horizontal hashing). Bars and thin error bars
represent the median estimate and 95% CI for a random subject, and reflect combined
uncertainty and variability. Circles and thick error bars represent the median estimate and 95%
CI for the population mean, and reflect uncertainty only 161
Figure 3-19. PBPK model predictions for the fraction of intake that is metabolized by GSH
conjugation (in the liver and kidney) under continuous inhalation (A) and oral (B) exposure
This document is a draft for review purposes only and does not constitute Agency policy.
xxvi DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
conditions in mice (dotted line), rats (dashed line), and humans (solid line). X-values are slightly
offset for clarity. Open circles (connected by lines) and thin error bars represent the median
estimate and 95% CI for a random subject, and reflect combined uncertainty and variability.
Filled circles and thick error bars represent the median estimate and 95% CI for the population
mean, and reflect uncertainty only 162
Figure 3-20. PBPK model predictions for the fraction of intake that is bioactivated DCVC in the
kidney under continuous inhalation (A) and oral (B) exposure conditions in rats (dashed line) and
humans (solid line). X-values are slightly offset for clarity. Open circles (connected by lines)
and thin error bars represent the median estimate and 95% CI for a random subject, and reflect
combined uncertainty and variability. Filled circles and thick error bars represent the median
estimate and 95% CI for the population mean, and reflect uncertainty only 163
Figure 3-21. PBPK model predictions for fraction of intake that is oxidized in the respiratory
tract under continuous inhalation (A) and oral (B) exposure conditions in mice (dotted line), rats
(dashed line), and humans (solid line). X-values are slightly offset for clarity. Open circles
(connected by lines) and thin error bars represent the median estimate and 95% CI for a random
subject, and reflect combined uncertainty and variability. Filled circles and thick error bars
represent the median estimate and 95% CI for the population mean, and reflect uncertainty only.
164
Figure 3-22. PBPK model predictions for the fraction of intake that is "untracked" oxidation of
TCE in the liver under continuous inhalation (A) and oral (B) exposure conditions in mice
(dotted line), rats (dashed line), and humans (solid line) X-values are slightly offset for clarity.
Open circles (connected by lines) and thin error bars represent the median estimate and 95% CI
for a random subject, and reflect combined uncertainty and variability. Filled circles and thick
error bars represent the median estimate and 95% CI for the population mean, and reflect
uncertainty only 165
Figure 3-23. PBPK model predictions for the weekly AUC of TCE in venous blood
(mg-hour/L-week) per unit exposure (ppm or mg/kg-day) under continuous inhalation (A) and
oral (B) exposure conditions in mice (dotted line), rats (dashed line), and humans (solid line).
X-values are slightly offset for clarity. Open circles (connected by lines) and thin error bars
represent the median estimate and 95% CI for a random subject, and reflect combined
uncertainty and variability. Filled circles and thick error bars represent the median estimate and
95% CI for the population mean, and reflect uncertainty only 166
Figure 3-24. PBPK model predictions for the weekly AUC of TCOH in blood
(mg-hour/L-week) per unit exposure (ppm or mg/kg-day) under continuous inhalation (A) and
oral (B) exposure conditions in mice (dotted line), rats (dashed line), and humans (solid line).
X-values are slightly offset for clarity. Open circles (connected by lines) and thin error bars
This document is a draft for review purposes only and does not constitute Agency policy.
xxvii DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
represent the median estimate and 95% CI for a random subject, and reflect combined
uncertainty and variability. Filled circles and thick error bars represent the median estimate and
95% CI for the population mean, and reflect uncertainty only 167
Figure 3-25. PBPK model predictions for the weekly AUC of TCA in the liver
(mg-hour/L-week) per unit exposure (ppm or mg/kg-day) under continuous inhalation (A) and
oral (B) exposure conditions in mice (dotted line), rats (dashed line), and humans (solid line).
X-values are slightly offset for clarity. Open circles (connected by lines) and thin error bars
represent the median estimate and 95% CI for a random subject, and reflect combined
uncertainty and variability. Filled circles and thick error bars represent the median estimate and
95% CI for the population mean, and reflect uncertainty only 168
Figure 3-26. Sensitivity analysis results: SC for mouse scaling parameters with respect to
dose-metrics following 100 ppm (light bars) and 600 ppm (dark bars), 7 h/day, 5 day/wk
inhalation exposures 176
Figure 3-27. Sensitivity analysis results: SC for mouse scaling parameters with respect to
dose-metrics following 300 mg/kg-day (light bars) and 1,000 mg/kg-day (dark bars), 5 day/wk
oral gavage exposures 177
Figure 3-28. Sensitivity analysis results: SC for rat scaling parameters with respect to
dose-metrics following 100 ppm (light bars) and 600 ppm (dark bars), 7 h/day, 5 day/wk
inhalation exposures 178
Figure 3-29. Sensitivity analysis results: SC for rat scaling parameters with respect to
dose-metrics following 300 mg/kg-day (light bars) and 1,000 mg/kg-day (dark bars), 5 day/wk
oral gavage exposures 179
Figure 3-30. Sensitivity analysis results: SC for female (light bars) and male (dark bars) human
scaling parameters with respect to dose-metrics following 0.001 ppm continuous inhalation
exposures 180
Figure 3-31. Sensitivity analysis results: SC for female (light bars) and male (dark bars) human
scaling parameters with respect to dose-metrics following 0.001 mg/kg-day continuous oral
exposures 181
Figure 4-1. Meta-analysis of kidney cancer and overall TCE exposure (the summary estimate is
in the bottom row, represented by the diamond). Random effects model; fixed effect model
same. Symbol sizes reflect relative weights of the studies 207
Figure 4-2. Meta-analysis of kidney cancer and TCE exposure—highest exposure groups. With
assumed null RR estimates for Antilla, Axelson, and Hansen (see Appendix C text). Random
effects model; fixed effect model same. The summary estimate is in the bottom row, represented
by the diamond. Symbol sizes reflect relative weights of the studies 209
This document is a draft for review purposes only and does not constitute Agency policy.
xxviii DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Figure 4-3. Relative risk estimates of liver and biliary tract cancer and overall TCE exposure.
Random effects model; fixed effect model same. The summary estimate is in the bottom row,
represented by the diamond. Symbol sizes reflect relative weights of the studies 284
Figure 4-4. Meta-analysis of liver cancer and TCE exposure—highest exposure groups, with
assumed null RR estimates for Hansen and Zhao (see Appendix C text). Random effects model;
fixed effect model same. The summary estimate is in the bottom row, represented by the
diamond. Symbol sizes reflect relative weights of the studies 285
Figure 4-5. Comparison of average fold-changes in relative liver weight to control and exposure
concentrations of 2 g/L or less in drinking water for TCA and DC A in male B6C3F1 mice for
14-30 days (Carter et al., 1995; DeAngelo et al., 1989; 2008; Kato-Weinstein et al., 2001;
Parrish et al., 1996; Sanchez and Bull, 1990) 334
Figure 4-6. Comparisons of fold-changes in average relative liver weight and gavage dose of
(top panel) male B6C3F1 mice for 10-28 days of exposure (Dees and Travis, 1993; Elcombe et
al., 1985; Goldsworthy and Popp, 1987; Merrick et al., 1989) and (bottom panel) in male
B6C3F1 and Swiss mice 336
Figure 4-7. Comparison of fold-changes in relative liver weight for data sets in male B6C3F1,
Swiss, and NRMI mice between TCE studies Kjellstrand et al., 1983a (Buben and O'Flaherty,
1985; Goel et al., 1992; Merrick et al., 1989) [duration 28-42 days]) and studies of direct oral
TCA administration to B6C3F1 mice (DeAngelo et al., 1989; DeAngelo et al., 2008; Kato-
Weinstein et al., 2001; Parrish et al., 1996)[duration 14-28 days]). Abscissa for TCE studies
consists of the median estimates of the internal dose of TCA predicted from metabolism of TCE
using the PBPK model described in Section 3.5 of the TCE risk assessment. Lines show linear
regression with intercept fixed at unity. All data were reported fold-change in mean liver
weight/body weight ratios, except for Kjellstrand et al. (1983b), with were the fold-change in the
ratio of mean liver weight to mean body weight. In addition, in Kjellstrand et al. (1983b), some
systemic toxicity as evidence by decreased total body weight was reported in the highest-dose
group 338
Figure 4-8. Comparison of hepatomegaly as a function of AUC of TCA in liver, using values for
the TCA drinking water fractional absorption (Fabs). Fold-changes in relative liver weight for
data sets in male B6C3F1, Swiss, and NRMI mice between TCE studies Kjellstrand et al., 1983a
(Buben and O'Flaherty, 1985; Goel et al., 1992; Merrick et al., 1989) [duration 28-42 days] and
studies of direct oral TCA administration to B6C3F1 mice (DeAngelo et al., 1989; DeAngelo et
al., 2008; Kato-Weinstein et al., 2001; Parrish et al., 1996) Green, 2003b [duration 14-28 days].
Linear regressions were compared using ANOVA to assess whether the TCE studies were
consistent with the TCA studies, using TCA as the dose-metric. For each analysis of drinking
water fraction absorption, ANOVA /^-values were <10 4 when comparing the assumption that all
This document is a draft for review purposes only and does not constitute Agency policy.
xxix DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
the data had a common slope with the assumption that TCE and TCA data had different slopes.
340
Figure 4-9. Fold-changes in relative liver weight for data sets in male B6C3F1, Swiss, and
NRMI mice reported by TCE studies of duration 28-42 days (Kjellstrand et al., 1983a(Buben
and O'Flaherty, 1985; Goel et al., 1992; Merrick et al., 1989) using internal dose-metrics
predicted by the PBPK model described in Section 3.5: (A) dose-metric is the median estimate of
the daily AUC of TCE in blood, (B) dose-metric is the median estimate of the total daily rate of
TCE oxidation. Lines show linear regression. Use of liver oxidative metabolism as a
dose-metric gives results qualitatively similar to (B), with R =0.86 341
Figure 4-10. Dose-response relationship, expressed as (A) percentage incidence and
(B) fold-increase over controls, for TCE hepatocarcinogenicity in NCI (1976). For comparison,
incidences of carcinomas for NTP (1990), Anna et al. (1994), and Bull et al. (2002) are included,
but without connecting lines since they are not appropriate for assessing the shape of the
dose-response relationship 357
Figure 4-11. Dose-response relationship, expressed as (A) incidence and (B) fold-increase over
controls, for TCE hepatocarcinogenicity in Maltoni et al. (1986). Note that the BT306
experiment reported excessive mortality due to fighting, and so the paradigm was repeated in
experiment BT306bis using mice from a different source 358
Figure 4-12. Dose-response data for hepatocellular carcinomas (HCC) (A) incidence and
(B) multiplicity, induced by DCA from DeAngelo et al. (1999). Drinking water concentrations
were 0, 0.05, 0.5, 1, 2, and 3.5 g/L, from which daily average doses were calculated using
observed water consumption in the study 359
Figure 4-13. Reported incidences of hepatocellular carcinomas (HCC) and hepatocellular
adenomas plus carcinomas (HCA + HCC) in various studies in B6C3F1 mice (DeAngelo et al.,
2008; Pereira, 1996). Combined HCA + HCC were not reported in Pereira (1996) 361
Figure 4-14. Reported incidence of hepatocellular carcinomas induced by DCA and TCA in
104-week studies (DeAngelo et al., 2008; DeAngelo et al., 1999). Only carcinomas were
reported in DeAngelo et al. (1999), so combined adenomas and carcinomas could not be
compared 363
Figure 4-15. Effects of dietary control on the dose-response curves for changes in liver tumor
incidences induced by CH in diet (Leakey et al., 2003a) 368
Figure 5-1. Flow-chart of the process used to derive the RfD and RfC for noncancer effects 2
Figure 5-2. Flow-chart for dose-response analyses of rodent noncancer effects using PBPK
model-based dose-metrics. Square nodes indicate point values, circle nodes indicate
distributions, and the inverted triangle indicates a (deterministic) functional relationship 68
This document is a draft for review purposes only and does not constitute Agency policy.
XXX DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Figure 5-3. Schematic of combined interspecies, intraspecies, and route-to-route extrapolation
from a rodent study LOAEL or NOAEL. In the case where BMD modeling is performed, the
applied dose values are replaced by the corresponding median internal dose estimate, and the
idPOD is the modeled BMDL in internal dose units 69
Figure 5-4. Flow-chart for uncertainty analysis of HECs and HEDs derived using PBPK
model-based dose-metrics. Square nodes indicate point values, circle nodes indicate
distributions, and the inverted triangle indicates a (deterministic) functional relationship 92
Figure 5-5. Flow-chart for dose-response analyses of rodent bioassays using PBPK model-based
dose-metrics. Square nodes indicate point values, circular nodes indicate distributions, and the
inverted triangles indicate a (deterministic) functional relationship 132
Figure 5-6. Flow-chart for uncertainty analysis of dose-response analyses of rodent bioassays
using PBPK model-based dose-metrics. Square nodes indicate point values, circular nodes
indicate distributions, and the inverted triangles indicate a (deterministic) functional relationship.
153
Figure 5-7. Flow-chart for route-to-route extrapolation of human site-specific cancer inhalation
unit risks to oral slope factors. Square nodes indicate point values, circle nodes indicate
distributions, and the inverted triangle indicates a (deterministic) functional relationship 175
This document is a draft for review purposes only and does not constitute Agency policy.
xxxi DRAFT—DO NOT CITE OR QUOTE
-------
LIST OF ABBREVIATIONS AND ACRONYMS
[14C]TCE
[14C]-radio labeled TCE
1,2-DCVC
S-(l,2-dichlorovinyl)-L-cysteine
17-P-HSD
17-P-hydroxy steroid dehydrogenase
8-OHdG
8-hydroxy-2' deoxyguanosine
ACO
acyl CoA oxidase
ADAF
age-dependent adjustment factor
ADME
absorption, distribution, metabolism, and excretion
AIC
Akaike Information Criteria
ALL
acute lymphoblastic leukemia
ALP
alkaline phosphatase
ALT
alanine aminotransferase
ANA
antinuclear antibodies
ANCA
antineutrophil -cytopl asmi c antib ody
AOAA
a beta-lyase inhibitor
ASD
autism spectrum disorder
ASPEN
Assessment System for Population Exposure Nationwide
AST
aspartate aminotrasferase
ATF-2
activating transcription factor 2
AT SDR
Agency for Toxic Substances and Disease Registry
AUC
area-under-the-curve
AV
atrioventricular
AVC
atrioventricular canal
AZDHS
Arizona Department of Health Services
BAER
brainstem auditory-evoked response
BAL
bronchoalveolar lavage
BMD
benchmark dose
BMDL
benchmark dose lower bound
BMDS
BenchMark Dose Software
BMI
body mass index
BMR
benchmark response
BUN
blood urea nitrogen
BW
body weight
CA DHS
California Department of Health Services
CH
chloral hydrate
CI
confidence interval
CLL
chronic lymphocytic leukemia
CNS
central nervous system
C02
carbon dioxide
CoA
coenzyme A
cRfC
candidate RfC
cRfD
candidate RfD
CRT
choice reaction time
CYP
cytochrome P450
This document is a draft for review purposes only and does not constitute Agency policy.
xxxii DRAFT—DO NOT CITE OR QUOTE
-------
LIST OF ABBREVIATIONS AND ACRONYMS (continued)
DAL
dichloroacetyl lysine
DAS02
diallyl sulfone
DBP
dibutyl phthalate
DCA
dichloroacetic acid
DCAA
dichloroacetic anhydride
DCAC
dichloroacetyl chloride
DCE
dichloroethane
DCVC
dichlorovinyl cysteine
DCVG
S-dichlorovinyl glutathione
DEHP
di(2-ethylhexyl) phthalate
DHEAS
dehydroepiandrosterone sulphate
DNP
dinitrophenol
DPM
disintegrations per minute
dsDNA
double-stranded DNA
ECX
concentration of the chemical at which x% of the maximal effect is produced
EEG
electroencephalogram
EPA
U.S. Environmental Protection Agency
ERG
el ectroreti nogram
FAA
fumarylacetoacetate
FDVE
fluoromethyl-2,2-difluoro-1 -(trifluoromethyl)vinyl ether
FMO
flavin mono-oxygenase
FOB
functional observational battery
FSH
follicle-stimulating hormone
G6PDH
glucose 6-p dehydrogenase
GA
glomerular antigen
GABA
gamma-amino butyric acid
G-CSF
granulocyte colony stimulating factor
GD
gestation day
GFR
glomerular filtration rate
GGT
y-glutamyl transpeptidase or y-transpeptidase
GI
gastro-intestinal
GIS
geographic information system
GSH
glutathione
GSSG
oxidized GSH
GST
glutathione-S-transferase
GT
glutamyl transferase
H&E
hematoxylin and eosin
h2o
water
HCC
hepatocellular carcinoma
hCG
human chorionic gonadotropin
HC1
hydrochloric acid
HDL-C
high density lipoprotein-cholesterol
HEC
human equivalent concentration
This document is a draft for review purposes only and does not constitute Agency policy.
xxxiii DRAFT—DO NOT CITE OR QUOTE
-------
LIST OF ABBREVIATIONS AND ACRONYMS (continued)
HED
human equivalent dose
HgCl2
mercuric chloride
HH
Hamberger and Hamilton
HPLC
high-performance liquid chromatography
HPT
hypothalami c-pituitary-testi s
i.a.
intra-arterial
i.p.
intraperitoneal
i.v.
intravenous
IARC
International Agency for Research on Cancer
ICC
intrahepatic cholangiocarcinoma
ICD
International Classification of Disease
ICRP
The International Commission on Radiological Protection
idPOD
internal dose points of departure
IDR
incidence density ratio
IFN
interferon
IgE
immunoglobulin E
IGF-II
insulin-like growth factor-II (gene)
IL
interleukin
IPCS
International Programme on Chemical Safety
IUGR
intrauterine growth restriction
JEM
j ob - exposure matrix
JTEM
job-task-exposure matrix
LC
lethal concentration
LCL
lower confidence limit
LDH
lactate dehydrogenase
LECX
lowest effective concentration corresponding to an extra risk of x%
LH
luteinizing hormone
InPBC
blood-air partition coefficient
InQCC
cardiac output
InVMAXC
VMAX for oxidation
InVPRC
ventilation-perfusion ratio
LOAEL
lowest observed adverse effect level
LOH
loss of heterozygosity
LORR
loss of righting reflex
MA
maleylacetone
MA DPH
Massachusetts Department of Public Health
MAA
mal eyl acetoacetate
MCA
monochloroacetic acid
MCMC
Markov chain Monte Carlo
MCP
methylclofenapate
MDA
malondialdehyde
MLE
maximum likelihood estimate
MNU
methyl nitrosourea
This document is a draft for review purposes only and does not constitute Agency policy.
xxxiv DRAFT—DO NOT CITE OR QUOTE
-------
LIST OF ABBREVIATIONS AND ACRONYMS (continued)
MOA
mode of action
MS
MSW
multistage Weibull
NAcDCVC
N-acetyl-S-(l,2-dichlorovinyl)-L-cysteine
NADH
nicotinamide adenine dinucleotide
NAG
N-acetyl-P-D-glucosaminidase
NAT
N-acetyl transferase
NCI
National Cancer Institute
NF
NHL
non-Hodgkin lymphoma
NK
natural killer
NOAEL
no-ob served-adverse-effect level
NOEC
no-observed-effect concentration
NOEL
no-ob served-effect level
NPMC
nonpurified rat peritoneal mast cells
NRC
National Research Council
NSATA
National-Scale Air Toxics Assessment
NTP
National Toxicology Program
NYS DOH
New York State Department of Health
ODE
ordinary differential equation
OECD
Organization for Economic Co-operation and Development
OFT
outflow tract
OP
oscillatory potential
OR
odds ratio
PAS
periodic acid-Schiff
PBPK
physiologically based pharmacokinetics
PC
partition coefficient
PCEs
polychromatic erythrocytes
PCNA
proliferating cell nuclear antigen
PCO
palmitoyl-CoA oxidase
PCR
polymerase chain reaction
p-cRfC
PBPK model-based candidate RfCs
p-cRfD
PBPK model-based candidate RfDs
PEG 400
polyethylene glycol 400
PFC
plaque-forming cell
PFU
plaque-forming units
PMR
proportionate mortality ratio
PND
postnatal day
P02
partial pressure oxygen
POD
point of departure
PPAR
peroxisome proliferator activated receptor
QC
quality control
RBL-2H3
rat basophilic leukemia
This document is a draft for review purposes only and does not constitute Agency policy.
XXXV DRAFT—DO NOT CITE OR QUOTE
-------
LIST OF ABBREVIATIONS AND ACRONYMS (continued)
RCC
renal cell carcinoma
RfC
inhalation reference concentration
RfD
oral reference dose
ROS
reactive oxygen species
RR
relative risk
RRm
summary RR
RT
reaction time
S9
metabolic activation system
SBA
serum bile acids
SC
sensitivity coefficient
SCEs
sister chromatid exchanges
S-D
Sprague-Dawley
SD
standard deviation
SDH
sorbitol dehydrogenase
SE
standard error
SEER
Surveillance, Epidemiology, and End Results
SES
socioeconomic status
SGA
small for gestational age
SHBG
sex-hormone binding globulin
SIR
standardized incidence ratio
SMR
standardized mortality ratio
SNP
single nucleotide polymorphism
SRBC
sheep red blood cells
SRT
simple reaction time
SSB
single-strand breaks
SSCP
single stand conformation polymorphism
ssDNA
single-stranded DNA
TaClo
tetrahydro-beta-carbolines
TBARS
thiobarbiturate acid-reactive substances
TCA
trichloroacetic acid
TCAA
tri chl oroacetal dehy de
TCAH
trichloroacetaldehyde hydrate
TCE
tri chl oroethyl ene
TCOG
tri chl oroethanol -glucuroni de conj ugate
TCOH
trichloroethanol
ThX
T-helper Type X
TNF
tumor necrosis factor
TNFAI
wild- type TNF-a (G —~ A substitution at position -238)
TNF All
TNF-a (G—~ A substitution at position -308)
TRI
Toxics Release Inventory
TSEP
trigeminal somatosensory evoked potential
TTC
total trichloro compounds
TWA
time-weighted average
This document is a draft for review purposes only and does not constitute Agency policy.
xxxvi DRAFT—DO NOT CITE OR QUOTE
-------
LIST OF ABBREVIATIONS AND ACRONYMS (continued)
U.S. EPA
UCL
UDS
UF
USGS
U-TCA
U-TTC
VEGF
VEP
VHL
VLivC
VOC
VSCC
W
WHO
YFF
This document is a draft for review purposes only and does not constitute Agency policy.
xxxvii DRAFT—DO NOT CITE OR QUOTE
U.S. Environmental Protection Agency
upper confidence limit
unscheduled DNA synthesis
uncertainty factor
United States Geological Survey
urinary-TCA
urinary total trichloro-compounds
vascular endothelial growth factor
visual evoked potential
von Hippel-Lindau
liver volume
volatile organic compound
voltage sensitive calcium channel
wakefulness
World Health Organization
fluorescent Y-bodies
-------
FOREWORD
The purpose of this Toxicological Review is to provide scientific support and rationale
for the hazard and dose-response assessment in IRIS pertaining to chronic exposure to
trichloroethylene. It is not intended to be a comprehensive treatise on the chemical or
toxicological nature of trichloroethylene.
The intent of Chapter 6, Major Conclusions in the Characterization of Hazard and Dose
Response, is to present the major conclusions reached in the derivation of the reference dose,
reference concentration and cancer assessment, where applicable, and to characterize the overall
confidence in the quantitative and qualitative aspects of hazard and dose response by addressing
the quality of the data and related uncertainties. The discussion is intended to convey the
limitations of the assessment and to aid and guide the risk assessor in the ensuing steps of the
risk assessment process.
For other general information about this assessment or other questions relating to IRIS,
the reader is referred to EPA's IRIS Hotline at (202) 566-1676 (phone), (202) 566-1749 (fax), or
hotline.iris@epa.gov (email address).
This document is a draft for review purposes only and does not constitute Agency policy.
xxxviii DRAFT—DO NOT CITE OR QUOTE
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
CHEMICAL MANAGER
Weihsueh A. Chiu
National Center for Environmental Assessment—Washington Office
U.S. Environmental Protection Agency
Washington, DC
AUTHORS
Ambuja Bale
National Center for Environmental Assessment—Immediate Office
U.S. Environmental Protection Agency
Washington, DC
Stanley Bar one
National Center for Environmental Assessment—Immediate Office
U.S. Environmental Protection Agency
Washington, DC
Rebecca Brown
National Center for Environmental Assessment—Washington Office
U.S. Environmental Protection Agency
Washington, DC
Jane C. Caldwell
National Center for Environmental Assessment—Washington Office
U.S. Environmental Protection Agency
Washington, DC
Chao Chen
National Center for Environmental Assessment—Washington Office
U.S. Environmental Protection Agency
Washington, DC
Weihsueh A. Chiu
National Center for Environmental Assessment—Washington Office
U.S. Environmental Protection Agency
Washington, DC
Glinda Cooper
National Center for Environmental Assessment—Immediate Office
U.S. Environmental Protection Agency
This document is a draft for review purposes only and does not constitute Agency policy.
xxxix DRAFT—DO NOT CITE OR QUOTE
-------
Washington, DC
AUTHORS, CONTRIBUTORS, AND REVIEWERS (continued)
Ghazi Dannan
National Center for Environmental Assessment—Washington Office
U.S. Environmental Protection Agency
Washington, DC
Marina Evans
National Health and Environmental Effects Research Laboratory
(on detail to National Center for Environmental Assessment—Washington Office)
U.S. Environmental Protection Agency
Research Triangle Park, NC
John Fox
National Center for Environmental Assessment—Washington Office
U.S. Environmental Protection Agency
Washington, DC
Kathryn Z. Guyton
National Center for Environmental Assessment—Washington Office
U.S. Environmental Protection Agency
Washington, DC
Maureen R. Gwinn
National Center for Environmental Assessment—Washington Office
U.S. Environmental Protection Agency
Washington, DC
Jennifer Jinot
National Center for Environmental Assessment—Washington Office
U.S. Environmental Protection Agency
Washington, DC
Nagalakshmi Keshava
National Center for Environmental Assessment—Washington Office
U.S. Environmental Protection Agency
Washington, DC
John Lipscomb
National Center for Environmental Assessment—Cincinnati Office
U.S. Environmental Protection Agency
Cincinnati, OH
This document is a draft for review purposes only and does not constitute Agency policy.
xl DRAFT—DO NOT CITE OR QUOTE
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS (continued)
Susan Makris
National Center for Environmental Assessment—Washington Office
U.S. Environmental Protection Agency
Washington, DC
Miles Okino
National Exposure Research Laboratory—Las Vegas Office
U.S. Environmental Protection Agency
Las Vegas, NV
Fred Power
National Exposure Research Laboratory—Las Vegas Office
U.S. Environmental Protection Agency
Las Vegas, NV
John Schaum
National Center for Environmental Assessment—Washington Office
Office of Research and Development
Washington, DC
Cheryl Siegel Scott
National Center for Environmental Assessment—Washington Office
Office of Research and Development
Washington, DC
REVIEWERS
This document has been reviewed by U.S. EPA scientists, reviewers from other Federal
agencies and White House offices, and the public, and peer reviewed by independent scientists
external to U.S. EPA. A summary and U.S. EPA's disposition of the comments received from
the independent external peer reviewers and from the public is included in Appendix I.
INTERNAL EPA REVIEWERS
Daniel Axelrad
National Center for Environmental Economics
Robert Benson
U.S. EPA Region 8
This document is a draft for review purposes only and does not constitute Agency policy.
xli DRAFT—DO NOT CITE OR QUOTE
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS (continued)
Ted Birner
National Center for Environmental Assessment—Immediate Office
Nancy Chiu
Office of Water
David Farrar
National Center for Environmental Assessment—Cincinnati Office
Lynn Flowers
National Center for Environmental Assessment—Immediate Office
Stiven Foster
Office of Solid Waste and Emergency Response
Susan Griffin
U.S. EPA Region 8
Samantha Jones
National Center for Environmental Assessment—Immediate Office
Leonid Kopylev
National Center for Environmental Assessment—Washington Office
Allan Marcus
National Center for Environmental Assessment—Immediate Office
Gregory Miller
Office of Children's Health Protection and Environmental Education
Deirdre Murphy
Office of Air Quality Planning and Standards
Marian 01 sen
U.S. EPA Region 2
Peter Preuss
National Center for Environmental Assessment—Immediate Office
Kathleen Raffaele
National Center for Environmental Assessment—Washington Office
This document is a draft for review purposes only and does not constitute Agency policy.
xlii DRAFT—DO NOT CITE OR QUOTE
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS (continued)
William Sette
Office of Solid Waste and Emergency Response
Bob Sonawane
National Center for Environmental Assessment—Washington Office
Suryanarayana Vulimiri
National Center for Environmental Assessment—Washington Office
Nina Ching Y. Wang
National Center for Environmental Assessment—Cincinnati Office
Paul White
National Center for Environmental Assessment—Washington Office
Marcia Bailey
U.S. EPA Region 10
ACKNOWLEDGMENTS
Drafts of Section 3.3 (TCE metabolism) were prepared for the U.S. EPA by Syracuse
Research Corporation under contract. Additional support, including literature searches and
retrievals and drafts of Appendix D were prepared for the U.S. EPA by the Oak Ridge Institute
for Science and Education (ORISE) through interagency agreement number DW-89939822-01-0
with the U.S. Department of Energy (DOE). ORISE is managed by Oak Ridge Associated
Universities under a contract with DOE. The PBPK modeling sections of this report are
dedicated to the memory of Fred Power (1938-2007). His keen analytical mind will be greatly
missed, but his gentle heart and big smile will be missed even more.
Additionally, we gratefully acknowledge Terri Konoza and Ellen Lorang of NCEA for
their management of the document production and reference/citation management processes, and
the following individuals for their support of these processes:
IntelliTech Systems, Inc. ECFlex, Inc.
Chris Broyles Heidi Glick
Stacey Lewis Debbie Kleiser
Kathleen Secor Crystal Lewis
Linda Tackett Lana Wood
This document is a draft for review purposes only and does not constitute Agency policy.
xliii DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
EXECUTIVE SUMMARY
There is substantial potential for human exposure to trichloroethylene (TCE), as it has a
widespread presence in ambient air, indoor air, soil, and groundwater. At the same time, humans
are likely to be exposed to a variety of compounds that are either metabolites of TCE or which
have common metabolites or targets of toxicity. Once exposed, humans, as well as laboratory
animal species, rapidly absorb TCE, which is then distributed to tissues via systemic circulation,
extensively metabolized, and then excreted primarily in breath as unchanged TCE or carbon
dioxide, or in urine as metabolites.
Based on the available human epidemiologic data and experimental and mechanistic
studies, it is concluded that TCE poses a potential human health hazard for noncancer toxicity to
the central nervous system, the kidney, the liver, the immune system, the male reproductive
system, and the developing fetus. The evidence is more limited for TCE toxicity to the
respiratory tract and female reproductive system. Following U.S. Environmental Protection
Agency (U.S. EPA, 2005c) Guidelines for Carcinogen Risk Assessment, TCE is characterized as
"carcinogenic in humans by all routes of exposure." This conclusion is based on convincing
evidence of a causal association between TCE exposure in humans and kidney cancer. The
human evidence of carcinogenicity from epidemiologic studies of TCE exposure is compelling
for non-Hodgkin Lymphoma but less convincing than for kidney cancer, and more limited for
liver and biliary tract cancer. Less human evidence is found for an association between TCE
exposure and other types of cancer, including bladder, esophageal, prostate, cervical, breast, and
childhood leukemia, breast. Further support for the characterization of TCE as "carcinogenic in
humans by all routes of exposure" is derived from positive results in multiple rodent cancer
bioassays in rats and mice of both sexes, similar toxicokinetics between rodents and humans,
mechanistic data supporting a mutagenic mode of action (MOA) for kidney tumors, and the lack
of mechanistic data supporting the conclusion that any of the MOA(s) for TCE-induced rodent
tumors are irrelevant to humans.
As TCE toxicity and carcinogenicity are generally associated with TCE metabolism,
susceptibility to TCE health effects may be modulated by factors affecting toxicokinetics,
including lifestage, gender, genetic polymorphisms, race/ethnicity, preexisting health status,
lifestyle, and nutrition status. In addition, while these some of these factors are known risk
factors for effects associated with TCE exposure, it is not known how TCE interacts with known
risk factors for human diseases.
For noncancer effects, the most sensitive types of effects, based either on human
equivalent concentrations/doses or on candidate inhalation reference concentrations (RfCs)/oral
This document is a draft for review purposes only and does not constitute Agency policy.
xliv DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
reference doses (RfDs), appear to be developmental, kidney, and immunological (adult and
developmental) effects. The neurological and reproductive effects appear to be about an order of
magnitude less sensitive, with liver effects another two orders of magnitude less sensitive. The
RfC estimate of 0.0004 ppm (0.4 ppb or 2 (j,g/m ) is based on route-to-route extrapolated results
from oral studies for the critical effects of heart malformations (rats) and immunotoxicity (mice).
This RfC value is further supported by route-to-route extrapolated results from an oral study of
toxic nephropathy (rats). Similarly, the RfD estimate for noncancer effects of 0.0005 mg/kg-day
is based on the critical effects of heart malformations (rats), adult immunological effects (mice),
and developmental immunotoxicity (mice), all from oral studies. This RfD value is further
supported by results from an oral study for the effect of toxic nephropathy (rats) and route-to-
route extrapolated results from an inhalation study for the effect of increased kidney weight
(rats). There is high confidence in these noncancer reference values, as they are supported by
moderate- to high-confidence estimates for multiple effects from multiple studies.
For cancer, the inhalation unit risk is 2 x 10~2 per ppm [4 x 10~6 per jig/m3], based on
human kidney cancer risks reported by Charbotel et al. (2006) and adjusted, using human
epidemiologic data, for potential risk for tumors at multiple sites. The oral unit risk for cancer is
2
5x10 per mg/kg-day, resulting from physiologically based pharmacokinetic model-based
route-to-route extrapolation of the inhalation unit risk estimate based on the human kidney
cancer risks reported in Charbotel et al. (2006) and adjusted, using human epidemiologic data,
for potential risk for tumors at multiple sites. There is high confidence in these unit risks for
cancer, as they are based on good quality human data, as well as being similar to unit risk
estimates based on multiple rodent bioassays. There is both sufficient weight of evidence to
conclude that TCE operates through a mutagenic MOA for kidney tumors and a lack of TCE-
specific quantitative data on early-life susceptibility. Generally, the application of age-
dependent adjustment factors (ADAFs) is recommended when assessing cancer risks for a
carcinogen with a mutagenic MOA. However, because the ADAF adjustment applies only to the
kidney cancer component of the total risk estimate, it is likely to have a minimal impact on the
total cancer risk except when exposures are primarily during early life.
This document is a draft for review purposes only and does not constitute Agency policy.
xlv DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
1. INTRODUCTION
This document presents background information and justification for the Integrated Risk
Information System (IRIS) Summary of the hazard and dose-response assessment of
trichloroethylene. IRIS Summaries may include oral reference dose (RfD) and inhalation
reference concentration (RfC) values for chronic and other exposure durations, and a
carcinogenicity assessment.
The RfD and RfC, if derived, provide quantitative information for use in risk assessments
for health effects known or assumed to be produced through a nonlinear (presumed threshold)
mode of action. The RfD (expressed in units of mg/kg/d) is defined as an estimate (with
uncertainty spanning perhaps an order of magnitude) of a daily exposure to the human
population (including sensitive subgroups) that is likely to be without an appreciable risk of
"3
deleterious effects during a lifetime. The inhalation RfC (expressed in units of ppm or (J,g/m ) is
analogous to the oral RfD, but provides a continuous inhalation exposure estimate. The
inhalation RfC considers toxic effects for both the respiratory system (portal-of-entry) and for
effects peripheral to the respiratory system (extrarespiratory or systemic effects). Reference
values are generally derived for chronic exposures (up to a lifetime), but may also be derived for
acute (<24 hours), short-term (>24 hours up to 30 days), and subchronic (>30 days up to 10% of
lifetime) exposure durations, all of which are derived based on an assumption of continuous
exposure throughout the duration specified. Unless specified otherwise, the RfD and RfC are
derived for chronic exposure duration.
The carcinogenicity assessment provides information on the carcinogenic hazard
potential of the substance in question and quantitative estimates of risk from oral and inhalation
exposure may be derived. The information includes a weight-of-evidence judgment of the
likelihood that the agent is a human carcinogen and the conditions under which the carcinogenic
effects may be expressed. Quantitative risk estimates may be derived from the application of a
low-dose extrapolation procedure. If derived, the oral slope factor is a plausible upper bound on
the estimate of risk per mg/kg/d of oral exposure. Similarly, an inhalation unit risk is a plausible
-3
upper bound on the estimate of risk per ppm or (J,g/m in air breathed.
Development of these hazard identification and dose-response assessments for
trichloroethylene has followed the general guidelines for risk assessment as set forth by the
National Research Council (1983). U.S. EPA Guidelines and Risk Assessment Forum Technical
Panel Reports that may have been used in the development of this assessment include the
following: EPA Guidelines and Risk Assessment Forum technical panel reports that may have
been used in the development of this assessment include the following: Guidelines for the Health
Risk Assessment of Chemical Mixtures (U.S. EPA, 1986a), Guidelines for Mutagenicity Risk
This document is a draft for review purposes only and does not constitute Agency policy.
1 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Assessment (U.S. EPA, 1986b), Recommendations for and Documentation of Biological Values
for Use in Risk Assessment (U.S. EPA, 1988), Guidelines for Developmental Toxicity Risk
Assessment (U.S. EPA, 1991), Interim Policy for Particle Size and Limit Concentration Issues in
Inhalation Toxicity (U.S. EPA, 1994a), Methods for Derivation of Inhalation Reference
Concentrations and Application of Inhalation Dosimetry (U.S. EPA, 1994b), Use of the
Benchmark Dose Approach in Health Risk Assessment (U.S. EPA, 1995), Guidelines for
Reproductive Toxicity Risk Assessment (U.S. EPA, 1996), Guidelines for Neurotoxicity Risk
Assessment (U.S. EPA, 1998), Science Policy Council Handbook. Risk Characterization
(U.S. EPA, 2000a), Benchmark Dose Technical Guidance Document (U.S. EPA, 2000b),
Supplementary Guidance for Conducting Health Risk Assessment of Chemical Mixtures
(U.S. EPA, 2000c), A Review of the Reference Dose and Reference Concentration Processes
(U.S. EPA, 2002), Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005a), Supplemental
Guidance for Assessing Susceptibility from Early-Life Exposure to Carcinogens (U.S. EPA,
2005b), Science Policy Council Handbook: Peer Review (U.S. EPA, 2006a), and A Framework
for Assessing Health Risks of Environmental Exposures to Children (U.S. EPA, 2006b).
The literature search strategy employed for this compound was based on the Chemical
Abstracts Service Registry Number and at least one common name. Any pertinent scientific
information submitted by the public to the IRIS Submission Desk was also considered in the
development of this document. The relevant literature was reviewed through December, 2010.
It should be noted that references have been added to the Toxicological Review after the external
peer review in response to peer reviewer's comments and for the sake of completeness. These
references have not changed the overall qualitative and quantitative conclusions.
This document is a draft for review purposes only and does not constitute Agency policy.
2 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
This document is a draft for review purposes only and does not constitute Agency policy.
1 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
2. EXPOSURE CHARACTERIZATION
The purpose of this exposure characterization is to summarize information about
trichloroethylene (TCE) sources, releases, media levels, and exposure pathways for the general
population (occupational exposure is also discussed to a lesser extent). It is not meant as a
substitute for a detailed exposure assessment for a particular risk assessment application. While
this section primarily addresses TCE, it also includes some information on a number of related
compounds. These related compounds include metabolites of TCE and other parent compounds
that produce similar metabolites as shown in Table 2-1. The first column in this table lists the
principal TCE metabolites in humans (trichloroethanol, trichloroethanol-glucuronide and
trichloroacetic acid) as well as a number of minor ones (ATSDR, 1997b). The subsequent
columns list parent compounds that can produce some of the same metabolites. The metabolic
reaction pathways are much more complicated than implied here and it should be understood that
this table is intended only to provide a general understanding of which parent compounds lead to
which TCE metabolites. Exposure to the TCE-related compounds can alter or enhance TCE's
metabolism and toxicity by generating higher internal metabolite concentrations than would
result from TCE exposure by itself. This characterization is based largely on earlier work by Wu
and Schaum (2000, 2001), but also provides updates in a number of areas.
Table 2-1. TCE metabolites and related parent compounds3
TCE metabolites
Parent compounds
Tetrachloro-
ethylene
1,1-Dichloro-
ethane
1,1,1-Tri-
chloroethane
l,l?l>2-Tetra-
chloroethane
1,2-Dichloro-
ethylene
Oxalic acid
X
X
Chloral
X
Chloral hydrate
X
Monochl oroaceti c
acid
X
X
X
X
X
Dichloroacetic acid
X
X
X
Trichloroacetic acid
X
X
X
Trichloroethanol
X
X
X
T ri chl oroethanol -
glucuronide
X
X
X
" X indicates that the parent compound can produce the corresponding metabolite (Hazardous Substances Data
Bank, http://toxnet.nlm.nih. gov./cgi-bin/sis/htmlgen?HSDB 1.
This document is a draft for review purposes only and does not constitute Agency policy.
1 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
2.1. ENVIRONMENTAL SOURCES
TCE is a stable, colorless liquid with a chloroform-like odor and chemical formula
C2CI3H as diagrammed in Figure 2-1 (Lewis, 2001). Its chemical properties are listed in
Table 2-2.
Figure 2-1. Molecular structure of TCE.
Table 2-2. Chemical properties of TCE
Property
Value
Reference
Molecular weight
131.39
Lide et al. (1998)
Boiling point
87.2°C
Lide et al. (1998)
Melting point
-84.7°C
Lide et al. (1998)
Density
1.4642 at 20°C
Budavari (1996)
Solubility
1,280 mg/L water at 25°C
Horvath et al. (1999)
Vapor pressure
69.8 mmHG @ 25°C
Boublik et al.(1984)
Vapor density
4.53 (air =1)
Budavari (1996)
Henry's Law Constant
9.85 x 10 3 atm-cu m/mol @ 25°C
Leighton and Calo
(1981)
Octanol/water partition
coefficient
log Kow = 2.61
Hansch et al. (1995)
Air concentration conversion
1 ppb = 5.38 (J,g/m3
HSDB (2002)
Trichloroethylene has been produced commercially since the 1920s in many countries by
chlorination of ethylene or acetylene. Its use in vapor degreasing began in the 1920s. In the
1930s, it was introduced for use in dry cleaning. This use was largely discontinued in the 1950s
and was replaced with tetrachloroethylene (ATSDR, 1997b). More recently, 80-90% of
trichloroethylene production worldwide is used for degreasing metals (IARC, 1995a). It is also
used in adhesives, paint-stripping formulations, paints, lacquers, and varnishes (SRI
This document is a draft for review purposes only and does not constitute Agency policy.
2 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
International, 1992). A number of past uses in cosmetics, drugs, foods, and pesticides have now
been discontinued including use as an extractant for spice oleoresins, natural fats and oils, hops,
and decaffeination of coffee (IARC, 1995a), and as a carrier solvent for the active ingredients of
insecticides and fungicides, and for spotting fluids (ATSDR, 1997b; WHO, 1985). The
production of TCE in the United States peaked in 1970 at 280 million kg (616 million pounds)
and declined to 60 million kg (132 million pounds) in 1998 (USGS, 2006). In 1996, the United
States imported 4.5 million kg (10 million pounds) and exported 29.5 million kg (65 million
pounds) ("Chemical Profile: Trichloroethylene," 1997). Table 2-3 summarizes the basic
properties and principal uses of the TCE related compounds.
Table 2-3. Properties and uses of TCE related compounds
Water
solubility
(mg/L)
Vapor
pressure
(mmll(j)
Uses
Source
s
T etrachl oroethyl en
e
150
18.5 @25°C
Dry cleaning, degreasing,
solvent
a
1,1,1-Trichloroetha
ne
4,400
124 @25°C
Solvents, degreasing
a
1,2-Dichloroethyle
ne
3,000-6,000
273-395
@30°C
Solvents, chemical
intermediates
a
1,1,1,2-
T etrachl oroethane
1,100
14 @25°C
Solvents, but currently not
produced in United States
a,b
1,1 -Dichl oroethane
5,500
234 @25°C
Solvents, chemical
intermediates
a
Chloral
High
35 @20°C
Herbicide production
a
Chloral hydrate
High
NA
Pharmaceutical production
a
Monochl oroaceti c
acid
High
1 @43°C
Pharmaceutical production
a
Dichl oroaceti c acid
High
<1 @20°C
Pharmaceuticals, not widely
used
a
Trichloroacetic
acid
High
1 @50°C
Herbicide production
a
Oxalic acid
220,000
0.54 @105°C
Scouring/cleaning agent,
degreasing
b
Dichlorovinyl
Not
Not available
Not available
This document is a draft for review purposes only and does not constitute Agency policy.
3 DRAFT—DO NOT CITE OR QUOTE
-------
cysteine
available
Trichloroethanol
Low
NA
Anesthetics and chemical
intermediate
C
1 " Wu and Schaum (2001).
2 bHSDB (2002).
3 c Lewis (2001).
This document is a draft for review purposes only and does not constitute Agency policy.
4 DRAFT—DO NOT CITE OR QUOTE
-------
1 Releases of TCE from nonanthropogenic activities are negligible (HSDB, 2002). Most of the
2 TCE used in the United States is released to the atmosphere, primarily from vapor degreasing
3 operations (ATSDR, 1997b). Releases to air also occur at treatment and disposal facilities, water
4 treatment facilities, and landfills (ATSDR, 1997b). TCE has also been detected in stack
5 emissions from municipal and hazardous waste incineration (ATSDR, 1997b). TCE is on the list
6 for reporting to U.S. Environmental Protection Agency (EPA)'s Toxics Release Inventory (TRI).
7 Reported releases into air predominate over other types and have declined over the period
8 1994-2004 (see Table 2-4).
9
10 Table 2-4. TRI releases of TCE (pounds/year)
11
Year
On-site
fugitive air
On-site
stack air
Total
on-site air
emissions
On-site
surface
water
discharges
Total on-site
underground
injection
Total
on-site
releases
to land
Total
off-site
disposal
or other
releases
Total on-
and
off-site
disposal
or other
releases
1994
15,018,818
15,929,943
30,948,761
1,671
288
4,070
96,312
31,051,102
1995
12,498,086
13,784,853
26,282,939
1,477
550
3,577
74,145
26,362,688
1996
10,891,223
10,995,228
21,886,451
541
1,291
9,740
89,527
21,987,550
1997
9,276,150
8,947,909
18,224,059
568
986
3,975
182,423
18,412,011
1998
6,769,810
6,504,289
13,274,099
882
593
800
136,766
13,413,140
1999
5,861,635
4,784,057
10,645,692
1,034
0
148,867
192,385
10,987,978
2000
5,485,493
4,375,516
9,861,009
593
47,877
9,607
171,952
10,091,038
2001
4,968,282
3,453,451
8,421,733
406
98,220
12,609
133,531
8,666,499
2002
4,761,104
3,436,289
8,197,393
579
140,190
230
139,398
8,477,790
2003
3,963,054
3,121,718
7,084,772
595
90,971
150,642
66,894
7,393,873
2004
3,040,460
3,144,980
6,185,440
216
123,637
2
71,780
6,381,075
2005
2,733,983
2,893,168
5,627,152
533
86,817
4,711
60,074
5,779,287
2006
2,816,241
2,795,184
5,611,425
482
0
77,339
90,758
5,780,004
12
13 Source: EPA TRI Explorer, http://www.epa.gov/triexplorer/trends.htm.
14
15
16 Under the National-Scale Air Toxics Assessment (NSATA) program, EPA has developed
17 an emissions inventory for TCE (U.S. EPA, 2006b). The inventory includes sources in the
18 United States plus the Commonwealth of Puerto Rico and the U.S. Virgin Islands. The types of
19 emission sources in the inventory include large facilities, such as waste incinerators and factories
20 and smaller sources, such as dry cleaners and small manufacturers. Figures 2-2 and 2-3
This document is a draft for review purposes only and does not constitute Agency policy.
5 DRAFT—DO NOT CITE OR QUOTE
-------
Trichloroethylene Emissions
1999
2% Aerospace Industries
2% Integrated Iron & Steel Manufacturing
2% Consumer and Commercial Products Use
4% Dry Cleaning
614 Miscellaneous Metal Parts & Products (Surface Coating)
2% Municipal Landfills
2% Pulp and Paper Production
2% Printing. Coating & Dyeing Of Fabrics
19% Other Categories (293 categories)
59% Halogenated Solvent Cleaners
Figure 2-2. Source contribution to TCE emissions.
>ni»rd
ieetwi
rovlefane®
hnapelle
Distribution of U.S. Ambient Concentrations
Highest In U.S.
95
90
Percentile 75
50
£5
Leweet In U.S.
0.019
0.000 13
0.000 013
0.000 000 74-
0.000 000 16
0.000 000 04-0
0
County Median Ambient Pollutant Concentration
( micrograms / cubic meter)
Source: U.S. EPA / MIPS
1999 UATA Natfonal-Seale Air Toxics Assessment
1999 Estimated County Median Ambient Concentrations
Ethyl acrylate — United States Counties
Figure 2-3. Annual emissions of TCE.
This document is a draft for review purposes only and does not constitute Agency policy.
6 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
show the results of the 1999 emissions inventory for TCE. Figure 2-2 shows the percent
contribution to total emissions by source category. A variety of sources have TCE emissions
with the largest ones identified as halogenated solvent cleaners and metal parts and products.
Figure 2-3 shows a national map of the emission density (tons/sq mi-yr) for TCE. This map
shows the highest densities in the far west and northeastern regions of the United States.
Emissions range from 0-4.12 tons/sq mi-yr.
2.2. ENVIRONMENTAL FATE
2.2.1. Fate in Terrestrial Environments
The dominant fate of trichloroethylene released to surface soils is volatilization. Because
of its moderate water solubility, trichloroethylene introduced into soil (e.g., landfills) also has the
potential to migrate through the soil into groundwater. The relatively frequent detection of
trichloroethylene in groundwater confirms this. Biodegradation in soil and groundwater may
occur at a relatively slow rate (half-lives on the order of months to years) (Howard et al., 1991).
2.2.2. Fate in the Atmosphere
In the atmosphere, trichloroethylene is expected to be present primarily in the vapor
phase, rather than sorbed to particulate, because of its high vapor pressure. Some removal by
scavenging during wet precipitation is expected because of its moderate water solubility. The
major degradation process affecting vapor phase trichloroethylene is photo-oxidation by
hydroxyl radicals. Photolysis in the atmosphere proceeds very slowly, if at all.
Trichloroethylene does not absorb ultraviolet light at wavelengths of less than 290 nm and thus
will not directly photolyze. Based on measured rate data for the vapor phase photo-oxidation
reaction with hydroxyl radicals, the estimated half-life of trichloroethylene in the atmosphere is
on the order of 1-11 days with production of phosgene, dichloroacetyl chloride, and formyl
chloride. Under smog conditions, degradation is more rapid (half-life on the order of hours)
(Howard et al., 1991; HSDB, 2002).
This document is a draft for review purposes only and does not constitute Agency policy.
1 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
2.2.3. Fate in Aquatic Environments
The dominant fate of trichloroethylene released to surface waters is volatilization
(predicted half-life of minutes to hours). Bioconcentration, biodegradation, and sorption to
sediments and suspended solids are not thought to be significant (HSDB, 2002).
Trichloroethylene is not hydrolyzed under normal environmental conditions. However, slow
photo-oxidation in water (half-life of 10.7 months) has been reported (Howard et al., 1991;
HSDB, 2002).
2.3. EXPOSURE CONCENTRATIONS
TCE levels in the various environmental media result from the releases and fate processes
discussed in Sections 2.1 and 2.2. No statistically based national sampling programs have been
conducted that would allow estimates of true national means for any environmental medium. A
substantial amount of air and groundwater data, however, has been collected as well as some
data in other media, as described below.
2.3.1. Outdoor Air—Measured Levels
TCE has been detected in the air throughout the United States. According to ATSDR
(1997b), atmospheric levels are highest in areas concentrated with industry and population, and
lower in remote and rural regions. Table 2-5 shows levels of TCE measured in the ambient air at
a variety of locations in the United States.
More recent ambient air measurement data for TCE were obtained from EPA's Air
Quality System database at the AirData Web site: http://www.epa.gov/air/data/index.html
(2007). These data were collected from a variety of sources including state and local
environmental agencies. The data are not from a statistically based survey and cannot be
assumed to provide nationally representative values. The most recent data (2006) come from
258 monitors located in 37 states. The means for these monitors range from 0.03-7.73 (J,g/m
and have an overall average of 0.23 (j,g/m . Table 2-6 summarizes the data for the years
1999-2006. The data suggest that levels have remained fairly constant since 1999 at about
"3
0.3 (J,g/m . Table 2-7 shows the monitoring data organized by land setting (rural, suburban, or
urban) and land use (agricultural, commercial, forest, industrial, mobile, and residential). Urban
air levels are almost four times higher than rural areas. Among the land use categories, TCE
levels are highest in commercial/industrial areas and lowest in forest areas.
This document is a draft for review purposes only and does not constitute Agency policy.
8 DRAFT—DO NOT CITE OR QUOTE
-------
2.3.2. Outdoor Air—Modeled Levels
1 Under the National-Scale Air Toxics Assessment program, EPA has compiled emissions
2 data and modeled air concentrations/exposures for the Criteria Pollutants and Hazardous Air
3 Pollutants (U.S. EPA, 2006b). The results of the 1999 emissions inventory for TCE were
4 discussed earlier and results presented in Figures 2-2 and 2-3. A computer simulation model
5 known as the Assessment System for Population Exposure Nationwide (ASPEN) is used to
6 estimate toxic air pollutant concentrations (http://www.epa.gov/ttnatw01/nata/aspen.htmn. This
7 model is based on the EPA's Industrial Source Complex Long Term model which simulates the
8 behavior of the pollutants after they are emitted into the atmosphere. ASPEN uses estimates of
9 toxic air pollutant emissions and meteorological data from National Weather Service Stations to
10 estimate
This document is a draft for review purposes only and does not constitute Agency policy.
9 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
Table 2-5. Concentrations of trichloroethylene in ambient air
Concentration (jig/m3)
Area
Year
Mean
Range
Rural
Whiteface Mountain, NYa
1974
0.5
<0.3-1.9
Badger Pass, CAa
1977
0.06
0.005-0.09
Reese River, NVa
1977
0.06
0.005-0.09
Jetmar, KSa
1978
0.07
0.04-0.11
All rural sites
1974-1978
0.005-1.9
Urban and Suburban
New Jerseya
1973-79
9.1
ND-97
New York City, NYa
1974
3.8
0.6-5.9
Los Angeles, CAa
1976
1.7
0.14-9.5
Lake Charles, LAa
1976-78
8.6
0.4-11.3
Phoenix, AZa
1979
2.6
0.06-16.7
Denver, COa
1980
1.07
0.15-2.2
St. Louis, MOa
1980
0.6
0.1-1.3
Portland, ORa
1984
1.5
0.6-3.9
Philadelphia, PAa
1983-1984
1.9
1.6-2.1
Southeast Chicago, ILb
1986-1990
1.0
East St. Louis, ILb
1986-1990
2.1
District of Columbia0
1990-1991
1.94
1-16.65
Urban Chicago, ILd
pre-1993
0.82-1.16
Suburban Chicago, ILd
pre-1993
0.52
300 cities in 42 states6
pre-1986
2.65
Several Canadian Citiesf
1990
0.28
Several United States Cities
1990
6.0
Phoenix, AZg
1994-1996
0.29
0-1.53
Tucson, AZg
1994-1996
0.23
0-1.47
All urban/sub urban sites
1973-1996
0-97
aIARC (1995a).
b Sweet (1992).
°Hendler (1992).
dScheff (1993).
e Shah (1988).
fBunce (1994).
8Zielinska-Psuja (1998).
ND = nondetect
This document is a draft for review purposes only and does not constitute Agency policy.
10 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 2-6. TCE ambient air monitoring data (jig/m3)
2
Year
Number of
monitors
Number of
states
Mean
Standard
deviation
Median
Range
1999
162
20
0.30
0.53
0.16
0.01-4.38
2000
187
28
0.34
0.75
0.16
0.01-7.39
2001
204
31
0.25
0.92
0.13
0.01-12.90
2002
259
41
0.37
1.26
0.13
0.01-18.44
2003
248
41
0.35
0.64
0.16
0.02-6.92
2004
256
37
0.32
0.75
0.13
0.00-5.78
2005
313
38
0.43
1.05
0.14
0.00-6.64
2006
258
37
0.23
0.55
0.13
0.03-7.73
3
4 Source: EPA's Air Quality System database at the AirData Web site:
5 http ://www. epa. gov/air/data/index .html.
6
7
8 Table 2-7. Mean TCE air levels across monitors by land setting and use
9 (1985-1998)
10
Rural
Subur-
ban
Urban
Agricul
-tural
Com-
mercia
1
Fores
t
Indus-
trial
Mobile
Resi-
dentia
1
Mean
concentration
(l^g/m3)
0.42
1.26
1.61
1.08
1.84
0.1
1.54
1.5
0.89
n
93
500
558
31
430
17
186
39
450
11
12 Source: EPA's Air Quality System database at the AirData Web site:
13 http ://www. epa. gov/air/data/index .html.
14
15
16 air toxics concentrations nationwide. The ASPEN model takes into account important
17 determinants of pollutant concentrations, such as
18
19
20 • rate of release;
21 • location of release;
22 • the height from which the pollutants are released;
This document is a draft for review purposes only and does not constitute Agency policy.
11 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
• wind speeds and directions from the meteorological stations nearest to the release;
• breakdown of the pollutants in the atmosphere after being released (i.e., reactive decay);
• settling of pollutants out of the atmosphere (i.e., deposition); and
• transformation of one pollutant into another (i.e., secondary formation).
The model estimates toxic air pollutant concentrations for every census tract in the
continental United States, the Commonwealth of Puerto Rico and the U.S. Virgin Islands.
Census tracts are land areas defined by the U.S. Bureau of the Census and typically contain about
4,000 residents each. Census tracts are usually smaller than 2 square miles in size in cities but
much larger in rural areas.
Figure 2-4 shows the results of the 1999 ambient air concentration modeling for TCE.
3 3
The county median air levels range from 0-3.79 |ig/m and an overall median of 0.054 |ig/m .
They have a pattern similar to the emission densities shown in Figure 2-3. These NSATA
modeled levels appear lower than the monitoring results presented above. For example, the 1999
"3
air monitoring data (see Table 2-6) indicates a median outdoor air level of 0.16 (j,g/m which is
-3
about three times as high as the modeled 1999 county median (0.054 |ig/m ). However, it should
be understood that the results from these two efforts are not perfectly comparable. The modeled
value is a median of county levels for the entire United States which includes many rural areas.
The monitors cover many fewer areas (n = 162 for 1999) and most are in nonrural locations. A
better analysis is provided by EPA (2006b) which presents a comparison of modeling results
from NSATA to measured values at the same locations. For 1999, it was found that
formaldehyde levels were underestimated at 79% of the sites (n = 92). Thus, while the NSATA
modeling results are useful for understanding geographic distributions, they may frequently
underestimate ambient levels.
2.3.3. Indoor Air
TCE can be released to indoor air from use of consumer products that contain it (i.e.,
adhesives and tapes), vapor intrusion (migration of volatile chemicals from the subsurface into
overlying buildings) and volatilization from the water supply. Where such sources are present, it
is likely that indoor levels will be higher than outdoor levels. A number of studies have
measured indoor levels of TCE:
This document is a draft for review purposes only and does not constitute Agency policy.
12 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
• The 1987 EPA Total Exposure Assessment Methodology study (U.S. EPA, 1987) showed
that the ratio of indoor to outdoor TCE concentrations for residences in Greensboro, NC,
was about 5:1.
1990 Estimated County Median Ambient Concentrations
Trichbioethylene — United States Counties
JiMeau
MarrtpiltB"
awes
SaeramwA
Defer
mm
i-ho*ri
HsnSMu
Distribution of U.S. Ambient Concentrations
Percentile
Hlghast In U.S.
95
90
75
50
25
Lowiet In U.S.
3.79
0.12
0.099 county Median Ambient Pollutant Concentration
C microqrarns / cubic meter)
QkQ54- * '
jSource; U.S. EPA / OAQPS
1999 NATA Natfonal-bcale Air Toxics Assessment
Figure 2-4. Modeled ambient air concentrations of TCE.
• In two homes using well water with TCE levels averaging 22-128 (J,g/L, the TCE levels
in bathroom air ranged from <500-40,000 (J,g/m3 when the shower ran less than
30 minutes (Andelman, 1985).
"3
• Shah and Singh (1988) report an average indoor level of 7.2 |ig/m based on over
2,000 measurements made in residences and workplaces during 1981-1984 from various
locations across the United States.
• Hers et al. (2001) provides a summary of indoor air TCE measurements at locations in
United States, Canada, and Europe with a range of <1-165 (j,g/m3.
• Sapkota et al. (2005) measured TCE levels inside and outside of the Baltimore Harbor
Tunnel toll booths during the summer of 2001. Mean TCE levels were 3.11 (J,g/m
indoors and 0.08 (J,g/m3 outdoors based on measurements on 7 days. The authors
This document is a draft for review purposes only and does not constitute Agency policy.
13 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
speculated that indoor sources, possibly dry cleaning residues on uniforms, were the
primary source of the indoor TCE.
• Sexton et al. (2005) measured TCE levels inside and outside residences in
Minneapolis/St. Paul metropolitan area. Two day samples were collected over
3 3
three seasons in 1999. Mean TCE levels were 0.5 (J,g/m indoors (n = 292), 0.2 (J,g/m
outdoors (n = 132) and 1.0 (J,g/m3 based on personal sampling (n = 288).
• Zhu et al. (2005) measured TCE levels inside and outside of residences in Ottawa,
Canada. Seventy-five homes were randomly selected and measurements were made
during the winter of 2002/2003. TCE was above detection limits in the indoor air of
33% of the residences and in the outdoor air of 19% of the residences. The mean levels
3 3
were 0.06 (J,g/m indoors and 0.08 (J,g/m outdoors. Given the high frequency of
nondetects, a more meaningful comparison can be made on basis of the 75th percentiles:
0.08 (J,g/m3 indoors and 0.01 (J,g/m3 outdoors.
TCE levels measured indoors have been directly linked to vapor intrusion at two sites in New
York:
• TCE vapor intrusion has occurred in buildings/residences near a former Smith Corona
manufacturing facility located in Cortlandville, NY. An extensive sampling program
-3
conducted in 2006 has detected TCE in groundwater (1-13 (J,g/L), soil gas (6-97 (J,g/m ),
subslab gas (2-1,600 (j.g/m3), and indoor air (1-17 (J,g/m3) (NYSDEC, 2006a).
• Evidence of vapor intrusion of TCE has also been reported in buildings and residences in
Endicott, NY. Sampling in 2003 showed total volatile organic compounds (VOCs) in
soil gas exceeding 10,000 (j,g/m3 in some areas. Indoor air sampling detected TCE levels
ranging from 1-140 (J,g/m3 (NYSDEC, 2006b).
Little et al. (1992) developed attenuation coefficients relating contaminants in soil gas
(assumed to be in chemical equilibrium with the groundwater) to possible indoor levels as a
result of vapor intrusion. On this basis they estimated that TCE groundwater levels of 540 (J,g/L,
"3
(a high contamination level) could produce indoor air levels of 5-500 (J,g/m . Vapor intrusion is
likely to be a significant source only in situations where residences are located near soils or
groundwater with high contamination levels. EPA (2002b) recommends considering vapor
intrusion when volatiles are suspected to be present in groundwater or soil at a depth of
<100 feet. Hers et al. (2001) concluded that the contribution of VOCs from subsurface sources
relative to indoor sources is small for most chemicals and sites.
This document is a draft for review purposes only and does not constitute Agency policy.
14 DRAFT—DO NOT CITE OR QUOTE
-------
2.3.4. Water
1 A number of early (pre-1990) studies measured TCE levels in natural water bodies
2 (levels in drinking water are discussed later in this section) as summarized in Table 2-8.
3 Table 2-8. Concentrations of trichloroethylene in water based on pre-1990
4 studies
5
Mean
Median
Range
Number of
Water type
Location
Year
(Hg/L)
(Hg/L)
(Hg/L)
samples
Ref.
Industrial effluent
U.S.
83
0.5
NR
IARC (1995a)
Surface waters
U.S.
83
0.1
NR
IARC (1995a)
Rainwater
Portland, OR
84
0.006
0.002-0.02
NR
Ligocki et al. (1985)
Groundwater
MN
83
0.2-144
NR
Sabel and Clark (1984)
NJ
76
<1,530
NR
Burmaster et al. (1982)
NY
80
<3,800
NR
Burmaster et al. (1982)
PA
80
<27,300
NR
Burmaster et al. (1982)
MA
76
<900
NR
Burmaster et al. (1982)
AZ
8.9-29
NR
IARC (1995a)
Drinking water
U.S.
76
0.2-49
IARC (1995a)
U.S.
77
0-53
IARC (1995a)
U.S.
78
0.5-210
IARC (1995a)
MA
84
max. 267
IARC (1995a)
NJ
84
23.4
max. 67
1130
Cohnetal. (1994a)
CA
85
8-12
486
EPA, (1987)
CA
84
66
486
EPA, (1987)
NC
84
5
48
EPA, (1987)
ND
84
5
48
EPA, (1987)
6
7 NR = Not Reported.
8
9
10 According to IARC (1995a), the reported median concentrations of TCE in 1983-1984 were
11 0.5 [j,g/L in industrial effluents and 0.1 [j,g/L in ambient water. Results from an analysis of the
12 EPA STORET Data Base (1980-1982) showed that TCE was detected in 28% of 9,295 surface
13 water reporting stations nationwide (ATSDR, 1997b). A more recent search of the STORET
14 database for TCE measurements nationwide during 2008 in streams, rivers and lakes indicated
15 three detects (0.03-0.04 |ig/L) out of 150 samples (STORET Database,
16 http://www.epa.gov/storet/dbtop.htmn.
17 ATSDR (1997b) has reported that TCE is the most frequently reported organic
18 contaminant in groundwater and the one present in the highest concentration in a summary of
19 ground water analyses reported in 1982. It has been estimated that between 9 and 34% of the
20 drinking water supply sources tested in the United States may have some trichloroethylene
21 contamination. This estimate is based on available Federal and State surveys (ATSDR, 1997b).
This document is a draft for review purposes only and does not constitute Agency policy.
15 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Squillace et al. (2004) reported TCE levels in shallow groundwater based on data from
the National Water Quality Assessment Program managed by USGS. Samples from 518 wells
were collected from 1996-2002. All wells were located in residential or commercial areas and
had a median depth of 10 m. The authors reported that approximately 8.3% of the well levels
were above the detection limit (level not specified), 2.3% were above 0.1 [j,g/L and 1.7% were
above 0.2 (J,g/L.
As part of the Agency's first Six-Year Review, EPA obtained analytical results for over
200,000 monitoring samples reported at 23,035 public water systems (PWS) in 16 states (U.S.
EPA, 2003a). Approximately 2.6% of the systems had at least one sample exceed a minimum
reporting level of 0.5 (J,g/L; almost 0.65% had at least one sample that exceeds the maximum
contaminant level (MCL) of 5 (J,g/L. Based on average system concentrations estimated by EPA,
54 systems (0.23%) had an average concentration that exceeded the MCL. EPA's statistical
analysis to extrapolate the sample result to all systems regulated for TCE resulted in an estimate
of 154 systems with average TCE concentrations that exceed the MCL.
TCE concentrations in ground water have been measured extensively in California. The
data were derived from a survey of water utilities with more than 200 service connections. The
survey was conducted by the California Department of Health Services (CDHS, 1986). From
January 1984 through December 1985, untreated water from wells in 819 water systems were
sampled for organic chemical contamination. The water systems use a total of 5,550 wells,
2,947 of which were sampled. TCE was found in 187 wells at concentrations up to 440 (J-g/L,
with a median concentration among the detects of 3.0 (J,g/L. Generally, the wells with the highest
concentrations were found in the heavily urbanized areas of the state. Los Angeles County
registered the greatest number of contaminated wells (149).
A second California study collected data on TCE levels in public drinking water
(Williams et al., 2002). The data were obtained from the CA DHS. The data spanned the years
1995-2001 and the number of samples for each year ranged from 3,447-4,226. The percent of
sources that were above the detection limit ranged from 9.6-11.7 per year (detection limits not
specified). The annual average detected concentrations ranged from 14.2-21.6 (J,g/L. Although
not reported, the overall average concentration of the samples (assuming an average of 20 [j,g/L
among the samples above the detection limit, 10% detection rate and 0 for the nondetects) would
be about 2 (J,g/L.
The USGS (2006) conducted a national assessment of 55 VOCs, including
trichloroethylene, in ground water. A total of 3,500 water samples were collected during
1985-2001. Samples were collected at the well head prior to any form of treatment. The types
of wells sampled included 2,400 domestic wells and 1,100 public wells. Almost 20% of the
This document is a draft for review purposes only and does not constitute Agency policy.
16 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
samples contained one or more of the VOCs above the assessment level of 0.2 (J,g/L. The
detection frequency increased to over 50% when a subset of samples was analyzed with a low
level method that had an assessment level of 0.02 (J,g/L. The largest detection frequencies were
observed in California, Nevada, Florida, the New England States, and Mid-Atlantic states. The
most frequently detected VOCs (>1% of samples) include TCE, tetrachloroethylene,
1,1,1-trichloroethane (methyl chloroform), 1,2 dichloroethylene, and 1,1-dichloroethane.
Findings specific to TCE include the following:
• Detection frequency was 2.6% at 0.2 [j,g/L and was 3.8% at 0.02 (J,g/L.
• The median concentration was 0.15 [j.g/L with a range of 0.02-100 (J,g/L.
• The number of samples exceeding the MCL (5 (J,g/L) was six at domestic wells and nine
at public wells.
USGS (2006) also reported that four solvents (TCE, tetrachloroethylene,
1,1,1-trichloroethane and methylene chloride) occurred together in 5% of the samples. The most
frequently occurring two-solvent mixture was TCE and tetrachloroethylene. The report stated
that the most likely reason for this co-occurrence is the reductive dechlorination of
tetrachloroethylene to TCE.
2.3.5. Other Media
Levels of TCE were found in the sediment and marine animal tissue collected in
1980-1981 near the discharge zone of a Los Angeles County waste treatment plant.
Concentrations were 17 [j,g/L in the effluent, <0.5 (J,g/kg in dry weight in sediment, and
0.3-7 (J,g/kg wet weight in various marine animal tissue (IARC, 1995a). TCE has also been
found in a variety of foods. U.S. Food and Drug Administration (FDA) has limits on TCE use as
a food additive in decaffeinated coffee and extract spice oleoresins (see Table 2-15). Table 2-9
summarizes data from two sources:
• IARC (1995a) reports average concentrations of TCE in limited food samples collected in
the United States.
• Jones and Smith (2003) measured VOC levels in over 70 foods collected from 1996-
2000 as part of the FDA's Total Diet Program. All foods were collected directly from
This document is a draft for review purposes only and does not constitute Agency policy.
17 DRAFT—DO NOT CITE OR QUOTE
-------
1 supermarkets. Analysis was done on foods in a ready-to-eat form. Sample sizes for most
2 foods were in the 2-5 range.
3
4 Table 2-9. Levels in food
5
IARC (1995a)
Fleming-Jones and Smith (2003)
Cheese 3.8 (J,g/kg
Butter and Margarine 73.6 (J,g/kg
Cheese 2-3 (J,g/kg
Butter 7-9 (J,g/kg
Margarine 2-21 (J,g/kg
Cheese Pizza 2 (J,g/kg
Peanut Butter 0.5 (J,g/kg
Nuts 2-5 (J,g/kg
Peanut Butter 4-70 (J,g/kg
Ground Beef 3-6 (J,g/kg
Beef Frankfurters 2-105 |ig/kg
Hamburger 5-9 (J,g/kg
Cheeseburger 7 (J,g/kg
Chicken Nuggets 2-5 (J,g/kg
Bologna 2-20 (J,g/kg
Pepperoni Pizza 2 (J,g/kg
Banana 2 (J,g/kg
Avocado 2-75 (J,g/kg
Orange 2 (J,g/kg
Chocolate Cake 3-57 (J,g/kg
Blueberry Muffin 3-4 (J,g/kg
Sweet Roll 3 (J,g/kg
Chocolate Chip Cookies 2-4 (J,g/kg
Apple Pie 2-4 (J,g/kg
Doughnuts 3 (J,g/kg
Tuna 9-11 (J,g/kg
Cereals 3 (J,g/kg
Grain-based Foods 0.9 (J,g/kg
Cereal 3 (J,g/kg
Popcorn 4-8 (J,g/kg
French Fries 3 (J,g/kg
Potato Chips 4-140 (J,g/kg
Coleslaw 3 (J,g/kg
6
7
This document is a draft for review purposes only and does not constitute Agency policy.
18 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
2.3.6. Biological Monitoring
Biological monitoring studies have detected TCE in human blood and urine in the United
States and other countries such as Croatia, China, Switzerland, and Germany (IARC, 1995a).
Concentrations of TCE in persons exposed through occupational degreasing operations were
most likely to have detectable levels (IARC, 1995a). In 1982, eight of eight human breastmilk
samples from four United States urban areas had detectable levels of TCE. The levels of TCE
detected, however, are not specified (ATSDR, 1997b; HSDB, 2002).
The Third National Health and Nutrition Examination Survey (NHANES III) examined
TCE concentrations in blood in 677 nonoccupationally exposed individuals. The individuals
were drawn from the general U.S. population and selected on the basis of age, race, gender and
region of residence (Ashley et al., 1994; IARC, 1995a). The samples were collected during
1988-1994. TCE levels in whole blood were below the detection limit of 0.01 [j,g/L for about
90% of the people sampled (see Table 2-10). Assuming that nondetects equal half of the
detection limit, the mean concentration was about 0.017 (J,g/L.
Table 2-10. TCE levels in whole blood by population percentile
Percentiles
10
20
30
40
50
60
70
80
90
Concentration
Og/L)
ND
ND
ND
ND
ND
ND
ND
ND
0.012
ND = Nondetect, i.e., below detection limit of 0.01 (ig/L.
Data from IARC (1995a) and Ashley et al. (1994).
2.4. EXPOSURE PATHWAYS AND LEVELS
2.4.1. General Population
Because of the pervasiveness of TCE in the environment, most people are likely to have
some exposure via one or more of the following pathways: ingestion of drinking water,
inhalation of outdoor/indoor air, or ingestion of food (ATSDR, 1997b). As noted earlier, the
NHANES survey suggests that about 10% of the population has detectable levels of TCE in
blood. Each pathway is discussed below.
This document is a draft for review purposes only and does not constitute Agency policy.
19 DRAFT—DO NOT CITE OR QUOTE
-------
2.4.1.1.1. Inhalation
1 As discussed earlier, EPA has estimated emissions and modeled air concentrations for the
2 Criteria Pollutants and Hazardous Air Pollutants under the National-Scale Air Toxics
3 Assessment program (U.S. EPA, 2006b). This program has also estimated inhalation exposures
4 on a nationwide basis. The exposure estimates are based on the modeled concentrations from
5 outdoor sources and human activity patterns (Rosenbaum, 2005). Table 2-11 shows the 1999
6 results for TCE.
7 These modeled inhalation exposures would have a geographic distribution similar to that
8 of the modeled air concentrations as shown in Figure 2-4. Table 2-11 indicates that TCE
9 inhalation exposures in urban areas are generally about twice as high as rural areas. While these
10 modeling results are useful for understanding the geographic distribution of exposures, they
11 Table 2-11. Modeled 1999 annual exposure concentrations (jig/m3) for
12 trichloroethylene
13
Percentile
Exposure concentration (
ug/m3)
Rural areas
Urban areas
Nationwide
5
0.030
0.048
0.038
10
0.034
0.054
0.043
25
0.038
0.065
0.056
50
0.044
0.086
0.076
75
0.053
0.122
0.113
90
0.070
0.189
0.172
95
0.097
0.295
0.262
Mean
0.058
0.130
0.116
14
15 Percentiles and mean are based on census tract values.
16 Source: http://www.epa.gOv/ttn/atw/nata/ted/exporisk.html#indb.
17
18
19 appear to underestimate actual exposures. This is based on the fact that, as discussed earlier, the
20 modeled ambient air levels are generally lower than measured values. Also, the modeled
21 exposures do not consider indoor sources. Indoor sources of TCE make the indoor levels higher
22 than ambient levels. This is particularly important to consider since people spend about 90% of
23 their time indoors (U.S. EPA, 1997a). A number of measurement studies were presented earlier
24 that showed higher TCE levels indoors than outdoors. Sexton et al. (2005) measured TCE levels
3 3
25 in Minneapolis/St. Paul area and found means of 0.5 (J,g/m indoors (n = 292) and 1.0 (J,g/m
This document is a draft for review purposes only and does not constitute Agency policy.
20 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
-3
based on personal sampling (n = 288). Using 1.0 (j,g/m and an average adult inhalation rate of
"3
13 m air/day (U.S. EPA, 1997a) yields an estimated intake of 13 (J,g/day. This is consistent with
ATSDR (1997b), which reports an average daily air intake for the general population of
11-33 (J,g/day.
2.4.1.1.2. Ingestion
The median value from the nationwide survey of domestic and public wells by USGS for
1985-2001 is 0.15 (J,g/L. This value was selected for exposure estimation purposes because it
was the most current and most representative of the national population. Using this value and an
average adult water consumption rate of 1.4 L/d yields an estimated intake of 0.2 (j,g/day. [This is
from U.S. EPA (1997a), but note that U.S. EPA (2004) indicates a mean per capita daily average
total water ingestion from all sources of 1.233 L], This is lower than the ATSDR (1997b)
estimate water intake for the general population of 2-20 (j,g/day. The use of the USGS survey to
represent drinking water is uncertain in two ways. First, the USGS survey measured only
groundwater and some drinking water supplies use surface water. Second, the USGS measured
TCE levels at the well head, not the drinking water tap. Further discussion about the possible
extent and magnitude of TCE exposure via drinking water is presented below.
According to ATSDR (1997b), TCE is the most frequently reported organic contaminant
in ground water (1997b), and between 9 and 34% of the drinking water supply sources tested in
the United States may have some TCE contamination. Approximately 90% of the 155,000
public drinking water systemsl in the United States are ground water systems. The drinking
water standard for TCE only applies to community water systems (CWSs) and approximately
78%) of the 51,972 CWSs in the United States are ground water systems (U.S. EPA, 2008c).
Although commonly detected in water supplies, the levels are generally low because, as
discussed earlier, MCL violations for TCE in public water supplies are relatively rare for any
extended period (U.S. EPA, 1998). The USGS (2006) survey found that the number of samples
exceeding the MCL (5 (^g/L) was six at domestic wells (n = 2,400) and nine at public wells (n =
1,100). Private wells, however, are often not closely monitored and if located near TCE
disposal/contamination sites where leaching occurs, may have undetected contamination levels.
About 10%o of Americans (27 million people) obtain water from sources other than public water
systems, primarily private wells (U.S. EPA, 1995). TCE is a common contaminant at Superfund
1 PWSs are defined as systems which provide water for human consumption through pipes or other constructed
conveyances to at least 15 service connections or serves an average of at least 25 people for at least 60 days a year.
EPA further specifies three types of PWSs, including CWS)—a PWS that supplies water to the same population
year-round.
This document is a draft for review purposes only and does not constitute Agency policy.
21 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
sites. It has been identified in at least 861 of the 1,428 hazardous waste sites proposed for
inclusion on the EPA National Priorities List (NPL) (ATSDR, 1997b). Studies have shown that
many people live near these sites: 41 million people live less than 4 miles from one or more of
the nation's NPL sites, and on average 3,325 people live within 1 mile of any given NPL site
(ATSDR, 1996a).
Table 2-12 presents preliminary estimates of TCE intake from food. They are based on
average adult food ingestion rates and food data from Table 2-9. This approach suggests a total
ingestion intake of about 5 j_ig/d. It is important to consider this estimate as preliminary because
it is derived by applying data from very limited food samples to broad classes of food.
Table 2-12. Preliminary estimates of TCE intake from food ingestion
Consumption
rate (g/kg-day)
Consumption
rate (g/day)
Concentration
in food ((ig/kg)
Intake
(Hg/day)
Fruit
3.4
238
2
0.48
Vegetables
4.3
301
3
0.90
Fish
20
10
0.20
Meat
2.1
147
5
0.73
Dairy products
8
560
3
1.68
Grains
4.1
287
3
0.86
Sweets
0.5
35
3
0.10
Total
4.96
a Consumption rates are per capita averages from EPA (1997a).
b Consumption rates in g/d assume 70 kg body weight.
2.4.1.1.3. Dermal
TCE in bathing water and consumer products can result in dermal exposure. A modeling
study has suggested that a significant fraction of the total dose associated with exposure to
volatile organics in drinking water results from dermal absorption (Brown et al., 1984). EPA
(2004) used a prediction model based on octanol-water partitioning and molecular weight to
derive a dermal permeability coefficient for TCE in water of 0.012 cm/hour. EPA used this
value to compute the dermally absorbed dose from a 35 minute shower and compared it to the
dose from drinking 2 L of water at the same concentration. This comparison indicated that the
This document is a draft for review purposes only and does not constitute Agency policy.
22 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
dermal dose would be 17% of the oral dose. Much higher dermal permeabilities were reported
by Nakai et al. (1999) based on human skin in vitro testing. For dilute aqueous solutions of
TCE, they measured a permeability coefficient of 0.12 cm/hour (26°C). Nakai et al. (1999) also
measured a permeability coefficient of 0.018 cm/hour for tetrachloroethylene in water. Poet
et al. (2000) measured dermal absorption of TCE in humans from both water and soil matrices.
The absorbed dose was estimated by applying a physiologically based pharmacokinetic model to
TCE levels in breath. The permeability coefficient was estimated to be 0.015 cm/hour for TCE
in water and 0.007 cm/hour for TCE in soil (Poet et al., 2000).
2.4.1.1.4. Exposure to Trichloroethylene (TCE) Related Compounds
Table 2-13 presents adult exposure estimates that have been reported for the TCE related
compounds. This table was originally compiled by Wu and Schaum (2001). The exposure/dose
estimates are taken directly from the listed sources or derived based on monitoring data
presented in the source documents. They are considered "preliminary" because they are
generally based on very limited monitoring data. These preliminary estimates suggest that
exposures to most of the TCE related compounds are comparable to or greater than TCE itself.
Table 2-13. Preliminary intake estimates of TCE and TCE-related chemicals
Chemical
Population
Media
Range of estimated
adult exposures
(jig/day)
Range of adult doses
(mg/kg-day)
Data sources"
T richloroethylene
General
Air
11-33
1.57E-04-4.71E-04
ATSDR (1997b)
General
Water
2-20b
2.86E-05-2.86E-04
ATSDR (1997b)
Occupational
Air
2,232-9,489
3.19E-02-1.36E-01
ATSDR (1997b)
T etrachloroethylene
General
Air
80-200
1.14E-03-2.86E-03
ATSDR (1997a)
General
Water
0.1-0.2
1.43E-06-2.86E-06
ATSDR (1997a)
Occupational
Air
5,897-219,685
8.43E-02-3.14
ATSDR (1997a)
1,1,1 -T richloroethane
General
Air
10.8-108
1.54E-04-1.54E-03
ATSDR (1995)
General
Water
0.38-4.2
5.5E-06-6.00E-05
ATSDR (1995)
1,2-Dichloroethylene
General
Air
1-6
1.43E-05-8.57E-05
ATSDR (1996b)
General
Water
2.2
3.14E -05
ATSDR (1996b)
Cis-l,2-Dichloroethylene
General
Air
5.4
7.71E -05
HSDB (1996)
General
Water
0.5-5.4
7.14E-06-7.71E-05
HSDB (1996)
1,1,1,2-Tetrachloroethane
General
Air
142
2.03E -03
HSDB (2002)
This document is a draft for review purposes only and does not constitute Agency policy.
23 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
1,1 -Dichloroethane
General
Air
4
5.71E -05
ATSDR (1990)
General
Water
2.47-469.38
3.53E-05-6.71E-03
ATSDR (1990)
Chloral
General
Water
0.02-36.4
2.86E-07-5.20E-04
HSDB (1996)
Monochloroacetic acid
General
Water
2-2.4
2.86E-05-3.43E-05
EPA (1994b)
Dichloroacetic acid
General
Water
10-266
1.43E-04-3.80E-03
IARC (1995a)
Trichloroacetic acid
General
Water
8.56-322
1.22E-03-4.60E-03
IARC (1995a)
"Originally compiled in Wu and Schaum (2001).
b New data from USGS (2006) suggests much lower water intakes, i.e., 0.2 jxg/d.
2.4.2. Potentially Highly Exposed Populations
Some members of the general population may have elevated TCE exposures. ATSDR
(1997b) has reported that TCE exposures may be elevated for people living near waste facilities
where TCE may be released, residents of some urban or industrialized areas, people exposed at
work (discussed further below) and individuals using certain products (also discussed further
below). Because TCE has been detected in breast milk samples of the general population,
infants who ingest breast milk may be exposed, as well. Increased TCE exposure is also a
possible concern for bottle-fed infants because they ingest more water on a bodyweight basis
than adults (the average water ingestion rate for adults is 21 mL/kg-day and for infants under one
year old it is 44 mL/kg-day) (U.S. EPA, 1997a). Also, because TCE can be present in soil,
children may be exposed through activities such as playing in or ingesting soil.
2.4.2.1.1. Occupational Exposure
Occupational exposure to TCE in the United States has been identified in various
degreasing operations, silk screening, taxidermy, and electronics cleaning (IARC, 1995a). The
major use of trichloroethylene is for metal cleaning or degreasing (IARC, 1995a). Degreasing is
used to remove oils, greases, waxes, tars, and moisture before galvanizing, electroplating,
painting, anodizing, and coating. The five primary industries using TCE degreasing are furniture
and fixtures; electronic and electric equipment; transport equipment; fabricated metal products;
and miscellaneous manufacturing industries (IARC, 1995a). Additionally, TCE is used in the
manufacture of plastics, appliances, jewelry, plumbing fixtures, automobile, textiles, paper, and
glass (IARC, 1995a).
This document is a draft for review purposes only and does not constitute Agency policy.
24 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Table 2-14 lists the primary types of industrial degreasing procedures and the years that
the associated solvents were used. Vapor degreasing has the highest potential for exposure
because vapors can escape into the work place. Hot dip tanks, where trichloroethylene is heated
to close to its boiling point of 87°C, are also major sources of vapor that can create exposures as
high as vapor degreasers. Cold dip tanks have a lower exposure potential, but they have a large
surface area which enhances volatilization. Small bench-top cleaning operations with a rag or
brush and open bucket have the lowest exposure potential. In combination with the vapor
source, the size and ventilation of the workroom are the main determinants of exposure intensity
(NRC, 2006).
Occupational exposure to TCE has been assessed in a number of epidemiologic and
industrial hygiene studies. Bakke et al. (2007) estimated that the arithmetic mean of TCE
occupational exposures across all industries and decades (mostly 1950s, 1970s, and 1980s)
"3
was38.2 ppm (210 mg/m ). They also reported that the highest personal and area air levels were
Table 2-14. Years of solvent use in industrial degreasing and cleaning
operations
Years
Vapor degreasers
Cold dip tanks
Rag or brush and bucket on bench
top
-1934-1954
T ri chl oroethylene
(poorly controlled)
Stoddard solventa
Stoddard solvent (general use),
alcohols (electronics shop), carbon
tetrachloride (instrument shop).
-1955-1968
Tri chl oroethylene
(poorly controlled,
tightened in 1960s)
Tri chl oroethyl ene
(replaced some
Stoddard solvent)
Stoddard solvent, trichloroethylene
(replaced some Stoddard solvent),
perchloroethylene,
1,1,1-trichloroethane (replaced carbon
tetrachloride, alcohols, ketones).
-1969-1978
Tri chl oroethylene,
(better controlled)
Tri chl oroethyl ene,
Stoddard solvent
T ri chl oroethyl ene, per chl oroethyl ene,
1,1,1-tri chl oroethane, alcohols,
ketones, Stoddard solvent.
-1979-1990
s
1,1,1 -Trichloroethan
e(replaced
tri chl oroethyl ene)
1,1,1-Tri chl oroetha
ne (replaced
tri chl oroethyl ene),
Stoddard solvent
1,1,1 -Tri chl oroethane,
perchloroethylene, alcohols, ketones,
Stoddard solvent.
a A mixture of straight and branched chain paraffins (48%), naphthenes (38%), and aromatic hydrocarbons (14%).
Source: Stewart and Dosemeci (2005) and Bakke et al. (2007).
-3
found in vapor degreasing operations (arithmetic mean of 44.6 ppm or 240 mg/m ). Hein et al.
(2010) developed and evaluated statistical models to estimate the intensity of occupational
This document is a draft for review purposes only and does not constitute Agency policy.
25 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
exposure to trichloroethylene (and other solvents) using a database of air measurement data and
associated exposure determinants. The measurement database was compiled from the published
literature and National Institute for Occupational Safety and Health (NIOSH) reports from 1940-
1998 (n = 484) and were split between personal (47%)and area (53%) measurements. The
predicted arithmetic mean exposure intensity levels for the evaluated exposure scenarios ranged
from 0.21-3,700 ppm (1.1-20,000 mg/m3) with a median of 30 ppm (160 mg/m3). Landrigan
et al. (1987) used air and biomonitoring techniques to quantify the exposure of degreasing
workers who worked around a heated, open bath of TRI. Exposures were found to be between
"3
22 and 66 ppm (117-357 mg/m ) on average, with short-term peaks between 76 and 370 ppm
(413-2,000 mg/m ). High peak exposures have also been reported for cardboard workers who
were involved with degreasing using a heated and open process (Henschler et al., 1995).
Lacking industrial hygiene data and making some assumptions about plant environment and TCE
usage, Cherrie et al. (2001) estimated that cardboard workers at a plant in Germany had peak
"3
exposures in the range of 200-4,000 ppm (1,100-22,000 mg/m ) and long-term average
exposures of 10-225 ppm (54-1,200 mg/m3). ATSDR (1997b) reports that the majority of
published worker exposure data show time-weighted average concentrations ranging from
<50 ppm-100 ppm (<270-540 mg/m ). NIOSH conducted a survey of various industries from
1981-1983 and estimated that approximately 401,000 U.S. employees in 23,225 plants in the
United States were potentially exposed to TCE during this timeframe (ATSDR, 1997b; IARC,
1995a). Occupational exposure to TCE has likely declined since the 1950s and 1960s due to
decreased usage, better release controls, and improvements in worker protection. Reductions in
TCE use are illustrated in Table 2-14, which shows that by about 1980, common degreasing
operations had substituted other solvents for TCE.
2.4.2.1.2. Consumer Exposure
Consumer products reported to contain TCE include wood stains, varnishes, and finishes;
lubricants; adhesives; typewriter correction fluids; paint removers; and cleaners (ATSDR,
1997b). Use of TCE has been discontinued in some consumer products (i.e., as an inhalation
anesthetic, fumigant, and an extractant for decaffeinating coffee) (ATSDR, 1997b).
2.4.3. Exposure Standards
Table 2-15 summarizes the federal regulations limiting TCE exposure.
This document is a draft for review purposes only and does not constitute Agency policy.
26 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
2.5. EXPOSURE SUMMARY
TCE is a volatile compound with moderate water solubility. Most TCE produced today
is used for metal degreasing. The highest environmental releases are to the air. Ambient air
monitoring data suggests that levels have remained fairly constant since 1999 at about 0.3 (J,g/m .
Indoor levels are commonly three or more times higher than outdoor levels due to releases from
building materials and consumer products. TCE is among the most common groundwater
contaminants and the median level based on a large survey by USGS for 1985-2001 is
0.15 (J,g/L. It has also been detected in a wide variety of foods in the 1-100 (J,g/kg range. None
of the environmental sampling has been done using statistically based national surveys.
However, a substantial amount of air and groundwater data have been collected allowing
reasonably well supported estimates of typical daily intakes by the general population:
inhalation—13 (j,g/day and water ingestion—0.2 (j,g/day. The limited food data suggests an
intake of about 5 (J,g/day, but this must be considered preliminary.
Much higher exposures have occurred to various occupational groups. For example, past
studies of aircraft workers have shown short term peak exposures in the hundreds of ppm
Table 2-15. TCE standards
Standard
Value
Reference
OSHA Permissible Exposure Limit: Table
Z-2 8-h time-weighted average.
100 ppm
(538 mg/m3)
29 CFR 1910.1000 (7/1/2000)
OSHA Permissible Exposure Limit: Table
Z-2 Acceptable ceiling concentration (this
cannot be exceeded for any time period
during an 8-h shift except as allowed in the
maximum peak standard below).
200 ppm
(1,076 mg/m3)
29 CFR 1910.1000 (7/1/2000)
OSHA Permissible Exposure Limit: Table
Z-2 Acceptable maximum peak above the
acceptable ceiling concentration for an 8-h
shift. Maximum Duration: 5 minutes in
any 2 h.
300 ppm
(1,614 mg/m3)
29 CFR 1910.1000 (7/1/2000)
MCL under the Safe Drinking Water Act.
5 ppb (5 ng/L)
40 CFR 141.161
FDA Tolerances for
decaffeinated ground coffee
decaffeinated soluble (instant) coffee
extract spice oleoresins.
25 ppm (25 p.g/g)
10 ppm (10 ng/g)
30 ppm (30 ng/g)
21 CFR 173.290 (4/1/2000)
OSHA = Occupational Safety and Health Administration.
This document is a draft for review purposes only and does not constitute Agency policy.
27 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
1
3 3
(>540,000 (J,g/m ) and long term exposures in the low tens of ppm (>54,000 (J,g/m ).
Occupational exposures have likely decreased in recent years due to better release controls and
improvements in worker protection.
Preliminary exposure estimates were presented for a variety of TCE related compounds
which include metabolites of TCE and other parent compounds that produce similar metabolites.
Exposure to the TCE related compounds can alter or enhance TCE's metabolism and toxicity by
generating higher internal metabolite concentrations than would result from TCE exposure by
itself. The preliminary estimates suggest that exposures to most of the TCE related compounds
are comparable to or greater than TCE itself.
This document is a draft for review purposes only and does not constitute Agency policy.
28 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
3. TOXICOKINETICS
Trichloroethylene (TCE) is a lipophilic compound that readily crosses biological
membranes. Exposures may occur via the oral, dermal, and inhalation route, with evidence for
systemic availability from each route. TCE is rapidly and nearly completely absorbed from the
gut following oral administration, and studies with animals indicate that exposure vehicle may
impact the time-course of absorption: oily vehicles may delay absorption whereas aqueous
vehicles result in a more rapid increase in blood concentrations.
Following absorption to the systemic circulation, TCE distributes from blood to solid
tissues by each organ's solubility. This process is mainly determined by the blood:tissue
partition coefficients (PCs), which are largely established by tissue lipid content. Adipose
partitioning is high, adipose tissue may serve as a reservoir for TCE, and accumulation into
adipose tissue may prolong internal exposures. TCE attains high concentrations relative to blood
in the brain, kidney, and liver—all of which are important target organs of toxicity. TCE is
cleared via metabolism mainly in three organs: the kidney, liver, and lungs.
The metabolism of TCE is an important determinant of its toxicity. Metabolites are
generally thought to be responsible for toxicity—especially for the liver and kidney. Initially,
TCE may be oxidized via cytochrome P450 (CYP) xenobiotic metabolizing isozymes or
conjugated with glutathione (GSH) by glutathione S-transferase enzymes. While CYP2E1 is
generally accepted to be the CYP form most responsible for TCE oxidation at low
concentrations, others forms may also contribute, though their contributions may be more
important at higher, rather than lower, environmentally-relevant exposures.
Once absorbed, TCE is excreted primarily either in breath as unchanged TCE or carbon
dioxide (CO2), or in urine as metabolites. Minor routes of elimination include excretion of
metabolites in saliva, sweat, and feces. Following oral administration or upon cessation of
inhalation exposure, exhalation of unmetabolized TCE is a major elimination pathway. Initially,
elimination of TCE upon cessation of inhalation exposure demonstrates a steep
concentration-time profile: TCE is rapidly eliminated in the minutes and hours postexposure, and
then the rate of elimination via exhalation decreases. Following oral or inhalation exposure,
urinary elimination of parent TCE is minimal, with urinary elimination of the metabolites
trichloroacetic acid (TCA) and trichloroethanol (TCOH) accounting for the bulk of the absorbed
dose of TCE.
This document is a draft for review purposes only and does not constitute Agency policy.
1 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Sections 3.1-3.4 below describe the absorption, distribution, metabolism, and excretion
(ADME) of TCE and its metabolites in greater detail. Section 3.5 then discusses physiologically
based pharmacokinetic(PBPK) modeling of TCE and its metabolites.
3.1. ABSORPTION
Trichloroethylene is a low-molecular-weight lipophilic solvent; these properties explain
its rapid transfer from environmental media into the systemic circulation after exposure. As
discussed below, it is readily absorbed into the bloodstream following exposure via oral
ingestion and inhalation, with more limited data indicating dermal penetration.
3.1.1. Oral
Available reports on human exposure to TCE via the oral route are largely restricted to
case reports of occupational or intentional (suicidal) ingestions and suggest significant gastric
absorption (e.g., Briining et al., 1998; Perbellini et al., 1991; Yoshida et al., 1996). Clinical
symptoms attributable to TCE or metabolites were observed in these individuals within a few
hours of ingestion (such as lack of consciousness), indicating absorption of TCE. In addition,
TCE and metabolites were measured in blood or urine at the earliest times possible after
ingestion, typically upon hospital admission, while urinary excretion of TCE metabolites was
followed for several days following exposure. Therefore, based on these reports, it is likely that
TCE is readily absorbed in the gastrointestinal tract; however, the degree of absorption cannot be
confidently quantified because the ingested amounts are not known.
Experimental evidence in mice and rats supports rapid and extensive absorption of TCE,
although variables such as stomach contents, vehicle, and dose may affect the degree of gastric
absorption. D'Souza et al. (1985) reported on bioavailability and blood kinetics in fasted and
nonfasted male Sprague-Dawley rats following intragastric administration of TCE at 5-25 mg/kg
in 50% polyethylene glycol (PEG 400) in water. TCE rapidly appeared in peripheral blood (at
the initial 0.5 minutes sampling) of fasted and nonfasted rats with peak levels being attained
shortly thereafter (6-10 minutes), suggesting that absorption is not diffusion limited, especially
in fasted animals. The presence of food in the gastro-intestinal (GI) tract, however, seems to
influence TCE absorption based on findings in the nonfasted animals of lesser bioavailability
(60-80% vs. 90% in fasted rats), smaller peak blood levels (two-threefold lower than nonfasted
animals), and a somewhat longer terminal half-life (ti/2) (174 vs. 112 minutes in fasted rats).
Studies by Prout et al. (1985) and Dekant et al. (1986b) have shown that up to 98% of
administered radiolabel was found in expired air and urine of rats and mice following gavage
This document is a draft for review purposes only and does not constitute Agency policy.
2 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
administration of [14C]-radio labeled TCE ([14C]TCE). Prout et al. (1985) and Green and Prout
(1985) compared the degree of absorption, metabolites, and routes of elimination among
two strains each of male rats (Osborne-Mendel and Park Wistar) and male mice (B6C3F1 and
Swiss-Webster) following a single oral administration of 10, 500, or 1,000 [14C]TCE. Additional
dose groups of Osborne-Mendel male rats and B6C3F1 male mice also received a single oral
dose of 2,000 mg/kg [14C]TCE. At the lowest dose of 10 mg/kg, there were no major differences
between rats and mice in routes of excretion with most of the administered radiolabel (nearly
60-70%) being in the urine. At this dose, the expired air from all groups contained 1-4% of
unchanged TCE and 9—14% CO2. Fecal elimination of the radiolabel ranged from 8.3% in
Osborne-Mendel rats to 24.1% in Park Wistar rats. However, at doses between 500 and
2,000 mg/kg, the rat progressively excreted a higher proportion of the radiolabel as unchanged
TCE in expired air such that 78% of the administered high dose was found in expired air (as
unchanged TCE) while only 13% was excreted in the urine.
Following exposure to a chemical by the oral route, distribution is determined by delivery
to the first organ encountered in the circulatory pathway—the liver (i.e., the first-pass effect),
where metabolism and elimination may limit the proportion that may reach extrahepatic organs.
Lee et al. (1996) evaluated the efficiency and dose-dependency of presystemic elimination of
TCE in male Sprague-Dawley rats following administration into the carotid artery, jugular vein,
hepatic portal vein, or the stomach of TCE (0.17, 0.33, 0.71, 2, 8, 16, or 64 mg/kg) in a
5% aqueous Alkamus emulsion (polyethoxylated vegetable oil) in 0.9% saline. The first-pass
elimination, decreased from 57.5 to <1% with increasing dose (0.17-16 mg/kg) which implied
that hepatic TCE metabolism may be saturated at doses above 16 mg/kg in the male rat. At
doses of 16 mg/kg or higher, hepatic first-pass elimination was almost nonexistent indicating
that, at relatively large doses, virtually all of TCE passes through the liver without being
extracted (Lee et al., 1996). In addition to the hepatic first-pass elimination findings, pulmonary
extraction, which was relatively constant (at nearly 5-8% of dose) over the dose range, also
played a role in eliminating TCE.
In addition, oral absorption appears to be affected by both dose and vehicle used. The
majority of oral TCE studies have used either aqueous solution or corn oil as the dosing vehicle.
Most studies that relied on an aqueous vehicle delivered TCE as an emulsified suspension in
Tween 80® or PEG 400 in order to circumvent the water solubility problems. Lee et al. (2000a;
2000b) used Alkamuls (a polyethoxylated vegetable oil emulsion) to prepare a 5% aqueous
emulsion of TCE that was administered by gavage to male Sprague-Dawley rats. The findings
confirmed rapid TCE absorption but reported decreasing absorption rate constants (i.e., slower
absorption) with increasing gavage dose (2-432 mg/kg). The time to reach blood peak
This document is a draft for review purposes only and does not constitute Agency policy.
3 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
concentrations increased with dose and ranged between 2 and 26 minutes postdosing. Other
pharmacokinetics data, including area under the blood concentration time curve (AUC) and
prolonged elevation of blood TCE levels at the high doses, indicated prolonged GI absorption
and delayed elimination due to metabolic saturation occurring at the higher TCE doses.
A study by Withey et al. (1983) evaluated the effect of dosing TCE with corn oil versus
pure water as a vehicle by administering four volatile organic compounds separately in each
dosing vehicle to male Wistar rats. Based on its limited solubility in pure water, the dose for
TCE was selected at 18 mg/kg (administered in 5 mL/kg). Times to peak in blood reported for
TCE averaged 5.6 minutes when water was used. In comparison, the time to peak in blood was
much longer (approximately 100 minutes) when the oil vehicle was used and the peaks were
smaller, below the level of detection, and not reportable.
Time-course studies reporting times to peak in blood or other tissues have been
performed using both vehicles (D'Souza et al., 1985; Dekant et al., 1984; Green and Prout, 1985;
Larson and Bull, 1992a; 1992b; Withey et al., 1983). Related data for other solvents (Chieco et
al., 1981; Dix et al., 1997; Kim et al., 1990b; Lilly et al., 1994) confirmed differences in TCE
absorption and peak height between the two administered vehicles. One study has also evaluated
the absorption of TCE from soil in rats (Kadry et al., 1991) and reported absorption within
16 hours for clay and 24 hours for sandy soil. In summary, these studies confirm that TCE is
relatively quickly absorbed from the stomach, and that absorption is dependent on vehicle used.
3.1.2. Inhalation
Trichloroethylene is a lipophilic volatile compound that is readily absorbed from inspired
air. Uptake from inhalation is rapid and the absorbed dose is proportional to exposure
concentration and duration, and pulmonary ventilation rate. Distribution into the body via
arterial blood leaving the lungs is determined by the net dose absorbed and eliminated by
metabolism in the lungs. Metabolic clearance in the lungs will be further discussed in
Section 3.3, below. In addition to metabolism, solubility in blood is the major determinant of the
TCE concentration in blood entering the heart and being distributed to the each body organ via
the arterial blood. The measure of TCE solubility in each organ is the partition coefficient, or the
concentration ratio between both organ phases of interest. The blood-to-air partition coefficient
quantifies the resulting concentration in blood leaving the lungs at equilibrium with alveolar air.
The value of the blood-to-air partition coefficient is used in PBPK modeling (see Section 3.5).
The blood-to-air partition has been measured in vitro using the same principles in different
This document is a draft for review purposes only and does not constitute Agency policy.
4 DRAFT—DO NOT CITE OR QUOTE
-------
1 studies and found to range between 8.1-11.7 in humans and somewhat higher values in mice and
2 rats (13.3-25.8) (see Tables 3-1-3-2, and references therein).
3 TCE enters the human body quickly by inhalation, and, at high concentrations, it may
4 lead to death (Coopman et al., 2003), narcosis, unconsciousness, and
5
This document is a draft for review purposes only and does not constitute Agency policy.
5 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-1. Bloodrair PC values for humans
2
Bloodrair partition
coefficient
Reference/notes
8.1 ± 1.8
Fiserova-Bergerova et al. (1984); mean ± SD (SD converted from
SE based on n = 5)
8.11
Gargas et al. (1989); (n = 3-15)
9.13 ± 1.73 [6.47-11]
Fisher et al. (1998); mean ± SD [range] of females (n = 6)
9.5
Sato and Nakajima (1979); (n = 1)
9.77
Koizumi (1989)
9.92
Sato et al. (1977); (n = 1)
11.15 + 0.74
[10.1-12.1]
Fisher et al. (1998); mean ± SD [range] of males (n = 7)
11.2 ± 1.8 [7.9-15]
Mahle et al. (2007); mean ± SD; 20 male pediatric patients aged 3-7
years (range; USAF, 2004)
11.0± 1.6 [6.6-13.5]
Mahle et al. (2007); mean ± SD; 18 female pediatric patients aged
3-17years (range; USAF, 2004)
11.7 ± 1.9 [6.7-16.8]
Mahle et al. (2007); mean ± SD; 32 male patients aged 23-82 years
(range; USAF, 2004)
10.6 ±2.3 [3-14.4]
Mahle et al., (2007); mean ± SD; 27 female patients aged 23-82
years (range; USAF, 2004)
3
4 SD = standard deviation, SE = standard error.
5
This document is a draft for review purposes only and does not constitute Agency policy.
6 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-2. Blood:air PC values for rats and mice
2
Bloodrair partition
coefficient
Reference/notes
Rat
15 + 0.5
Fisher et al. (1998); mean + SD (SD converted from SE based on
n = 3)
17.5
Rodriguez et al. (2007)
20.5 + 2.4
Barton et al. (1995); mean + SD (SD converted from SE based on
n = 4)
20.69 + 3.3
Simmons et al. (2002); mean + SD (n = 7-10)
21.9
Gargas et al. (1989) (n = 3-15)
25.8
Koizumi (1989) (pooled n = 3)
25.82+ 1.7
Sato et al. (1977); mean + SD (n = 5)
13.3 + 0.8 [11.6-15]
Mahle et al. (2007); mean + SD; 10 PND 10 male rat pups (range;
USAF, 2004)
13.4+1.8 [11.8-17.2]
Mahle et al. (2007); mean + SD; 10 PND 10 female rat pups (range;
USAF, 2004)
17.5 + 3.6 [11.7-23.1]
Mahle et al. (2007); mean + SD; 9 adult male rats (range; USAF,
2004)
21.8+1.9 [16.9-23.5]
Mahle et al. (2007); mean + SD; 11 aged male rats (range; USAF,
2004)
Mouse
13.4
Fisher et al. (1991); male
14.3
Fisher et al. (1991); female
15.91
Abbas and Fisher (1997)
3
4 SD = standard deviation, SE = standard error, PND = postnatal day.
5
6
7 acute kidney damage (Carried et al., 2007). Controlled exposure studies in humans have shown
8 absorption of TCE to approach a steady state within a few hours after the start of inhalation
9 exposure (Fernandez et al., 1977; Monster et al., 1976; Vesterberg and Astrand, 1976;
10 Vesterberg et al., 1976). Several studies have calculated the net dose absorbed by measuring the
11 difference between the inhaled concentration and the exhaled air concentration. Soucek and
12 Vlachova (1960) reported between 58-70% absorption of the amount inhaled for 5-hour
This document is a draft for review purposes only and does not constitute Agency policy.
1 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
exposures between 93-158 ppm. Bartonicek (1962) obtained an average retention value of 58%
after 5 hours of exposure to 186 ppm. Monster et al. (1976) also took into account minute
ventilation measured for each exposure, and calculated between 37-49% absorption in subjects
exposed to 70 and 140 ppm. The impact of exercise, the increase in workload, and its effect on
breathing has also been measured in controlled inhalation exposures. Astrand and Ovrum (1976)
reported 50—58% uptake at rest and 25-46% uptake during exercise from exposure at 100 or
200 ppm (540 or 1,080 mg/m3, respectively) of TCE for 30 minutes (see Table 3-3). These
authors also monitored heart rate and pulmonary ventilation. In contrast, Jakubowski and
Wieczorek (1988) calculated about 40% retention in their human volunteers exposed to TCE at 9
"3
ppm (mean inspired concentration of 48-49 mg/m ) for 2 hours at rest, with no change in
retention during increase in workload due to exercise (see Table 3-4).
Environmental or occupational settings may results from a pattern of repeated exposure
to TCE. Monster et al. (1979) reported 70-ppm TCE exposures in volunteers for 4 hours for
5 consecutive days, averaging a total uptake of 450 mg per 4 hours exposure (see Table 3-5). In
dry-cleaning workers, Skender et al. (1991) reported initial blood concentrations of 0.38 |imol/L,
increasing to 3.4 |imol/L 2 days after. Results of these studies support rapid absorption of TCE
via inhalation.
Direct measurement of retention after inhalation exposure in rodents is more difficult
because exhaled breath concentrations are challenging to obtain. The only available data are
from Dallas et al. (1991), who designed a nose-only exposure system for rats using a facemask
equipped with one-way breathing valves to obtain measurements of TCE in inspired and exhaled
air. In addition, indwelling carotid artery cannulae were surgically implanted to facilitate the
simultaneous collection of blood. After a 1-hour acclimatization period, rats were exposed to
50- or 500-ppm TCE for 2 hours and the time course of TCE in blood and expired air was
measured during and for 3 hours following exposure. When air concentration data were
analyzed to reveal absorbed dose (minute volume multiplied by the concentration difference
between inspired and exhaled breath), it was demonstrated that the fractional absorption of either
concentration was more than 90% during the initial 5 minutes of exposure. Fractional absorption
then decreased to 69 and 71% for the 50 and 500-ppm groups during the second hour of
exposure. Cumulative uptake appeared linear with respect to time over the 2-hour exposure,
resulting in absorbed doses of 8.4 mg/kg and 73.3 mg/kg in rats exposed to 50 and 500 ppm,
respectively. Given the 10-fold difference in inspired concentration and the 8.7-fold difference
in uptake, the authors interpreted this information to indicate that metabolic saturation occurred
at some concentration below 500 ppm. In comparing the absorbed doses to those developed for
This document is a draft for review purposes only and does not constitute Agency policy.
8 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-3. Air and blood concentrations during exposure to TCE in humans
2 (Astrand and Ovrum, 1976)
3
TCE
conc.
(mg/m3)
Work
load
(watt)
Exposure
series
TCE concentration in
Uptake as
% of
amount
available
Amount
taken up
(mg)
Alveolar
air
(mg/m3)
Arterial
blood
(mg/kg)
Venous
blood
(mg/kg)
540
0
I
124 ±9
1.1 ±0.1
0.6 ±0.1
53 ±2
79 ±4
540
0
II
127 ± 11
1.3 ±0.1
0.5 ±0.1
52 ±2
81 ±7
540
50
I
245 ± 12
2.7 ±0.2
1.7 ± 0.4
40 ±2
160 ±5
540
50
II
218 ± 7
2.8 ±0.1
1.8 ± 0.3
46 ± 1
179 ±2
540
50
II
234 ± 12
3.1 ±0.3
2.2 ±0.4
39 ± 2
157 ± 2
540
50
II
244 ± 16
3.3 ±0.3
2.2 ±0.4
37 ± 2
147 ±9
1,080
0
I
280 ± 18
2.6 ±0.0
1.4 ± 0.3
50 ±2
156 ±9
1,080
0
III
212 ± 7
2.1 ±0.2
1.2 ± 0.1
58 ± 2
186 ±7
1,080
50
I
459 ±44
6.0 ±0.2
3.3 ±0.8
45 ±2
702 ±31
1,080
50
III
407 ±30
5.2 ±0.5
2.9 ±0.7
51 ±3
378 ± 18
1,080
100
III
542 ±33
7.5 ±0.7
4.8 ±1.1
36 ±3
418 ± 39
1,080
150
III
651 ±53
9.0± 1.0
7.4 ±1.1
25 ±5
419 ±84
4
5 Series I consisted of 30-minute exposure periods of rest, rest, 50W and 50W; Series II consisted of
6 30-minute exposure periods of rest, 50W, 50W, 50W; Series III consisted of 30-minute exposure
7 periods of rest, 50W, 100W, 150W.
8
This document is a draft for review purposes only and does not constitute Agency policy.
9 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-4. Retention of inhaled TCE vapor in humans (Jakubowski and
2 Wieczorek, 1988)
3
Workload
Inspired concentration
(mg/m3)
Pulmonary
ventilation (m3/hour)
Retention
Uptake
(mg/hour)
Rest
48 ± 3a
0.65 ±0.07
0.40 ±0.05
12 ± 1.1
25 W
49 ± 1.3
1.30 ± 0.14
0.40 ±0.05
25 ±2.9
50 W
49 ± 1.6
1.53 ±0.13
0.42 ±0.06
31 ±2.8
75 W
48 ± 1.9
1.87 ± 0.14
0.41 ±0.06
37 ± 4.8
4
5 a Mean ± standard deviation, n = 6 adult males.
6
7 W = watts.
8
9
10 Table 3-5. Uptake of TCE in human volunteers following 4 hour exposure to
11 70 ppm (Monster et al., 1979)
12
BW
(kg)
MV (L/min)
% Retained
Uptake
(mg/day)
Uptake (mg/kg-day)
A
80
9.8 ±0.4
45 ±0.8
404 ± 23
5.1
B
82
12.0 ±0.7
44 ±0.9
485 ±35
5.9
C
82
10.9 ±0.8
49 ± 1.2
493 ± 28
6.0
D
67
11.8 ± 0.8
35 ± 2.6
385 ± 38
5.7
E
90
11.0 ± 0.7
46 ± 1.1
481 ±25
5.3
Mean
5.6 ±0.4
13
14 MV = minute-volume.
15 B W = body weight.
16
17
18 the 70-ppm-exposed human (see Monster et al., 1979), Dallas et al. (1991) concluded that on a
19 systemic dose (mg/kg) basis, rats receive a much higher TCE dose from a given
20 inhalationexposure than do humans. In particular, using the results cited above, the absorption
21 per ppm-hour was 0.084 and 0.073 mg/kg-ppm-hour at 50 and 500 ppm in rats (Dallas et al.,
22 1991) and 0.019 mg/kg-ppm-hour at 70 ppm in humans (Monster et al., 1979)—a difference of
23 around fourfold. However, rats have about a 10-fold higher alveolar ventilation rate per unit
24 body weight (BW) than humans (Brown et al., 1997), which more than accounts for the observed
25 increase in absorption.
This document is a draft for review purposes only and does not constitute Agency policy.
10 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
Other experiments, such as closed-chamber gas uptake experiments or blood
concentration measurements following open-chamber (fixed concentration) experiments,
measure absorption indirectly but are consistent with significant retention. Closed-chamber
gas-uptake methods (Gargas et al., 1988) place laboratory animals or in vitro preparations into
sealed systems in which a known amount of TCE is injected to produce a predetermined
chamber concentration. As the animal retains a quantity of TCE inside its body, due to
metabolism, the closed-chamber concentration decreases with time when compared to the start of
exposure. Many different studies have made use of this technique in both rats and mice to
calculate total TCE metabolism (i.e., Andersen et al., 1987a; Fisher et al., 1991; Simmons et al.,
2002). This inhalation technique is combined with PBPK modeling to calculate metabolic
parameters, and the results of these studies are consistent with rapid absorption of TCE via the
respiratory tract. Figure 3-1 shows and example from Simmons et al. (2002), in Long Evans rats,
that demonstrates an immediate decline in chamber concentrations of TCE indicating absorption,
with multiple initial concentrations needed for each metabolic calculation. At concentrations
below metabolic saturation, a secondary phase of uptake appears, after 1 hour from starting the
exposure, indicative of metabolism. At concentrations above 1,000 ppm, metabolism appears
saturated, with time course curves having a flat phase after absorption. At intermediate
concentrations, between 100-1,000 ppm, the secondary phase of uptake appears after
distribution as continued decreases in chamber concentration as metabolism proceeds. Using a
combination of experiments that include both metabolic linear decline and saturation obtained by
using different initial concentrations, both components of metabolism can be estimated from the
gas uptake curves, as shown in Figure 3-1.
Several other studies in humans and rodents have measured blood concentrations of TCE
or metabolites and urinary excretion of metabolites during and after inhalation exposure (e.g.,
Filser and Bolt, 1979; 1991; Fisher et al., 1998; 1990). While qualitatively indicative of
absorption, blood concentrations are also determined by metabolism, distribution, and excretion,
so comparisons between species may reflect similarities or differences in any of the absorption,
distribution, metabolism, and excretion processes.
3.1.3. Dermal
Skin membrane is believed to present a diffusional barrier for entrance of the chemical into the
body, and TCE absorption can be quantified using a permeability rate or permeability constant,
though not all studies performed such a calculation. Absorption through the skin has been shown
This document is a draft for review purposes only and does not constitute Agency policy.
11 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
to be rapid by both vapor and liquid TCE contact with the skin. Human dermal absorption of
TCE vapors was investigated by Kezic et al. (2000). Human volunteers were exposed to
10000 q
?
Q.
a.
c
o
"iS 1000 i
i-
c
a)
o
c
o
o
a> 100 i
.Q
E
(0
-C
o
HI
o
10
¦ + +
+ ++++
++ +++
++++
+ 3000 ppm
~ 1000 ppm
a 500 ppm
o 100 ppm
+++++
+++++++++
++
AaaAa aDDn°ann
A'iA, D°an°°ona
'AA
Aa
aaa
D°nnn
Aa
aaa
aAa
>0<>Oo
AA/
Oo
Oo,
Oo
~ ~
'-o
Ooo
OOOo
1 1 1 1 1 r~
0 1 2 3 4 5 6
Time (hr)
Figure 3-1. Gas uptake data from closed-chamber exposure of rats to TCE.
Symbols represent measured chamber concentrations. Source: Simmons et al.
(2002).
3.18 x 10 ppm around each enclosed arm for 20 minutes. Adsorption was found to be rapid
(within 5 minutes), reaching a peak in exhaled breath around 30 minutes, with a calculated
dermal penetration rate averaging 0.049 cm/hour for TCE vapors.
With respect to dermal penetration of liquid TCE, Nakai et al. (1999) used surgically
removed skin samples exposed to TCE in aqueous solution in a chamber designed to measure the
difference between incoming and outgoing [14C]TCE. The in vitro permeability constant
calculated by these researchers averaged 0.12 cm/hour. In vivo, Sato and Nakajima (1978)
exposed adult male volunteers dermally to liquid TCE for 30 minutes, with exhaled TCE
appearing at the initial sampling time of 5 minutes after start of exposure, with a maximum
observed at 15 minutes. In Kezic et al. (2001), human volunteers were exposed dermally for
3 minutes to neat liquid TCE, with TCE detected in exhaled breath at the first sampling point of
This document is a draft for review purposes only and does not constitute Agency policy.
12 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
3 minutes, and maximal concentrations observed at 5 minutes. Skin irritancy was reported in all
subjects, which may have increased absorption. A dermal flux of 430 ± 295 (mean ± standard
error [SE]) nmol/cm /minute was reported in these subjects, suggesting high interindividual
variability.
Another species where dermal absorption for TCE has been reported is in guinea pigs.
Jakobson et al. (1982) applied liquid TCE to the shaved backs of guinea pigs and reported peak
blood TCE levels at 20 minutes after initiation of exposure. Bogen et al. (1992) estimated
permeability constants for dermal absorption of TCE in hairless guinea pigs between
0.16-0.47 mL/cm /hour across a range of concentrations (19-100,000 ppm).
3.2. DISTRIBUTION AND BODY BURDEN
TCE crosses biological membranes and quickly results in rapid systemic distribution to
tissues—regardless of the route of exposure. In humans, in vivo studies of tissue distribution are
limited to tissues taken from autopsies following accidental poisonings or from surgical patients
exposed environmentally, so the level of exposure is typically unknown. Tissue levels reported
after autopsy show wide systemic distribution across all tested tissues, including the brain,
muscle, heart, kidney, lung, and liver (Coopman et al., 2003; De Baere et al., 1997; Dehon et al.,
2000; Ford et al., 1995). However, the reported levels themselves are difficult to interpret
because of the high exposures and differences in sampling protocols. In addition, human
populations exposed environmentally show detectable levels of TCE across different tissues,
including the liver, brain, kidney, and adipose tissues (Kroneld, 1989; McConnell et al., 1975;
Pellizzari et al., 1982).
In addition, TCE vapors have been shown to cross the human placenta during childbirth
(Laham, 1970), with experiments in rats confirming this finding (Withey and Karpinski, 1985).
In particular, Laham (1970) reported determinations of TCE concentrations in maternal and fetal
blood following administration of TCE vapors (concentration unreported) intermittently and at
birth (see Table 3-6). TCE was present in all samples of fetal blood, with ratios of
concentrations in fetal:maternal blood ranging from approximately 0.5 to approximately 2. The
concentration ratio was less than 1.0 in six pairs, greater than one in three pairs, and
approximately one in one pair; in general, higher ratios were observed at maternal concentrations
below 2.25 mg/100 mL. Because no details of exposure concentration, duration, or time
postexposure were given for samples taken, these results are not suitable for use in PBPK
modeling, but they do demonstrate the placental transfer of TCE in humans. Withey and
Karpinski (1985) exposed pregnant rats to TCE vapors (302, 1,040, 1,559, or 2,088 ppm for
This document is a draft for review purposes only and does not constitute Agency policy.
13 DRAFT—DO NOT CITE OR QUOTE
-------
1 5 hours) on gestation Day 17 and concentrations of TCE in maternal and fetal blood were
2 determined. At all concentrations, TCE concentration in fetal blood was approximately one-third
3 the concentration in corresponding maternal blood. Maternal blood concentrations approximated
4
This document is a draft for review purposes only and does not constitute Agency policy.
14 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Table 3-6. Concentrations of TCE in maternal and fetal blood at birth
TCE concentration in
blood (mg/100 mL)
Ratio of
concentrations
fetalrmaternal
Maternal
Fetal
4.6
2.4
0.52
3.8
2.2
0.58
8
5
0.63
5.4
3.6
0.67
7.6
5.2
0.68
3.8
3.3
0.87
2
1.9
0.95
2.25
3
1.33
0.67
1
1.49
1.05
2
1.90
Source: Laham (1970).
15, 60, 80, and 110 |ig/gram blood. When the position along the uterine horn was examined,
TCE concentrations in fetal blood decreased toward the tip of the uterine horn. TCE appears to
also distribute to mammary tissues and is excreted in milk. Pellizzari et al. (1982) conducted a
survey of environmental contaminants in human milk using samples from cities in the
northeastern region of the United States and one in the southern region. No details of times
postpartum, milk lipid content, or TCE concentration in milk or blood are reported, but TCE was
detected in 8 milk samples taken from 42 lactating women. Fisher et al. (1990) exposed lactating
rats to 600-ppm TCE for 4 hours and collected milk immediately following the cessation of
exposure. TCE was clearly detectable in milk, and, from a visual interpretation of the graphic
display of their results, concentrations of TCE in milk approximated 110 |ag/m L milk.
In rodents, detailed tissue distribution experiments have been performed using different
routes of administration (Abbas and Fisher, 1997; Greenberg et al., 1999; Keys et al., 2003;
Pfaffenberger et al., 1980; Savolainen et al., 1977; Simmons et al., 2002). Savolainen et al.
(1977) exposed adult male rats to 200-ppm TCE for 6 hours/day for a total of 5 days.
Concentrations of TCE in the blood, brain, liver, lung, and perirenal fat were measured 17 hours
after cessation of exposure on the fourth day and after 2, 3, 4, and 6 hours of exposure on the
This document is a draft for review purposes only and does not constitute Agency policy.
15 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
fifth day (see Table 3-7). TCE appeared to be rapidly absorbed into blood and distributed to
brain, liver, lungs, and perirenal fat. TCE concentrations in these tissues reached near-maximal
values within 2 hours of initiation of exposure on the fifth day. Pfaffenberger et al. (1980) dosed
rats by gavage with 1 or 10 mg TCE/kg/day in corn oil for 25 days to evaluate the distribution
from serum to adipose tissue. During the exposure period, concentrations of TCE in serum were
below the limit of detection (1 |ig/L) and were 280 and 20,000 ng per gram of fat in the 1 and
10 mg/day dose groups, respectively. Abbas and Fisher (1997) and Greenberg et al. (1999)
measured tissue concentrations in the liver, lung, kidney, and fat of mice administered TCE by
gavage (300-2,000 mg/kg) and by inhalation exposure (100 or 600 ppm for 4 hours). In a study
to investigate the effects of TCE on neurological function, Simmons et al. (2002) conducted
pharmacokinetic experiments in rats exposed to 200, 2,000, or 4,000 ppm TCE vapors for 1 hour.
Time-course data were collected on blood, liver, brain, and fat. The data were used to develop a
PBPK model to explore the relationship between internal dose and neurological effect. Keys
et al. (2003), exposed groups of rats to TCE vapors of 50 or 500 ppm for 2 hours and sacrificed
at different time points during exposure. In addition to inhalation, this study also includes oral
gavage and intra-arterial dosing, with the following time course measured: liver, fat, muscle,
blood, GI, brain, kidney, heart, lung, and spleen. These pharmacokinetic data were presented
with an updated PBPK model for all routes.
Besides the route of administration, another important factor contributing to body
distribution is the individual solubility of the chemical in each organ, as measured by a partition
coefficient. For volatile compounds, partition coefficients are measured in vitro using the vial
equilibration technique to determine the ratio of concentrations between organ and air at
equilibrium. Table 3-8 reports values developed by several investigators from mouse, rat, and
human tissues. In humans, partition coefficients in the following tissues have been measured:
brain, fat, kidney, liver, lung, and muscle; but the organ having the highest TCE partition
coefficient is fat (63-70), while the lowest is the lung (0.5-1.7). The adipose tissue also has the
highest measured value in rodents, and is one of the considerations needed to be accounted for
when extrapolating across species. However, the rat adipose partition coefficient value is
smaller (23-36), when compared to humans, that is, TCE is less lipophilic in rats than humans.
For the mouse, the measured fat partition coefficient averages 36, ranging between rats and
humans. The value of the partition coefficient plays a role in distribution for each organ and is
computationally described in computer simulations using a PBPK model. Due to its high
lipophilicity in fat, as compared to blood, the adipose tissue behaves as a storage compartment
This document is a draft for review purposes only and does not constitute Agency policy.
16 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-7. Distribution of TCE to rat tissues3 following inhalation exposure
2 (Savolainen et al., 1977)
3
Exposure
on 5th day
Tissue (concentration in nmol/gram tissue)
Cerebrum
Cerebellum
Lung
Liver
Perirenal fat
Blood
0b
0
0
0.08
0.04
0.23 ± 0.09
0.35 ±0.1
2
9.9 ±2.7
11.7 ± 4.2
4.9 ±0.3
3.6
65.9± 1.2
7.5 ± 1.6
3
7.3 ±2.2
8.8 ±2.1
5.5 ± 1.4
5.5 ± 1.7
69.3 ±3.3
6.6 ±0.9
4
7.2 ± 1.7
7.6 ±0.5
5.8 ± 1.1
2.5 ± 1.4
69.5 ±6.3
6.0 ±0.2
6
7.4 ±2.1
9.5 ±2.5
5.6 ± 0.5
2.4 ±0.2
75.4 ± 14.9
6.8 ± 1.2
4
5 "Data presented as mean of two determinations ± range.
6 bSample taken 17 hours following cessation of exposure on Day 4.
7
8
9
This document is a draft for review purposes only and does not constitute Agency policy.
17 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-8. Tissue:blood partition coefficient values for TCE
2
Species
/
tissue
TCE partition
coefficient
References
Tissuerbloo
d
Tissuerair
Human
Brain
2.62
21.2
Fiserova-Bergerova et al. (1984)
Fat
63.8-70.2
583-674.4
Sato et al. (1977); Fiserova-Bergerova et al. (1984);
Fisher et al. (1998)
Kidney
1.3-1.8
12-14.7
Fiserova-Bergerova et al. (1984); Fisher et al. (1998)
Liver
3.6-5.9
29.4-54
Fiserova-Bergerova et al. (1984); Fisher et al. (1998)
Lung
0.48-1.7
4.4-13.6
Fiserova-Bergerova et al. (1984); Fisher et al. (1998)
Muscle
1.7-2.4
15.3-19.2
Fiserova-Bergerova et al. (1984); Fisher et al. (1998)
Rat
Brain
0.71-1.29
14.6-33.3
Sato et al. (1977); Simmons et al. (2002); Rodriguez
et al. (2007)
Fat
22.7-36.1
447-661
Gargas et al. (1989); Sato et al. (1977); Simmons et al.
(2002); Rodriguez et al. (2007); Fisher et al. (1989);
Koizumi (1989); Barton et al. (1995)
Heart
1.1
28.4
Sato et al. (1977)
Kidney
1.0-1.55
17.7-40
Sato et al., (1977); Barton et al., (1995); Rodriguez et al.,
(2007)
Liver
1.03-2.43
20.5-62.7
Gargas et al. (1989); Sato et al. (1977); Simmons et al.
(2002); Rodriguez et al. (2007); Fisher et al. (1989);
Koizumi, (1989); Barton et al. (1995)
Lung
1.03
26.6
Sato et al. (1977)
Muscle
0.46-0.84
6.9-21.6
Gargas et al. (1989); Sato et al. (1977); Simmons et al.
(2002); Rodriguez et al. (2007); Fisher et al. (1989);
Koizumi, (1989); Barton et al. (1995)
Spleen
1.15
29.7
Sato et al. (1977)
Testis
0.71
18.3
Sato et al. (1977)
Milk
7.10
NR.
Fisher et al. (1990)
Mouse
Fat
36.4
578.8
Abbas and Fisher (1997)
This document is a draft for review purposes only and does not constitute Agency policy.
18 DRAFT—DO NOT CITE OR QUOTE
-------
Kidney
2.1
32.9
Abbas and Fisher (1997)
Liver
1.62
23.2
Fisher et al. (1991)
Lung
2.6
41.5
Abbas and Fisher (1997)
Muscle
2.36
37.5
Abbas and Fisher (1997)
1 N.R. = not reported.
This document is a draft for review purposes only and does not constitute Agency policy.
19 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
for this chemical, affecting the slower component of the chemical's distribution. For example
Monster et al. (1979) reported that, following repeated inhalation exposures to TCE, TCE
concentrations in expired breath postexposure were highest for the subject with the greatest
amount of adipose tissue (adipose tissue mass ranged 3.5-fold among subjects). The intersubject
range in TCE concentration in exhaled breath increased from approximately twofold at 20 hours
to approximately 10-fold 140 hours postexposure. Notably, they reported that this difference
was not due to differences in uptake, as body weight and lean body mass were most closely
associated with TCE retention. Thus, adipose tissue may play an important role in postexposure
distribution, but does not affect its rapid absorption.
Mahle et al. (2007) reported age-dependent differences in partition coefficients in rats,
(see Table 3-9) that can have implications as to life-stage-dependent differences in tissue TCE
distribution. To investigate the potential impact of these differences, Rodriguez et al. (2007)
developed models for the postnatal Day 10 rat pup; the adult and the aged rat, including
age-specific tissue volumes and blood flows; and age-scaled metabolic constants. The models
predict similar uptake profiles for the adult and the aged rat during a 6-hour exposure to
500 ppm; uptake by the postnatal day (PND) 10 rat was higher (see Table 3-10). The effect was
heavily dependent on age-dependent changes in anatomical and physiological parameters
(alveolar ventilation rates and metabolic rates); age-dependent differences in partition coefficient
values had minimal impact on predicted differences in uptake.
Finally, TCE binding to tissues or cellular components within tissues can affect overall
pharmacokinetics. The binding of a chemical to plasma proteins, for example, affects the
availability of the chemical to other organs and the calculation of the total half-life. However,
most studies have evaluated binding using [14C]TCE, from which one cannot distinguish
covalent binding of TCE from that of TCE metabolites. Nonetheless, several studies have
demonstrated binding of TCE-derived radiolabel to cellular components (Mazzullo et al., 1992;
Moslen et al., 1977). Bolt and Filser (1977) examined the total amount irreversibly bound to
tissues following 9-, 100-, and 1,000-ppm exposures via inhalation in closed-chambers. The
largest percent of in vivo radioactivity taken up occurred in the liver; albumin is the protein
favored for binding (see Table 3-11). Banerjee and van Duuren (1978) evaluated the in vitro
binding of TCE to microsomal proteins from the liver, lung, kidney, and stomachs in rats and
mice. In both rats and mice, radioactivity was similar in stomach and lung, but about 30% lower
in kidney and liver.
Based on studies of the effects of metabolizing enzyme induction on binding, there is
some evidence that a major contributor to the observed binding is from TCE metabolites rather
than from TCE itself. Dekant et al. (1986b) studied the effect of enzyme modulation on the
This document is a draft for review purposes only and does not constitute Agency policy.
20 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-9. Age-dependence of tissuerair partition coefficients in rats
2
Age
Liver
Kidney
Fat
Muscle
Brain
PND 10 male
22.1 ±2.3
15.2 ±1.3
398.7 ±89.2
43 .9 ± 11.0
11.0 ± 0.6
PND 10 female
21.2± 1.7
15.0 ±1.1
424.5 ±67.5
48.6 ± 17.3
11.6± 1.2
Adult male
20.5 ±4.0
17.6 ± 3.9a
631.4 ± 43.la
12.6 ±4.3
17.4 ± 2.6
Aged male
34.8 ± 8.7a'b
19.9 ± 3.4a
757.5 ±48.3a'b
26.4 ± 10.3a'b
25.0 ± 2.0a'b
3
4 a Statistically significant (p < 0.05) difference between either the adult or aged partition coefficient and the PND10
5 male partition coefficient.
6 b Statistically significant (p < 0.05) difference between aged and adult partition coefficient.
7
8 Data are mean ± standard deviation; n = 10, adult male and pooled male and female litters; 11, aged males.
9 Source: Mahle et al. (2007).
10
11
12 Table 3-10. Predicted maximal concentrations of TCE in rat blood
13 following a 6-hour inhalation exposure (Rodriguez et al., 2007)
14
Age
Exposure concentration
50 ppm
500 ppm
Predicted peak
concentration
(mg/L) in:a
Predicted
time to reach
90% of steady
state (hour)b
Predicted peak
concentration
(mg/L) in:a
Predicted
time to reach
90% of steady
state (hour)b
Venous
blood
Brain
Venous
blood
Brain
PND 10
3.0
2.6
4.1
33
28
4.2
Adult
0.8
1.0
3.5
22
23
11.9
Aged
0.8
1.2
6.7
21
26
23.3
15
16 a During a 6 hour exposure.
17 b Under continuous exposure.
18
This document is a draft for review purposes only and does not constitute Agency policy.
21 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-11. Tissue distribution of TCE metabolites following inhalation
2 exposure
3
Tissue"
Percent of radioactivity taken up/g tissue
TCE = 9 ppm,
n = 4
TCE = 100 ppm,
n = 4
TCE = 1,000 ppm,
n = 3
Total
metabolites
Irreversibly
bound
Total
metabolites
Irreversibly
bound
Total
metabolites
Irreversibly
bound
Lung
0.23 ±
0.026
0.06 ±0.002
0.24 ±
0.025
0.06 ±0.006
0.22 ±
0.055
0.1 ±0.003
Liver
0.77 ±
0.059
0.28 ±0.027
0.68 ±
0.073
0.27 ±0.019
0.88 ±
0.046
0.48 ±0.020
Spleen
0.14 ±
0.015
0.05 ±0.002
0.15 ±
0.001
0.05 ±0.004
0.15 ±
0.006
0.08 ±0.003
Kidney
0.37 ±
0.005
0.09 ±0.007
0.40 ±
0.029
0.09 ±0.007
0.39 ±
0.045
0.14 ± 0.016
Small
intestine
0.41 ±
0.058
0.05 ±0.010
0.38 ±
0.062
0.07 ±0.008
0.28 ±
0.015
0.09 ±0.015
Muscle
0.11 ±
0.005
0.014 ±
0.001
0.11 ±
0.013
0.012 ±
0.001
0.10±
0.011
0.027 ±
0.003
4
5 a Male Wistar rats, 250 g.
6
7 n = number of animals.
8 Values shown are means ± standard deviation.
9 Source: Bolt and Filser (1977).
10
11
12 binding of radiolabel from [14C]TCE by comparing tissue binding after administration of
13 200 mg/kg via oral gavage in corn oil between control (naive) rats and rats pretreated with
14 phenobarbital (a known inducer of CYP2B family) or Aroclor 1254 (a known inducer of both
15 CYP1A and CYP2B families of isoenzymes) (see Table 3-12). The results indicate that
16 induction of total cytochromes P-450 content by three- to fourfold resulted in nearly 10-fold
17 increase in radioactivity (disintegrations per minute; [DPM]) bound in liver and kidney. By
18 contrast, Mazzullo et al. (1992) reported that, phenobarbital pretreatment did not result in
19 consistent or marked alterations of in vivo binding of radiolabel to DNA, RNA, or protein in rats
20 and mice at 22 hours after an intraperitoneal (i.p.) injection of [14C]TCE. On the other hand, in
21 vitro experiments by Mazzullo et al. (1992) reported reduction of TCE-radiolabel binding to calf
22 thymus DNA with introduction of a CYP inhibitor into incubations containing rat liver
This document is a draft for review purposes only and does not constitute Agency policy.
22 DRAFT—DO NOT CITE OR QUOTE
-------
1 microsomal protein. Moreover, increase/decrease of glutathione levels in incubations containing
2 lung cytosolic protein led to a parallel increase/decrease in TCE-radiolabel binding to calf
3 thymus DNA.
4
5
This document is a draft for review purposes only and does not constitute Agency policy.
23 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Table 3-12. Binding of 14C from [14C]TCE in rat liver and kidney at 72 hours
after oral administration of 200 mg/kg [14C]TCE (Dekant et al., 1986b)
Tissue
DPM/gram tissue
Untreated
Phenobarbital
Arochlor 1254
Liver
850± 100
9,300 ± 1,100
8,700 ± 1,000
Kidney
680± 100
5,700 ± 900
7,300 ±800
3.3. METABOLISM
This section focuses on both in vivo and in vitro studies of the biotransformation of
trichloroethylene, identifying metabolites that are deemed significant for assessing toxicity and
carcinogenicity. In addition, metabolism studies may be used to evaluate the flux of parent
compound through the known metabolic pathways. Sex-, species-, and interindividual
differences in the metabolism of TCE are discussed, as are factors that possibly contribute to this
variability. Additional discussion of variability and susceptibility is presented in Section 4.10.
3.3.1. Introduction
The metabolism of TCE has been studied mostly in mice, rats, and humans and has been
extensively reviewed (IARC, 1995d; Lash et al., 2000a; U.S. EPA, 1985). It is now well
accepted that TCE is metabolized in laboratory animals and in humans through at least
two distinct pathways: (1) oxidative metabolism via the cytochrome P450 mixed-function
oxidase system and (2) GSH conjugation followed by subsequent further biotransformation and
processing, either through the cysteine conjugate beta lyase pathway or by other enzymes (Lash
et al., 2000b). While the flux through the conjugative pathway is less, quantitatively, than the
flux through oxidation (Bloemen et al., 2001), GSH conjugation is an important route
toxicologically, giving rise to relatively potent toxic biotransformation products (Elfarra et al.,
1987; Elfarra et al., 1986).
Information about metabolism is important because, as discussed extensively in
Chapter 4, certain metabolites are thought to cause one or more of the same acute and chronic
toxic effects, including carcinogenicity, as TCE. Thus, in many of these cases, the toxicity of
TCE is generally considered to reside primarily in its metabolites rather than in the parent
compound itself.
This document is a draft for review purposes only and does not constitute Agency policy.
24 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
3.3.2. Extent of Metabolism
TCE is extensively metabolized in animals and humans. The most comprehensive
mass-balance studies are in mice and rats (Dekant et al., 1984, 1986a; Dekant et al., 1986b;
Green and Prout, 1985; Prout et al., 1985) in which [14C]TCE is administered by oral gavage at
doses of 2 -2,000 mg/kg, the data from which are summarized in Figure 3-2 and Figure 3-3. In
both mice and rats, regardless of sex and strain, there is a general trend of increasing exhalation
of unchanged TCE with dose, suggesting saturation of a metabolic pathway. The increase is
smaller in mice (from 1-6% to 10-18%) than in rats (from 1-3% to 43—78%), suggesting
greater overall metabolic capacity in mice. The dose at which apparent saturation occurs appears
to be more sex- or strain-dependent in mice than in rats. In particular, the marked increase in
exhaled TCE occurred between 20 and 200 mg/kg in female NMRI mice, between 500 and
1,000 mg/kg in B6C3F1 mice, and between 10 and 500 mg/kg in male Swiss-Webster mice.
However, because only one study is available in each strain, interlot or interindividual variability
might also contribute to the observed differences. In rats, all three strains tested showed marked
increase in unchanged TCE exhaled between 20 and 200 mg/kg or 10 and 500 mg/kg.
Recovered urine, the other major source of excretion, had mainly TCA, trichloroethanol, and
trichloroethanol-glucuronide conjugate (TCOG), but revealed no detectable TCE. The source of
radioactivity in feces was not analyzed, but it is presumed not to include substantial TCE given
the complete absorption expected from the corn oil vehicle. Therefore, at all doses tested in
mice, and at doses <200 mg/kg in rats, the majority of orally administered TCE is metabolized.
Pretreatment of rats with P450 inducers prior to a 200 mg/kg dose did not change the pattern of
recovery, but it did increase the amount recovered in urine by 10-15%), with a corresponding
decrease in the amount of exhaled unchanged TCE (Dekant et al., 1986b).
Comprehensive mass balance studies are not available in humans, but several studies
have measured or estimated recovery of TCE in exhaled breath and/or TCA and TCOH in urine
following controlled inhalation exposures to TCE (Monster et al., 1976; Opdam, 1989b; Soucek
and Vlachova, 1960). Opdam (1989b) only measured exhaled breath, and estimated that, on
average, 15—20% of TCE uptake (retained dose) was exhaled after exposure to 5.8-38 ppm for
29-62 minutes. Soucek and Vlachova (1960) and Bartonicek (1962) did not measure exhaled
breath but did report 69-73% of the retained dose excreted in urine as TCA and TCOH
"3
following exposure to 93-194 ppm (500-1,043 mg/m ) for 5 hours. Soucek and Vlachova
(1960) additionally reported 4% of the retained dose excreted in urine as monochloroacetic acid
(MCA). Monster et al. (1976) reported that an average of 10% of the retained TCE dose was
This document is a draft for review purposes only and does not constitute Agency policy.
25 DRAFT—DO NOT CITE OR QUOTE
-------
100
90 ¦
80
70 ¦
60 ¦
~ 50 ¦
T3
O
O
>
o
o
o
&
">
o
.2 40 -\
T>
TO
^ 30 ¦
20 ¦
10 ¦
2 mg/kg 20 200
mg/kg mg/kg
F/NMRI
l l
10 500 1000 2000
mg/kg mg/kg mg/kg mg/kg
M/B6C3F1
I
10 500 1000
mg/kg mg/kg mg/kg
M/Swiss-Webster
~ Cage
wash
¦ Carcass
¦ C02
Exhaled
~ Feces
~ Urine
~ TCE
Exhaled
Mouse Sex/Strain and Dose
3
1 eliminated unchanged following 6 hour exposures to 70-140 ppm (376-752 mg/m ) TCE, along
2 with an average of 57% of the retained dose excreted in urine as TCA and free or conjugated
3
4
5 Figure 3-2. Disposition of [14C]TCE administered by oral gavage in mice
6 (Dekant et al., 1984; Dekant et al., 1986b; Green and Prout, 1985; Prout et
7 al., 1985).
8
9
This document is a draft for review purposes only and does not constitute Agency policy.
26 DRAFT—DO NOT CITE OR QUOTE
-------
100
90 -
80
70 -
60 -
¦a
CD
CD
>
o
o
a>
J
'>
u
ro
° 40 H
2
^ 30 -
£ 50
20
10 -
2 mg/kg 20 200
mg/kg mg/kg
F/Wistar
10 500 1000 2000
mg/kg mg/kg mg/kg mg/kg
M/Osborne-Mendel
10 500 1000
mg/kg mg/kg mg/kg
M/Alderley-Park Wistar
~ Cage
wash
¦ Carcass
~ C02
Exhaled
~ Feces
~ Urine
~ TCE
Exhaled
Rat Sex/Strain and Dose
1
2
3
4
5
6
Figure 3-3. Disposition of [14C]TCE administered by oral gavage in rats
(Dekant et al., 1984; Dekant et al., 1986b; Green and Prout, 1985; Prout et
al., 1985).
This document is a draft for review purposes only and does not constitute Agency policy.
27 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
TCOH. The differences among these studies may reflect a combination of interindividual
variability and errors due to the difficulty in precisely estimating dose in inhalation studies, but
in all cases less than 20% of the retained dose was exhaled unchanged and greater than 50% was
excreted in urine as TCA and TCOH. Therefore, it is clear that TCE is extensively metabolized
in humans. No saturation was evident in any of these human recovery studies at the exposure
levels tested.
3.3.3. Pathways of Metabolism
As mentioned in Section 3.3.1, TCE metabolism in animals and humans has been
observed to occur via two major pathways: P450-mediated oxidation and GSH conjugation.
Products of the initial oxidation or conjugation step are further metabolized to a number of other
metabolites. For P450 oxidation, all steps of metabolism occur primarily in the liver, although
limited oxidation of TCE has been observed in the lungs of mice, as discussed below. The GSH
conjugation pathway also begins predominantly in the liver, but toxicologically significant
metabolic steps occur extrahepatically—particularly in the kidney (Lash et al., 1999a; 2006;
Lash et al., 1998b; Lash et al., 1995). The mass-balance studies cited above found that at
exposures below the onset of saturation, >50% of TCE intake is excreted in urine as oxidative
metabolites (primarily as TCA and TCOH), so TCE oxidation is generally greater than TCE
conjugation. This is discussed in detail in Section 3.3.3.3.
3.3.3.1.1. Cytochrome P450-Dependent Oxidation
Oxidative metabolism by the cytochrome P450, or CYP-dependent, pathway is
quantitatively the major route of TCE biotransformation (IARC, 1995d; 2000a; Lash et al.,
2000b; U.S. EPA, 1985). The pathway is operative in humans and rodents and leads to several
metabolic products, some of which are known to cause toxicity and carcinogenicity (IARC,
1995d; U.S. EPA, 1985). Although several of the metabolites in this pathway have been clearly
identified, others are speculative or questionable. Figure 3-4 depicts the overall scheme of TCE
P450 metabolism.
In brief, TCE oxidation via P450, primarily CYP2E1 (Guengerich and Shimada, 1991),
yields an oxygenated TCE-P450 intermediate. The TCE-P450 complex is a transition state that
goes on to form chloral or TCE oxide. In the presence of water, chloral rapidly equilibrates with
chloral hydrate (CH), which undergoes reduction and oxidation by alcohol dehydrogenase and
This document is a draft for review purposes only and does not constitute Agency policy.
28 DRAFT—DO NOT CITE OR QUOTE
-------
H (TCE) CI
P450
ohch2 —
(W-(Hydroxyacetyl)- N (CH2)2OH
aminoethanol) |-|
TCE-O-P450
(TCE-O) ^ci
OH
CI, CH
cue
CI, c
(CHL)
(DCAC)
ALDH
WDH or
P450
CI, c
cue
P450
CI, CH
(TCOH) QH
(DCA) OH
GST-
zeta>
EHR
UGT
cue
CO-
CI CH
(TCOG) o-gluc
(Glyoxylic
acid)
HO (OA) OH
(MCA) OH
aldehyde dehydrogenase or aldehyde oxidase to form TCOH and TCA, respectively (Dekant et
al., 1986b; Green and Prout, 1985; Miller and Guengerich, 1983). TCE oxide can rearrange to
Figure 3-4. Scheme for the oxidative metabolism of TCE.
Adapted from: Clewell et al. (2000); Cummings et al. (2001); Forkert et al. (2006); Lash et al.
(2000b); Lash et al. (2000a); Tong et al. (1998).
This document is a draft for review purposes only and does not constitute Agency policy.
29 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-13. In vitro TCE oxidative metabolism in hepatocytes and
2 microsomal fractions
3
In vitro
system
Km
Vmax
1,000 X
Vmax/Km
Source
(iM in
medium
nmol TCE
oxidized/min/mg
MSPa or 106
hepatocytes
Human
hepatocytes
210± 159
(45-403)
0.268 ±0.215
(0.101-0.691)
2.45 ±2.28
(0.46-5.57)
Lipscomb et al. (1998a)
Human liver
microsomal
protein
16.7 ±2.45
(13.3-19.7)
1.246 ±0.805
(0.490-3.309)
74.1 ±44.1
(38.9-176)
Lipscomb et al. (1997) (Low
Km)
30.9 ±3.3
(27.0-36.3)
1.442 ±0.464
(0.890-2.353)
47.0 ± 16.0
(30.1-81.4)
Lipscomb et al. (1997) (Mid
Km)
51.1 ±3.77
(46.7-55.7)
2.773 ±0.577
(2.078-3.455)
54.9 ± 14.1
(37.3-69.1)
Lipscomb et al. (1997) (High
Km)
24.6
1.44
58.5
Lipscomb et al. (1998b)
(pooled)
12 ±3
(9-14)
0.52 ±0.17
(0.37-0.79)
48 ±23
(26-79)
Elfarra et al. (1998) (males,
high affinity)
26 ± 17
(13-45)
0.33 ±0.15
(0.19-0.48)
15 ± 10
(11-29)
Elfarra et al. (1998) (females,
high affinity)
Rat liver
microsomal
protein
55.5
4.826
87.0
Lipscomb et al. (1998b)
(pooled)
72 ±82
0.96 ±0.65
24 ±21
Elfarra et al. (1998) (males,
high affinity)
42 ±21
2.91 ±0.71
80 ±34
Elfarra et al. (1998) (females,
high affinity)
Rat kidney
microsomal
protein
940
0.154
0.164
Cummings et al. (2001)
Mouse liver
microsomal
protein
35.4
5.425
153
Lipscomb et al. (1998b)
(pooled)
378 ± 414
8.6 ±4.5
42 ±29
Elfarra et al. (1998) (males)
161 ±29
26.06 ±7.29
163 ±37
Elfarra et al. (1998) (females)
4
5 a MSP = Microsomal protein.
6
This document is a draft for review purposes only and does not constitute Agency policy.
3 0 DRAFT—DO NOT CITE OR QUOTE
-------
1 Notes: Results presented as mean ± standard deviation (minimum-maximum). KM for human hepatocytes
2 converted from ppm in headspace to (iM in medium using reported hepatocyte:air partition coefficient (Lipscomb et
3 al., 1998a).
4
This document is a draft for review purposes only and does not constitute Agency policy.
31 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
dichloroacetyl chloride. Table 3-13 summarizes available in vitro measurements of TCE
oxidation, as assessed by the formation of CH, TCOH, and TCA. Glucuronidation of TCOH
forms TCOG, which is readily excreted in urine. Alternatively, TCOG can be excreted in bile
and passed to the small intestine where it is hydrolyzed back to TCOH and reabsorbed (Bull,
2000). TCA is poorly metabolized but may undergo dechlorination to form dichloroacetic
acid(DCA). However, TCA is predominantly excreted in urine, albeit at a relatively slow rate as
compared to TCOG. Like the TCE-P450 complex, TCE oxide also seems to be a transient
metabolite. Recent data suggest that it is transformed to dichloroactyl chloride, which
subsequently decomposes to form DCA (Cai and Guengerich, 1999). As shown in Figure 3-4,
several other metabolites, including oxalic acid and A-(hydroxyacetyl) aminoethanol, may form
from the TCE oxide or the TCE-O-P450 intermediate and have been detected in the urine of
rodents and humans following TCE exposure. Pulmonary excretion of CO2 has been identified
in exhaled breath from rodents exposed to 14C-labeled TCE and is thought to arise from
metabolism of DCA. The following sections provide details as to pathways of TCE oxidation,
including discussion of inter- and intraspecies differences in metabolism.
3.3.3.1.2. Formation of trichloroethylene oxide
In previous studies of halogenated alkene metabolism, the initial step was the generation
of a reactive epoxide (Anders and Jakobson, 1985). Early studies in anesthetized human patients
(Powell, 1945), dogs (Butler, 1949), and later reviews (e.g., Goeptar et al., 1995) suggest that the
TCE epoxide may be the initial reaction product of TCE oxidation.
Epoxides can form acyl chlorides or aldehydes, which can then form aldehydes,
carboxylic acids, or alcohols, respectively. Thus, earlier studies suggesting the appearance of
CH, TCA, and TCOH as the primary metabolites of TCE were considered consistent with the
oxidation of TCE to an epoxide intermediate (Butler, 1949; Powell, 1945). Following in vivo
exposures to 1,1-dichloroethylene, a halocarbon very similar in structure to TCE, mouse liver
cytosol and microsomes and lung Clara cells exhibited extensive P450-mediated epoxide
formation (Dowsley et al., 1996; Forkert, 1999a, b; Forkert et al., 1999) Indeed, TCE oxide
inhibits purified CYP2E1 activity (Cai and Guengerich, 2001a) similarly to TCE inhibition of
CYP2E1 in human liver microsomes (Lipscomb et al., 1997).
Conversely, cases have been made against TCE oxide as an obligate intermediate to the
formation of chloral. Using liver microsomes and reconstituted P450 systems (Miller and
Guengerich, 1982, 1983) or isolated rat hepatocytes (Miller and Guengerich, 1983), it has been
suggested that chlorine migration and generation of a TCE-O-P450 complex (via the heme
This document is a draft for review purposes only and does not constitute Agency policy.
32 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
oxygen) would better explain the observed destruction of the P450 heme, an outcome not likely
to be epoxide-mediated. Miller and Guengerich (1982) found CYP2E1 to generate an epoxide
but argued that the subsequent production of chloral was not likely related to the epoxide. Green
and Prout (1985) argued against epoxide (free form) formation in vivo in mice and rats,
suggesting that the expected predominant metabolites would be carbon monoxide, CO2, MCA,
and DCA, rather than the observed predominant appearance of TCA, TCOH, and TCOG.
It appears likely that both a TCE-O-P450 complex and a TCE oxide are formed, resulting
in both CH and dichloroacetyl chloride, respectively, though it appears that the former
predominates. In particular, it has been shown that dichloroacetyl chloride can be generated
from TCE oxide, dichloracetyl chloride can be trapped with lysine (Cai and Guengerich, 1999),
and that dichloracetyl-lysine adducts are formed in vivo (Forkert et al., 2006). Together, these
data strongly suggest TCE oxide as an intermediate metabolite, albeit short-lived, from TCE
oxidation in vivo.
3.3.3.1.3. Formation of chloral hydrate (CH), trichloroethanol (TCOH) and trichloroacetic
acid (TCA)
CH (in equilibrium with chloral) is a major oxidative metabolite produced from TCE as
has been shown in numerous in vitro systems, including human liver microsomes and purified
P450 CYP2E1 (Guengerich et al., 1991) as well as recombinant rat, mouse, and human P450s
including CYP2E1 (Forkert et al., 2005). However, in rats and humans, in vivo circulating CH is
generally absent from blood following TCE exposure. In mice, CH is detectable in blood and
tissues but is rapidly cleared from systemic circulation (Abbas and Fisher, 1997). The low
systemic levels of CH are because of its rapid transformation to other metabolites.
CH is further metabolized predominantly to TCOH (Sellers et al., 1972; Shultz and
Weiner, 1979) and/or CYP2E1 (Ni et al., 1996). The role for alcohol dehydrogenase was
suggested by the observation that ethanol inhibited CH reduction to TCOH (Larson and Bull,
1989; Muller et al., 1975; Sellers et al., 1972). For instance, Sellers et al. (1972) reported that
coexposure of humans, to ethanol and CH resulted in a higher percentage of urinary TCOH (24%
of CH metabolites) compared to TCA (19%). When ethanol was absent, 10 and 11% of CH was
metabolized to TCOH and TCA, respectively. However, because ethanol can be oxidized by
both alcohol dehydrogenase and CYP2E1, there is some ambiguity as to whether these
observations involve competition with one or the other of these enzymes. For instance, Ni et al.
(1996) reported that CYP2E1 expression was necessary for metabolism of CH to mutagenic
metabolites in a human lymphoblastoid cell line, suggesting a role for CYP2E1. Furthermore, Ni
This document is a draft for review purposes only and does not constitute Agency policy.
3 3 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
et al. (1996) reported that cotreatment of mice with CH and pyrazole, a specific CYP2E1
inducer, resulted in enhanced liver microsomal lipid peroxidation, while treatment with DPEA,
an inhibitor of CYP2E1, suppressed lipid peroxidation, suggesting CYP2E1 as a primary enzyme
for CH metabolism in this system. Lipscomb et al. (1996) suggested that two enzymes are likely
responsible for CH reduction to TCOH based on observation of bi-phasic metabolism for this
pathway in mouse liver microsomes. This behavior has also been observed in mouse liver
cytosol, but was not observed in rat or human liver microsomes. Moreover, CH metabolism to
TCOH increased significantly both in the presence of nicotinamide adenine dinucleotide
(NADH) in the 700 x g supernatant of mouse, rat, and human liver homogenate as well as with
the addition of NADPH in human samples, suggesting two enzymes may be involved (Lipscomb
et al., 1996).
TCOH formed from CH is available for oxidation to TCA (see below) or glucuronidation
via uridine 5'-diphospho-glucuronyltransferase to TCOG, which is excreted in urine or in bile
(Stenner et al., 1997). Biliary TCOG is hydrolyzed in the gut and available for reabsorption to
the liver as TCOH, where it can be glucuronidated again or metabolized to TCA. This
enterohepatic circulation appears to play a significant role in the generation of TCA from TCOH
and in the observed lengthy residence time of this metabolite, compared to TCE. Using jugular-,
duodenal-, and bile duct-cannulated rats, Stenner et al. (1997) showed that enterohepatic
circulation of TCOH from the gut back to the liver and subsequent oxidation to TCA was
responsible for 76% of TCA measured in the systemic blood.
Oxidation of CH and TCOH to TCA has been demonstrated in vivo in mice (Dekant et
al., 1986b; Green and Prout, 1985; Larson and Bull, 1992a), rats (Dekant et al., 1986b; Green
and Prout, 1985; Larson and Bull, 1992a; Pravecek et al., 1996; Stenner et al., 1997; Templin et
al., 1995b), dogs (Templin et al., 1995b), and humans (Sellers et al., 1978). Urinary metabolite
data in mice and rats exposed to 200 mg/kg TCE (Dekant et al., 1986b; Larson and Bull, 1992a);
and humans following oral CH exposure (Sellers et al., 1978) show greater TCOH production
relative to TCA production. However, because of the much longer urinary half-life in humans of
TCA relative to TCOH, the total amount of TCA excreted may be similar to TCOH (Fisher et al.,
1998; Monster et al., 1976). This is thought to be primarily due to conversion of TCOH to TCA,
either directly or via "back-conversion" of TCOH to CH, rather than due to the initial formation
of TCA from CH (Owens and Marshall, 1955).
In vitro data are also consistent with CH oxidation to TCA being much less than CH
reduction to TCOH. For instance, Lipscomb et al. (1996) reported 1,832-fold differences in Km
values and 10-195-fold differences in clearance efficiency (Vmax/Km) for TCOH and TCA in all
three species (see Table 3-14). Clearance efficiency of CH to TCA in mice is very similar to
This document is a draft for review purposes only and does not constitute Agency policy.
34 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
humans but is 13-fold higher than rats. Interestingly, Bronley-DeLancey et al. (2006) recently
reported that similar amounts of TCOH and TCA were generated from CH using cryopreserved
human hepatocytes. However, the intersample variation was extremely high, with measured
Vmax ranging from eightfold greater TCOH to fivefold greater TCA and clearance (Vmax/Km)
ranging from 13-fold greater TCOH to 17-fold greater TCA. Moreover, because a comparison
with fresh hepatocytes or microsomal protein was not made, it is not clear to what extent these
differences are due to population heterogeneity or experimental procedures.
The metabolism of CH to TCA and TCOH involves several enzymes including CYP2E1,
alcohol dehydrogenase, and aldehyde dehydrogenase enzymes (Guengerich et al., 1991; Miller
and Guengerich, 1983; Ni et al., 1996; Shultz and Weiner, 1979; Wang et al., 1993). Because
these enzymes have preferred cofactors (NADPH, NADH, and NAD+), cellular cofactor ratio
and redox status of the liver may have an impact on the preferred pathway (Kawamoto et al.,
1988a; Lipscomb et al., 1996).
3.3.3.1.4. Formation of dichloroacetic acid (DCA) and other products
As discussed above, DCA could hypothetically be formed via multiple pathways. The
work reviewed by Guengerich (2004) has suggested that one source of DCA may be through a
TCE oxide intermediary. Miller and Guengerich (1983) report evidence of formation of the
epoxide, and Cai and Guengerich (1999) report that a significant amount (about 35%) of DCA is
formed from aqueous decomposition of TCE oxide via hydrolysis in an almost pH-independent
manner. Because this reaction forming DCA from TCE oxide is a chemical process rather than a
process mediated by enzymes, and because evidence suggests that some epoxide was formed
from TCE oxidation, Guengerich (2004) notes that DCA would be an expected product of TCE
oxidation (see also Yoshioka et al., 2002). Alternatively, dechlorination of TCA and oxidation
of TCOH have been proposed as sources of DCA (Lash et al., 2000a). Merdink et al. (2000)
investigated dechlorination of TCA and reported trapping a DCA radical with the spin-trapping
agent phenyl-tert-butyl nitroxide, identified by gas chromatography/mass spectroscopy, in both a
chemical Fenton system and rodent microsomal incubations with TCA as substrate.
Dose-dependent catalysis of TCA to DCA was observed in cultured microflora from B6C3F1
mice (Moghaddam et al., 1996). However, while antibiotic-treated mice lost the ability to
produce DCA in the gut, plasma DCA levels were unaffected by antibiotic treatment, suggesting
that the primary site of murine DCA production is
This document is a draft for review purposes only and does not constitute Agency policy.
3 5 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-14. In vitro kinetics of trichloroethanol and trichloroacetic acid
2 formation from chloral hydrate in rat, mouse, and human liver homogenates
3
Species
TCOH
TCA
Kma
VMAXb
V\1\\/Km(
Kma
VMAXb
V\i\\/Kmc
Rat
0.52
24.3
46.7
16.4
4
0.24
Moused
0.19
11.3
59.5
3.5
10.6
3.0
High affinity
0.12
6.3
52.5
nae
na
na
Low affinity
0.51
6.1
12.0
na
na
na
Human
1.34
34.7
25.9
23.9
65.2
2.7
4
5 a Km presented as mM CH in solution.
6 b Vmax presented as nmoles/mg supernatant protein/min.
7 0 Clearance efficiency represented by VMax/Km-
8 d Mouse kinetic parameters derived for observations over the entire range of CH exposure as well as discrete,
9 bi-phasic regions for CH concentrations below (high affinity) and above (low affinity) 1.0 mM.
10 e na = not applicable.
11
12 Source: Lipscomb et al. (1996).
13
14
15 other than the gut (Moghaddam et al., 1997).
16 However, direct evidence for DC A formation from TCE exposure remains equivocal. In
17 vitro studies in human and animal systems have demonstrated very little DC A production in the
18 liver (James et al., 1997). In vivo, DCA was detected in the blood of mice (Larson and Bull,
19 1992a; Templin et al., 1993) and humans (Fisher et al., 1998) and in the urine of rats and mice
20 (Larson and Bull, 1992b) exposed to TCE by aqueous oral gavage. However, the use of strong
21 acids in the analytical methodology produces ex vivo conversion of TCA to DCA in mouse blood
22 (Ketcha et al., 1996). This method may have resulted in the appearance of DCA as an artifact in
23 human plasma (Fisher et al., 1998) and mouse blood in vivo (Templin et al., 1995b). Evidence
24 for the artifact is suggested by DCA AUCs that were larger than would be expected from the
25 available TCA (Templin et al., 1995b). After the discovery of these analytical issues, Merdink
26 et al. (1998) reevaluated the formation of DCA from TCE, TCOH, and TCA in mice, with
27 particular focus on the hypothesis that DCA is formed from dechlorination of TCA. They were
28 unable to detect blood DCA in naive mice after administration of TCE, TCOH, or TCA. Low
29 levels of DCA were detected in the blood of children administered therapeutic doses of CH
30 (Henderson et al., 1997), suggesting TCA or TCOH as the source of DCA. Oral TCE exposure
31 in rats and dogs failed to produce detectable levels of DCA (Templin et al., 1995b).
This document is a draft for review purposes only and does not constitute Agency policy.
3 6 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Another difficulty in assessing the formation of DC A is its rapid metabolism at low
exposure levels. Degradation of DCA is mediated by glutathione-S-transferase (GST)-zeta
(Saghir and Schultz, 2002; Tong et al., 1998), apparently occurring primarily in the hepatic
cytosol. DCA metabolism results in suicide inhibition of the enzyme, evidenced by decreased
DCA metabolism in DCA-treated animals (Gonzalez-Leon et al., 1999) and humans (Shroads et
al., 2008) and loss of DCA metabolic activity and enzymatic protein in liver samples from
treated animals (Schultz et al., 2002). This effect has been noted in young mice exposed to DCA
in drinking water at doses approximating 120 mg/kg-day (Schultz et al., 2002). The
experimental data and pharmacokinetic model simulations of several investigators (Jia et al.,
2006; Keys et al., 2004; Li et al., 2008; Merdink et al., 1998; Shroads et al., 2008) suggest that
several factors prevent the accumulation of measurable amounts of DCA: (1) its formation as a
short-lived intermediate metabolite, and (2) its rapid elimination relative to its formation from
TCA. While DCA elimination rates appear approximately one order of magnitude higher in rats
and mice than in humans (James et al., 1997) (see Table 3-15), they still may be rapid enough so
that even if DCA were formed in humans, it would be metabolized too quickly to appear in
detectable quantities in blood.
A number of other metabolites, such as oxalic acid, MCA, glycolic acid, and glyoxylic
acid, are formed from DCA (Lash et al., 2000a; Saghir and Schultz, 2002). Unlike other
oxidative metabolites of TCE, DCA appears to be metabolized primarily via hepatic cytosolic
proteins. Since P450 activity resides almost exclusively in the microsomal and mitochondrial
cell fractions, DCA metabolism appears to be independent of P450. Rodent microsomal and
mitochondrial metabolism of DCA was measured to be <10% of cytosolic metabolism
(Lipscomb et al., 1995). DCA in the liver cytosol from rats and humans is transformed to
glyoxylic acid via a GSH-dependent pathway (James et al., 1997). In rats, the Km for GSH was
0.075 mM with a Vmax for glyoxylic acid formation of 1.7 nmol/mg protein/minute. While this
pathway may not involve GST (as evidenced by very low GST activity in this study), Tong et al.
(1998) showed GST-zeta, purified from rat liver, to be involved in metabolizing DCA to
glyoxylic acid, with a Vmax of 1,334 nmol/mg protein/minute and Km of 71.4 |iM for glyoxylic
acid formation and a GSH Km of 59 |iM,
This document is a draft for review purposes only and does not constitute Agency policy.
3 7 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Table 3-15. In vitro kinetics of DCA metabolism in hepatic cytosol
of mice, rats, and humans
Vmax
Km
Species
(nmol/min/mg protein)
(liM)
Vmax/Km
Mouse
13.1
350
37.4
Rat
11.6
280
41.4
Human
0.37
71
5.2
Source: James et al. (1997).
3.3.3.1.5. Tissue distribution of oxidative metabolism and metabolites
Oxidative metabolism of TCE, irrespective of the route of administration, occurs
predominantly in the liver, but TCE metabolism via the P450 (CYP) system also occurs at other
sites because CYP isoforms are present to some degree in most tissues of the body. For
example, both the lung and kidneys exhibit cytochrome P450 enzyme activities (Cummings et
al., 2001; Forkert et al., 2005; 1997a; Green et al., 1997b). Green et al. (1997b) detected TCE
oxidation to chloral in microsomal fractions of whole-lung homogenates from mice, rats, and
humans, with the activity in mice the greatest and in humans the least. The rates were slower
than in the liver (which also has a higher microsomal protein content as well as greater tissue
mass) by 1.8-, 10-, and > 10-fold in mice, rats, and humans, respectively. While qualitatively
informative, these rates were determined at a single concentration of about 1 mM TCE. A full
kinetic analysis was not performed, so clearance and maximal rates of metabolism could not be
determined. With the kidney, Cummings et al. (2001) performed a full kinetic analysis using
kidney microsomes, and found clearance rates (Vmax/Km) for oxidation were more than 100-fold
smaller than average rates that were found in the liver (see Table 3-13). In human kidney
microsomes, Amet et al.(1997) reported that CYP2E1 activity was weak and near detection
limits, with no CYP2E1 detectable using immunoblot analysis. Cummings and Lash (2000)
reported detecting oxidation of TCE in only one of four kidney microsome samples, and only at
the highest tested concentration of 2 mM, with a rate of 0.13 nmol/minute/mg protein. This rate
contrasts with the Vmax values for human liver microsomal protein of 0.19-3.5 nmol/minute/mg
protein reported in various experiments (see Table 3-13, above). Extrahepatic oxidation of TCE
may play an important role for generation of toxic metabolites in situ. The roles of local
This document is a draft for review purposes only and does not constitute Agency policy.
3 8 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
metabolism in kidney and lung toxicity are discussed in detail in Sections 4.4 and 4.7,
respectively.
With respect to further metabolism beyond oxidation of TCE, CH has been shown to be
metabolized to TCA and TCOH in lysed whole blood of mice and rats and fractionated human
blood (Lipscomb et al., 1996) (see Table 3-16). TCOH production is similar in mice and rats and
is approximately twofold higher in rodents than in human blood. However, TCA formation in
human blood is two- or threefold higher than in mouse or rat blood, respectively. In human
blood, TCA is formed only in the erythrocytes. TCOH formation occurs in both plasma and
erythrocytes, but fourfold more TCOH is found in plasma than in an equal volume of packed
erythrocytes. While blood metabolism of CH may contribute further to its low circulating levels
in vivo., the metabolic capacity of blood (and kidney) may be substantially lower than liver.
Regardless, any CH reaching the blood may be rapidly metabolized to TCA and TCOH. DCA
and TCA are known to bind to plasma proteins. Schultz et al. (1999) measured DCA binding in
rats at a single concentration of about 100 |iM and found a binding fraction of less than 10%.
However, these data are not greatly informative for TCE exposure in which DCA levels are
significantly lower, and limitation to a single concentration precludes fitting to Templin et al.
(1993, 1995a; 1995b), Schultz et al. (1999), Lumpkin et al. (2003), and Yu et al. (2003) all
measured TCA binding in various species and at various concentration ranges. Of these,
Templin et al. (1995a; 1995b) and Lumpkin et al. (2003) measured levels in humans, mice, and
rats. Lumpkin et al. (2003) studied the widest concentration range, spanning reported TCA
plasma concentrations from experimental studies. Table 3-17 shows derived binding parameters.
However, these data are not entirely consistent among researchers; two- to fivefold differences in
BMax and Kd are noted in some cases, although some differences existed in the rodent strains and
experimental protocols used. In general, however, at lower concentrations, the bound fraction
appears greater.
This document is a draft for review purposes only and does not constitute Agency policy.
3 9 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-16. TCOH and TCA formed from CH in vitro in lysed whole blood
2 of rats and mice or fractionated blood of humans (nmoles formed in 400 jiL
3 samples over 30 minutes)
4
Rat
Mouse
Human
Erythrocytes
Plasma
TCOH
45.4 ±4.9
46.7 ± 1.0
15.7 ±1.4
4.48 ±0.2
TCA
0.14 ± 0.2
0.21 ±0.3
0.42 ±0.0
not detected
5
6 Source: Lipscomb et al. (1996).
7
8
9 in humans than in rats and mice. Typical human TCE exposures, even in controlled experiments
10 with volunteers, lead to TCA blood concentrations well below the reported Kd (see Table 3-17,
11 below), so the TCA binding fraction should be relatively constant. However, in rats and mice,
12 experimental exposures may lead to peak concentrations similar to, or above, the reported Kd
13 (e.g., Templin et al., 1993; Yu et al., 2000), meaning that the bound fraction should temporarily
14 decrease following such exposures.
15 Limited data are available on tissue:blood partitioning of the oxidative metabolites CH,
16 TCA, TCOH and DCA, as shown in Table 3-18. As these chemicals are all water soluble and
17 not lipophilic, it is not surprising that their partition coefficients are close to one (within about
18 twofold). It should be noted that the TCA tissue:blood partition coefficients reported in
19 Table 3-18 were measured at concentrations 1.6-3.3 M, over 1,000-fold higher than the reported
20 Kd. Therefore, these partition coefficients should reflect the equilibrium between tissue and free
21 blood concentrations. In addition, only one in vitro measurement has been reported of
22 blood:plasma concentration ratios for TCA: Schultz et al. (1999) reported a value of 0.76 in rats.
23
3.3.3.1.6. Species-, sex-, and age-dependent differences of oxidative metabolism
24 The ability to describe species- and sex-dependent variations in TCE metabolism is
25 important for species extrapolation of bioassay data and identification of human populations that
26 are particularly susceptible to TCE toxicity. In particular, information on the variation in the
27 initial oxidative step of CH formation from TCE is desirable, because this is the rate-limiting
28 step in the eventual formation and distribution of the putative toxic metabolites TCA and DCA
29 (Lipscomb et al., 1997).
This document is a draft for review purposes only and does not constitute Agency policy.
40 DRAFT—DO NOT CITE OR QUOTE
-------
1 Inter- and intraspecies differences in TCE oxidation have been investigated in vitro using
2 cellular or subcellular fractions, primarily of the liver. The available in vitro metabolism data on
3 TCE oxidation in the liver (see Table 3-13) show substantial inter and intraspecies variability.
4 Table 3-17. Reported TCA plasma binding parameters
5
A
Bmax
(HM)
Ka
(HM)
A+
Bmax/K,
Concentration
range (jiM
bound+free)
Human
Templin et al. (1995b)
-
1,020
190
5.37
3-1,224
Lumpkin et al. (2003)
-
708.9
174.6
4.06
0.06-3,065
Rat
Templin et al. (1995b)
-
540
400
1.35
3-1,224
Yu et al. (2000)
0.602
312
136
2.90
3.8-1,530
Lumpkin et al. (2003)
-
283.3
383.6
0.739
0.06-3,065
Mouse
Templin et al. (1993)
-
310
248
1.25
3-1,224
Lumpkin et al. (2003)
-
28.7
46.1
0.623
0.06-1,226
6
7 Notes: Binding parameters based on the equation CboUnd = A x Cfree + Bmax x Cfree/(Kd + Cfree), where CboUnd is the
8 bound concentration, Cfree is the free concentration, and A = 0 for Templin et al. (1993; 1995b) and Lumpkin et al.
9 (2003). The quantity A+ BMAv/K,i is the ratio of bound-to-free at low concentrations.
10
11 Table 3-18. Partition coefficients for TCE oxidative metabolites
12
Species/tissue
Tissue:blood partition coefficient
CH
TCA
TCOH
DCA
Human"
Kidney
-
0.66
2.15
-
Liver
-
0.66
0.59
-
Lung
-
0.47
0.66
-
Muscle
-
0.52
0.91
-
Mouseb
Kidney
0.98
0.74
1.02
0.74
Liver
1.42
1.18
1.3
1.08
Lung
1.65
0.54
0.78
1.23
Muscle
1.35
0.88
1.11
0.37
This document is a draft for review purposes only and does not constitute Agency policy.
41 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
a Fisher et al. (1998).
b Abbas and Fisher (1997).
Note: TCA and TCOH partition coefficients have not been reported for rats.
Across species, microsomal data show that mice apparently have greater capacity (Vmax) than
rat or humans, but the variability within species can be 2- to 10-fold. Part of the explanation may
be related to CYP2E1 content. Although liver P450 content is similar across species, mice and
rats exhibit higher levels of CYP2E1 content (0.85 and 0.89 nmol/mg protein, respectively)
(Davis et al., 2002; Nakajima et al., 1993) than humans (approximately 0.25-0.30 nmol/mg
protein) (Davis et al., 2002; Elfarra et al., 1998). Thus, the data suggest that rodents would have
a higher capacity than humans to metabolize TCE, but this is difficult to verify in vivo because
very high exposure concentrations in humans would be necessary to assess the maximum
capacity of TCE oxidation.
With respect to the KM of liver microsomal TCE oxidative metabolism, where KM is
indicative of affinity (the lower the numerical value of KM, the higher the affinity), the trend
appears to be mice and rats have higher Km values (i.e., lower affinity) than humans, but with
substantial overlap due to interindividual variability. Note that, as shown in Table 3-13, the
ranking of rat and mouse liver microsomal KM values between the two reports by Lipscomb et al.
(1998b) and Elfarra et al. (1998) is not consistent. However, both studies clearly show that KM is
the lowest (i.e., affinity is highest) in humans. Because clearance at lower concentrations is
determined by the ratio Vmax to Km, the lower apparent Km in humans may partially offset the
lower human Vmax, and lead to similar oxidative clearances in the liver at environmentally
relevant doses. However, differences in activity measured in vitro may not translate into in vivo
differences in metabolite production, as the rate of metabolism in vivo depends also on the rate
of delivery to the tissue via blood flow (Lipscomb et al., 2003). The interaction of enzyme
activity and blood flow is best investigated using PBPK models and is discussed, along with
descriptions of in vivo data, in Section 3.5.
Data on sex- and age-dependence in oxidative TCE metabolism are limited but suggest
relatively modest differences in humans and animals. In an extensive evaluation of
CYP-dependent activities in human liver microsomal protein and cryopreserved hepatocytes,
Parkinson et al. (2004) identified no age or gender-related differences in CYP2E1 activity. In
liver microsomes from 23 humans, the Km values for females was lower than males, but Vmax
values were very similar (Lipscomb et al., 1997). Appearance of total trichloro compounds
(TTC) in urine following intrapertoneal dosing with TCE was 28% higher in female rats than in
males (Verma and Rana, 2003). The oxidation of TCE in male and female rat liver microsomes
was not significantly different; however, pregnancy resulted in a decrease of 27-39% in the rate
This document is a draft for review purposes only and does not constitute Agency policy.
42 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
of CH production in treated microsomes from females (Nakajima et al., 1992b). Formation of
CH in liver microsomes in the presence of 0.2 or 5.9 mM TCE exhibited some dependency on
age of rats, with formation rates in both sexes of 1.1-1.7 nmol/mg protein/minute in 3-week-old
animals and 0.5-1.0 nmol/mg protein/minute in 18-week-old animals (Nakajima et al., 1992b).
Fisher et al. (1991) reviewed data available at that time on urinary metabolites to
characterize species differences in the amount of urinary metabolism accounted for by TCA (see
Table 3-19). They concluded that TCA seemed to represent a higher percentage of urinary
metabolites in primates than in other mammalian species, indicating a greater proportion of
oxidation leading ultimately to TCA relative to TCOG.
3.3.3.1.7. Cytochrome P450 (CYP) isoforms and genetic polymorphisms
A number of studies have identified multiple P450 isozymes as having a role in the oxidative
metabolism of TCE. These isozymes include CYP2E1 (Guengerich et al., 1991; Guengerich and
Shimada, 1991; Nakajima et al., 1988; Nakajima et al., 1992a; Nakajima et al., 1990), CYP3A4
(Shimada et al., 1994), CYP1A1/2, CYP2C11/6 (Nakajima et al., 1992a, 1993), CYP2F, and
CYP2B1 (Forkert et al., 2005). Recent studies in CYP2E1-knockout mice have shown that in
the absence of CYP2E1, mice still have substantial capacity for TCE oxidation (Forkert et al.,
2006; Kim and Ghanayem, 2006). However, CYP2E1 appears to be the predominant (i.e.,
higher affinity) isoform involved in oxidizing TCE (Forkert et al., 2005; Guengerich et al., 1991;
Guengerich and Shimada, 1991; Nakajima et al., 1992a). In rat liver, CYP2E1 catalyzed TCE
oxidation more than CYP2C11/6 (Nakajima et al., 1992a). In rat recombinant-derived P450s,
the CYP2E1 had a lower KM (higher affinity) and higher Vmax/Km ratio (intrinsic clearance)
than CYP2B1 or CYP2F4 (Forkert et al., 2005). Interestingly, there was substantial differences
in Km between rat and human CYP2Els and between rat CYP2F4 and mouse CYP2F2,
suggesting that species-specific isoforms have different kinetic behavior (see Table 3-20).
The presence of multiple P450 isoforms in human populations affects the variability in
individuals' ability to metabolize TCE. Studies using microsomes from human liver or from
human lymphoblastoid cell lines expressing CYP2E1, CYP1A1, CYP1A2, or CYP3A4 have
shown that CYP2E1 is responsible for greater than 60% of oxidative TCE metabolism
(Lipscomb et al., 1997). Similarities between metabolism of chlorzoxazone (a CYP2E1
substrate) in liver microsomes from 28 individuals (Peter et al., 1990) and TCE metabolism
helped identify CYP2E1 as the predominant (high affinity) isoform for TCE oxidation.
Additionally, Lash et al. (2000a) suggested that, at concentrations above the KM value for
CYP2E1, CYP1A2 and CYP2A4 may also metabolize TCE in humans; however, their
This document is a draft for review purposes only and does not constitute Agency policy.
43 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Table 3-19. Urinary excretion of trichloroacetic acid by various species
exposed to trichloroethylene [based on data reviewed in (Fisher et al., 1991)]
Species
Percentage of
urinary
excretion of TCA
Dose route
TCE dose
(mg TCE/kg)
References, comments
Male
Female
Baboon3'0
16
—
Intramuscular
injection
50
Mueller et al. (1982)
Chimpanzee3
24
22
Intramuscular
injection
50
Mueller et al. (1982)
Monkey,
Rhesus3'0
19
—
Intramuscular
injection
50
Mueller et al. (1982)
Mice, NMRIb
—
8-20
Oral
intubation
2-200
Dekant et al. (1986b)
Mice, B6C3Fla
7-12
—
Oral
intubation
10-2,000
Green and Prout (1985)
Rabbit,
Japanese
White3'0
0.5
Intraperitoneal
injection
200
Nomiyama and
Nomiyama (1979)
Rat, Wistarb
—
14-17
Oral
intubation
2-200
Dekant et al. (1986b)
Rat,
Osborne-Me
ndef
6-7
Oral
intubation
10-2,000
Green and Prout (1985)
Rat, Holtzman3
7
—
Intraperitoneal
injection
10 mg TCE/rat
Nomiyama and
Nomiyama (1979)
a Percentage urinary excretion determined from accumulated amounts of TCOH and TCA in urine 3-6 days
postexposure.
b Percentage urinary excretion determined from accumulated amounts of TCOH, dichloroacetic acid, oxalic acid,
and iV-(hydroxyacetyl)aminoethanol in urine 3 days postexposure.
0 Sex is not specified.
Note: The human data tabulated in Fisher et al. (1991) from Nomiyama and Nomiyama (1971) were not included
here because they were relative to urinary excretion of total trichloro-compounds—not as fraction of intake as was
the case for the other data included here.
This document is a draft for review purposes only and does not constitute Agency policy.
44 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Table 3-20. P450 isoform kinetics for metabolism of TCE to CH in human,
rat, and mouse recombinant P450s
Experiment
Km
jiM
Vmax
pmol/min/pmol P450
Vmax/Km
Human rCYP2El
196 ±40
4 ±0.2
0.02
Rat rCYP2El
14 ±3
11 ±0.3
0.79
Rat rCYP2B 1
131 ±36
9 ±0.5
0.07
Rat rCYP2F4
64 ±9
17 ± 0.5
0.27
Mouse rCYP2F2
114 ± 17
13 ±0.4
0.11
Source: Forkert et al. (2005)
contribution to the overall TCE metabolism was considered low compared to that of CYP2E1.
Given the difference in expression of known TCE-metabolizing P450 isoforms (see Table 3-21)
and the variability in P450-mediated TCE oxidation (Lipscomb et al., 1997), significant
variability may exist in individual human susceptibility to TCE toxicity.
Differences in content and/or intrinsic catalytic properties (Km, Vmax) of specific enzymes
among species, strains, and individuals may play an important role in the observed differences in
TCE metabolism and resulting toxicities. Lipscomb et al. (1997) reported observing three
statistically distinct groups of Km values for TCE oxidation using human microsomes. The mean
± standard deviation [SD] (|iM TCE) for each of the three groups was 16.7 ± 2.5 (// = 10), 30.9 ±
3.3 (n = 9), and 51.1 ± 3.8 (n = 4). Within each group, there were no significant differences in
sex or ethnicity. However, the overall observed KM values in female microsomes (21.9 ±
3.5 [xM, n = 10) were significantly lower than males (33.1 ±3.5 [jM, n = 13). Interestingly, in
human liver microsomes, different groups of individuals with different affinities for TCE
oxidation appeared to also have different activities for other substrates not only with respect to
CYP2E1 but also CYP1A2 (Lash et al., 2000a) (see Table 3-21). Genetic polymorphisms in
humans have been identified in the CYP isozymes thought to be responsible for TCE metabolism
(Pastino et al., 2000),but no data exist correlating these polymorphisms with enzyme activity. It
is relevant to note that repeat polymorphism (Hu et al., 1999) or polymorphism in the regulatory
sequence (McCarver et al., 1998) were not involved in the constitutive expression of human
CYP2E1; however, it is unknown if these types of polymorphisms may play a role in the
inducibility of the respective gene.
Table 3-21. P450 isoform activities in human liver microsomes exhibiting
different affinities for TCE
This document is a draft for review purposes only and does not constitute Agency policy.
45 DRAFT—DO NOT CITE OR QUOTE
-------
1
Affinity group
CYP isoform activity (pmol/min/mg protein)
CYP2E1
CYP1A2
CYP3A4
Low Km
520 ±295
241± 146
2.7 ±2.7
Mid Km
820 ± 372
545 ±200
2.9 ± 2.8
High Km
1,317 ±592
806 ± 442
1.8 ± 1.1
2
3 Activities of CYP1A2, CYP2E1, and CYP3 A4 were measured with phenacetin, chlorzoxazone, and testosterone as
4 substrates, respectively. Data are means ± standard deviation from 10, 9, and four samples for the low-, mid-, and
5 high-KM groups, respectively. Only CYP3 A4 activities are not significantly different (p < 0.05) from one another
6 by Kruskal-Wallis one-way analysis of variance.
7 Source: Lash et al. (2000a).
8
9
10 Individual susceptibilities to TCE toxicity may also result from variations in enzyme
11 content, either at baseline or due to enzyme induction/inhibition, which can lead to alterations in
12 the amounts of metabolites formed. Certain physiological and pathological conditions or
13 exposure to other chemicals (e.g., ethanol and acetominophen) can induce, inhibit, or compete
14 for enzymatic activity. Given the well established (or characterized) role of the liver to
15 oxidatively metabolize TCE (by CYP2E1), increasing the CYP2E1 content or activity (e.g., by
16 enzyme induction) may not result in further increases in TCE oxidation. Indeed, Kaneko et al.
17 (1994a) reported that enzyme induction by ethanol consumption in humans increased TCE
"3
18 metabolism only at high concentrations (500 ppm, 2,687 mg/m ) in inspired air. However, other
19 interactions between ethanol and the enzymes that oxidatively metabolize TCE metabolites can
20 result in altered metabolic fate of TCE metabolites. In addition, enzyme inhibition or
21 competition can decrease TCE oxidation and subsequently alter the TCE toxic response via, for
22 instance, increasing the proportion undergoing GSH conjugation Lash et al. (2000a). TCE itself
23 is a competitive inhibitor of CYP2E1 activity (Lipscomb et al., 1997), as shown by reduced
24 /;-nitrophenol hydroxylase activity in human liver microsomes, and so may alter the toxicity of
25 other chemicals metabolized through that pathway. On the other hand, suicidal CYP heme
26 destruction by the TCE-oxygenated CYP intermediate has also been shown (Miller and
27 Guengerich, 1983).
28
3.3.3.1.8. Glutathione (GSH) Conjugation Pathway
29 Historically, the conjugative metabolic pathways have been associated with xenobiotic
30 detoxification. This is true for GSH conjugation of many compounds. However, several
This document is a draft for review purposes only and does not constitute Agency policy.
46 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
halogenated alkanes and alkenes, including TCE, are bioactivated to cytotoxic metabolites by the
GSH conjugate processing pathway (mercapturic acid) pathways (Elfarra et al., 1987; Elfarra et
al., 1986). In the case of TCE, production of reactive species several steps downstream from the
initial GSH conjugation is believed to cause cytotoxicity and carcinogenicity, particularly in the
kidney. Since the GSH conjugation pathway is in competition with the P450 oxidative pathway
for TCE biotransformation, it is important to understand the role of various factors in
determining the flux of TCE through each pathway. Figure 3-5 depicts the present
understanding of TCE metabolism via GSH conjugation.
3.3.3.1.9. Formation of S-(l,2-dichlorovinyl)glutathione or S-(2,2-dichlorovinyl)glutathione
(DCVG)
The conjugation of TCE to GSH produces S-(l,2-dichlorovinyl)glutathione or its isomer
S-(2,2-dichlorovinyl)glutathione (DCVG). There is some uncertainty as to which GST isoforms
mediate TCE conjugation. Lash and colleagues studied TCE conjugation in renal tissue
preparations, isolated renal tubule cells from male F344 rats and purified GST alpha-class
isoforms 1-1, 1-2 and 2-2 (Cummings et al., 2000a; Cummings and Lash, 2000; Lash et al.,
2000b). The results demonstrated high conjugative activity in renal cortex and in proximal
tubule cells. Although the isoforms studied had similar Vmax values, the Km value for GST 2-2
was significantly lower than the other forms, indicating that this form will catalyze TCE
conjugation at lower (more physiologically relevant) substrate concentrations. In contrast, using
purified rat and human enzymes, Hissink et al. (2002) reported in vitro activity for DCVG
formation only for mu- and pi-class GST isoforms, and none towards alpha-class isoforms;
however, the rat mu-class GST 3-3 was several folds more active than the human mu-class
GST Ml-1. Although GSTs are present in tissues throughout the body, the majority of TCE
GSH conjugation is thought to occur in the liver (Lash et al., 2000a). Using in vitro
studies with renal preparations, it has been demonstrated that GST catalyzed conjugation of TCE
is increased following the inhibition of CYP-mediated oxidation (Cummings and Lash, 2000).
In F344 rats, following gavage doses of 263-1,971 mg/kg TCE in 2 mL corn oil, DCVG
was observed in the liver and kidney of females only, in blood of both sexes (Lash et al., 2006),
and in bile of males (Dekant, 1990). The data from Lash et al. (2006) are difficult to interpret
because the time courses seem extremely erratic, even for the oxidative metabolites TCOH and
TCA. Moreover, a comparison of blood levels of TCA and TCOH with other studies
This document is a draft for review purposes only and does not constitute Agency policy.
47 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
CI
CI
>=<
H (TCE) CI
CI
H
>=<
GST
SG
CI
GGT
Cl2 C2 H
FMO-3
CL C, H
(DCVCS)
CGDP
(DCVC)
B-lyase
(DCVT)
NAT
Acylase ^'2 ^2 ^
(NAcDCVC)
Cl2 C2 H
CYP3A
(NAcDCVCS)
Figure 3-5. Scheme for GSH-dependent metabolism of TCE.
Adapted from: Lash et al. (2000a); Cummings and Lash (2000); NRC (2006).
in rats at similar doses reveals differences of over 1,000-fold in reported concentrations. For
instance, at the lowest dose of 263 mg/kg, the peak blood levels of TCE and TCA in male F344
rats were 10.5 and 1.6 |ig/L, respectively (Lash et al., 2006). By contrast, Larson and Bull
(1992a) reported peak blood TCE and TCA levels in male Sprague-Dawley rats over 1,000-fold
higher—around 10 and 13 mg/L, respectively—following oral doses of 197 mg/kg as a
suspension in 1% aqueous Tween 80®. The results of Larson and Bull (1992a) are similar to
This document is a draft for review purposes only and does not constitute Agency policy.
48 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Lee et al. (2000b), who reported peak blood TCE levels of 20-50 mg/L after male
Sprague-Dawley rats received oral doses of 144-432 mg/kg in a 5% aqueous Alkamus emulsion
(polyethoxylated vegetable oil), and to Stenner et al. (1997), who reported peak blood levels of
TCA in male F344 rats of about 5 mg/L at a slightly lower TCE oral dose of 100 mg/kg
administered to fasted animals in 2% Tween 80®. Thus, while useful qualitatively as an
indicator of the presence of DCVG in rats, the quantitative reliability of reported concentrations,
for metabolites of either oxidation or GSH conjugation, may be questionable.
In humans, DCVG was readily detected at in human blood following onset of a 4-hour
TCE inhalation exposure to 50 or 100 ppm (269 or 537 mg/m3) (Lash et al., 1999a). At 50 ppm,
peak blood levels ranged from 2.5-30 [xM, while at 100 ppm, the mean (± SE, n = 8) peak blood
levels were 46.1 ± 14.2 [xM in males and 13.4 ± 6.6 [xM in females. Although on average, male
subjects had threefold higher peak blood levels of DCVG than females, DCVG blood levels in
half of the male subjects were similar to or lower than those of female subjects. This suggests a
polymorphism in GSH conjugation of TCE rather than a true gender difference (Lash et al.,
1999a) as also has been indicated by Hissink et al. (2002) for the human mu-class GST Ml-1
enzyme. Interestingly, as shown in Table 3-22, the peak blood levels of DCVG are similar on a
molar basis to peak levels of TCE, TCA, and TCOH in the same subjects, as reported in
Fisher et al. (1998).
Tables 3-23-3-25 summarize DCVG formation from TCE conjugation from in vitro
studies of liver and kidney cellular and subcellular fractions in mouse, rat, and human
(tissue-distribution and species-and gender-differences in DCVG formation are discussed
below). As shown by these tables, different investigators have reported considerably different
rates for TCE conjugation in human liver and kidney cell fractions. For instance, values in
Table 3-23 from Lash et al. (1999a) are between two and five orders of magnitude higher than
those reported by Green et al. (1997a) or Dekant et al. (1990) (see Table 3-25). In addition,
Green et al. (1997a) and Dekant et al. (1990) reported a difference in the relative importance of
rat liver cytosol and rat liver microsomes for GSH conjugation, with Green et al. (1997a)
reporting activity in the cytosol and none in the microsomes and Dekant et al. (1990) reporting
the opposite.
The reasons for such discrepancies are unclear, but they may be related to different
analytical methods (Lash et al., 2000a). In particular, Lash et al. (1999a) employed the "Reed
method," which used ion-exchange high-performance liquid chromatography (HPLC) of
derivatized analytes. This HPLC method is characterized by variability and an overall decline in
retention times over the life of the HPLC column due to derivatization of amine groups on the
This document is a draft for review purposes only and does not constitute Agency policy.
49 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-22. Comparison of peak blood concentrations in humans exposed to
2 100 ppm (537 mg/m3) TCE for 4 hours (Fisher et al., 1998; Lash et al., 1999a)
3
Chemical species
Peak blood concentration (mean ± SD, jiM)
Males
Females
TCE
23 ± 11
14 ±4.7
TCA
56 ±9.8
59 ± 12
TCOH
21 ±5.0
15 ± 5.6
DCVG
46.1 ± 14.2
13.4 ± 6.6
4
5
6
This document is a draft for review purposes only and does not constitute Agency policy.
50 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-23. GSH conjugation of TCE (at 1-2 mM) in liver and kidney
2 cellular fractions in humans, male F344 rats, and male B6C3F1 mice from
3 Lash laboratory
4
Species and cellular/subcellular fraction (TCE
concentration)
DCVG formation
(nmol/hour/mg protein or 106 cells)
Male
Female
Human
Hepatocytes (0.9 mM) [pooled]
11 ±3
Liver cytosol (1 mM) [individual samples]
156± 16
174 ± 13
Liver cytosol (2 mM) [pooled]
346
Liver microsomes (1 mM) [individual samples]
108 ±24
83 ± 11
Liver microsomes (1 mM) [pooled]
146
Kidney cytosol (2 mM) [pooled]
42
Kidney microsomes (1 mM) [pooled]
320
Rat
Liver cytosol (2 mM)
7.30 ± 2.8
4.86 ±0.14
Liver microsomes (2 mM)
10.3 ±2.8
7.24 ± 0.24
Kidney cortical cells (2 mM)
0.48 ±0.02
0.65 ±0.15
Kidney cytosol (2 mM)
0.45 ±0.22
0.32 ±0.02
Kidney microsomes (2 mM)
not detected
0.61 ±0.06
Mouse
Liver cytosol (2 mM)
24.5 ±2.4
21.7 ± 0.9
Liver microsomes (2 mM)
40.0 ±3.1
25.6 ±0.8
Kidney cytosol (2 mM)
5.6 ±0.24
3.7 ±0.48
Kidney microsomes (2 mM)
5.47 ± 1.41
16.7 ± 4.7
5
6 Mean ± SE. Source: Lash et al. (1999b; 1998a; 1995); Cummings and Lash (2000); Cummings et al. (2000a).
7
8
This document is a draft for review purposes only and does not constitute Agency policy.
51 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-24. Kinetics of TCE metabolism via GSH conjugation in male F344
2 rat kidney and human liver and kidney cellular and subcellular fractions
3 from Lash laboratory
4
Tissue and cellular fraction
Km
(jiM TCE)
Vmax
(nmol
DCVG/min/mg
protein or 10
hepatocytes)
1,000 X
Vmax/Km
Rat
Kidney proximal tubular cells: low affinity
2,910
0.65
0.22
Kidney proximal tubular cells: high affinity
460
0.47
1.0
Human
Liver hepatocytesa
37-106
0.16-0.26
2.4-4.5
Liver cytosol: low affinity
333
8.77
2.6
Liver cytosol: high affinity
22.7
4.27
190
Liver microsomes: low affinity
250
3.1
12
Liver microsomes: high affinity
29.4
1.42
48
Kidney proximal tubular cells: low affinity
29,400
1.35
0.046
Kidney proximal tubular cells: high affinity
580
0.11
0.19
Kidney cytosol
26.3
0.81
31
Kidney microsomes
167
6.29
38
5
6 a Kinetic analyses of first six-nine (out of 10) data points from Figure 1 from Lash et al. (1999a) using
7 Lineweaver-Burk or Eadie-Hofstee plots and linear regression (R2 = 0.50-0.95). Regression with best R2 used
8 first 6 data points and Eadie-Hofstee plot, with resulting KM and Vmax of 106 and 0.26, respectively.
9
10 Source: Lash et al. (1999a); Cummings and Lash (2000); Cummings et al. (2000a).
11
This document is a draft for review purposes only and does not constitute Agency policy.
52 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-25. GSH conjugation of TCE (at 1.4-4 mM) in liver and kidney
2 cellular fractions in humans, male F344 rats, and male B6C3F1 mice from
3 Green and Dekant laboratories
4
Species and cellular/subcellular fraction (TCE
concentration)
DCVG formation
(nmol/hour/mg protein) [substrate
concentration in mM]
Dekant et al.
(1990)
Green et al.
(1997a)
Human
Liver cytosol
-
0.00019 ±0.00014
Liver microsomes
-
n.d.
Kidney cytosol
-
n.d.
Kidney microsomes
-
n.d.
Rat
Liver cytosol
<0.002
0.00162 ±0.00002
Liver microsomes
0.002
n.d.
Kidney cytosol
-
n.d.
Kidney microsomes
-
n.d.
Mouse
Liver cytosol
-
0.0025
Liver microsomes
-
n.d.
Kidney cytosol
-
n.d.
Kidney microsomes
-
n.d.
5
6 n.d. = not determined
7 where available, mean ± SD. Source: Dekant etal. (1990), Green etal. (1997a)
8
9
This document is a draft for review purposes only and does not constitute Agency policy.
53 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
column (Lash et al., 1999b). Although data are limited, the GSH pathway metabolite levels
reported by methods that utilize 14C TCE and radiochemical detection followed by mass
spectrometry identification of the metabolites are lower. In particular, Green et al. (1997a) and
Dekant et al. (1990) both used HPLC with radiochemical detection. Peak identity was confirmed
by Green et al. (1997a) using liquid chromatography/mass spectrometry (LC/MS) and by
GC/MS following hydrolysis by Dekant et al. (1990). In addition, studies using HPLC-MS/MS
techniques with stable isotope-labeled DCVG and dichlorovinyl cysteine (DCVC) standards
have also been used to detect GSH pathway metabolite levels Kim et al. (2009). Based on the in
vitro work presented in Table 3-23 using the "Reed method," one would expect mouse serum
DCVG levels to be -4-6 times lower than humans. However, using the HPLC-MS/MS
technique of Kim et al. (2009), the peak DCVG serum levels are -1,000 times lower in mouse
serum than determined by Lash et al. (1999a) in human serum. Although advances in LC
technology, and differences in exposure routes (inhalation versus oral, with different first pass),
exposure doses, and the degree of competition with TCE oxidation (greater in mouse than in
human) should be considered, this much-larger-than-expected difference is consistent with the
suggestion that the "Reed method" provides an overestimation of DCVG levels in humans. This
could occur if the "Reed method" identifies nonspecific derivatives as DCVG or other GSH
pathway metabolites. However, the degree of overestimation is unclear, and differing results in
humans may be attributable to true interindividual variation (especially since GSTs are known to
be polymorphic). However, overall, there remains significant uncertainty in the quantitative
estimation of DCVG formation from TCE both in vivo and in vitro.
3.3.3.1.10. Formation of S-(l,2-dichlorovinyl) cysteine or S-(2,2-dichlorovinyl) cysteine
(DCVC)
The cysteine conjugate, isomers S-(l,2-dichlorovinyl) cysteine (1,2-DCVC) or
S-(2,2-dichlorovinyl) cysteine (2,2-DCVC), is formed from DCVG in a two-step sequence.
DCVG is first converted to the cysteinylglycine conjugate
S-(l,2-dichlorovinyl)-L-cysteinylglycine or its isomer S-(2,2-dichlorovinyl)-L-cysteinylglycine
by y-glutamyl transpeptidase (GGT) in the renal brush border (Elfarra and Anders, 1984; Lash et
al., 1988).
Cysteinylglycine dipeptidases in the renal brush border and basolateral membrane
convert DCVG to DCVC via glycine cleavage (Goeptar et al., 1995; Lash et al., 1995). This
reaction can also occur in the bile or gut, as DCVG excreted into the bile is converted to DCVC
and reabsorbed into the liver where it may undergo further acetylation.
This document is a draft for review purposes only and does not constitute Agency policy.
54 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
3.3.3.1.11. Formation of N-Acetyl-S-(l,2-dichlorovinyl)-L-cysteine or N-Acetyl-S-(2,2-
dichlorovinyl)-L-cysteine (NAcDCVC)
N-acetylation of DCVC can either occur in the kidney, as demonstrated in rat kidney
microsomes (Duffel and Jakoby, 1982), or in the liver (Birner et al., 1997). Subsequent release
of DCVC from the liver to blood may result in distribution to the kidney resulting in increased
internal kidney exposure to the acetylated metabolite over and above what the kidney already is
capable of generating. In the kidney, N-Acetyl-S-(l,2-dichlorovinyl)-L-cysteine or N-Acetyl-S-
(2,2-dichlorovinyl)-L-cysteine (NAcDCVC) may undergo deacetylation, which is considered a
rate-limiting-step in the production of proximal tubule damage (Wolfgang et al., 1989b; Zhang
and Stevens, 1989). As a polar mercapturate, NAcDCVC may be excreted in the urine as
evidenced by findings in mice (Birner et al., 1993), rats (Bernauer et al., 1996; Commandeur and
Vermeulen, 1990), and humans who were exposed to TCE (Bernauer et al., 1996; Birner et al.,
1993), suggesting a common glutathione-mediated metabolic pathway for DCVC among species.
3.3.3.1.12. Beta lyase metabolism of S-(l,2-dichlorovinyl) cysteine (DCVC)
The enzyme cysteine conjugate P-lyase catalyzes the breakdown of DCVC to reactive
nephrotoxic metabolites (Goeptar et al., 1995). This reaction involves removal of pyruvate and
ammonia and production of S-(l,2-dichlorovinyl) thiol (DCVT), an unstable intermediate, which
rearranges to other reactive alkylation metabolites that form covalent bonds with cellular
nucleophiles (Dekant et al., 1988; Goeptar et al., 1995). The rearrangement of DCVT to
enethiols and their acetylating agents has been described in trapping experiments (Dekant et al.,
1988) and proposed to be responsible for nucleophilic adduction and toxicity in the kidney. The
quantification of acid-labile adducts was proposed as a metric for TCE flux through the GSH
pathway. However, the presence of analytical artifacts precluded such analysis. In fact,
measurement of acid-labile adduct products resulted in higher values in mice than in rats (Eyre et
al., 1995a, b).
DCVC metabolism to reactive species via a P-lyase pathway has been observed in vitro
by Green et al. (1997a), who reported greater P-lyase activity in rats than in mice or humans.
However, in vitro DCVC metabolism by the competing enzyme A'-acetyl transferase was also
reported to be greater in rats than mice and humans. In vivo, P-lyase activity in humans and rats
(reaction rates were not reported) was demonstrated using a surrogate substrate,
2-(fluoromethoxy)-l,l,3,3,3-pentafluoro-l-propene (Iyer et al., 1998). P-lyase -mediated
This document is a draft for review purposes only and does not constitute Agency policy.
5 5 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
reactive adducts have been described in several extrarenal tissues, including rat and human liver
and intestinal microflora (Dohn and Anders, 1982; Larsen and Stevens, 1986; Stevens, 1985;
Stevens and Jakoby, 1983; Tateishi et al., 1978; Tomisawa et al., 1986; Tomisawa et al.,
1984)and rat brain (Alberati-Giani et al., 1995; Malherbe et al., 1995).
In the kidneys, glutamine transaminase K appears to be primarily responsible for P-lyase
metabolism of DCVC (Jones et al., 1988; Lash et al., 1986; Lash et al., 1990b; Perry et al., 1993;
Stevens et al., 1988; Stevens et al., 1986). P-lyase transformation of DCVC appears to be
regulated by 2-keto acids. DCVC toxicity in isolated rat proximal tubular cells was significantly
increased with the addition of a-keto-y-methiolbutyrate or phenylpyruvate (Elfarra et al., 1986).
The presence of a-keto acid cofactors is necessary to convert the inactive form of the P-lyase
enzyme (containing pyridoxamine phosphate) to the active form (containing pyridoxal
phosphate) (Goeptar et al., 1995).
Both low- and high-molecular-weight enzymes with P-lyase activities have been
identified in rat kidney cytosol and mitochondria (Abraham et al., 1995a; Abraham et al., 1995b;
Lash et al., 1986; Stevens et al., 1988). While glutamine transaminase K and
kynureninase-associated P-lyase activities have been identified in rat liver (Alberati-Giani et al.,
1995; Stevens, 1985), they are quite low compared to renal glutamine transaminase K activity
and do not result in hepatotoxicity in DCVG- or DCVC-treated rats (Elfarra and Anders, 1984).
Similar isoforms of P-lyase have also been reported in mitochondrial fractions of brain tissue
(Cooper, 2004).
The kidney enzyme L-a-hydroxy (L-amino) acid oxidase is capable of forming an
iminium intermediate and keto acid analogues (pyruvate or
S-(l,2-dichlorovinyl)-2-oxo-3-mercaptopropionate) of DCVC, which decomposes to
dichlorovinylthiol (Lash et al., 1990a; Stevens et al., 1989). In rat kidney homogenates, this
enzyme activity resulted in as much as 35% of GSH pathway-mediated bioactivation. However,
this enzyme is not present in humans, an important consideration for extrapolation of renal
effects across species.
3.3.3.1.13. Sulfoxidation of S-(l,2-dichlorovinyl) cysteine (DCVC) and N-Acetyl-S-
(l,2-dichlorovinyl)-L-cysteine (NAcDCVC)
A second pathway for bioactivation of TCE S-conjugates involves sulfoxidation of either
the cysteine or mercapturic acid conjugates (Lash et al., 2003; Park et al., 1992; Sausen and
Elfarra, 1990) (Birner et al., 1998; Krause et al., 2003; Lash et al., 1994; Werner et al., 1995a,
1996; Werner et al., 1995b). Sulfoxidation of DCVC was mediated mainly by flavin
This document is a draft for review purposes only and does not constitute Agency policy.
56 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
monooxygenase 3 (FM03), rather than CYP, in rabbit liver microsomes (Ripp et al., 1997) and
human liver microsomes (Krause et al., 2003). Krause et al., (2003) also reported DCVC
sulfoxidation by human cDNA-expressed FM03, as well as detection of FM03 protein in human
kidney samples. While Krause et al. (2003) were not able to detect sulfoxidation in human
kidney microsomes, the authors noted FM03 expression in the kidney was lower and more
variable than that in the liver. However, sulfoxidation products in tissues or urine have not been
reported in vivo.
Sulfoxidation of NAcDCVC, by contrast, was found to be catalyzed predominantly, if not
exclusively, by CYP3A enzymes (Werner et al., 1996), whose expressions are highly
polymorphic in humans. Sulfoxidation of other haloalkyl mercapturic acid conjugates has also
been shown to be catalyzed by CYP3A (Altuntas et al., 2004; Werner et al., 1995a; Werner et al.,
1995b). While Lash et al. (2000a) suggested that this pathway would be quantitatively minor
because of the relatively low CYP3 A levels in the kidney, no direct data exist to establish the
relative toxicological importance of this pathway relative to bioactivation of DCVC by P-lyase or
FM03. However, the contribution of CYP3A in S-conjugate sulfoxidation to nephrotoxicity in
vivo was recently demonstrated by Sheffels et al. (2004) with
fluoromethyl-2,2-difluoro-l-(trifluoromethyl)vinyl ether (FDVE). In particular, in vivo
production and urinary excretion of FDVE-mercapturic acid sulfoxide metabolites were
unambiguously established by mass spectrometry, and CYP inducers/inhibitors
increased/decreased nephrotoxicity in vivo while having no effect on urinary excretion of
metabolites produced through P-lyase (Sheffels et al., 2004). These data suggest that, by
analogy, sulfoxidation of NAcDCVC may be an important bioactivating pathway.
3.3.3.1.14. Tissue distribution of glutathione (GSH) metabolism
The sites of enzymatic metabolism of TCE to the various GSH pathway-mediated
metabolites are significant in determining target tissue toxicity along this pathway. Figure 3-6
presents a schematic of interorgan transport and metabolism of TCE along the glutathione
pathway. TCE is taken up either by the liver or kidney and conjugated to DCVG. The primary
factors affecting TCE flux via this pathway include high hepatic GST activity, efficient transport
of DCVG from the liver to the plasma or bile, high renal brush border and low hepatic GGT
activities, and the capability for GSH conjugate uptake into the renal basolateral membranes with
limited or no uptake into liver cell plasma membranes.
This document is a draft for review purposes only and does not constitute Agency policy.
57 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
TCE
DCVG
DCVC
NAcDCVC
¦¦~Blood flow
-~Bile flow
-~Glomerular filtration
-Metabolism
Blood/
Blood/
Blood/
Blood/
Plasma
Plasma
Plasma
Plasma
Rest of
Rest of
Rest of
Rest of
Small
Intestine
Small
Intestine
Small
Intestine
Small
Intestine
Kidney
Kidney
Kidney
Kidney
DCVT DCVCS NAcDCVCS
As discussed previously, GST activity is present in many different cell types. However,
the liver is the major tissue for GSH conjugation. GST activities in rat and mouse cytosolic
fractions were measured using l-chloro-2,4-dinitrobenzene, a GST substrate that is nonspecific
Figure 3-6. Interorgan TCE transport and metabolism via the GSH pathway. See
Figure 3-5 for enzymes involved in metabolic steps. Source: Lash et al.(2000a; 2000b);
NRC (2006).
for particular isoforms (Lash et al., 1998b). Specific activities (normalized for protein content)
in whole kidney cytosol were slightly less than those in the liver (0.64 compared to 0.52 mU/mg
protein for males and females). However, the much larger mass of the liver compared to the
kidney indicates that far more total GST activity resides in the liver. This is consistent with in
vitro data on TCE conjugation to DCVG, discussed previously (see Tables 3-23 and 3-24). For
This document is a draft for review purposes only and does not constitute Agency policy.
58 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
instance, in humans, rats, and mice, liver cytosol exhibits greater DCVG production than kidney
cytosol. Distinct high- and low-affinity metabolic profiles were observed in the liver but not in
the kidney (see Table 3-24). In microsomes, human liver and kidney had similar rates of DCVG
production, while for rats and mice, the production in the liver was substantially greater.
According to studies by Lash et al.(1998a; 1998b), the activity of GGT, the first step in the
conversion of DCVG to DCVC, is much higher in the kidney than the liver of mice, rats, and
humans, with most of the activity being concentrated in the microsomal, rather than the
cytosolic, fraction of the cell (see Table 3-26). In rats, this activity is quite high in the kidney but
is below the level of detection in the liver while the relative kidney to liver levels in humans and
mice were higher by 18- and up to 2,300-fold, respectively. Similar qualitative findings were
also reported in another study (Hinchman and Ballatori, 1990) when total organ GGT levels were
compared in several species (see Table 3-27). Cysteinylglycine dipeptidase was also
preferentially higher in the kidney than the liver of all tested species although the interorgan
differences in this activity (one-ninefold) seemed to be less dramatic than for GGT (see
Table 3-27). High levels of both GGT and dipeptidases have also been reported in the small
intestine of rat (Kozak and Tate, 1982) and mouse (Habib et al., 1996), as well as GGT in the
human jejunum (Fairman et al., 1977). No specific human intestinal cysteinylglycine
dipeptidase has been identified; however, a related enzyme (EC 3.4.13.11) from human kidney
microsomes has been purified and studied (Adachi et al., 1989) while several human intestinal
dipeptidases have been characterized including a membrane dipeptidase (EC 3.4.13.19) which
has a wide dipeptide substrate specificity including cysteinylglycine (Hooper et al., 1994; Ristoff
and Larsson, 2007).
3.3.3.1.15. Sex- and species-dependent differences in glutathione (GSH) metabolism
Diverse sex and species differences appear to exist in TCE metabolism via the
glutathione pathway. In rodents, rates of TCE conjugation to GSH in male rats and mice are
higher than females (see Table 3-23). Verma and Rana (2003) reported twofold higher GST
activity values in liver cytosol of female rats, compared to males, given 15 intraperitoneal
injections of TCE over 30 days period. This effect may be due to sex-dependent variation in
induction, as GST activities in male and female controls were similar. DCVG formation rates by
liver and kidney subcellular fractions were much higher in both sexes of mice than in rats and,
except for mouse kidney microsomes, the rates were generally higher in males than in females of
the same species (see Table 3-23).
This document is a draft for review purposes only and does not constitute Agency policy.
59 DRAFT—DO NOT CITE OR QUOTE
-------
1 In terms of species differences, comparisons at 1-2 mM TCE concentrations (see
2 Table 3-23) suggest that, in liver and kidney cytosol, the greatest DCVG production rate was in
3 humans, followed by mice and then rats. However, different investigators have reported
4 considerably different rates for TCE conjugation in human liver and kidney cell fractions. For
5 instance,
This document is a draft for review purposes only and does not constitute Agency policy.
60 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-26. GGT activity in liver and kidney subcellular fractions of mice,
2 rats, and humans
3
Species
Sex
Tissue
Cellular fraction
Activity (mU/mg)
Mouse
Male
Liver
Cytosol
0.07 ±0.04
Microsomes
0.05 ±0.04
Kidney
Cytosol
1.63 ±0.85
Microsomes
92.6 ± 15.6
Female
Liver
Cytosol
0.10±0.10
Microsomes
0.03 ±0.03
Kidney
Cytosol
0.79 ±0.79
Microsomes
69.3 ± 14.0
Rat
Male
Liver
Cytosol
<0.02
Microsomes
<0.02
Kidney
Cytosol
<0.02
Microsomes
1,570 ± 100
Female
Liver
Cytosol
<0.02
Microsomes
<0.02
Kidney
Cytosol
<0.02
Microsomes
1,840 ±40
Human
Male
Liver
Cytosol
8.89 ±3.58
Microsomes
29
Kidney
Cytosol
13.2 ±1.0
Microsomes
960 ± 77
4
5 Source: Lashetal. (1999b; 1998a)
6
7
8
9
This document is a draft for review purposes only and does not constitute Agency policy.
61 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Table 3-27. Multispecies comparison of whole-organ activity levels of GGT
and dipeptidase
Species
Whole organ enzyme activity (^imol substrate/organ)
Kidney
Liver
GGT
Dipeptidase
GGT
Dipeptidase
Rat
1,010 ± 41
20.2 ± 1.1
7.1 ± 1.4
6.1 ±0.4
Mouse
60.0 ±4.2
3.0 ± 0.3
0.47 ±0.05
1.7 ± 0.2
Rabbit
1,119± 186
112 ± 17
71.0 ±9.1
12.6 ± 1.0
Guinea pig
148 ± 13
77 ± 10
46.5 ±4.2
13.2 ±1.5
Pig
3,800 ±769
2,428 ± 203
1,600 ±255
2,178 ±490
Macaque
988
136
181
71
Source: Hinchman and Ballatori (1990).
values in Table 3-23 from Lash et al. (1999a) are between two and five orders of magnitude
higher than those reported by Green et al. (1997a). (The rates of DCVG formation by liver
cytosol from male F344 rat, male B6C3F1 mouse, and human were 1.62, 2.5, and
0.19 pmol/minute/mg protein, respectively, while there was no measurable activity in liver
microsomes or subcellular kidney fractions (Green et al., 1997a). The reasons for such
discrepancies are unclear but may be related to different analytical methods employed such as
detection of radiolabled substrate vs. derivatized analytes (Lash et al., 2000a).
Expression of GGT activity does not appear to be influenced by sex (see Table 3-26); but
species differences in kidney GGT activity are notable with rat subcellular fractions exhibiting
the highest levels and mice and humans exhibiting about 4-6% and 50%, respectively, of rat
levels (Lash et al., 1999b; Lash et al., 1998a). Table 3-27 shows measures of whole-organ GGT
and dipeptidase activities in rats, mice, guinea pigs, rabbits, pigs, and monkeys. These data show
that the whole kidney possesses higher activities than liver for these enzymes, despite the
relatively larger mass of the liver.
As discussed above, the three potential bioactivating pathways subsequent to the
formation of DCVC are catalyzed by P-lyase, FM03 or CYP3A. Lash et al. (2000a) compared
in vitro P-lyase activities and kinetic constants (when available) for kidney of rats, mice, and
humans. They reported that variability of these values spans up to two orders of magnitude
depending on substrate, analytical method used, and research group. Measurements of rat,
This document is a draft for review purposes only and does not constitute Agency policy.
62 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
mouse, and human P-lyase activities collected by the same researchers following
tetrachloroethylene exposure (Green et al., 1990) resulted in higher KM and lower Vmax values
for mice and humans than rats. Further, female rats exhibited higher KM and lower Vmax values
than males.
With respect to FM03, Ripp et al. (1999) found that this enzyme appeared catalytically
similar across multiple species, including humans, rats, dogs, and rabbits, with respect to several
substrates, including DCVC, but that there were species differences in expression. Specifically,
in male liver microsomes, rabbits had threefold higher methionine S-oxidase activity than mice
and dogs had 1.5-fold higher activity than humans and rats. Species differences were also noted
in male and female kidney microsomes; rats exhibited two- to sixfold higher methionine
S-oxidase activity than the other species. Krause et al. (2003) detected DCVC sulfoxidation in
incubations with human liver microsomes but did not in an incubation with a single sample of
human kidney microsomes. However, FM03 expression in the 26 human kidney samples was
found to be highly variable, with a range of five-sixfold (Krause et al., 2003).
No data on species differences in CYP3A-mediated sulfoxidation of NAcDCVC are
available. However, Altuntas et al. (2004) examined sulfoxidation of cysteine and mercapturic
acid conjugates of FDVE (fluoromethyl-2,2-difluoro-l-(trifluoromethyl)vinyl ether) in rat and
human liver and kidney microsomes. They reported that the formation of sulfoxides from the
mercapturates A-Ac-FFVC and (Z)-A'-Ac-FFVC (FFVC is
(I\Z)-S-(\ -fluoro-2-fluoromethoxy-2-(trifluoromethyl)vinyl-Lcysteine) were greatest in rat liver
microsomes, and 2- to 30-fold higher than in human liver microsomes (which had high
variability). Sulfoxidation of A-Ac-FFVC could not be detected in either rat or human kidney
microsomes, but sulfoxidation of (Z)-A-Ac-FFVC was detected in both rat and human kidney
microsomes at rates comparable to human liver microsomes. Using human- and rat-expressed
CYP3A, Altuntas et al. (2004) reported that rates of sulfoxidation of (Z)-A'-Ac-FFVC were
comparable in human CYP3 A4 and rat CYP3 Al and CYP3 A2, but that only rat CYP3 Al and
A2 catalyzed sulfoxidation of A-Ac-FFVC. As the presence or absence of the species
differences in mercapturate sulfoxidation appears to be highly chemical-specific, no clear
inferences can be made as to whether species differences exist for sulfoxidation of NAcDCVC
Also relevant to assess the flux through the various pathways are the rates of
A'-acetylation and de-acetylation of DCVC. This is demonstrated by the results of Elfarra and
Hwang (1990) using S-(2-benzothiazolyl)-L-cysteine as a marker for P-lyase metabolism in rats,
mice, hamsters, and guinea pigs. Guinea pigs exhibited about twofold greater flux through the
P-lyase pathway, but this was not attributable to higher P-lyase activity. Rather, guinea pigs
have relatively low iV-acetylation and high deacetylation activities, leading to a high level of
This document is a draft for review purposes only and does not constitute Agency policy.
63 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
substrate recirculation (Lau et al., 1995). Thus, a high A'-deacetylase: A-acetyl ase activity ratio
may favor DCVC recirculation and subsequent metabolism to reactive species. In human,
Wistar rat, Fischer rat, and mouse cytosol, deacetylation rates for NAcDCVC varied less than
threefold (0.35, 0.41, 0.61, and 0.94 nmol DCVC formed/minute/mg protein in humans, rats, and
mice) (Birner et al., 1993). However, similar experiments have not been carried out for
A'-acetylation of DCVC, so the balance between its A-acetyl ati on and de-acetylation has not been
established.
3.3.3.1.16. Human variability and susceptibility in glutathione (GSH) conjugation
Knowledge of human variability in metabolizing TCE through the glutathione pathway is
limited to in vitro comparisons of variance in GST activity rates. Unlike CYP-mediated
oxidation, quantitative differences in the polymorphic distribution or activity levels of GST
isoforms in humans are nxot presently known. However, the available data (Lash et al., 1999a;
Lash et al., 1999b) do suggest that significant variation in GST-mediated conjugation of TCE
exists in humans. In particular, at a single substrate concentration of 1 mM, the rate of GSH
conjugation of TCE in human liver cytosol from 9 male and 11 females spanned a range of
2.4-fold (34.7-83.6 nmol DCVG formed/20-minute/mg protein) (Lash et al., 1999b). In liver
microsomes from 5 males and 15-females, the variation in activity was 6.5-fold (9.9-64.6 nmol
DCVG formed/20 minute/mg protein). No sex-dependent variation was identified. Despite
being less pronounced than the known variability in human CYP-mediated oxidation, the impact
on risk assessment of the variability in GSH conjugation to TCE is currently unknown especially
in the absence of data on variability for A'-acetylation and bioactivation via P-lyase, FM03, or
CYP3 A in the human kidney.
3.3.3.1.17. Relative Roles of the Cytochrome P450 (CYP) and Glutathione (GSH) Pathways
In vivo mass balance studies in rats and mice, discussed above, have shown
unequivocally that in these species, CYP oxidation of TCE predominates over GSH conjugation.
In these species, at doses from 2-2,000 mg/kg of [14C]TCE, the sum of radioactivity in exhaled
TCE, urine, and exhaled CO2 constitutes 69-94% of the dose, with the vast majority of the
radioactivity in urine (95-99%) attributable to oxidative metabolites (Dekant et al., 1984; Dekant
et al., 1986b; Green and Prout, 1985; Prout et al., 1985). The rest of the radioactivity was found
mostly in feces and the carcass. More rigorous quantitative limits on the amount of GSH
conjugation based on in vivo data such as these can be obtained using PBPK models, discussed
in Section 3.5.
This document is a draft for review purposes only and does not constitute Agency policy.
64 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Comprehensive mass-balance studies are unavailable in humans. DCVG and DCVC in
urine have not been detected in any species, while the amount of urinary NAcDCVC from
human exposures is either below detection limits or very small from a total mass balance point of
view (Bernauer et al., 1996; Birner et al., 1993; Bloemen et al., 2001; Lash et al., 1999b). For
instance, the ratio of primary oxidative metabolites (TCA + TCOH) to NAcDCVC in urine of
"3
rats and humans exposed to 40-160 ppm (215-860 mg/m ) TCE heavily favored oxidation,
resulting in ratios of 986-2,562:1 in rats and 3,292-7,163:1 in humans (Bernauer et al., 1996).
Bloemen et al. (2001) reported that at most 0.05% of an inhaled TCE dose would be excreted as
NAcDCVC, and concluded that this suggested TCE metabolism by GSH conjugation was of
minor importance. While it is a useful biomarker of exposure and an indicator of GSH
conjugation, NAcDCVC may capture only a small fraction of TCE flux through the GSH
conjugation pathway due to the dominance of bioactivating pathways (Lash et al., 2000a).
A number of lines of evidence suggest that the amount of TCE conjugation to GSH in
humans, while likely smaller than the amount of oxidation, may be much more substantial than
analysis of urinary mercapturates would suggest. In Table 3-28, in vitro estimates of the Vmax,
Km, and clearance (Vmax/Km) for hepatic oxidation and conjugation of TCE are compared in a
manner that accounts for differences in cytosolic and microsomal partitioning and protein
content. Surprisingly, the range of in vitro kinetic estimates for oxidation and conjugation of
TCE substantially overlap, suggesting similar flux through each pathway, though with high
interindividual variation. The microsomal and cytosolic protein measurements of GSH
conjugation should be caveated by the observation by Lash et al. (1999a) that GSH conjugation
of TCE was inhibited by -50% in the presence of oxidation. Note that this comparison cannot be
made in rats and mice because in vitro kinetic parameters for GSH conjugation in the liver are
not available in those species (only activity at 1 or 2 mM have been measured).
Furthermore, as shown earlier in Table 3-22, the human in vivo data of Lash et al. (1999a) show
blood concentrations of DCVG similar, on a molar basis, to that of TCE, TCA, or TCOH,
suggesting substantial conjugation of TCE. In addition, these data give a lower limit as to the
amount of TCE conjugated. In particular, by multiplying the peak blood concentration of DCVG
by the blood volume, a minimum amount of DCVG in the body at that time can be derived (i.e.,
assuming the minimal empirical distribution volume equal to the blood volume). As shown in
Table 3-29, this lower limit amounts to about 0.4-3.7%) of the inhaled TCE dose. Since this is
the minimum amount of DCVG in the body at a single time point, the total amount of DCVG
formed is likely to be substantially greater owing to possible distribution outside of the blood as
well as the metabolism and/or excretion of DCVG. Lash et al. (1999b) found levels of urinary
mercapturates were near or below the level of detection of 0.19 |iuM, results that are consistent
This document is a draft for review purposes only and does not constitute Agency policy.
65 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
with those of Bloemen et al. (2001), who reported urinary concentrations below 0.04 |iM at two-
to fourfold lower cumulative exposures. Taken together, these results confirm the suggestion by
Lash et al. (2000a) that NAcDCVC is a poor quantitative marker for the flux through the GSH
pathway.
However, as discussed in Section 3.3.3.2.1, data from other laboratories have reported
substantially lower amounts of GSH conjugation in vitro. The reasons for such discrepancies are
unclear, but they may be related to different analytical methods (Lash et al., 2000a). More recent
in vivo data from Kim et al. (2009) in mice reported -1000 times lower DCVG in mouse serum
as compared to the levels of DCVG reported by Lash et al. (1999a) in human blood. These data
are consistent with the suggestion that the "Reed method" employed by Lash et al. (1999a)
overestimated DCVG levels in humans. However, the degree of overestimation is unclear, as is
the degree to which differences may be attributable to true inter-species or inter-individual
variability.
In summary, TCE oxidation is likely to be greater quantitatively than conjugation with
GSH in mice, rats, and humans. Some evidence suggests that the flux through the GSH pathway,
particularly in humans, may be greater by an order of magnitude or more than the <0.1%
typically excreted of NAcDCVC in urine. This is evidenced both by a direct comparison of in
vitro rates of oxidation and conjugation, as well as by in vivo data on the amount of DCVG in
blood. PBPK models can be used to more quantitatively synthesize these data and put more
rigorous limits on the relative amounts of TCE oxidation and conjugation with GSH. Such
analyses are discussed in Section 3.5. However, these data are not consistent with studies in
other laboratories using different analytical methods, which report two to five orders of
magnitude lower estimates of GSH conjugation. Because the reason for these differences have
not been fully determined, substantial uncertainty remains in the degree of GSH conjugation,
particularly in humans.
This document is a draft for review purposes only and does not constitute Agency policy.
66 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Table 3-28. Comparison of hepatic in vitro oxidation and conjugation of TCE
Cellular or
subcellular
fraction
Vmax
(nmol TCE
metabolized/min/g tissue)
Km
(jiM in blood)
Vmax/Km
(mL/min/g tissue)
Oxidation
GSH
conjugation
Oxidation
GSH
conjugation
Oxidation
GSH
conjugation
Hepatocytes
10.0-68.4
16-25
22.1-198
16-47
0.087-1.12
0.55-1.0
Liver
microsomes
6.1-111
45
2.66-11.la
5.9a
1.71-28.2a
7.6a
71.0-297b
157b
0.064-1.06b
0.29b
Liver
cytosol
-
380
-
4.5a
-
84a
-
-
22.7b
-
16.7b
Note: When triphasic metabolism was reported, only high affinity pathway is shown here.
Conversion assumptions for VMAX:
Hepatocellularity of 99 million cells/g liver (Barter et al., 2007);
Liver microsomal protein content of 32 mg protein/g tissue (Barter et al., 2007); and
Liver cytosolic protein content of 89 mg protein/g tissue (based on rats: (Prasanna et al., 1989; van Bree et al.,
1990).
Conversion assumptions for KM:
For hepatocytes, KM in headspace converted to KM in blood using blood:air partition coefficient of 9.5 (reported
range of measured values 6.5-12.1, Table 3-1);
For microsomal protein, option (a) assumes KM in medium is equal to KM in tissue, and converts to KM in blood
by using a livenblood partition coefficient of 5 (reported ranges of measured values 3.6-5.9, Table 3-8), and option
(b) converts KM in medium to KM in air using the measured microsomal protein:air partition coefficient of 1.78
(Lipscomb et al., 1997), and then converts to KM in blood by using the blood:air partition coefficient of 9.5; and
For cytosolic protein, option (a) assumes KM in medium is equal to KM in tissue, and converts to KM in blood by
using a livenblood partition coefficient of 5 (reported ranges of measured values 3.6-5.9, Table 3-8), and option (b)
assumes KM in medium is equal to KM in blood, so no conversion is necessary.
This document is a draft for review purposes only and does not constitute Agency policy.
67 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-29. Estimates of DCVG in blood relative to inhaled TCE dose in
2 humans exposed to 50 and 100 ppm (269 and 537 mg/m3) (Fisher et al., 1998;
3 Lash et al., 1999b)
4
Sex exposure
Estimated inhaled TCE dose
(mmolf
Estimated peak amount of DCVG in
blood (mmol)b
Males
50 ppm x 4 h
3.53
0.11+0.08
100 ppm x 4 h
7.07
0.26 + 0.08
Females
50 ppm x 4 h
2.36
0.010 + 0
100 ppm x 4 h
4.71
0.055 + 0.027
5
6 a Inhaled dose estimated by (50 or 100 ppm)/(24,450 ppm/mM) x (240 min) x QP, where alveolar ventilation rate QP
7 is 7.2 L/min for males and 4.8 L/min for females. QP is calculated as (VT - VD) x fR with the following
8 respiratory parameters: tidal volume VT (0.75 L for males, 0.46 L for females), dead space VD (0.15 L for males,
9 0.12 L for females), and respiration frequency fR (12 min1 for males, 14 min"1 for females) (assumed sitting,
10 awake from The International Commission on Radiological Protection (ICRP, 2003).
11 b Peak amount of DCVG in blood estimated by multiplying the peak blood concentration by the estimated blood
12 volume: 5.6 L in males and 4.1 L infemales (ICRP, 2003).
13
14
15
3.4. TRICHLOROETHYLENE (TCE) EXCRETION
16 This section discusses the major routes of excretion of TCE and its metabolites in exhaled
17 air, urine, and feces. Unmetabolized TCE is eliminated primarily via exhaled air. As discussed
18 in Section 3.3, the majority of TCE absorbed into the body is eliminated by metabolism. With
19 the exception of CO2, which is eliminated solely via exhalation, most TCE metabolites have low
20 volatility and, therefore, are excreted primarily in urine and feces. Though trace amounts of TCE
21 metabolites have also been detected in sweat and saliva (Bartonicek, 1962), these excretion
22 routes are likely to be relatively minor.
23
3.4.1. Exhaled Air
24 In humans, pulmonary elimination of unchanged trichloroethylene and other volatile
25 compounds is related to ventilation rate, cardiac output, and the solubility of the compound in
26 blood and tissue, which contribute to final exhaled air concentration of TCE. In their study of
This document is a draft for review purposes only and does not constitute Agency policy.
68 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
the impact of workload on TCE absorption and elimination, Astrand and Ovrum (1976)
characterized the postexposure elimination of TCE in expired breath. TCE exposure (540 or
-3
1,080 mg/m ; 100 or 200 ppm) was for a total of 2 hours, at workloads from 0-150 Watts.
Elimination profiles were roughly equivalent among groups, demonstrating a rapid decline in
TCE concentrations in expired breath postexposure (see Table 3-30).
The lung clearance of TCE represents the volume of air from which all TCE can be
removed per unit time, and is a measure of the rate of excretion via the lungs. Monster et al.
(1976) reported lung clearances ranging from 3.8-4.9 L/minute in four adults exposed at rest to
70 ppm and 140 ppm of trichloroethylene for 4 hours. Pulmonary ventilation rates in these
individuals at rest ranged from 7.7-12.3 L/minute. During exercise, when ventilation rates
increased to 29-30 L/minute, lung clearance was correspondingly higher, 7.7-12.3 L/minute.
Under single and repeated exposure conditions, Monster et al. (1979; 1976) reported from
7-17% of absorbed TCE excreted in exhaled breath. Pulmonary elimination of unchanged
trichloroethylene at the end of exposure is a first-order diffusion process across the lungs from
blood into alveolar air, and it can be thought of as the reversed equivalent of its uptake from the
lungs. Exhaled pulmonary excretion occurs in several distinct (delayed) phases corresponding to
release from different tissue groups, at different times. Sato et al. (1977) detected 3 first-order
phases of pulmonary excretion in the first 10 hours after exposure to 100 ppm for 4 hours, with
fitted half-times of pulmonary elimination of 0.04 hour, 0.67 hour, and 5.6 hours, respectively.
Opdam (1989a) sampled alveolar air up to 20-310 hours after 29-62 minute exposures to 6-38
ppm, and reported terminal half-lives of 8-44 hours at rest. Chiu et al. (2007)sampled alveolar
air up to 100 hours after 6-hour exposures to 1 ppm and reported terminal half-lives of 14-23
hours. The long terminal half-time of TCE pulmonary excretion indicates that a considerable
time is necessary to completely eliminate the compound, primarily due to the high partitioning to
adipose tissues (see Section 3.2).
As discussed above, several studies (Dekant et al., 1984; Green and Prout, 1985; Prout et
al., 1985) have investigated the disposition of [14C]TCE in rats and mice following gavage
administrations (see Section 3.3.2). These studies have reported CO2 as an exhalation excretion
product in addition to unchanged TCE. With low doses, the amount of TCE excreted unchanged
in exhaled breath is relatively low. With increasing dose in rats, a disproportionately increased
amount of radiolabel is expired as unchanged TCE. This may indicate saturation of metabolic
activities in rats at doses 200 mg/kg and above, which is perhaps only minimally apparent in the
data from mice. In addition, exhaled air TCE concentration has been measured after constant
inhalation exposure for 2 hours to 50 or 500 ppm in rats (Dallas et al., 1991), and after dermal
This document is a draft for review purposes only and does not constitute Agency policy.
69 DRAFT—DO NOT CITE OR QUOTE
-------
1 exposure in rats and humans (Poet et al., 2000). Exhaled TCE data from rodents and humans
2 have been integrated into the PBPK model presented in Section 3.5.
This document is a draft for review purposes only and does not constitute Agency policy.
70 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-30. Concentrations of TCE in expired breath from
2 inhalation-exposed humans (Astrand and Ovrum, 1976)
3
Time
postexposure
Alveolar air
Ia
II
III
0
459 ±44
244 ± 16
651 ±53
30
70 ±5
51 ±3
105 ± 18
60
40 ±4
28 ±2
69 ±8
90
35 ± 9
21 ± 1
55 ± 2
120
31 ±8
16 ± 1
45 ± 1
300
8 ± 1
9 ± 2
14 ± 2
420
5 ±0.5
4 ±0.5
8± 1.3
19 hours
2 ±0.3
2 ±0.2
4 ±0.5
4
5 a Roman numerals refer to groups assigned different workloads.
6
7 Concentrations are in mg/m3 for expired air.
8
9
10 Finally, TCOH is also excreted in exhaled breath, though at a rate about 10,000-fold
11 lower than unmetabolized TCE (Monster, 1979; Monster et al., 1976).
12
3.4.2. Urine
13 Urinary excretion after TCE exposure consists predominantly of the metabolites TCA
14 and TCOH, with minor contributions from other oxidative metabolites and GSH conjugates.
15 Measurements of unchanged TCE in urine have been at or below detection limits (e.g., Chiu et
16 al., 2007; Fisher et al., 1998). The recovery of urinary oxidative metabolites in mice, rats, and
17 humans was addressed earlier (see Section 3.3.2) and will not be discussed here. Because of
18 their relatively long elimination half-life, urinary oxidative metabolites have been used as an
19 occupational biomarker of TCE exposure for many decades (Carrieri et al., 2007; Ikeda and
20 Imamura, 1973). Ikeda and Imamura (1973) measured total trichloro compounds TCOH and
21 TCA, in urine over 3 consecutive postexposure days for 4 exposure groups totaling 24 adult
22 males and one exposure group comprising 6 adult females. The elimination half-life for TTC
23 ranged 26.1-48.8 hours in males and was 50.7 hours in females. The elimination half-life for
24 TCOH was 15.3 hours in the only group of males studied and was 42.7 hours in females. The
This document is a draft for review purposes only and does not constitute Agency policy.
11 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
elimination half-life for TCA was 39.7 hours in the only group of males studied and was 57.6
hours in females. These authors compared their results to previously published elimination
half-lives for TTC, TCOH, and TCA. Following experimental exposures of groups of two-five
adults, elimination half-lives ranged from 31-50 hours for TTC; 19-29 hours for TCOH; and
36-55 hours for TCA (Bartonicek, 1962; Nomiyama andNomiyama, 1971; Ogata et al., 1971;
Stewart et al., 1970). The urinary elimination half-life of TCE metabolites in a subject who
worked with and was addicted to sniffing TCE for 6-8 years approximated 49.7 hours for
TCOH, 72.6 hours for TCA, and 72.6 hours for TTC (Ikeda et al., 1971).
The quantitative relationship between urinary concentrations of oxidative metabolites and
exposure in an occupational setting was investigated by Ikeda (1977). This study examined the
urinary elimination of TCE and metabolites in urine of 51 workers from 10 workshops. The
concentration of TCA and TCOH in urine demonstrated a marked concentration-dependence,
with concentrations of TCOH being approximately twice as high as those for TCA. Urinary
half-life values were calculated for six males and six females from five workshops; males were
intermittently exposed to 200 ppm and females were intermittently exposed to 50 ppm
(269 mg/m3). Urinary elimination half-lives for TTC, TCOH and TCA were 26.1, 15.3, and
39.7 hours; and 50.7, 42.7 and 57.6 hours in males and females, respectively, which were similar
to the range of values previously reported. These authors estimated that urinary elimination of
parent TCE during exposure might account for one-third of the systemically absorbed dose.
Importantly, urinary TCA exhibited marked saturation at exposures higher than 50 ppm.
Because neither TTC nor urinary TCOH (in the form of the glucuronide TCOG) showed such an
effect, this saturation cannot be due to TCE oxidation itself, but must rather be from one of the
metabolic processes forming TCA from TCOH. Unfortunately, since biological monitoring
programs usually measure only urinary TCA, rather than TTC, urinary TCA levels above around
150 mg/L cannot distinguish between exposures at 50 ppm and at much higher concentrations.
It is interesting to attempt to extrapolate on a cumulative exposure basis the Ikeda (1977)
results for urinary metabolites obtained after occupational exposures at 50 ppm to the controlled
exposure study by Chiu et al. (2007) at 1.2 ppm for 6 hours (the only controlled exposure study
for which urinary concentrations, rather than only cumulative excretion, are available). Ikeda
(1977) reported that measurements were made during the second half of the week, so one can
postulate a cumulative exposure duration of 20-40 hours. At 50 ppm, Ikeda (1977) report a
urinary TCOH concentration of about 290 mg/L, so that per ppm-hour, the expected urinary
concentration would be 290/(50 20 ~ 40) = 0.145 ~ 0.29 mg/L-ppm-hour. The cumulative
exposure in Chiu et al. (2007) is 1.2 x 6 = 7.2 ppm-hour, so the expected urinary TCOH
concentration would be 7.2 x (0.145 ~ 0.29) = 1.0-2.1 mg/L. This estimate is somewhat
This document is a draft for review purposes only and does not constitute Agency policy.
72 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
surprisingly consistent with the actual measurements of Chiu et al. (2007) during the first day
postexposure, which ranged from 0.8-1.2 mg/L TCOH in urine.
On the other hand, extrapolation of TCA concentrations was less consistent. At 50 ppm,
Ikeda (1977) report a urinary TCA concentration of about 140 mg/L, so that per ppm-hour, the
expected urinary concentration would be 140/(50 x 20 ~ 40) = 0.07 ~ 0.14 mg/L-ppm-hour. The
cumulative exposure in Chiu et al. (2007) is 1.2 x 6 = 7.2 ppm-hour, so the expected urinary
TCA concentration would be 7.2 x (0.07 ~ 0.14) = 0.5 -1.0 mg/L, whereas Chiu et al. (2007)
reported urinary TCA concentrations on the first day after exposure of 0.03-0.12 mg/L.
However, as noted in Chiu et al. (2007), relative urinary excretion of TCA was 3- to 10-fold
lower in Chiu et al. (2007) than other studies at exposures of 50-140 ppm, which may explain
part of the discrepancies. However, this may be due in part to saturation of many urinary TCA
measurements, and, furthermore, interindividual variance, observed to be substantial in Fisher
et al.(1998), cannot be ruled out.
Urinary elimination kinetics have been reported to be much faster in rodents than in
humans. For instance, adult rats were exposed to 50, 100, or 250 ppm (269, 537, or
-3
1,344 mg/m ) via inhalation for 8 hours or were administered an i.p. injection (1.47 g/kg) and the
urinary elimination of total trichloro compounds was followed for several days (Ikeda and
Imamura, 1973). These authors calculated urinary elimination half-lives of 14.3-15.6 hours for
female rats and 15.5-16.6 hours for male rats; the route of administration did not appear to
influence half-life value. In other rodent experiments using orally administered radiolabeled
TCE, urinary elimination was complete within one or two days after exposure (Dekant et al.,
1984; Green and Prout, 1985; Prout et al., 1985).
3.4.3. Feces
Fecal elimination accounts for a small percentage of TCE as shown by limited
information in the available literature. Bartonicek (1962) exposed 7 human volunteers to
1.042 mg TCE/L air for 5 hours and examined TCOH and TCA in feces on the 3rd and 7th day
following exposure. The mean amount of TCE retained during exposure was 1,107 mg,
representing 51-64% (mean 58%) of administered dose. On the third day following TCE
exposure, TCOH and TCA in feces demonstrated mean concentrations of 17.1 and
18.5 mg/100 grams feces, similar to concentrations in urine. However, because of the 10-fold
smaller daily rate of excretion of feces relative to urine, this indicates fecal excretion of these
metabolites is much less significant than urinary excretion. Neither TCOH nor TCA was
detected in feces on the seventh day following exposure.
This document is a draft for review purposes only and does not constitute Agency policy.
73 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
In rats and mice, total radioactivity has been used to measure excretion in feces after oral
gavage TCE administration in corn oil, but since the radiolabel was not characterized it is not
possible to determine whether the radiolabel in feces represented unabsorbed parent compound,
excreted parent compound, and/or excreted metabolites. Dekant et al. (1984) reported mice
eliminated 5% of the total administered TCE, while rats eliminated 2% after oral gavage.
Dekant et al. (1986b) reported a dose response related increase in fecal elimination with dose,
ranging between 0.8-1.9% in rats and 1.6-5% in mice after oral gavage in corn oil. Due to the
relevant role of CYP2E1 in the metabolism of TCE (see Section 3.3.3.1.6), Kim and Ghanayem
(2006) compared fecal elimination in both wild-type and CYP2E1 knockout mice and reported
fecal elimination ranging between 4.1-5.2%) in wild-type and 2.1-3.8%) in knockout mice
exposed by oral gavage in aqueous solution.
3.5. PHYSIOLOGICALLY BASED PHARMACOKINETIC (PBPK) MODELING OF
TRICHLOROETHYLENE (TCE) AND ITS METABOLITES
3.5.1. Introduction
PBPK models are extremely useful tools for quantifying the relationship between
external measures of exposure and internal measures of toxicologically relevant dose. In
particular, for the purposes of this assessment, PBPK models are evaluated for the following:
(1) providing additional quantitative insights into the ADME of TCE and metabolites described
in the sections above; (2) cross-species pharmacokinetic extrapolation of rodent studies of both
cancer and noncancer effects, (3) exposure-route extrapolation; and (4) characterization of
human pharmacokinetic variability. The following sections first describe and evaluate previous
and current TCE PBPK modeling efforts, then discuss the insights into ADME (1, above), and
finally present conclusions as to the utility of the model to predict internal doses for use in
dose-response assessment (2-4, above).
3.5.2. Previous Physiologically Based Pharmacokinetic (PBPK) Modeling of
Trichloroethylene (TCE) for Risk Assessment Application
TCE has an extensive number of both in vivo pharmacokinetic and PBPK modeling
studies (see Chiu et al., 2006, supplementary material, for a review). Models previously
developed for occupational or industrial hygiene applications are not discussed here but are
reviewed briefly in Clewell et al. (2000). Models designed for risk assessment applications have
This document is a draft for review purposes only and does not constitute Agency policy.
74 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
focused on descriptions of TCE and its major oxidative metabolites TCA, TCOH, and TCOG.
Most of these models were extensions of the "first generation" of models developed by Fisher
and coworkers (Allen and Fisher, 1993; Fisher et al., 1991) in rats, mice, and humans. These
models, in turn, are based on a Ramsey and Andersen (1984) structure with flow-limited tissue
compartments and equilibrium gas exchange, saturable Michaelis-Menten kinetics for oxidative
metabolism, and lumped volumes for the major circulating oxidative metabolites TCA and
TCOH. Fisher and coworkers updated their models with new in vivo and in vitro experiments
performed in mice (Abbas and Fisher, 1997; Greenberg et al., 1999) and human volunteers
(Fisher et al., 1998) and summarized their findings in Fisher (2000). Clewell et al. (2000) added
enterohepatic recirculation of TCOG and pathways for local oxidative metabolism in the lung
and GST metabolism in the liver. While Clewell et al. (2000) does not include the updated
Fisher data, they have used a wider set of in vivo and in vitro mouse, rat, and human data than
previous models. Finally, Bois (2000a, b) performed reestimations of PBPK model parameters
for the Fisher and Clewell models using a Bayesian population approach (Gelman et al., 1996,
and discussed further below).
As discussed in Rhomberg (2000), the choice as to whether to use the Fisher, Clewell,
and Bois models for cross-species extrapolation of rodent cancer bioassays led to quantitative
results that differed by as much as an order of magnitude. There are a number of differences in
modeling approaches that can explain their differing results. First, the Clewell et al. (2000)
model differed structurally in its use of single-compartment volume-of-distribution models for
metabolites as opposed to the Fisher (2000) models' use of multiple physiologic compartments.
Also, the Clewell et al. (2000) model, but not the Fisher models, includes enterohepatic
recirculation of TCOH/TCOG (although reabsorption was set to zero in some cases). In addition
to structural differences in the models, the input parameter values for these various models were
calibrated using different subsets of the overall in vivo database (see Chiu et al., 2006,
supplementary material, for a review). The Clewell et al. (2000) model is based primarily on a
variety of data published before 1995; the Fisher (2000) models were based primarily on new
studies conducted by Fisher and coworkers (after 1997); and the Bois (2000a, b) reestimations of
the parameters for the Clewell et al. (2000) and Fisher (2000) models used slightly different data
sets than the original authors. The Bois (2000a, b) reanalyses also led to somewhat different
parameter estimates than the original authors, both because of the different data sets used as well
as because the methodology used by Bois allowed many more parameters to be estimated
simultaneously than were estimated in the original analyses.
Given all these methodological differences, it is not altogether surprising that the
different models led to different quantitative results. Even among the Fisher models themselves,
This document is a draft for review purposes only and does not constitute Agency policy.
75 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
Fisher (2000) noted some inconsistencies, including differing estimates for metabolic parameters
between mouse gavage and inhalation experiments. These authors included possible
explanations for these inconsistencies: the impact of corn oil vehicle use during gavage (Staats et
al., 1991) and the impact of a decrease in ventilation rate in mice due to sensory irritation during
the inhalation of solvents [e.g., Stadler and Kennedy (1996)].
As discussed in a report by the National Research Council (NRC, 2006), several
additional PBPK models relevant to TCE pharmacokinetics have been published since 2000 and
are reviewed briefly here. Poet et al. (2000) incorporated dermal exposure to TCE in PBPK
models in rats and humans, and published in vivo data in both species from dermal exposure
(Poet et al., 2000; Thrall and Poet, 2000). Albanese et al. (2002) published a series of models
with more complex descriptions of TCE distribution in adipose tissue but did not show
comparisons with experimental data. Simmons et al. (2002) developed a PBPK model for TCE
in the Long-Evans rat that focused on neurotoxicity endpoints and compared model predictions
with experimentally determined TCE concentrations in several tissues—including the brain.
Keys et al. (2003) investigated the lumping and unlumping of various tissue compartments in a
series of PBPK models in the rat and compared model predictions with TCE tissue
concentrations in a multitude of tissues. Although none of these TCE models included
metabolite descriptions, the experimental data was available for either model or evaluation.
Finally, Keys et al. (2004) developed a model for DCA in the mouse that included a description
of suicide inhibition of GST-zeta, but this model was not been linked to TCE.
3.5.3. Development and Evaluation of an Interim "Harmonized" Trichloroethylene (TCE)
Physiologically Based Pharmacokinetic (PBPK) Model
Throughout 2004, EPA and the U.S. Air Force jointly sponsored an integration of the
Fisher, Clewell, and Bois modeling efforts (Hack et al., 2006). In brief, a single interim PBPK
model structure combining features from both the Fisher and Clewell models was developed and
used for all three species of interest (mice, rats, and humans). An effort was made to combine
structures in as simple a manner as possible; the evaluation of most alternative structures was left
for future work. The one level of increased complexity introduced was inclusion of species- and
dose-dependent TCA plasma binding, although only a single in vitro study of Lumpkin et al.
(2003) was used as parameter inputs. As part of this joint effort, a hierarchical Bayesian
population analysis using Markov chain Monte Carlo (MCMC) sampling (similar to the Bois
[2000a, b] analyses) was performed on the revised model with a cross-section of the combined
database of kinetic data to provide estimates of parameter uncertainty and variability (Hack et al.,
This document is a draft for review purposes only and does not constitute Agency policy.
76 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
2006). Particular attention was given to using data from each of the different efforts, but owing
to time and resource constraints, a combined analysis of all data was not performed. The results
from this effort suggested that a single model structure could provide reasonable fits to a variety
of data evaluated for TCE and its major oxidative metabolites TCA, TCOH, and TCOG.
However, in many cases, different parameter values—particularly for metabolism—were
required for different studies, indicating significant interindividual or interexperimental
variability. In addition, these authors concluded that dosimetry of DCA, conjugative
metabolites, and metabolism in the lung remained highly uncertain (Hack et al., 2006).
Subsequently, EPA conducted a detailed evaluation of the Hack et al. (2006) model that
included (1) additional model runs to improve convergence; (2) evaluation of posterior
distributions for population parameters; and (3) comparison of model predictions both with the
data used in the Hack et al. (2006) analysis as well as with additional data sets identified in the
literature. Appendix A provides the details and conclusions of this evaluation, briefly
summarized in Table 3-31, along with their pharmacokinetic implications.
3.5.4. Physiologically Based Pharmacokinetic (PBPK) Model for Trichloroethylene (TCE)
and Metabolites Used for This Assessment
3.5.4.1.1. Introduction
Based on the recommendations of the NRC (2006) as well as additional analysis and
evaluation of the Hack et al. (2006) PBPK model, an updated PBPK model for TCE and
metabolites was developed for use in this risk assessment. The updated model is reported in
Evans et al. (2009) and Chiu et al. (2009), and the discussion below provides some details in
additional to the information in the published articles.
This updated model included modification of some aspects of the Hack et al. (2006)
PBPK model structure, incorporation of additional in vitro and in vivo data for estimating model
parameters, and an updated hierarchical Bayesian population analysis of PBPK model
uncertainty and variability. In the subsections below, the updated PBPK model and baseline
This document is a draft for review purposes only and does not constitute Agency policy.
77 DRAFT—DO NOT CITE OR QUOTE
-------
L/l
K>
Table 3-31. Conclusions from evaluation of Hack et al. (2006), and implications for PBPK model development
Conclusion from evaluation of Hack et al. (2006) model
Implications for PBPK model parameters, structure, or data
For some model parameters, posterior distributions were somewhat inconsistent
with the prior distributions.
• For parameters with strongly informative priors (e.g., tissue volumes and
flows), this may indicate errors in the model.
• For many parameters, the prior distributions were based on visual fits to the
same data. If the posteriors are inconsistent, then that means the priors
were "inappropriately" informative, and, thus, the same data was used
twice.
Reevaluation of all prior distributions
• Update priors for parameters with independent data (physiological
parameters, partition coefficients, in vitro metabolism), looking across
all available data sets.
• For priors without independent data (e.g., many metabolism
parameters), use less informative priors (e.g., log-uniform distributions
with wide bounds) so as prevent bias.
Evaluate modifications to the model structure, as discussed below.
A number of data sets involve TCE (i.a., portal vein), TCA (oral, i.v.), and
TCOH (oral, i.v.) dosing routes that are not currently in the model, but could be
useful for calibration.
• Additional dosing routes can be added easily.
TCE concentrations in blood, air, and tissues well-predicted only in rats, not in
mice and humans. Specifically:
• In mice, the oral uptake model could not account for the time-course of
several data sets. Blood TCE concentrations after inhalation consistently
over-predicted.
• In rats, tissue concentrations measured in data not used for calibration were
accurately predicted.
• In humans, blood and air TCE concentrations were consistently
over-predicted in the majority of (but not all) data sets.
• In mice, uptake from the stomach compartment (currently zero), but
previously included in Abbas and Fisher (1997), may improve the
model fit.
• In mice and humans, additional extrahepatic metabolism, either
presystemic (e.g., in the lung) or postsystemic (e.g., in the kidney)
and/or a wash-in/wash-out effect may improve the model fit.
Total metabolism appears well-predicted in rats and mice based on
closed-chamber data, but required significantly different Vmax values between
dose groups. Total recovery in humans (60-70%) is less than the model would
predict. In all three species, the ultimate disposition of metabolism is uncertain.
In particular, there are uncertainties in attributing the "missing" metabolism to
• GSH pathway (e.g., urinary mercapturates may only capture a fraction of
the total flux; moreover, in Bernauer et al. (1996), excretion was still
on-going at end of collection period; model does not accurately depict
time-course of mercapturate excretion).
• Other hepatic oxidation (currently attributed to DCA).
• Extrahepatic systemic metabolism (e.g., kidney).
• Presystemic metabolism in the lung.
• Additional metabolism of TCOH or TCA (see below).
• Calibration of GSH pathway may be improved by utilizing in vitro data
on liver and kidney GSH metabolism, adding a DCVG compartment to
improve the prediction of the time-course for mercapturate excretion,
and/or using the Lash et al. (1999b) blood DCVG in humans
(necessitating the addition of a DCVG compartment).
• Presystemic lung metabolism can only be evaluated if added to the
model (in vitro data exists to estimate the Vmax for such metabolism).
In addition, a wash-in/wash-out effect [e.g., suggested by Greenberg
et al. (1999)] can be evaluated using a continuous breathing model that
separately tracks inhaled and exhaled air, with adsorption/desporption in
the respiratory tract.
• Additional elimination pathways for TCOH and TCA can be added for
evaluation.
3^
rs
sj
3
a
>!
ca
fs*
si
<1 ^
00 §
Grj
>!
o
a
a
§•
>!
o
rs
o
a
>!
0
p
H
1
o
o
z;
H S.
W K-
oy
o
c
o
H
W
s
-------
Table 3-31. Conclusions from evaluation of Hack et al. (2006), and implications for PBPK model development
(continued)
Conclusion from evaluation of Hack et al. (2006) model
Implications for PBPK model parameters, structure, or data
TCA blood/plasma concentrations well predicted following TCE exposures in
all species. However, there may be inaccuracies in the total flux of TCA
production, as well as its disposition.
• In TCA dosing studies, the majority (>50%), but substantially <100%, was
recovered in urine, suggesting significant metabolism of TCA. Although
urinary TCA was well predicted in mice and humans (but not in rats), if
TCA metabolism is significant, then this means that the current model
underestimates the flux of TCE metabolism to TCA.
• An improved TCOH/TCOG model may also provide better estimates of
TCA kinetics (see below).
TCOH/TCOG concentrations and excretion were inconsistently predicted,
particularly after TCOH dosing.
• In mice and rats, first-order clearance for TCOH glucuronidation was
predicted to be greater than hepatic blood flow, which is consistent with a
first pass effect that is not currently accounted for.
• In humans, the estimated clearance rate for TCOH glucuronidation was
substantially smaller than hepatic blood flow. However, the presence of
substantial TCOG in blood (as opposed to free TCOH) in the Chiu et al.
(2007) data are consistent with greater glucuronidation than predicted by
the model.
• In TCOH dosing studies, substantially <100% was recovered in urine as
TCOG and TCA, suggesting another metabolism or elimination pathway.
• Additional elimination pathways for TCOH and TCA can be added for
evaluation.
• The addition of a liver compartment for TCOH and TCOG would
permit hepatic first-pass effects to be accounted for, as appears
necessary for mice and rats.
i.a. = intra-arterial, i.v. = intravenous.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
parameter values are described, as well as the approach and results of the analysis of PBPK
model uncertainty and variability. Appendix A provides more detailed descriptions of the model
and parameters, including background on hierarchical Bayesian analyses, model equations,
statistical distributions for parameter uncertainty and variability, data sources for these parameter
values, and the PBPK model code. Additional computer codes containing input files to the
MCSim program are available electronically.
3.5.4.1.2. Updated Physiologically Based Pharmacokinetic (PBPK) Model Structure
The updated TCE PBPK model is illustrated in Figure 3-7, with detailed descriptions of
the model structure, equations, and parameters found in Appendix A (see Section A.4), and the
major changes from the Hack et al. (2006) model described here. The TCE submodel was
augmented by the addition of kidney and venous blood compartments, and an updated
respiratory tract model that included both metabolism and the possibility of local storage in the
respiratory tissue. In particular, in the updated lung, separate processes describing inhalation and
exhalation allowed for adsorption and desorption from tracheobronchial epithelium
(wash-in/wash-out), with the possibility of local metabolism as well. In addition, conjugative
metabolism in the kidney was added, motivated by the in vitro data on TCE conjugation
described in Sections 3.3.3.2-3.3.3.3. With respect to oxidation, a portion of the lung
metabolism was assumed to produce systemically available oxidative metabolites, including
TCOH and TCA, with the remaining fraction assumed to be locally cleared. This is clearly a
lumping of a multistep process, but the lack of data precludes the development of a more
sequential model. TCE oxidation in the kidney was not included because it was not likely to
constitute a substantial flux of total TCE oxidation given the much lower CYP activity in the
kidney relative to the liver (Cummings and Lash, 2000; Cummings et al., 1999) and the greater
tissue mass of the liver.2 In addition, liver compartments were added to the TCOH and TCOG
2 The extraction ratio for kidney oxidation is likely to be very low, as shown by the following calculation in rats and
humans. In rats, the in vitro kidney oxidative clearance (Vmax/Km) rate (see Table 3-13, converting units) is
1.64 x 10 L/min/mg microsomal protein. Converting units using 16 mg microsomal protein to g tissue (Bong
et al., 1985) gives a clearance rate per unit tissue mass of 2.6 x 10 6 L/min/g kidney. This is more than a 1000-fold
smaller than the kidney specific blood flow rate of 6.3 x 10 L/min/g kidney (Brown etal.. 1997). In humans, an in
vitro clearance rate of 6.5 x 10 x L/min/mg microsomal protein is derived from the only detectable in vitro oxidation
rate from Cummings and Lash (2000) of 0.13 nmol/minute/mg protein at 2 mM. Using the same conversion from
microsomal protein to tissue mass gives a clearance rate of 1.0 x 10 6 L/min/g kidney, more than 1000-fold smaller
than the kidney specific blood flow of 3.25 x 10 L/min/g kidney (Brown etal.. 1997). No data on kidney
metabolism are available in mice, but the results are likely to be similar. Therefore, even accounting for
uncertainties of up to an order of magnitude in the in vitro-to-in vivo conversion, kidney oxidation should contribute
negligibly to total metabolism of TCE.
This document is a draft for review purposes only and does not constitute Agency policy.
80 DRAFT—DO NOT CITE OR QUOTE
-------
Cjnhaled aiP^
t
TCE
Exhaled air
Respiratory
Tract Lumen
Clnhalationl
Jt
Respiratory
Tract Tissue
^ _Oxidatjon _ |
Respiratory
Tract Lumen
CExhalationl
(Dead space)
. Gas Ex change
Venous
Blood
Rapidly
Perfused
Slowly
Perfused
(j3raT)
Fat
Gut
Liver
Stomach
Duodenum
Kidney
«_r>i^2.r
PV^
Oxidation & "J
Conjugation }
V. ConjUigation_ i
Oxidative Metabolism
' Luna ^
. Oxidation .
Local
Clearance_ f
I Liver
v _ Oxidation _ t
•/
\ TCOH 1
b i oiai
I Systemic
A Oxidation ,,
jjK TCA J
\ Other J
Conjugative Metabolism
(rat and human only)
, Liver fc=^
v Conjugation t r
DCVG
1 ®'°" I
activation .
' Kidnev ^
^ Conjugation f r
DCVC
£
Urine ,
v 1NAcDCVC]_ ,
TCOH
!-~
Blood
Body
Liver
TCOG
TCOG
Other
TCOG
B ood
TCOH
TCOH
TCA
cv>*n
!-~
Plasma
I Urine '
4—
Body
,v,CoraD
Liver
Other
TCOH
Legend
CD Input (exposure/dose)
~ "Dynamic" Compartment (solved by ODEs)
T [ "Static" Compartment (at local steady-state)
Transformation or Excretion
1 submodels to account properly for first-pass hepatic metabolism, which is important for
2 consistency across
3 Figure 3-7. Overall structure of PBPK model for TCE and metabolites used
4 in this assessment. Boxes with underlined labels are additions or modifications
5 of the Hack et al. (2006) model, which are discussed in Table 3-32.
This document is a draft for review purposes only and does not constitute Agency policy.
81 DRAFT—DO NOT CITE OR QUOTE
-------
1
This document is a draft for review purposes only and does not constitute Agency policy.
82 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-32. Discussion of changes to the Hack et al. (2006) PBPK model
2 implemented for this assessment
3
Change to Hack
et al. (2006) PBPK
model
Discussion
TCE respiratory
tract compartments
and metabolism
In vitro data indicate that the lung (at least in the mouse) has a
significant capacity for oxidizing TCE. However, in the Hack et al.
(2006) model, respiratory metabolism was blood flow-limited. The
model structure used was inconsistent with other PBPK models in
which the same mechanism for respiratory metabolism is assumed [e.g.,
styrene, Sarangapani et al. (2003)]. In these models, the main source of
exposure in the respiratory tract tissue is from the respiratory lumen—
not from the tracheobronchial blood flow. In addition, a
wash-in/wash-out effect has also been postulated. The current structure,
which invokes a "continuous breathing" model with separate "inhaled"
and "exhaled" respiratory lumens, can accommodate both respiratory
metabolism due to exposure from the respiratory lumen as well as a
wash-in/wash-out effect in which there is temporary storage in the
respiratory tract tissue.
Moreover, preliminary analyses indicated that these changes to the
model structure allowed for a substantially better fit to mouse
closed-chamber data under the requirement that all the dose levels are
modeled using the same set of parameters.
TCE kidney
compartment
In vitro data indicate that the kidney has a significant capacity for
conjugating TCE with GSH.
TCE venous blood
compartment
Many PBPK models have used a separate blood compartment. It was
believed to be potentially important and feasible to implement here
because (1) TCE blood concentrations were often not well predicted by
the Hack et al. (2006) model; (2) the TCA submodel has a plasma
compartment, which is a fraction of the blood volume based on the
blood volume; (3) adequate independent information on blood volume
is available; and (4) the updated model was to include the intravenous
route of exposure.
TCOH and TCOG
liver compartments
In mice and rats, the Hack et al. (2006) model estimated a rate of TCOH
glucuronidation that exceeded hepatic blood flow (all glucuronidation is
assumed to occur in the liver), which indicated a significant first-pass
effect. Therefore, a separate liver compartment is necessary to account
properly for hepatic first-pass.
TCOH and TCA
"other" elimination
pathways
Mass-balance studies with TCOH and TCA dosing indicated that,
although the majority of TCOH and TCA are excreted in urine, the
amount is still substantially less than 100%. Therefore, additional
elimination of TCOH and TCA must exist and should be accounted for.
This document is a draft for review purposes only and does not constitute Agency policy.
83 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
DCVG compartment
(human model only)
Blood DCVG data in humans exist as part of the Fisher et al. (1998)
experiments, reported in Lash et al. (1999b), and a DCVG compartment
is necessary in order to utilize those data.
routes of exposure. Furthermore, additional clearance pathways of TCOH and TCA were added
to their respective submodels. With respect to TCE conjugation, in humans, an additional
DCVG compartment was added between TCE conjugation and production of DCVC. In
addition, it should be noted that the urinary clearance of DCVC represents a lumping of
A-acetyl ati on of DCVC, deacetylation ofNAcDCVC, and urinary excretion NAcDCVC, and that
the bioactivation of DCVC represents a lumping of thiol production from DCVC by beta-lyase,
sulfoxidation of DCVC by FM03, and sulfoxidation ofNAcDCVC by CYP3A. Such lumping
was used because these processes are not individually identifiable given the available data.
3.5.4.1.3. Specification of Baseline Physiologically Based Pharmacokinetic (PBPK) Model
Parameter
Point estimates for PBPK model parameters ("baseline values"), used as central estimates
in the prior distributions for population mean parameters in the hierarchical Bayesian statistical
model (see Appendix A), were developed using standard methodologies to ensure biological
plausibility, and were a refinement of those used in Hack et al. (2006). Because the Bayesian
parameter estimation methodology utilizes the majority of the useable in vivo data on TCE
pharmacokinetics, all baseline parameter estimates were based solely on measurements
independent of the in vivo data. This avoids using the same data in both the prior and the
likelihood. These parameters were, in turn, given truncated normal or lognormal distributions
for the uncertainty in the population mean. If no independent data were available, as is the case
for many "downstream" metabolism parameters, then no baseline value was specified, and a
noninformative prior was used. Section 3.5.5.4, below, discusses the updating of these
noninformative priors using interspecies scaling.
In keeping with standard practice, many of the PBPK model parameters were "scaled" by
body or organ weights, cardiac output, or allometrically by an assumed (fixed) power of body
weight. Metabolic capacity and cardiac output were scaled by the % power of body weight and
rate coefficients were scaled by the—Vi power of body weight, in keeping with general
expectations as to the relationship between metabolic rates and body size (U.S. EPA, 1992; West
et al., 2002). So as to ensure a consistent model structure across species as well as improve the
performance of the MCMC algorithm, parameters were further scaled to the baseline
This document is a draft for review purposes only and does not constitute Agency policy.
84 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
point-estimates where available, as was done by Hack et al. (2006). For example, to obtain the
actual liver volume (VLivC) in liters, a point estimate is first obtained by multiplying the fixed,
species-specific baseline point estimate for the fractional liver volume by a fixed body weight
(measured or species-specific default) with density of 1 kg per liter assumed to convert from kg
to liters. Then, any deviation from this point estimate is represented by multiplying by a separate
"scaled" parameter VLivC that has a value of one if there is no deviation from the point estimate.
These "scaled" parameters are those estimated by the MCMC algorithm, and for which
population means and variances are estimated.
Baseline physiological parameters were reestimated based on the updated tissue lumping
(e.g., separate blood and kidney compartments) using the standard references International
Commission on Radiological Protection (ICRP, 2003) and Brown et al. (1997). For a few of
these parameters, such as hematocrit and respiratory tract volumes in rodents, additional
published sources were used as available, but no attempt was made to compile a comprehensive
review of available measurements. In addition, a few parameters, such as the slowly perfused
volume, were calculated rather than sampled in order to preserve total mass or flow balances.
For chemical-specific distribution and metabolism parameters, in vitro data from various
sources were used. Where multiple measurements had been made, as was the case for many
partition coefficients, TCA plasma protein binding parameters, and TCE metabolism, different
results were pooled together, with their uncertainty reflected appropriately in the prior
distribution. Such in vitro measurements were available for most chemical partition coefficients,
except for those for TCOG (TCOH used as a proxy) and DCVG. There were also such data to
develop baseline values for the oxidative metabolism of TCE in the liver (Vmax and KM), the
relative split in TCE oxidation between formation of TCA and TCOH, and the Vmax for TCE
oxidation in the lung. For GSH conjugation, the geometric means of the in vitro data from Lash
et al. (1999a) and Green et al. (1997a) were used as central estimates, with a wide enough
uncertainty range to encompass both (widely disparate) estimates. Thus, the prior distribution
for these parameters was only mildly informative, and the results are primarily determined by the
available in vivo data. All other metabolism parameters were not given baseline values and
needed to be estimated from the in vivo data.
3.5.4.1.4. Dose-Metric Predictions
The purpose of this PBPK model is to make predictions of internal dose in rodents used in
toxicity studies or in humans in the general population, and not in the groups or individuals for
which pharmacokinetic data exist. Therefore, to evaluate its predictive utility for risk
This document is a draft for review purposes only and does not constitute Agency policy.
85 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
assessment, a number of dose-metrics were selected for simulation in a "generic" mouse, rat, or
human, summarized in Table 3-33. The parent dose-metric was AUC in blood. TCE
metabolism dose-metrics (i.e., related to the amount metabolized) included both total
metabolism, metabolism splits between oxidation versus conjugation, oxidation in the liver
versus the lung, the amount of oxidation in the liver to products other than TCOH and TCA, and
the amount of TCA produced. These metabolism rate dose-metrics are scaled by body weight in
the case of TCA produced, by the metabolizing tissue volume and by body weight to the
3/4 power in the cases of the lung and "other" oxidation in the liver, and by body weight to the
3/4 power only in other cases. With respect to the oxidative metabolites, liver concentrations of
TCA and blood concentrations of free TCOH were used. With respect to conjugative
metabolites, the dose-metrics considered were total GSH metabolism scaled by body weight to
the 3/4 power, and the amount of DCVC bioactivated (rather than excreted in urine) per unit body
weight to the 3/4 power and per unit kidney mass.
All dose-metrics are converted to daily or weekly averages based on simulations lasting
10 weeks for rats and mice and 100 weeks for humans. These simulation times were the shortest
for which additional simulation length did not add substantially to the average (i.e., less than a
few percent change with a doubling of simulation time).
3.5.5. Bayesian Estimation of Physiologically Based Pharmacokinetic (PBPK) Model
Parameters, and Their Uncertainty and Variability
3.5.5.1.1. Updated Pharmacokinetic Database
An extensive search was made for data not previously considered in the PBPK modeling
of TCE and metabolites, with a few studies identified or published subsequent to the review by
Chiu et al. (2006b). The studies considered for analysis are listed in Tables 3-34-3-35, along
with an indication of whether and how they were used. 3
The least amount of data was available for mice, so an effort was made to include as
many studies as feasible for use in calibrating the PBPK model parameters. Exceptions include
mouse studies with CH or DCA dosing, since those metabolites are not included in the PBPK
model. In addition, the Birner et al. (1993) data only reported urine concentrations, not the
amount excreted in urine. Because there is uncertainty as to total volume of urine excreted, and
over what time period, these data were not used. Moreover, many other studies had urinary
3 Additional in vivo data on TCE or metabolites published after the PBPK modeling was completed (Sweeney et al..
2009')(Kim et al.. 2009: Liu et al.. 20091 were evaluated separately, and discussed in Appendix A.
This document is a draft for review purposes only and does not constitute Agency policy.
86 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
excretion data, so this exclusion should have minimal impact. Several data sets not included by
Hack et al. (2006) were used here. Of particular importance was the inclusion of TCA and
TCOH dosing data from Abbas et al. (1997), Green and Prout (1985), Larson and Bull (1992a),
and Templin et al. (1993). A substantial amount of data is available in rats, so some data that
appeared to be redundant were excluded from the calibration set and saved for comparison with
posterior predictions (a "validation" set). In particular, those used for "validation" are one
closed-chamber experiment (Andersen et al., 1987b), several data sets with only TCE blood data
(D'Souza et al., 1985; Jakobson et al., 1986; Lee et al., 1996), and selected time courses from
Fisher et al. (1991) and Lee et al. (2000a; 2000b), and one unpublished data set (Bruckner et al.,
unpublished). The
Table 3-33. PBPK model-based dose-metrics
Abbreviation
Description
ABioactDCVCBW3
4
Amount of DCVC bioactivated in the kidney (mg) per unit body weight74
(kgy4)
ABioactDCVCKid
Amount of DCVC bioactivated in the kidney (mg) per unit kidney mass
(kg)
AMetGSHB W 3 4
Amount of TCE conjugated with GSH (mg) per unit body weight 4 (kg 4)
AMetLivlBW34
Amount of TCE oxidized in the liver per unit body weight74 (kgy4)
AMetLivOtherB W 3
4
Amount of TCE oxidized to metabolites other than TCA and TCOH in the
liver (mg) per unit body weight74 (kg3/4)
AMetLivOtherLiv
Amount of TCE oxidized to metabolites other than TCA and TCOH in the
liver (mg) per unit liver mass (kg)
AMetLngB W 3 4
Amount of TCE oxidized in the respiratory tract (mg) per unit body
weight74 (kg3/4)
AMetLngResp
Amount of TCE oxidized in the respiratory tract (mg) per unit respiratory
tract tissue mass (kg)
AUCCBld
Area under the curve of the venous blood concentration of TCE
(mg-hour/L)
AUCCTCOH
Area under the curve of the blood concentration of TCOH (mg-hour/L)
AUCLivTCA
Area under the curve of the liver concentration of TCA (mg-hour/L)
TotMetabBW34
Total amount of TCE metabolized (mg) per unit body weight74 (kg3/4)
T otOxMetabB W 3 4
Total amount of TCE oxidized (mg) per unit body weight74 (kg'74)
TotTCAInBW
Total amount of TCA produced (mg) per unit body weight (kg)
This document is a draft for review purposes only and does not constitute Agency policy.
87 DRAFT—DO NOT CITE OR QUOTE
-------
Table 3-34. Rodent studies with pharmacokinetic data considered for analysis
Reference
Species
(strain)
Sex
TCE exposures
Other
exposures
Calibration
Validation
Not
used
Comments
Mouse studies
Abbas et al. (1996)
Mouse
(B6C3F1)
M
--
CH i.v.
V
CH not in model.
Abbas and Fisher
(1997)
Mouse
(B6C3F1)
M
Oral (corn oil)
--
Va
Abbas et al. (1997)
Mouse
(B6C3F1)
M
--
TCOH, TCA
i.v.
V
Barton etal. (1999)
Mouse
(B6C3F1)
M
--
DCA i.v. and
oral (aqueous)
V
DCA not in model.
Birneretal. (1993)
Mouse
(NMRI)
M+F
Gavage
--
V
Only urine concentrations
available, not amount.
Fisher and Allen,
(1993)
Mouse
(B6C3F1)
M+F
Gavage (corn oil)
--
V
Fisher etal. (1991)
Mouse
(B6C3F1)
M+F
Inhalation
--
Va
Green and Prout
(1985)
Mouse
(B6C3F1)
M
Gavage (corn oil)
TCA i.v.
V
Greenberg et al.
(1999)
Mouse
(B6C3F1)
M
Inhalation
--
Va
Larson and Bull
(1992a)
Mouse
(B6C3F1)
M
DCA, TCA oral
(aqueous)
V
Only data on TCA dosing
was used, since DCA is not
in the model.
Larson and Bull
(1992b)
Mouse
(B6C3F1)
M
Oral (aqueous)
--
V
Merdink et al.
(1998)
Mouse
(B6C3F1)
M
i.v.
CH i.v.
V
Only data on TCE dosing
was used, since CH is not in
the model.
-------
Table 3-34. Rodent studies with pharmacokinetic data considered for analysis (continued)
Reference
Species
(strain)
Sex
TCE exposures
Other
exposures
Calibration
Validation
Not
used
Comments
Prout et al. (1985)
Mouse
(B6C3F1,
Swiss)
M
Gavage (corn oil)
Va
Templin et al.
(1993)
Mouse
(B6C3F1)
M
Oral (aqueous)
TCA oral
Va
Rat studies
Andersen et al.
(1997)
Rat (F344)
M
Inhalation
--
Barton etal. (1995)
Rat (S-D)
M
Inhalation
V
Initial chamber
concentrations unavailable,
so not used.
Bernauer et al.
(1996)
Rat (Wistar)
M
Inhalation
--
Va
Birneretal. (1993)
Rat (Wistar,
F344)
M+F
Gavage (ns)
--
V
Only urine concentrations
available, not amount.
Birneretal. (1997)
Rat (Wistar)
M+F
DCVC i.v.
V
Single dose, route does not
recapitulate how DCVC is
formed from TCE, excreted
NAcDCVC ~100-fold
greater than that from
relevant TCE exposures
(Bernauer et al., 1996).
Bruckner et al.
unpublished
Rat (S-D)
M
Inhalation
V
Not published, so not used
for calibration. Similar to
Keys et al. (2003) data.
Dallas etal. (1991)
Rat (S-D)
M
Inhalation
--
V
D'Souza et al.
(1985)
Rat (S-D)
M
i.v., oral
(aqueous)
V
Only TCE blood
measurements, and > 10-fold
greater than other similar
studies.
Fisher etal. (1989)
Rat (F344)
F
Inhalation
--
V
-------
L/l
K>
Table 3-34. Rodent studies with pharmacokinetic data considered for analysis (continued)
Reference
Species
(strain)
Sex
TCE exposures
Other
exposures
Calibration
Validation
Not
used
Comments
Fisher etal. (1991)
Rat (F344)
M+F
Inhalation
--
Va
V
Experiment with blood only
data not used for calibration.
Green and Prout
(1985)
Rat
(Osborne-Me
ndel)
M
Gavage (corn oil)
TCA gavage
(aqueous)
V
Hissink et al.
(2002)
Rat (Wistar)
M
Gavage (corn oil),
i.v.
--
V
Jakobson et al.
(1986)
Rat (S-D)
F
Inhalation
Various
pretreatments
(oral)
V
Pretreatments not included.
Only blood TCE data
available.
Kaneko et al.
(1994a)
Rat (Wistar)
M
Inhalation
Ethanol
pretreatment
(oral)
V
Pretreatments not included.
Keys et al. (2003)
Rat (S-D)
M
Inhalation,
oral (aqueous),
i.a.
V
Kimmerle and
Eben (1973a)
Rat (Wistar)
M
Inhalation
--
V
Larson and Bull
(1992a)
Rat (F344)
M
DCA, TCA oral
(aqueous)
V
Only TCA dosing data used,
since DCA is not in the
model.
Larson and Bull
(1992b)
Rat (S-D)
M
Oral (aqueous)
--
Va
Lash et al. (2006)
Rat (F344)
M+F
Gavage (corn oil)
--
V
Highly inconsistent with
other studies.
Lee et al. (1996)
Rat (S-D)
M
Arterial, venous,
portal, stomach
injections
V
Only blood TCE data
available.
Lee et al. (2000a;
2000b)
Rat (S-D)
M
Stomach
injection, i.v., p.v.
p-nitrophenol
pretreatment
(i.a.)
V
V
Pretreatments not included.
Only experiments with blood
and liver data used for
calibration.
3^
rs
sj
3
a
>!
&
fs*
si
O o
Grj
>!
o
a
a
s
§•
>!
a
o
0
&
fe
H
1
o
o
2; . _
H J
H S.
W|
p
o
c
o
H
W
rs
o
5
Grj
-------
Table 3-34. Rodent studies with pharmacokinetic data considered for analysis (continued)
Reference
Species
(strain)
Sex
TCE exposures
Other
exposures
Calibration
Validation
Not
used
Comments
Merdink et al.
(1999)
Rat (F344)
M
--
CH, TCOH i.v.
V
TCOH dosing used; CH not
in model.
Poet et al. (2000)
Rat (F344)
M
Dermal
--
V
Dermal exposure not in
model.
Prout et al. (1985)
Rat
(Osborne-Me
ndel, Wistar)
M
Gavage (corn oil)
Va
Saghir et al. (2002)
Rat (F344)
M
--
DCA i.v., oral
(aqueous)
V
DCA not in model.
Simmons et al.
(2002)
Rat
(Long-Evans)
M
Inhalation
--
V
Stenner et al.
(1997)
Rat (F344)
M
intraduodenal
TCOH, TCA
i.v.
V
Templin et al.
(1995b)
Rat (F344)
M
Oral (aqueous)
--
Va
Thrall et al. (2000)
Rat (F344)
M
i.v., i.p.
with toluene
V
Only exhaled breath data
available from i.v. study, i.p.
dosing not in model.
Yu et al. (2000)
Rat (F344)
M
-
TCA i.v.
V
a Part or all of the data in the study was used for calibration in Hack et al. (2006).
i.a. = intra-arterial, i.p. = intraperitoneal, i.v. = intravenous, p.v. = intraperivenous.
-------
Table 3-35. Human studies with pharmacokinetic data considered for analysis
Reference
Species
(number of
individuals)
Sex
TCE
exposures
Other
exposures
Calibration
Validation
Not
used
Comments
Bartonicek (1962)
Human (n = 8)
M+F
Inhalation
V
Sparse data, so not included for
calibration to conserve computational
resources.
Bernauer et al. (1996)
Human
M
Inhalation
--
Va
Grouped data, but unique in that
includes NAcDCVC urine data.
Bloemenetal. (2001)
Human (n = 4)
M
Inhalation
V
Sparse data, so not included for
calibration to conserve computational
resources.
Chiu et al. (2007)
Human (n = 6)
M
Inhalation
-
V
Ertleetal. (1972)
Human
M
Inhalation
CH oral
V
Very similar to Muller data.
Fernandez et al. (1977)
Human
M
Inhalation
-
V
Fisher etal. (1998)
Human
(n = 17)
M+F
Inhalation
--
Va
Kimmerle and Eben
(1973b)
Human
(n = 12)
M+F
Inhalation
--
V
Lapare et al. (1995)
Human (n = 4)
M+F
Inhalation
Complex exposure patterns, and only
grouped data available for urine, so
used for validation.
Lash etal. (1999b)
Human
M+F
Inhalation
V
Grouped only, but unique in that
DCVG blood data available (same
individuals as Fisher et al. (1998)].
Monster et al. (1976)
Human (n = 4)
M
Inhalation
--
Experiments with exercise not
included.
Monster etal. (1979)
Human
M
Inhalation
-
Va
Grouped data only.
Muller et al. (1972)
Human
ns
Inhalation
--
V
Same data also included in Muller
etal. (1975).
-------
Table 3-35. Human studies with pharmacokinetic data considered for analysis (continued)
Reference
Species
(number of
individuals)
Sex
TCE
exposures
Other
exposures
Calibration
Validation
Not
used
Comments
Muller et al. (1974)
Human
M
Inhalation
CH, TCA,
TCOH oral
V
Va
TCA and TCOH dosing data used for
calibration, since it is rare to have
metabolite dosing data. TCE dosing
data used for validation, since only
grouped data available. CH not in
model.
Muller et al. (1975)
Human
M
Inhalation
Ethanol oral
Va
Grouped data only.
Paykoc et al. (1945)
Human (n = 3)
ns
-
TCA i.v.
V
Poet et al. (2000)
Human
M+F
Dermal
--
Dermal exposure not in model.
Sato et al. (1977)
Human
M
Inhalation
-
V
Stewart et al. (1970)
Human
ns
Inhalation
-
Va
Treibigetal. (1976)
Human
ns
Inhalation
-
Va
Vesterberg and Astrand
(1976)
Human
M
Inhalation
--
V
All experiments included exercise, so
were not included.
"Part or all of the data in the study was used for calibration in Hack et al. (2006).
b Grouped data from this study was used for calibration in Hack et al. (2006), but individual data was used here.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Andersen et al. (1987b) data were selected randomly from the available closed-chamber data,
while the other data sets were selected because they were unpublished or because they were
more limited in scope (e.g., TCE blood only) and so were not as efficient for use in the
computationally-intensive calibration stage. As with the mouse analyses, TCA and TCOH
dosing data were incorporated to better calibrate those pathways.
The human pharmacokinetic database of controlled exposure studies is extensive but also
more complicated. For the majority of the studies, only grouped or aggregated data were
available, and most of those data were saved for "validation" since there remained a large
number of studies for which individual data were available. However, some data that may be
uniquely informative are only available in grouped form, in particular DCVG blood
concentrations, NAcDCVC urinary excretion, and data from TCA and TCOH dosing. While
there are analytic uncertainties as to the DCVG blood measurements, discussed above in Section
3.3.3.2.1, they were nonetheless included here because they are the only in vivo data available on
this measurement in humans. The uncertainty associated with their use is discussed below (see
Section 3.5.7.3.2).
In addition, several human data sets, while having individual data, involved sparse
collection at only one or a few time points per exposure (Bartonicek, 1962; Bloemen et al., 2001)
and were subsequently excluded to conserve computational resources. Lapare et al. (1995),
which involved multiple, complex exposure patterns over the course of a month and was missing
the individual urine data, was also excluded due to the relatively low amount of data given the
large computational effort required to simulate it. Several studies also investigated the effects of
exercise during exposure on human TCE toxicokinetics. The additional parameters in a model
including exercise would include those for characterizing the changes in cardiac output, alveolar
ventilation, and regional blood flow as well as their interindividual variability, and would have
further increased the computational burden. Therefore, it was decided that such data would be
excluded from this analysis. Even with these exclusions, data on a total of 42 individuals, some
involving multiple exposures, were included in the calibration.
3.5.5.1.2. Updated Hierarchical Population Statistical Model and Prior Distributions
While the individual animals of a common strain and sex within a study are likely to vary
to some extent, this variability was not included as part of the hierarchical population model for
several reasons. First, generally, only aggregated pharmacokinetic data (arithmetic mean and
standard deviation or standard error) are available from rodent studies. While methods exist for
addressing between-animal variability with aggregated data (e.g., Chiu and Bois, 2007), they
This document is a draft for review purposes only and does not constitute Agency policy.
94 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
require a higher level of computational intensity. Second, dose-response data are generally also
only separated by sex and strain, and otherwise aggregated. Thus, in analyzing dose-response
data (see Chapter 5), one usually has no choice but to treat all the animals in a particular study of
a particular strain and sex as identical units. In the Hack et al. (2006) model, each simulation
was treated as a separate observational unit, so different dosing levels within the same study
were treated separately and assigned different PBPK model parameters. However, the animals
within a study are generally inbred and kept under similarly controlled conditions, whereas
animals in different studies—even if of the same strain and sex—likely have differences in
genetic lineage, diet, and handling. Thus, animals within a study are likely to be much more
homogeneous than animals between studies. As a consequence, in the revised model, for
rodents, different animals of the same sex and strain in the same study (or series of studies
conducted simultaneously) were treated as identical, and grouped together as a single "subject."
Thus, the predictions from the population model in rodents simulate "average" pharmacokinetics
for a particular "lot" of rodents of a particular species, strain, and sex. Between-animal
variability is not explicitly modeled, but it is incorporated in a "residual" error term as part of the
likelihood function (see Appendix A, Section A.4.3.4). Therefore, a high degree of within-study
variability would be reflected in a high posterior value in the variance of the residual-error.
In humans, however, interindividual variability is of interest, and, furthermore,
substantial individual data are available in humans. However, in some studies, the same
individual was exposed more than once, so those data should be grouped together (in the Hack
et al. [2006] model, they were treated as different "individuals"). Because the primary interest
here is chronic exposure, and because it would add substantially to the computational burden,
interoccasion variability—changes in pharmacokinetic parameters in a single individual over
time—is not addressed. Therefore, each individual is considered a single "subject," and the
predictions from the population model in humans are the "average" across different occasions for
a particular individual (adult). Between-occasion variability is not explicitly modeled, but it is
incorporated in a "residual" error term as part of the likelihood function (see Appendix A,
Section A.4.3.4). Therefore, a high degree of between-occasion variability would be reflected in
a high posterior value in the variance of the residual-error.
As discussed in Section 3.3.3.1, sex and (in rodents) strain differences in oxidative
metabolism were modest or minimal. While some sex-differences have been noted in GSH
metabolism (see Sections 3.3.3.2.7-3.3.3.2.8), almost all of the available in vivo data is in males,
making it more difficult to statistically characterize that difference with PBPK modeling.
Therefore, within a species, different sexes and (in rodents) strains were considered to be drawn
This document is a draft for review purposes only and does not constitute Agency policy.
95 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
from a single, species-level population. For humans, each individual was considered to be drawn
from a single (adult) human population.
Thus, from here forward, the term "subject" will be used to refer to both a particular "lot"
of a particular rodents' species, strain, and sex for, and a particular human individual. The term
"population" will, therefore, refer to the collection of rodent "lots" of the same species and the
collection of human individuals.
Figure A-l in Appendix A illustrates the hierarchical structure. Informative prior
distributions reflecting the uncertainty in the population mean and variance, detailed in
Appendix A, were updated from those used in Hack et al. (2006) based on an extensive analysis
of the available literature. The population variability of the scaling parameter across subjects is
assumed to be distributed as a truncated normal distribution, a standard assumption in the
absence of specific data suggesting otherwise. Because of the truncation of extreme values, the
sensitivity to this choice is expected to be small as long as the true underlying distribution is
uni-modal and symmetric. In addition, most scaling parameters, being strictly positive in their
original units, were log-transformed—so these parameters have lognormal distributions in their
original units. The uncertainty distribution for the population parameters was assumed to be a
truncated normal distribution for population mean parameters and an inverse gamma distribution
for population variance parameters—both standard choices in hierarchical models.
Section 3.5.5.3, next, discusses specification of prior distributions in the case where no data
independent of the calibration data exist.
3.5.5.1.3. Use of Interspecies Scaling to Update Prior Distributions in the Absence of Other
Data
For many metabolic parameters, little or no in vitro or other prior information is available
to develop prior distributions. Initially, for such parameters, noninformative priors in the form of
log-uniform distributions with a range spanning at least 104 were specified. However, in the
time available for analysis (up to about 100,000 iterations), only for the mouse did all these
parameters achieve adequate convergence. This suggests that some of these parameters are
poorly identified for the rat and human. Additional preliminary runs indicated replacing the
log-uniform priors with lognormal priors and/or requiring more consistency between species
could improve identifiability sufficiently for adequate convergence. However, an objective
method of "centering" the lognormal distributions that did not rely on the in vivo data (e.g., via
visual fitting or limited optimization) being calibrated against was necessary in order to
minimize potential bias.
This document is a draft for review purposes only and does not constitute Agency policy.
96 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
Therefore, the approach taken was to consider three species sequentially, from mouse to
rat to human, and to use interspecies scaling to update the prior distributions across species. This
sequence was chosen because the models are essentially "nested" in this order, the rat model
adds to the mouse model the "downstream" GSH conjugation pathways, and the human model
adds to the rat model the intermediary DCVG compartment. Therefore, for those parameters
with little or no independent data only, the mouse posteriors were used to update the rat priors,
and both the mouse and rat posteriors were used to update the human priors. Table 3-36 contains
a list of the parameters for which this scaling was used to update prior distributions. The scaling
relationship is defined by the "scaled parameters" listed in Appendix A (see Section A.4.1,
Table A-4), and generally follows standard practice. For instance, Vmax and clearance rates
scale by body weight to the 3/4 power, whereas KM values are assumed to not scale, and rate
constants (inverse time units) scale by body weight to the -Vi power.
The scaling model is given explicitly as follows. If 0, are the "scaled" parameters
(usually also natural-log-transformed) that are actually estimated, and A is the "universal"
(species-independent) parameter, then 0, = A + eh where 8, is the species-specific "departure"
from the scaling relationship, assumed to be normally distributed with variance oe . Therefore,
the mouse model gives an initial estimate of "A," which is used to update the priordistribution
for 9r = A + sr in the rat. The rat and mouse together then give a "better" estimate of A, which is
used to update the prior distribution for 0/, = A + s/, in the human, with the assumed distribution
for 8h- The mathematical details are given in Appendix A, but three key points in this model are
worth noting here:
• It is known that interspecies scaling is not an exact relationship, and that, therefore, in
any particular case it may either over- or underestimate. Therefore, the variance in the
new priors reflect a combination of (1) the uncertainty in the "previous" species'
posteriors as well as (2) a "prediction error" that is distributed lognormally withgeometric
standard deviation (GSD) of 3.16-fold, so that the 95% confidence range about the
central estimate spans 100-fold. This choice was dictated partially by practicality, as
larger values of the GSD used in preliminary runs did not lead to adequate convergence
within the time available for analysis.
• The rat posterior is a product of its prior (which is based on the mouse posterior) and its
likelihood. Therefore, using the rat and mouse posteriors together to update the human
priors would use the mouse posterior "twice." Therefore, the rat posterior is
disaggregated into its prior and its likelihood using a lognormal approximation (since the
prior is lognormal), and only the (approximate) likelihood is used along with the mouse
posterior to develop the human prior.
• The model transfers the marginal distributions for each parameter across species, so
correlations between parameters are not retained. This is a restriction on the software
This document is a draft for review purposes only and does not constitute Agency policy.
97 DRAFT—DO NOT CITE OR QUOTE
-------
1 used for conducting MCMC analyses. However, assuming independence will lead to a
2 "broader" joint distribution, given the same marginal distributions. Therefore, this
3 assumption tends to reduce the weight of the interspecies scaling as compared to the
4 species-specific calibration data.
This document is a draft for review purposes only and does not constitute Agency policy.
98 DRAFT—DO NOT CITE OR QUOTE
-------
a 2
o S.
§ ^
>! St
S ^5
S o
oq o
^ s
a c>
.cj ^
^ fc
§•
>1
Table 3-36. Parameters for which scaling from mouse to rat, or from mouse and rat to human, was used to
update the prior distributions
Parameter with no or highly uncertain a priori data
Mouse
Rat
Rat ->
Human
Mouse+
Rat
Human
Comments
Respiratory lumen->tissue diffusion flow rate
V
V
No a priori information
TCOG body/blood partition coefficient
V
V
Prior centered on TCOH data, but highly uncertain
TCOG liver/body partition coefficient
V
V
Prior centered on TCOH data, but highly uncertain
Fraction of hepatic TCE oxidation not to TCA+TCOH
V
V
No a priori information
VMax for hepatic TCE GSH conjugation
V
Rat data on at 1 and 2 mM. Human data at more
concentrations, so VMax and KM can be estimated
Km for hepatic TCE GSH conjugation
V
VMax for renal TCE GSH conjugation
V
Rat data on at 1 and 2 mM. Human data at more
concentrations, so VMax and KM can be estimated
Km for renal TCE GSH conjugation
V
VMax for Tracheo-bronchial TCE oxidation
V
V
Prior based on activity at a single concentration
Km for Tracheo-bronchial TCE oxidation
V
V
No a priori information
Fraction of respiratory oxidation entering systemic circulation
V
V
No a priori information
VMAX for hepatic TCOH^TCA
V
V
No a priori information
Km for hepatic TCOH^TCA
V
V
No a priori information
VMAX for hepatic TCOH^TCOG
V
V
No a priori information
Km for hepatic TCOH^TCOG
V
V
No a priori information
Rate constant for hepatic TCOH->other
V
V
No a priori information
Rate constant for TCA plasma->urine
V
V
Prior centered at GFR, but highly uncertain
Rate constant for hepatic TCA->other
V
V
No a priori information
Rate constant for TCOG liver->bile
V
V
No a priori information
Lumped rate constant for TCOG bile->TCOH liver
V
V
No a priori information
Rate constant for TCOG->urine
V
V
Prior centered at GFR, but highly uncertain
Lumped rate constant for DCVC->Urinary NAcDCVC
V
Not included in mouse model
Rate constant for DCVC bioactivation
V
Not included in mouse model
<3
vo
vo
O
O
GFR = glomerular filtration rate.
See Appendix A, Table A-4 for scaling relationships.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
To summarize, in order to improve rate of the convergence of the MCMC analyses in rats
and humans, a sequential approach was used for fitting scaling parameters without strong prior
species-specific information. In particular, an additional assumption was made that across
species, these scaling parameters were, in absence of other information, expected to have a
common underlying value. These assumptions are generally based on allometric scaling
principles—with partition coefficients and concentrations scaling directly and rate constants
scaling by BW ~4 (so clearances and maximum metabolic capacities would scale by BW/4).
These assumptions are used consistently throughout the parameter calibration process.
Therefore, after running the mouse model, the posterior distribution for these parameters was
used, with an additional error term, as priors for the rat model. Subsequently, after the mouse
and rat model were run, their posterior distributions were combined, with an additional error
term, to use a priors for the human model. With this methodology for updating the prior
distributions, adequate convergence was achieved for the rat and human after
110,000-140,000 iterations (discussed further below).
3.5.5.1.4. Implementation
The PBPK model was coded in for use in the MCSim software (version 5.0.0), which was
developed particularly for implementing MCMC simulations. As a quality control (QC) check,
results were checked against the original Hack et al. (2006) model, with the original structures
restored and parameter values made equivalent, and the results were within the error tolerances
of the ordinary differential equation (ODE) solver after correcting an error in the Hack et al.
(2006) model for calculating the TCA liver plasma flow. In addition, the model was translated to
MatLab (version 7.2.0.232) with simulation results checked and found to be within the error
tolerances of the ODE solver used ("odel5s"). Mass balances were also checked using the
baseline parameters, as well as parameters from preliminary MCMC simulations, and found to
be within the error tolerances of the ODE solver. Appendix A contains the MCSim model code.
3.5.6. Evaluation of Updated Physiologically Based Pharmacokinetic (PBPK) Model
3.5.6.1.1. Convergence
As in previous similar analyses (Bois, 2000a, b; David et al., 2006; Gelman et al., 1996;
Hack et al., 2006), the potential scale reduction factor "i?" is used to determine whether different
This document is a draft for review purposes only and does not constitute Agency policy.
100 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
independent MCMC chains have converged to a common distribution. The R diagnostic is
calculated for each parameter in the model, and represents the factor by which the standard
deviation or other measure of scale of the posterior distribution (such as a confidence interval
[CI]) may potentially be reduced with additional samples (Gelman et al., 2004). This
convergence diagnostic declines to 1 as the number of simulation iterations approaches infinity,
so values close to 1 indicate approximate convergence, with values of 1.1 and below commonly
considered adequate (Gelman et al., 2004). However, as an additional diagnostic, the
convergence of model dose-metric predictions was also assessed. Specifically, dose-metrics for
a number of generic exposure scenarios similar to those used in long-term bioassays were
generated, and their natural log (due to their approximate lognormal posterior distributions) was
assessed for convergence using the potential scale reduction factor "R" This is akin to the idea
of utilizing sensitivity analysis so that effort is concentrated on calibrating the most sensitive
parameters for the purpose of interest. In addition, predictions of interest which do not
adequately converge can be flagged as such, so that the statistical uncertainty associated with the
limited sample size can be considered.
The mouse model had the most rapid reduction in potential scale reduction factors.
Initially, four chains of 42,500 iterations each were run, with the first 12,500 discarded as
"burn-in" iterations. The initial decision for determining "burn-in" was determined by visual
inspection. At this point, evaluating the 30,000 remaining iterations, all the population
parameters except for the Vmax for DCVG formation had R< 1.2, with only the first-order
clearance rate for DCVG formation and the Vmax and Km for TCOH glucuronidation having
R> 1.1. For the samples used for inference, all of these initial iterations were treated as
"burn-in" iterations, and each chain was then restarted and run for an additional
68,700-71,400 iterations (chains were terminated at the same time, so the number of iterations
per chains was slightly different). For these iterations, all values of R were <1.03. Dose-metric
predictions calculated for exposure scenarios 10-600 ppm either continuously or 7 hour/day,
5 day/week and 10-3,000 mg/kg-day either continuously or by gavage 5 day/week. These
predictions were all adequately converged, with all values of R < 1.03.
As discussed above, for parameters with little or no a priori information, the posterior
distributions from the mouse model were used to update prior distributions for the rat model,
accounting for both the uncertainty reflected in the mouse posteriors as well as the uncertainty in
interspecies extrapolation. Four chains were run to 111,960-128,000 iterations each (chains
were terminated at the same time and run on computers with slightly different processing speeds,
so the number of iterations per chains was slightly different). As is standard, about the
first "half' of the chains—i.e., the first 64,000 iterations—were discarded as "burn-in" iterations,
This document is a draft for review purposes only and does not constitute Agency policy.
101 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
and the remaining iterations were used for inferences. For these remaining iterations, the
diagnostic R was <1.1 for all population parameters except the fraction of oxidation not
producing TCA or TCOH (R = 1.44 for population mean, R= 1.35 for population variance), the
Km for TCOH -> TCA (R = 1.19 for population mean), the Vmax and Km for TCOH
glucuronidation (R = 1.23 and 1.12, respectively for population mean, and R= 1.13 for both
population variances), and the rate of "other" metabolism of TCOH (R = 1.29 for population
mean and R= 1.18 for population variance). Due to resource constraints, chains needed to be
stopped at this point. However, these are similar to the degree of convergence reported in Hack
et al. (2006). Dose-metric predictions calculated for two inhalation exposure scenarios
(10-600 ppm continuously or 7 hours/day, 5 day/week) and two oral exposure scenarios
(10-3,000 mg/kg-day continuously or by gavage 5 day/week).
All dose-metric predictions had R < 1.04, except for the amount of "other" oxidative
metabolism (i.e., not producing TCA or TCOH), which had R = 1.12-1.16, depending on the
exposure scenario. The poorer convergence of this dose-metric is expected given that a key
determining parameter, the fraction of oxidation not producing TCA or TCOH, had the poorest
convergence among the population parameters.
For the human model, a set of four chains was run for 74,160-84,690 iterations using
"preliminary" updated prior distributions based on the mouse posteriors and preliminary runs of
the rat model. Once the rat chains were completed, final updated prior distributions were
calculated and the last iteration of the preliminary runs were used as starting points for the final
runs. The center of the final updated priors shifted by less than 25% of the standard deviation of
either the preliminary or revised priors, so that the revised median was between the
40th percentile and 60th percentile of the preliminary median, and vice versa. The standard
deviations changed by less than 5%. Therefore, the use of the preliminary chains as a starting
point should introduce no bias, as long as an appropriate burn-in period is used for the final runs.
The final chains were run for an additional 59,140-61,780 iterations, at which point, due
to resource constraints, chains needed to be stopped. After the first 20,000 iterations, visual
inspection revealed the chains were no longer dependent on the starting point. These iterations
were therefore discarded as "burn-in" iterations, and for the remaining -40,000 iterations used
for inferences. All population mean parameters had R < 1.1 except for the respiratory tract
diffusion constant (R = 1.20), the liver:blood partition coefficient for TCOG (R = 1.23), the rate
of TCE clearance in the kidney producing DCVG (R = 1.20), and the rate of elimination of
TCOG in bile (R= 1.46). All population variances also had R < 1.1 except for the variance for
the fraction of oxidation not producing TCOH or TCA (R= 1.10). Dose-metric predictions were
This document is a draft for review purposes only and does not constitute Agency policy.
102 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
assessed for continuous exposure scenarios at 1-60 ppm in air or 1-300 mg/kg-day orally.
These predictions were all adequately converged with all values of R < 1.02.
3.5.6.1.2. Evaluation of Posterior Parameter Distributions
Posterior distributions of the population parameters need to be checked as to whether
they appear reasonable given the prior distributions. Inconsistency between the prior and
posterior distributions may indicate insufficiently broad (i.e., due to overconfidence) or
otherwise incorrectly specified priors, a misspecification of the model structure (e.g., leading to
pathological parameter estimates), or an error in the data. As was done with the evaluation of
Hack et al. (2006) in Appendix A, parameters were flagged if the interquartile regions of their
prior and posterior distributions did not overlap.
Appendix A contains detailed tables of the "sampled" parameters, and their prior and
posterior distributions. Because these parameters are generally scaled one or more times to
obtain a physically meaningful parameter, they are difficult to interpret. Therefore, in
Tables 3-37-3-39, the prior and posterior population distributions for the PBPK model
parameters obtained after scaling are summarized. Since it is desirable to characterize the
contributions from both uncertainty in population parameters and variability within the
population, the following procedure is adopted. First, 500 sets of population parameters (i.e.,
population mean and variance for each scaling parameter) are either generated from the prior
distributions via Monte Carlo or extracted from the posterior MCMC samples—these represent
the uncertainty in the population parameters. To minimize autocorrelation, for the posteriors, the
samples were obtained by "thinning" the chains to the appropriate degree. From each of these
sets of population parameters, 100 sets of "subject"-level parameters were generated by Monte
Carlo—each of these represents the population variability, given a particular set of population
parameters. Thus, a total of 50,000 subjects, representing 100 (variability) each for 500 different
populations (uncertainty), were generated. For each of the 500 populations, the scaling
parameters are converted to PBPK model parameters, and the population median and GSD is
calculated—representing the central tendency and variability for that population. Then, the
median and the 95% CIs for the population median and GSD are calculated, and presented in the
tables that follow. Thus, these tables summarize separately the uncertainty in population
distribution as well as the variability in the population, while also accounting for correlations
among the population-level parameters. Finally, Table 3-40 shows the change in the CI in the
population median for the PBPK model parameters between the prior and posterior distributions,
This document is a draft for review purposes only and does not constitute Agency policy.
103 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
as well as the shift in the central estimate (median) of the population median PBPK model
parameter.
The prior and posterior distributions for most physiological parameters were similar. The
posterior distribution was substantially narrower (i.e., less uncertainty) than the prior distribution
only in the case of the diffusion rate from the respiratory lumen to the respiratory tissue, which
also was to be expected given the very wide, noninformative prior for that parameter.
For distribution parameters, there were only relatively minor changes between prior and
posterior distributions for TCE and TCOH partition coefficients. The posterior distributions for
several TCA partition coefficients and plasma binding parameters were substantially narrower
than their corresponding priors, but the central estimates were similar, meaning that values at the
high and low extremes were not likely. For TCOG as well, partition coefficient posterior
distributions were substantially narrower, which was expected given the greater uncertainty in
the prior distributions (TCOH partition coefficients were used as a proxy).
Again, posterior distributions indicated that the high and low extremes were not likely.
Finally, posterior distribution for the distribution volume for DCVG was substantially narrower
than the prior distribution, which only provided a lower bound given by the blood volume. In
this case, the upper bounds were substantially lower in the posterior.
Posterior distributions for oral absorption parameters in mice and rats (there were no oral
studies in humans) were also informed by the data, as reflected in their being substantially more
narrow than the corresponding priors. Finally, with a few exceptions, TCE and metabolite
kinetic parameters showed substantially narrower posterior distributions than prior distributions,
indicating that they were fairly well specified by the in vivo data. The exceptions were the Vmax
for hepatic oxidation in humans (for which there was substantial in vitro data) and the Vmax for
respiratory metabolism in mice and rats (although the posterior distribution for the Km for this
pathway was substantially narrower than the corresponding prior).
However, for some parameters, the posterior distributions in the population medians had
CIs greater than 100-fold. In mice, the absorption parameter for TCA still had posterior CI of
400-fold, reflecting the fact that the absorption rate is poorly estimated from the few available
studies with TCA dosing. In addition, mouse metabolism parameters for GSH conjugation have
posterior CIs greater than 10,000-fold, reflecting the lack of any direct data on GSH conjugation
in mice. In rats, two parameters related to TCOH and TCOG had CIs between 100- and
1,000-fold, reflecting the poor identifiability of these parameters given the available data. In
humans, only the oral absorption parameters for TCA and TCOH had CIs greater than 100-fold,
reflecting the fact that the absorption rate is poorly estimated from the few available studies with
TCOH and TCA dosing.
This document is a draft for review purposes only and does not constitute Agency policy.
104 DRAFT—DO NOT CITE OR QUOTE
-------
1 In terms of general consistency between prior and posterior distributions, in most cases,
2 the central estimate of the population median shifted by less than threefold. In almost all the
3 cases that the shift was greater (see bold entries in Table 3-40), the prior distribution had a wide
4 distribution, with CI greater (sometimes substantially greater) than 100-fold. The only exception
This document is a draft for review purposes only and does not constitute Agency policy.
105 DRAFT—DO NOT CITE OR QUOTE
-------
s 2 ho
& ?
rs ^ <*>
o ' a,
a g"
5 ^
s
s o S
^ | §
Oq o %¦
Uo
^ a §•
^ S vS
o a. ~b
O
^ to ^
Table 3-37. Prior and posterior uncertainty and variability in mouse PBPK model parameters
Parameter description
PBPK
Parameter
Prior Population
Median: Median
(2.5%, 97.5%)
Posterior Population
Median: Median (2.5%,
97.5%)
Prior Population GSD:
Median (2.5%, 97.5%)
Posterior Population
GSD: Median (2.5%,
97.5%)
Cardiac output (L/hour)
QC
0.84 (0.59, 1.2)
1 (0.79, 1.3)
1.17(1.1, 1.4)
1.35 (1.15, 1.54)
Alveolar ventilation
(L/hour)
QP
2.1 (1.3,3.5)
2.1 (1.5,2.7)
1.27(1.17, 1.54)
1.45 (1.28, 1.66)
Scaled fat blood flow
QFatC
0.07(0.03,0.11)
0.072(0.044,0.1)
1.65 (1.22, 2.03)
1.64 (1.3, 1.99)
Scaled gut blood flow
QGutC
0.14(0.11,0.17)
0.16(0.14,0.17)
1.15 (1.09, 1.19)
1.12(1.07, 1.19)
Scaled liver blood flow
QLivC
0.02 (0.016, 0.024)
0.021 (0.017, 0.024)
1.15 (1.09, 1.19)
1.15 (1.09, 1.19)
Scaled slowly perfused
blood flow
QSlwC
0.22(0.14,0.29)
0.21 (0.15,0.28)
1.3 (1.15, 1.38)
1.3 (1.17, 1.39)
Scaled rapidly perfused
blood flow
QRapC
0.46 (0.37, 0.56)
0.45 (0.37, 0.52)
1.15 (1.11, 1.2)
1.17(1.12, 1.2)
Scaled kidney blood flow
QKidC
0.092 (0.054,0.13)
0.091 (0.064,0.12)
1.34 (1.14, 1.45)
1.34 (1.18, 1.44)
Respiratory lumen:tissue
diffusive clearance rate
(L/hour)
DResp
0.017 (3.2e-05, 15)
2.5 (1.4, 5.1)
1.37 (1.25, 1.62)
1.53 (1.37, 1.73)
Fat fractional
compartment volume
VFatC
0.071 (0.032,0.11)
0.089 (0.061,0.11)
1.59 (1.19, 1.93)
1.4 (1.19, 1.78)
Gut fractional
compartment volume
VGutC
0.049 (0.041, 0.057)
0.048 (0.042, 0.055)
1.11 (1.07, 1.14)
1.11 (1.08, 1.14)
Liver fractional
compartment volume
VLivC
0.054 (0.038, 0.071)
0.047 (0.037, 0.06)
1.22(1.12, 1.29)
1.23 (1.17, 1.3)
Rapidly perfused
fractional compartment
volume
VRapC
0.1 (0.087,0.11)
0.099(0.09,0.11)
1.08 (1.05, 1.11)
1.09(1.06, 1.11)
Fractional volume of
respiratory lumen
VRespLumC
0.0047 (0.004, 0.0053)
0.0047 (0.0041, 0.0052)
1.09(1.06, 1.12)
1.09(1.07, 1.12)
Fractional volume of
respiratory tissue
VRespEffC
7e-04 (6e-04, 0.00079)
7e-04 (0.00062, 0.00078)
1.09(1.06, 1.12)
1.1 (1.07, 1.12)
Kidney fractional
compartment volume
VKidC
0.017(0.015,0.019)
0.017(0.015,0.019)
1.08 (1.05, 1.11)
1.09(1.06, 1.11)
Blood fractional
compartment volume
VBldC
0.049 (0.042, 0.056)
0.048 (0.043, 0.054)
1.1 (1.06, 1.13)
1.1 (1.08, 1.13)
o
On
I
o
-------
s 2 ho
& ?
rs ^ <*>
o ' a,
a g"
5 ^
s
s o S
^ | §
Oq o %¦
Uo
^ a §•
^ S vS
o a. ~b
O
^ to ^
Table 3-37. Prior and posterior uncertainty and variability in mouse PBPK model parameters (contuined)
Parameter description
PBPK
Parameter
Prior Population
Median: Median
(2.5%, 97.5%)
Posterior Population
Median: Median (2.5%,
97.5%)
Prior Population GSD:
Median (2.5%, 97.5%)
Posterior Population
GSD: Median (2.5%,
97.5%)
Slowly perfused fractional
compartment volume
VSlwC
0.55 (0.5, 0.59)
0.54 (0.51,0.57)
1.05 (1.04, 1.07)
1.05 (1.04, 1.07)
Plasma fractional
compartment volume
VPlasC
0.026 (0.016, 0.036)
0.022 (0.016, 0.029)
1.24 (1.15, 1.35)
1.27 (1.19, 1.36)
TCA body fractional
compartment volume [not
incl. blood+liver]
VBodC
0.79 (0.77, 0.8)
0.79 (0.78, 0.81)
1.01 (1.01, 1.02)
1.01 (1.01, 1.02)
TCOH/G body fractional
compartment volume [not
incl. liver]
VBodTCOHC
0.84 (0.82, 0.85)
0.84 (0.83, 0.85)
1.01 (1.01, 1.02)
1.01 (1.01, 1.02)
TCE blood:air partition
coefficient
PB
15 (10, 23)
14 (11, 17)
1.22(1.12, 1.42)
1.44 (1.28, 1.53)
TCE fat:blood partition
coefficient
PFat
36 (21, 62)
36 (26, 49)
1.26(1.14, 1.52)
1.32 (1.16, 1.56)
TCE gut:blood partition
coefficient
PGut
1.9(0.89,3.8)
1.5 (0.94, 2.6)
1.36 (1.2, 1.75)
1.36 (1.2, 1.79)
TCE livenblood partition
coefficient
PLiv
1.7(0.89,3.5)
2.2 (1.3, 3.3)
1.37 (1.2, 1.75)
1.39 (1.21, 1.84)
TCE rapidly
per£used:blood partition
coefficient
PRap
1.8 (0.98, 3.7)
1.8(1.1,3)
1.37 (1.2, 1.76)
1.37 (1.2, 1.77)
TCE respiratory tissue:air
partition coefficient
PResp
2.7(1.2, 5)
2.5 (1.5,4.2)
1.36 (1.19, 1.78)
1.37 (1.19, 1.74)
TCE kidney :blood
partition coefficient
PKid
2.2 (0.96, 4.6)
2.6(1.7, 4)
1.36 (1.2, 1.77)
1.51 (1.25, 1.88)
TCE slowly
per£used:blood partition
coefficient
PSlw
2.4(1.2,4.9)
2.2 (1.4, 3.5)
1.38 (1.2, 1.78)
1.39(1.21, 1.8)
TCA blood:plasma
concentration ratio
TCAPlas
0.76 (0.4, 16)
1.1 (0.75, 1.8)
1.21 (1.09, 1.58)
1.23 (1.1, 1.73)
o
O
O
-------
s 2 ho
& ?
rs ^ <*>
o ' a,
a g"
5 ^
s
s o S
^ | §
Oq o %¦
Uo
^ a §•
^ S vS
o a. ~b
O
^ to ^
Table 3-37. Prior and posterior uncertainty and variability in mouse PBPK model parameters (contuined)
Parameter description
PBPK
Parameter
Prior Population
Median: Median
(2.5%, 97.5%)
Posterior Population
Median: Median (2.5%,
97.5%)
Prior Population GSD:
Median (2.5%, 97.5%)
Posterior Population
GSD: Median (2.5%,
97.5%)
Free TCAbody:blood
plasma partition
coefficient
PBodTCA
0.77 (0.27, 17)
0.87 (0.59, 1.5)
1.41 (1.23, 1.8)
1.39 (1.24, 1.9)
Free TCA livenblood
plasma partition
coefficient
PLivTCA
1.1 (0.36,21)
1.1 (0.64, 1.9)
1.41 (1.23, 1.8)
1.4 (1.24, 1.87)
Protein:TCA dissociation
constant (|imolc/L)
kDissoc
100 (13,790)
130 (24, 520)
2.44 (1.73, 5.42)
2.64 (1.75,5.45)
Maximum binding
concentration (|imolc/L)
Bmax
87 (9.6, 790)
140 (28, 690)
2.72 (1.92, 5.78)
2.88 (1.93,5.89)
TCOH body:blood
partition coefficient
PBodTCOH
1.1 (0.61,2.1)
0.89 (0.65, 1.3)
1.29(1.16, 1.66)
1.31 (1.17, 1.61)
TCOH livenbody
partition coefficient
PLivTCOH
1.3 (0.73,2.3)
1.9 (1.2,2.6)
1.3 (1.16, 1.61)
1.35 (1.18, 1.68)
TCOG body:blood
partition coefficient
PBodTCOG
0.95 (0.016, 77)
0.48 (0.18, 1.1)
1.36 (1.19, 2.05)
1.41 (1.22,2.19)
TCOG liver:body
partition coefficient
PLivTCOG
1.3 (0.019, 92)
1.3 (0.64, 2.6)
1.36(1.18, 2.13)
1.56 (1.28,2.52)
DCVG effective volume
of distribution
VDCVG
0.033 (0.0015, 15)
0.027 (0.0016,4.1)
1.28(1.08, 1.97)
1.31 (1.1,2.19)
TCE stomach absorption
coefficient (/hour)
kAS
1.7 (0.0049, 450)
1.7 (0.37, 13)
4.74 (2.29, 23.4)
4.28 (2.39, 13.4)
TCE stomach-duodenum
transfer coefficient (/hour)
kTSD
1.4 (0.043, 51)
4.5 (0.51,26)
3.84 (2.09, 10.6)
4.79 (2.53, 10.9)
TCE duodenum
absorption coefficient
(/hour)
kAD
1.2 (0.0024, 200)
0.27 (0.067, 1.6)
4.33 (2.14, 26)
4.17(2.34, 14.4)
TCA stomach absorption
coefficient (/hour)
kASTCA
0.63 (0.0027, 240)
4 (0.2, 74)
4.26 (2.27, 23.4)
5.15 (2.56, 22)
Vmax for hepatic TCE
oxidation (mg/hour)
Vmax
3.9 (1.4, 15)
2.5 (1.6,4.2)
2.02 (1.56, 2.85)
1.86 (1.59,2.47)
Km for hepatic TCE
oxidation (mg/L)
Km
34 (1.6, 620)
2.7(1.4, 8)
1.25 (1.15, 1.61)
2.08 (1.48,3.49)
o
oo
o
o
-------
s 2 ho
& ?
rs ^ <*>
o ' a,
a g"
5 ^
s
s o S
^ | §
Oq o %¦
Uo
^ a §•
^ S vS
o a. ~b
O
^ to ^
Table 3-37. Prior and posterior uncertainty and variability in mouse PBPK model parameters (contuined)
Parameter description
PBPK
Parameter
Prior Population
Median: Median
(2.5%, 97.5%)
Posterior Population
Median: Median (2.5%,
97.5%)
Prior Population GSD:
Median (2.5%, 97.5%)
Posterior Population
GSD: Median (2.5%,
97.5%)
Fraction of hepatic TCE
oxidation not to
TCA+TCOH
FracOther
0.43 (0.0018, 1)
0.023 (0.0037, 0.15)
1.23 (1,2.13)
1.49 (1.25,2.83)
Fraction of hepatic TCE
oxidation to TCA
FracTCA
0.086 (0.00022, 0.66)
0.13 (0.084,0.21)
1.48(1.12, 2.56)
1.4 (1.21, 1.96)
Vmax for hepatic TCE
GSH conjugation
(mg/hour)
VmaxDCVG
3.7 (0.0071,2800)
0.6 (0.01, 480)
1.55 (1.33,2.52)
1.61 (1.37,2.91)
Km for hepatic TCE GSH
conjugation (mg/L)
kmdcvg
250 (0.0029, 6500000)
2200(0.17, 2300000)
1.81 (1.47, 3.62)
1.93 (1.49,3.68)
Vmax for renal TCE GSH
conjugation (mg/hour)
Vmax KidDCVG
0.34 (0.00051, 180)
0.027 (0.0012, 13)
1.49 (1.26, 2.49)
1.54(1.28,2.72)
KM for renal TCE GSH
conjugation (mg/L)
KMKidDCVG
150 (0.0053, 6200000)
160 (0.078, 280000)
1.79 (1.43,3.45)
1.91 (1.5, 3.91)
Vmax for
tracheo-bronchial TCE
oxidation (mg/hour)
Vmax Clara
0.24 (0.03, 3.9)
0.42(0.1, 1.5)
2.32 (1.74, 3.66)
4.13 (2.27,6.79)
Km for tracheo-bronchial
TCE oxidation (mg/L)
KMClara
1.5 (0.0018, 630)
0.011 (0.0024, 0.09)
1.47 (1.25,2.58)
1.63 (1.28,5.02)
Fraction of respiratory
metabolism to systemic
circ.
FracLungSys
0.34 (0.0016, 1)
0.78 (0.18,0.99)
1.24(1,2.1)
1.11 (1, 1.72)
Vmax for hepatic
TCOH—>TCA (mg/hour)
Vmax TCOH
0.064 (1.4e-05, 380)
0.12(0.048,0.28)
1.5(1.24,2.61)
1.6 (1.28, 2.92)
Km for hepatic
TCOH—>TCA (mg/L)
kmtcoh
1.4 (0.00018, 5300)
0.92 (0.26, 2.7)
1.48(1.24, 2.41)
1.49 (1.26, 2.4)
Vmax for hepatic
TCOH—>TCOG
(mg/hour)
Vmax Glue
0.11 (1.3e-05, 310)
4.6 (1.9, 16)
1.48 (1.26, 2.53)
1.47(1.26,2.14)
Km for hepatic
TCOH—>TCOG (mg/L)
KmG1uc
1.8(0.0018,610)
30 (5.3, 130)
1.48 (1.25,2.48)
1.8(1.3,4.72)
Rate constant for hepatic
TCOH—mother (/hour)
kMetTCOH
0.19 (3.9e-05, 1400)
8.8 (1.9, 23)
1.47 (1.25,2.36)
1.54(1.26,2.92)
o
vo
O
O
-------
a
o
<
g
cs
w
o
^3
§•
>!
St
Cb
o
St
3
>1
>!
o
Co
£
Si
s
O
sis
§-
c>
Table 3-37. Prior and posterior uncertainty and variability in mouse PBPK model parameters (contuined)
Parameter description
PBPK
Parameter
Prior Population
Median: Median
(2.5%, 97.5%)
Posterior Population
Median: Median (2.5%,
97.5%)
Prior Population GSD:
Median (2.5%, 97.5%)
Posterior Population
GSD: Median (2.5%,
97.5%)
Rate constant for TCA
plasma—murine (/hour)
kUrnTCA
32 (0.38, 1700)
3.2 (1.2,7.1)
1.57 (1.34, 2.61)
1.84(1.44,2.94)
Rate constant for hepatic
TCA—mother (/hour)
kMetTCA
0.12 (4e-04, 130)
1.5 (0.63, 2.9)
1.48 (1.25,2.32)
1.51 (1.26,2.27)
Rate constant for TCOG
liver—>bile (/hour)
kBile
0.3 (4e-04, 160)
2.4 (0.74, 8.4)
1.48 (1.24, 2.29)
1.51 (1.26,2.39)
Lumped rate constant for
TCOG bile—>TCOH liver
(/hour)
kEHR
0.21 (0.00036, 150)
0.039(0.0026, 0.11)
1.47 (1.23,2.29)
1.53 (1.28,2.94)
Rate constant for
TCOG—mrine (/hour)
kUrnTCOG
1 (0.00015, 6200)
12 (2.6, 77)
1.71 (1.4, 3.13)
3.44 (1.89,9.49)
Rate constant for hepatic
DCVG—>DCVC (/hour)
kDCVG
0.24 (4e-04, 160)
0.81 (0.0033, 46)
1.48 (1.25,2.39)
1.52 (1.25,2.5)
Lumped rate constant for
DCVC—>urinary
NAcDCVC (/hour)
kNAT
0.29 (4e-04, 160)
0.37 (0.0021, 34)
1.5 (1.25,2.49)
1.53 (1.25,2.77)
Rate constant for DCVC
bioactivation (/hour)
kKidBioact
0.18 (4e-04, 150)
0.23 (0.0024, 33)
1.48 (1.25,2.51)
1.53 (1.25,3.03)
O
O
-------
s 2 ho
& ?
rs ^ <*>
o ' a,
a g"
5 ^
s
s o S
^ | §
Oq o %¦
Uo
^ a §•
^ S vS
o a. ~b
O
^ to ^
Table 3-38. Prior and posterior uncertainty and variability in rat PBPK model parameters
Parameter description
PBPK
Parameter
Prior Population
Median: Median
(2.5%, 97.5%)
Posterior Population
Median: Median (2.5%,
97.5%)
Prior Population GSD:
Median (2.5%, 97.5%)
Posterior Population
GSD: Median (2.5%,
97.5%)
Cardiac output (L/hour)
QC
5.3 (4.2,6.9)
6.1 (5.2,7.4)
1.12(1.07, 1.28)
1.26 (1.12, 1.36)
Alveolar ventilation
(L/hour)
QP
10(5.1, 18)
7.5 (5.8, 10)
1.32(1.18, 1.71)
1.52 (1.33, 1.84)
Scaled fat blood flow
QFatC
0.071 (0.032,0.11)
0.081 (0.06,0.1)
1.66(1.21,2.02)
1.5 (1.3, 1.86)
Scaled gut blood flow
QGutC
0.15(0.12,0.18)
0.17(0.15,0.19)
1.15 (1.09, 1.19)
1.13 (1.08, 1.18)
Scaled liver blood flow
QLivC
0.021 (0.017, 0.026)
0.022 (0.018, 0.025)
1.15 (1.09, 1.2)
1.15 (1.1, 1.19)
Scaled slowly perfused
blood flow
QSlwC
0.33 (0.21, 0.46)
0.31 (0.23,0.4)
1.31 (1.15, 1.4)
1.32 (1.22, 1.41)
Scaled rapidly perfused
blood flow
QRapC
0.28(0.15,0.42)
0.28 (0.18,0.36)
1.38 (0.0777, 1.72)
1.42 (0.0856, 1.75)
Scaled kidney blood flow
QKidC
0.14(0.12,0.16)
0.14(0.12,0.16)
1.11 (1.07, 1.14)
1.11 (1.08, 1.14)
Respiratory lumen:tissue
diffusive clearance rate
(L/hour)
DResp
9.9 (0.48, 85)
21 (9.5, 46)
1.41 (1.26, 1.77)
1.59(1.41, 1.9)
Fat fractional
compartment volume
VFatC
0.069 (0.031,0.11)
0.069 (0.046, 0.091)
1.61 (1.2, 1.93)
1.59 (1.34, 1.88)
Gut fractional
compartment volume
VGutC
0.032 (0.027, 0.037)
0.032 (0.028, 0.036)
1.11 (1.07, 1.14)
1.11 (1.08, 1.14)
Liver fractional
compartment volume
VLivC
0.034 (0.026, 0.042)
0.033 (0.028, 0.039)
1.16(1.09, 1.21)
1.17(1.12, 1.21)
Rapidly perfused
fractional compartment
volume
VRapC
0.087 (0.076,0.1)
0.088 (0.079, 0.097)
1.1 (1.06, 1.13)
1.1 (1.07, 1.13)
Fractional volume of
respiratory lumen
VRespLumC
0.0046 (0.0037, 0.0057)
0.0047 (0.0039, 0.0055)
1.16(1.1, 1.21)
1.16(1.11, 1.21)
Fractional volume of
respiratory tissue
VRespEffC
5e-04 (0.00039,
0.00061)
5e-04 (0.00041, 0.00058)
1.16(1.09, 1.21)
1.16(1.11, 1.2)
Kidney fractional
compartment volume
VKidC
0.0069 (0.0056, 0.0082)
0.007 (0.006, 0.008)
1.13 (1.08, 1.17)
1.13 (1.09, 1.17)
O
O
-------
s 2 ho
& ?
rs ^ <*>
o ' a,
a g"
5 ^
s
s o S
^ | §
Oq o %¦
Uo
^ a §•
^ S vS
o a* ~b
O
^ to ^
Table 3-38 Prior and posterior uncertainty and variability in rat PBPK model parameters (contuined)
Parameter description
PBPK
Parameter
Prior Population
Median: Median
(2.5%, 97.5%)
Posterior Population
Median: Median (2.5%,
97.5%)
Prior Population GSD:
Median (2.5%, 97.5%)
Posterior Population
GSD: Median (2.5%,
97.5%)
Blood fractional
compartment volume
VBldC
0.073 (0.063, 0.085)
0.074 (0.066, 0.082)
1.1 (1.06, 1.13)
1.1 (1.07, 1.13)
Slowly perfused fractional
compartment volume
VSlwC
0.6 (0.55, 0.63)
0.6 (0.57, 0.62)
1.05 (1.04, 1.06)
1.05 (1.04, 1.06)
Plasma fractional
compartment volume
VPlasC
0.039 (0.025, 0.054)
0.04 (0.032, 0.049)
1.24 (1.15, 1.35)
1.22 (1.16, 1.33)
TCA body fractional
compartment volume [not
incl. blood+liver]
VBodC
0.79 (0.78, 0.81)
0.79 (0.78, 0.8)
1.01 (1.01, 1.01)
1.01 (1.01, 1.01)
TCOH/G body fractional
compartment volume [not
incl. liver]
VBodTCOHC
0.87 (0.86, 0.87)
0.87 (0.86, 0.87)
1.01 (1, 1.01)
1.01 (1, 1.01)
TCE blood:air partition
coefficient
PB
22 (14, 33)
19 (16, 24)
1.26 (1.19, 1.35)
1.3 (1.22, 1.38)
TCE fat:blood partition
coefficient
PFat
27 (16, 46)
31 (24, 42)
1.32(1.22, 1.44)
1.32 (1.23, 1.43)
TCE gut:blood partition
coefficient
PGut
1.3 (0.69, 3)
1.1 (0.79, 1.7)
1.36 (1.21, 1.79)
1.36 (1.2, 1.68)
TCE livenblood partition
coefficient
PLiv
1.5 (1.2, 1.9)
1.6 (1.3, 1.8)
1.15 (1.11, 1.2)
1.15 (1.11, 1.2)
TCE rapidly
per£used:blood partition
coefficient
PRap
1.3 (0.66, 2.7)
1.3 (0.82,2.1)
1.35 (1.18, 1.82)
1.37 (1.2, 1.76)
TCE respiratory tissue:air
partition coefficient
PResp
0.97 (0.48, 2.1)
1 (0.62, 1.6)
1.37 (1.19, 1.77)
1.36 (1.19, 1.78)
TCE kidney :blood
partition coefficient
PKid
1.3 (0.77, 2.2)
1.2 (0.9, 1.7)
1.31 (1.19, 1.5)
1.3 (1.2, 1.45)
TCE slowly
per£used:blood partition
coefficient
PSlw
0.57 (0.35, 0.97)
0.73 (0.54, 0.97)
1.32 (1.23, 1.43)
1.33 (1.25, 1.46)
TCA blood:plasma
concentration ratio
TCAPlas
0.78 (0.6, 0.96)
0.78 (0.71, 0.86)
1.12(1.06, 1.22)
1.11 (1.07, 1.17)
to
O
O
-------
s 2 ho
& ?
rs ^ <*>
o ' a,
a g"
5 ^
s
s o S
^ | §
Oq o %¦
Uo
^ a §•
^ S vS
o a* ~b
O
^ to ^
Table 3-38 Prior and posterior uncertainty and variability in rat PBPK model parameters (contuined)
Parameter description
PBPK
Parameter
Prior Population
Median: Median
(2.5%, 97.5%)
Posterior Population
Median: Median (2.5%,
97.5%)
Prior Population GSD:
Median (2.5%, 97.5%)
Posterior Population
GSD: Median (2.5%,
97.5%)
Free TCAbody:blood
plasma partition
coefficient
PBodTCA
0.7(0.18,2.2)
0.76 (0.46, 1.3)
1.72 (1.39, 2.81)
1.65(1.4,2.19)
Free TCA livenblood
plasma partition
coefficient
PLivTCA
0.84 (0.25, 3.3)
1.1 (0.61,2.1)
1.71 (1.39, 2.78)
1.66 (1.38,2.37)
Protein:TCA dissociation
constant (|imolc/L)
kDissoc
270 (95, 790)
280 (140, 530)
1.62 (1.31,2.43)
1.6 (1.31,2.31)
Maximum binding
concentration (|imolc/L)
Bmax
320 (80, 1300)
320 (130, 750)
1.89 (1.5,2.64)
1.84 (1.49,2.57)
TCOH body:blood
partition coefficient
PBodTCOH
1 (0.33, 4)
1.1 (0.51,2.1)
1.71 (1.37, 2.69)
1.76 (1.38,2.45)
TCOH liver:body
partition coefficient
PLivTCOH
1.3 (0.39,4.5)
1.2 (0.59, 2.8)
1.71 (1.37,2.8)
1.78 (1.37,2.75)
TCOG body:blood
partition coefficient
PBodTCOG
0.48 (0.021, 14)
1.6(0.091, 16)
1.39 (1.2, 1.97)
1.42(1.21,2.52)
TCOG liver:body
partition coefficient
PLivTCOG
1.3 (0.078, 39)
10 (2.7, 41)
1.4 (1.2,2.14)
1.42(1.21,2.3)
DCVG effective volume
of distribution
VDCVG
0.27 (0.27, 0.27)
0.27 (0.27, 0.27)
1 (1, 1)
1(1, 1)
TCE stomach absorption
coefficient (/hour)
kAS
0.73 (0.0044, 400)
2.5 (0.32, 19)
4.16(2.21,20)
9.3 (4.07, 31.1)
TCE stomach-duodenum
transfer coefficient (/hour)
kTSD
1.4 (0.04, 45)
3.2 (0.31, 19)
3.92 (2.13, 10.4)
5.54 (2.77, 10.7)
TCE duodenum
absorption coefficient
(/hour)
kAD
0.96 (0.0023, 260)
0.17 (0.038, 1)
4.17(2.15,20.8)
4.07 (2.51, 11.9)
TCA stomach absorption
coefficient (/hour)
kASTCA
0.83 (0.0024, 240)
1.4 (0.13, 13)
4.15 (2.2, 18.7)
4.21 (2.4, 11.4)
Vmax for hepatic TCE
oxidation (mg/hour)
Vmax
5.8 (2, 19)
5.3 (3.9, 7.7)
1.97 (1.54, 2.92)
1.69(1.47,2.15)
Km for hepatic TCE
oxidation (mg/L)
Km
18 (1.9, 240)
0.74 (0.54, 1.4)
2.76 (1.89, 6.46)
1.84(1.51,2.7)
Fraction of hepatic TCE
FracOther
0.027 (0.0018, 0.59)
0.29 (0.047, 0.56)
1.42 (1.15,2.33)
2.15 (1.32,5.06)
O
O
-------
s 2 ho
& ?
rs ^ <*>
o ' a,
a g"
5 ^
s
s o S
^ | §
Oq o %¦
Uo
^ a §•
^ S vS
o a* ~b
O
^ to ^
Table 3-38 Prior and posterior uncertainty and variability in rat PBPK model parameters (contuined)
Parameter description
PBPK
Parameter
Prior Population
Median: Median
(2.5%, 97.5%)
Posterior Population
Median: Median (2.5%,
97.5%)
Prior Population GSD:
Median (2.5%, 97.5%)
Posterior Population
GSD: Median (2.5%,
97.5%)
oxidation not to
TCA+TCOH
Fraction of hepatic TCE
oxidation to TCA
FracTCA
0.2 (0.027, 0.76)
0.046 (0.023, 0.087)
1.35(1.11,2.14)
1.84 (1.36, 2.8)
Vmax for hepatic TCE
GSH conjugation
(mg/hour)
VmaxDCVG
2(0.015, 1100)
5.8 (0.16, 340)
1.52 (1.3,2.67)
1.57 (1.32,2.93)
Km for hepatic TCE GSH
conjugation (mg/L)
kmdcvg
1500 (1.2, 1800000)
6300 (120, 720000)
1.83 (1.45, 3.15)
1.88 (1.48,3.49)
Vmax for renal TCE GSH
conjugation (mg/hour)
Vmax KidDCVG
0.038 (0.00027, 13)
0.0024 (5e-04, 0.014)
1.52(1.3,2.81)
1.56(1.29,2.72)
Km for renal TCE GSH
conjugation (mg/L)
KMKidDCVG
470 (0.47, 530000)
0.25 (0.038, 2.2)
1.84 (1.47, 4.27)
1.93 (1.49,3.57)
Vmax for
tracheo-bronchial TCE
oxidation (mg/hour)
Vmax Clara
0.2 (0.0077, 2.4)
0.17(0.042,0.69)
2.26(1.71,3.3)
4.35 (1.99, 6.7)
Km for tracheo-bronchial
TCE oxidation (mg/L)
KMClara
0.016 (0.0014, 0.58)
0.025 (0.005,0.15)
1.47 (1.26, 2.39)
1.65 (1.28, 10.5)
Fraction of respiratory
metabolism to systemic
circ.
FracLungSys
0.82 (0.027, 1)
0.73 (0.06, 0.98)
1.09(1, 1.71)
1.13 (1.01, 1.86)
Vmax for hepatic
TCOH—>TCA (mg/hour)
Vmax TCOH
0.75 (0.037, 20)
0.71 (0.27, 2.2)
1.51 (1.25,2.64)
1.68 (1.3,3.23)
Km for hepatic
TCOH—>TCA (mg/L)
kmtcoh
1 (0.029, 23)
19 (3.6, 94)
1.52(1.26,2.7)
1.72 (1.26,3.93)
Vmax for hepatic
TCOH—>TCOG
(mg/hour)
Vmax Glue
27 (0.83, 620)
11 (4.1, 32)
1.5 (1.25,2.59)
2.3 (1.41, 5.19)
Km for hepatic
TCOH—>TCOG (mg/L)
KmG1uc
31 (1, 570)
6.3 (1.2, 20)
1.5 (1.25,2.74)
2.04(1.3,8.4)
Rate constant for hepatic
TCOH—mother (/hour)
kMetTCOH
4.2 (0.17, 150)
3 (0.57, 15)
1.49 (1.27, 2.67)
1.72(1.3, 8.31)
Rate constant for TCA
plasma—murine (/hour)
kUrnTCA
1.9 (0.21, 47)
0.92 (0.51, 1.7)
1.56 (1.33,2.81)
1.58 (1.36,2.25)
O
O
-------
a
o
<
g
cs
w
o
^3
§•
>!
St
Cb
o
St
3
>1
>!
o
Co
£
Si
s
O
sis
§-
c>
Table 3-38 Prior and posterior uncertainty and variability in rat PBPK model parameters (contuined)
Parameter description
PBPK
Parameter
Prior Population
Median: Median
(2.5%, 97.5%)
Posterior Population
Median: Median (2.5%,
97.5%)
Prior Population GSD:
Median (2.5%, 97.5%)
Posterior Population
GSD: Median (2.5%,
97.5%)
Rate constant for hepatic
TCA—mother (/hour)
kMetTCA
0.76 (0.037, 19)
0.47 (0.17, 1.2)
1.5 (1.26, 2.74)
1.52 (1.27,2.45)
Rate constant for TCOG
liver—>bile (/hour)
kBile
1.4 (0.052, 31)
14 (2.7, 39)
1.5 (1.25,2.8)
1.63 (1.29, 4.1)
Lumped rate constant for
TCOG bile—>TCOH liver
(/hour)
kEHR
0.013 (0.00055, 0.64)
1.7 (0.34, 7.4)
1.5 (1.25,2.49)
1.67(1.26,5.91)
Rate constant for
TCOG—mrine (/hour)
kUrnTCOG
11 (0.063, 1000)
12 (0.45, 370)
1.74 (1.42, 2.99)
1.86 (1.43,3.54)
Rate constant for hepatic
DCVG—>DCVC (/hour)
kDCVG
30000 (30000, 30000)
30000 (30000, 30000)
1 (1, 1)
1(1, 1)
Lumped rate constant for
DCVC—>urinary
NAcDCVC (/hour)
kNAT
0.15 (0.00024, 84)
0.0029 (0.00066, 0.015)
1.49(1.24,2.8)
1.54 (1.26,2.45)
Rate constant for DCVC
bioactivation (/hour)
kKidBioact
0.12 (0.00023,83)
0.0092 (0.0012,0.043)
1.48 (1.24, 2.68)
1.52 (1.25,2.5)
O
O
-------
s 2 ho
& ?
rs ^ <*>
o ' a,
a g"
5 ^
s
s o S
^ | §
Oq o %¦
Uo
^ a §•
^ S vS
o a. ~b
O
^ to ^
Table 3-39. Prior and posterior uncertainty and variability in human PBPK model parameters
Parameter description
PBPK
Parameter
Prior Population
Median: Median
(2.5%, 97.5%)
Posterior Population
Median: Median (2.5%,
97.5%)
Prior Population GSD:
Median (2.5%, 97.5%)
Posterior Population
GSD: Median (2.5%,
97.5%)
Cardiac output (L/hour)
QC
390 (280, 560)
330 (280, 390)
1.17(1.1, 1.39)
1.39 (1.26, 1.54)
Alveolar ventilation
(L/hour)
QP
380 (220, 640)
440 (360, 530)
1.27(1.17, 1.52)
1.58 (1.44, 1.73)
Scaled fat blood flow
QFatC
0.051 (0.021,0.078)
0.043 (0.033, 0.055)
1.64 (1.23,2)
1.92(1.72,2.09)
Scaled gut blood flow
QGutC
0.19(0.15,0.23)
0.16(0.14,0.18)
1.16(1.1, 1.21)
1.16(1.12, 1.2)
Scaled liver blood flow
QLivC
0.063 (0.029, 0.099)
0.039 (0.026, 0.055)
1.62 (1.22, 1.92)
1.8 (1.62, 1.98)
Scaled slowly perfused
blood flow
QSlwC
0.22 (0.13,0.3)
0.17(0.14,0.21)
1.34 (1.18, 1.45)
1.39 (1.31, 1.46)
Scaled rapidly perfused
blood flow
QRapC
0.29 (0.18, 0.4)
0.39 (0.34, 0.43)
1.31 (1.14, 1.57)
1.22(1.16, 1.3)
Scaled kidney blood flow
QKidC
0.19(0.16,0.22)
0.19(0.18,0.21)
1.1 (1.07, 1.13)
1.1 (1.07, 1.12)
Respiratory lumen:tissue
diffusive clearance rate
(L/hour)
DResp
560 (44, 3300)
270 (130, 470)
1.37 (1.25, 1.61)
1.71 (1.52,2.35)
Fat fractional
compartment volume
VFatC
0.19 (0.088,0.31)
0.16(0.12,0.21)
1.66 (1.23, 1.93)
1.65 (1.4, 1.9)
Gut fractional
compartment volume
VGutC
0.02 (0.018, 0.022)
0.02 (0.019,0.021)
1.07 (1.04, 1.08)
1.06(1.05, 1.08)
Liver fractional
compartment volume
VLivC
0.026 (0.018, 0.032)
0.026 (0.022, 0.03)
1.21 (1.12, 1.28)
1.2 (1.13, 1.26)
Rapidly perfused
fractional compartment
volume
VRapC
0.087 (0.079, 0.096)
0.088 (0.083, 0.093)
1.07(1.05, 1.09)
1.06(1.05, 1.08)
Fractional volume of
respiratory lumen
VRespLumC
0.0024 (0.0018,0.003)
0.0024 (0.0021, 0.0027)
1.18(1.1, 1.23)
1.17(1.12, 1.22)
Fractional volume of
respiratory tissue
VRespEffC
0.00018 (0.00014,
0.00022)
0.00018 (0.00015,
0.00021)
1.18(1.1, 1.24)
1.17(1.13, 1.23)
Kidney fractional
compartment volume
VKidC
0.0043 (0.0034, 0.0052)
0.0043 (0.0038, 0.0048)
1.15 (1.09, 1.19)
1.14(1.1, 1.19)
On
O
O
-------
s 2 ho
& ?
rs ^ <*>
o ' a,
a g"
5 ^
s
s o S
^ | §
Oq o %¦
Uo
^ a §•
^ S vS
o a* ~b
O
^ to ^
Table 3-39 Prior and posterior uncertainty and variability in human PBPK model parameters (contuined)
Parameter description
PBPK
Parameter
Prior Population
Median: Median
(2.5%, 97.5%)
Posterior Population
Median: Median (2.5%,
97.5%)
Prior Population GSD:
Median (2.5%, 97.5%)
Posterior Population
GSD: Median (2.5%,
97.5%)
Blood fractional
compartment volume
VBldC
0.077 (0.066, 0.088)
0.078 (0.072, 0.084)
1.1 (1.06, 1.13)
1.1 (1.07, 1.13)
Slowly perfused fractional
compartment volume
VSlwC
0.45 (0.33, 0.55)
0.48 (0.43, 0.52)
1.18(1.1, 1.24)
1.16(1.12, 1.22)
Plasma fractional
compartment volume
VPlasC
0.044 (0.037,0.051)
0.044 (0.04, 0.048)
1.11 (1.08, 1.14)
1.11 (1.08, 1.14)
TCA body fractional
compartment volume [not
incl. blood+liver]
VBodC
0.75 (0.74, 0.77)
0.75 (0.74, 0.76)
1.01 (1.01, 1.01)
1.01 (1.01, 1.01)
TCOH/G body fractional
compartment volume [not
incl. liver]
VBodTCOHC
0.83 (0.82, 0.84)
0.83 (0.83, 0.83)
1.01 (1, 1.01)
1.01 (1, 1.01)
TCE blood:air partition
coefficient
PB
9.6 (6.5, 13)
9.2 (8.2, 10)
1.18(1.13, 1.26)
1.21 (1.16, 1.28)
TCE fat:blood partition
coefficient
PFat
68 (46, 98)
57 (49, 66)
1.18(1.11, 1.33)
1.18(1.11, 1.3)
TCE gut:blood partition
coefficient
PGut
2.6(1.3,5.3)
2.9 (1.9,4.1)
1.37 (1.2, 1.78)
1.41 (1.21, 1.77)
TCE livenblood partition
coefficient
PLiv
4 (1.9, 8.5)
4.1 (2.7, 5.9)
1.37 (1.22, 1.81)
1.33 (1.19, 1.6)
TCE rapidly
per£used:blood partition
coefficient
PRap
2.6(1.2,5.7)
2.4 (1.8, 3.2)
1.37 (1.21, 1.78)
1.5 (1.25, 1.87)
TCE respiratory tissue:air
partition coefficient
PResp
1.3 (0.65, 2.7)
1.3 (0.9, 1.9)
1.36(1.19, 1.81)
1.32 (1.2, 1.56)
TCE kidney :blood
partition coefficient
PKid
1.6(1.1,2.3)
1.6 (1.3, 1.9)
1.17(1.1, 1.33)
1.15 (1.09, 1.25)
TCE slowly
per£used:blood partition
coefficient
PSlw
2.1 (1.2,3.5)
2.3 (1.9,2.8)
1.28 (1.14, 1.53)
1.51 (1.36, 1.66)
TCA blood:plasma
concentration ratio
TCAPlas
0.78 (0.55, 15)
0.65 (0.6, 0.77)
1.08 (1.03, 1.53)
1.52 (1.23,2.03)
O
O
-------
s 2 ho
& ?
rs ^ <*>
o ' a,
a g"
5 ^
s
s o S
^ | §
Oq o %¦
Uo
^ a §•
^ S vS
o a* ~b
O
^ to ^
Table 3-39 Prior and posterior uncertainty and variability in human PBPK model parameters (contuined)
Parameter description
PBPK
Parameter
Prior Population
Median: Median
(2.5%, 97.5%)
Posterior Population
Median: Median (2.5%,
97.5%)
Prior Population GSD:
Median (2.5%, 97.5%)
Posterior Population
GSD: Median (2.5%,
97.5%)
Free TCAbody:blood
plasma partition
coefficient
PBodTCA
0.45 (0.19, 8.1)
0.44 (0.33, 0.55)
1.36 (1.19, 1.75)
1.67 (1.38, 2.2)
Free TCA livenblood
plasma partition
coefficient
PLivTCA
0.59 (0.24, 10)
0.55 (0.39, 0.77)
1.36 (1.18, 1.76)
1.65 (1.37,2.16)
Protein:TCA dissociation
constant (|imolc/L)
kDissoc
180 (160, 200)
180 (170, 190)
1.05 (1.03, 1.09)
1.04(1.03, 1.07)
Maximum binding
concentration (|imolc/L)
Bmax
830 (600, 1100)
740 (630, 880)
1.17(1.1, 1.3)
1.16(1.1, 1.28)
TCOH body:blood
partition coefficient
PBodTCOH
0.89 (0.51, 1.7)
1.5 (1.3, 1.7)
1.29(1.16, 1.64)
1.34 (1.25, 1.47)
TCOH livenbody
partition coefficient
PLivTCOH
0.58 (0.32, 1.1)
0.63 (0.45, 0.87)
1.29 (1.16, 1.65)
1.29(1.17, 1.5)
TCOG body:blood
partition coefficient
PBodTCOG
0.67 (0.036, 16)
0.72 (0.3, 1.8)
1.38 (1.2,2.42)
7.83 (4.86, 12.6)
TCOG livenbody
partition coefficient
PLivTCOG
1.8(0.11,28)
3.1 (0.87, 8.1)
1.38 (1.19, 2.04)
4.94 (2.73, 8.58)
DCVG effective volume
of distribution
VDCVG
73 (5.2, 36000)
6.1 (5.4,7.3)
1.27 (1.08, 1.95)
1.1 (1.07, 1.16)
TCE stomach absorption
coefficient (/hour)
kAS
1.4(1.4, 1.4)
1.4 (1.4, 1.4)
1 (1, 1)
1(1, 1)
TCE stomach-duodenum
transfer coefficient (/hour)
kTSD
1.4(1.4, 1.4)
1.4 (1.4, 1.4)
1 (1, 1)
1(1, 1)
TCE duodenum
absorption coefficient
(/hour)
kAD
0.75 (0.75, 0.75)
0.75 (0.75, 0.75)
1 (1, 1)
1(1, 1)
TCA stomach absorption
coefficient (/hour)
kASTCA
0.58 (0.0022, 210)
3 (0.061, 180)
4.26 (2.13, 17.6)
5.16 (2.57, 22.3)
TCOH stomach
absorption coefficient
(/hour)
kASTCOH
0.49 (0.0024, 210)
7.6(0.11, 150)
4.19(2.22,21.5)
5.02 (2.44, 18.5)
Vmax for hepatic TCE
Vmax
430 (130, 1500)
190 (130, 290)
1.98 (1.69, 2.31)
2.02(1.77,2.38)
00
o
O
-------
s 2 ho
& ?
rs ^ <*>
o ' a,
a g"
5 ^
s
s o S
^ | §
Oq o %¦
Uo
^ a §•
^ S vS
o a* ~b
O
^ to ^
Table 3-39 Prior and posterior uncertainty and variability in human PBPK model parameters (contuined)
Parameter description
PBPK
Parameter
Prior Population
Median: Median
(2.5%, 97.5%)
Posterior Population
Median: Median (2.5%,
97.5%)
Prior Population GSD:
Median (2.5%, 97.5%)
Posterior Population
GSD: Median (2.5%,
97.5%)
oxidation (mg/hour)
Km for hepatic TCE
oxidation (mg/L)
Km
3.7 (0.22, 63)
0.18(0.078,0.4)
2.74 (2.1,5.62)
4.02 (2.9, 5.64)
Fraction of hepatic TCE
oxidation not to
TCA+TCOH
FracOther
0.12(0.0066,0.7)
0.11 (0.024,0.23)
1.4(1.11,2.38)
2.71 (1.37,5.33)
Fraction of hepatic TCE
oxidation to TCA
FracTCA
0.19 (0.036,0.56)
0.035 (0.024, 0.05)
2.55 (1.51, 3.96)
2.25 (1.89,2.87)
Vmax for hepatic TCE
GSH conjugation
(mg/hour)
VmaxDCVG
100 (0.0057, 690000)
340(110, 1100)
1.91 (1.55, 3.76)
6.18(3.35, 11.3)
Km for hepatic TCE GSH
conjugation (mg/L)
kmdcvg
3.1 (0.21,42)
3.6 (1.2, 11)
1.52(1.26, 2.91)
4.2 (2.48, 8.01)
Vmax for renal TCE GSH
conjugation (mg/hour)
VMAxKidDCVG
220 (0.028, 6700000)
2.1 (0.17, 9.3)
1.86 (1.51,3.33)
4.02 (1.57,33.9)
KM for renal TCE GSH
conjugation (mg/L)
KMKidDCVG
2.7(0.14,41)
0.76 (0.29, 5.8)
1.5 (1.27,2.56)
1.49 (1.27,2.32)
Vmax for
tracheo-bronchial TCE
oxidation (mg/hour)
Vmax Clara
25 (1, 260)
18(3.8,41)
2.25 (1.85, 3.25)
2.9(2.12,6.49)
Km for tracheo-bronchial
TCE oxidation (mg/L)
KMClara
0.019(0.0017, 0.5)
0.31 (0.057, 1.4)
1.48 (1.25,2.39)
10.8 (1.99,37.6)
Fraction of respiratory
metabolism to systemic
circ.
FracLungSys
0.75 (0.051,0.99)
0.96 (0.86, 0.99)
1.12(1, 1.75)
1.02(1, 1.1)
Vmax for hepatic
TCOH—>TCA (mg/hour)
Vmax TCOH
42 (0.77, 2200)
9.2 (5.5, 20)
1.83 (1.46, 3.43)
3.15 (2.3, 5.44)
Km for hepatic
TCOH—>TCA (mg/L)
kmtcoh
5(0.23,81)
2.2 (1.3,4.5)
1.49 (1.25,2.57)
2.58 (1.75,4.5)
Vmax for hepatic
TCOH—>TCOG
(mg/hour)
Vmax Glue
720 (12, 50000)
900 (340, 2000)
1.83 (1.48, 3.5)
2.29(1.84,4.57)
Km for hepatic
KmG1uc
10 (0.53, 190)
130 (47, 290)
1.5 (1.25,2.6)
1.58 (1.26,3.69)
vo
O
O
-------
a
o
<
g
cs
w
o
^3
§•
>!
St
Cb
*3
o
St
3
>1
>!
o
Co
£
Si
s
O
sis
.C?"
§-
c>
Table 3-39 Prior and posterior uncertainty and variability in human PBPK model parameters (contuined)
Parameter description
PBPK
Parameter
Prior Population
Median: Median
(2.5%, 97.5%)
Posterior Population
Median: Median (2.5%,
97.5%)
Prior Population GSD:
Median (2.5%, 97.5%)
Posterior Population
GSD: Median (2.5%,
97.5%)
TCOH—>TCOG (mg/L)
Rate constant for hepatic
TCOH—mother (/hour)
kMetTCOH
0.83 (0.035, 10)
0.25 (0.042, 0.7)
1.5 (1.26, 3)
5.13 (2.72, 16.7)
Rate constant for TCA
plasma—murine (/hour)
kUrnTCA
0.26 (0.038, 4)
0.11 (0.083,0.15)
1.48 (1.29, 2.29)
1.86 (1.58,2.28)
Rate constant for hepatic
TCA—mother (/hour)
kMetTCA
0.19(0.01,2.6)
0.096 (0.038,0.19)
1.48 (1.26, 2.57)
2.52 (1.79,4.34)
Rate constant for TCOG
liver—>bile (/hour)
kBile
1.2 (0.059, 16)
2.5 (1.1,6.9)
1.47 (1.25,2.75)
1.56 (1.27,3.21)
Lumped rate constant for
TCOG bile—>TCOH liver
(/hour)
kEHR
0.074 (0.004, 1.4)
0.053 (0.033, 0.087)
1.52 (1.26, 2.64)
1.72 (1.35,2.51)
Rate constant for
TCOG—mrine (/hour)
kUrnTCOG
2.9 (0.061, 260)
2.4 (0.83, 7)
1.75 (1.4, 3.31)
18.7(11.6,31.8)
Rate constant for hepatic
DCVG—>DCVC (/hour)
kDCVG
0.044 (6.3e-05, 22)
2.5 (1.9, 3.4)
1.48 (1.25,2.83)
1.51 (1.3, 1.86)
Lumped rate constant for
DCVC—^urinary
NAcDCVC (/hour)
kNAT
0.00085 (5.5e-05,
0.041)
le-04 (4.7e-05, 7e-04)
1.51 (1.25,2.34)
1.47(1.24,2.48)
Rate constant for DCVC
bioactivation (/hour)
kKidBioact
0.0022 (9.5e-05, 0.079)
0.023 (0.0062, 0.061)
1.51 (1.25,2.57)
1.52 (1.25,2.69)
to
o
O
O
-------
a 2
o S
a ^3
>! St
££¦ ^
OQ o
<* s
a c?
.rs ^
^ O
^ S
o ^
s?
>!
§•
cs
st
>!
Si
Table 3-40. Confidence interval (CI) widths (ratio of 97.5% to 2.5% estimates) and fold-shift in median estimate
for the PBPK model population median parameters, sorted in order of decreasing CI width. Shifts in the median
estimate greater than threefold are in bold to denote larger shifts between the prior and posterior distributions
Mouse
Rat
Human
Width of CI on population
median
Fold-shift
in pop.
median
Width of CI on
population median
Fold-shift
in pop.
median
Width of CI on
population median
Fold-shift
in pop.
median
PBPK
Parameter
Prior
Posterior
PBPK
Parameter
Prior
Posterior
PBPK
Parameter
Prior
Posterior
KviDCVG
2230000000
13400000
x8.8
KviDCVG
1500000
5800
x4.29
kASTCA
94300
3040
x5.18
KMKidDCVG
1170000000
3540000
xl.05
VmaxDCVG
71100
2130
x2.86
kASTCOH
85900
1420
xl5.6
VmaxDCVG
400000
46200
-6.18
kUrnTCOG
16700
822
xl.04
Vmax
KidDCVG
236000000
55.1
-105
Vmax KidDCVG
357000
11000
-12.8
PBodTCOG
666
172
x3.43
KMClara
289
23.9
xl6.2
kASTCA
89300
374
x6.3
kASTCA
98200
95.7
xl.69
KMKidDCVG
287
20
-3.48
kTSD
1190
51.1
x3.26
kTSD
1130
61.8
x2.29
kMetTCOH
289
16.6
-3.28
kEHR
412000
42.1
-5.43
kAS
91000
60.2
x3.41
kNAT
756
15.1
-8.14
FracOther
567
39.5
-18.5
KMKidDCVG
1130000
58.6
-1880
Vmax Clara
255
10.6
-1.41
KMClara
351000
37.5
-134
kKidBioact
366000
35.6
-13.3
kKidBioact
833
9.91
xl0.5
kAS
91900
35.9
Xl
KMClara
406
29.9
xl.53
VmaxDCVG
122000000
9.78
x3.29
kUrnTCOG
40500000
29.9
xll.8
Vmax KidDCVG
48500
27.5
-15.6
FracOther
106
9.75
-1.09
Bmax
81.8
24.4
xl.66
kMetTCOH
891
26.4
-1.41
PLivTCOG
253
9.32
xl.77
KiG1lic
344000
24.3
xl6.3
kAD
115000
26.3
-5.53
kmdcvg
198
9.13
xl.18
kAD
84900
23.8
-4.53
kmtcoh
781
26
xl8.7
kUrnTCOG
4290
8.5
-1.19
kDissoc
60.3
21.8
xl.33
kNAT
351000
22.7
-50.2
kBile
274
6.54
x2.01
Vmax Clara
131
15
xl.75
kEHR
1160
21.9
xl34
KmG1uc
365
6.07
xl3.4
kMetTCOH
35500000
12.1
x47.4
K|Gluc
562
17.1
-4.98
PBodTCOG
454
5.85
xl.08
kBile
390000
11.3
x8.23
Vmax Clara
305
16.5
-1.21
Vmax Glue
4330
5.71
xl.25
Kv,TCOH
29600000
10.5
-1.47
FracLungSys
36.7
16.3
-1.12
Km
288
5.1
-20.5
Vmax Glue
23600000
8.28
x41.1
PLivTCOG
501
14.8
x8.07
kMetTCA
248
4.89
-1.94
§•
<3
to
-------
a 2
o S
a ^3
>! St
££¦ ^
OQ o
<* s
a c?
.rs ^
^ O
^ S
o ^
s?
>!
§•
cs
st
>!
Si
Table 3-40. Confidence interval (CI) widths (ratio of 97.5% to 2.5% estimates) and fold-shift in median estimate for
the PBPK model population median parameters, sorted in order of decreasing CI width (contuined)
Mouse
Rat
Human
Width of CI on population
median
Fold-shift
in pop.
median
Width of CI on
population median
Fold-shift
in pop.
median
Width of CI on
population median
Fold-shift
in pop.
median
PBPK
Parameter
Prior
Posterior
PBPK
Parameter
Prior
Posterior
PBPK
Parameter
Prior
Posterior
PBodTCOG
4770
6.27
-1.95
kBile
588
14.8
X
9.67
DResp
74.3
3.71
-2.06
VmaxTCOH
27100000
5.78
xl.8
FracOther
331
11.9
X
10.7
VmaxTCOH
2900
3.62
-4.56
Km
386
5.76
-12.5
Vmax TCOH
550
8.25
1.06
kmtcoh
359
3.48
-2.33
kUrnTCA
4540
5.76
-10.2
Vmax Glue
740
7.79
2.4
kEHR
339
2.62
-1.39
FracLungSys
608
5.55
x2.27
kMetTCA
507
6.93
1.61
Vmax
11.5
2.27
-2.33
kMetTCA
316000
4.59
xl2
Bmax
16.2
5.79
X
1
PResp
4.1
2.16
-1.01
PLivTCOG
4860
3.99
xl.04
DResp
180
4.81
X
2.12
PLiv
4.44
2.14
xl.02
DResp
475000
3.64
xl47
PLivTCOH
11.5
4.7
1.09
QLivC
3.46
2.11
-1.62
PLivTCA
58.3
2.88
xl
PBodTCOH
12.1
4.03
X
1.03
PGut
4.21
2.1
xl.ll
PResp
4
2.85
-1.07
kDissoc
8.38
3.85
X
1.04
FracTCA
15.5
2.06
-5.37
PRap
3.78
2.79
-1.03
FracTCA
28.1
3.85
4.27
PLivTCA
42.6
1.98
-1.07
PGut
4.33
2.77
-1.25
PLivTCA
13.3
3.49
X
1.37
PLivTCOH
3.52
1.93
xl.08
Vmax
10.7
2.67
-1.58
kUrnTCA
219
3.28
2
kDCVG
344000
1.8
x55.7
PBodTCA
62.6
2.55
xl.14
PBodTCA
12
2.8
X
1.09
kUrnTCA
105
1.79
-2.32
PSlw
4.04
2.54
-1.06
PResp
4.32
2.6
X
1.04
VFatC
3.49
1.76
-1.21
PLiv
3.87
2.5
xl.26
Km
123
2.56
24
PRap
4.66
1.74
-1.09
FracTCA
3060
2.49
xl.49
PRap
4.01
2.53
1.01
QFatC
3.7
1.7
-1.19
TCAPlas
40.6
2.38
xl.46
PGut
4.35
2.16
1.17
PBodTCA
42.9
1.7
-1.04
PKid
4.78
2.37
xl.2
Vmax
9.5
1.98
1.11
PSlw
2.9
1.5
Xl.ll
QFatC
3.62
2.26
xl.02
QRapC
2.77
1.97
1
PKid
2.05
1.49
-1.01
PLivTCOH
3.19
2.13
xl.48
VFatC
3.58
1.96
1
QP
2.97
1.48
xl.16
PBodTCOH
3.41
2.01
-1.27
PKid
2.89
1.85
1.11
QSlwC
2.25
1.48
-1.26
QKidC
2.39
1.91
-1.01
QP
3.59
1.79
1.38
QC
2.04
1.39
-1.19
PFat
3.01
1.89
-1.01
PSlw
2.76
1.79
X
1.28
Bmax
1.92
1.38
-1.12
QSlwC
2.04
1.88
-1.02
PFat
2.91
1.77
X
1.16
VLivC
1.79
1.36
xl.01
VPlasC
2.18
1.87
-1.17
QSlwC
2.19
1.69
1.06
PFat
2.13
1.34
-1.2
§•
<3
to
to
o
O
-------
a 2
o S
a ^3
>! St
££¦ ^
OQ o
<* s
a c?
.rs ^
^ O
^ S
o ^
s?
>!
§•
cs
st
>!
Si
Table 3-40. Confidence interval (CI) widths (ratio of 97.5% to 2.5% estimates) and fold-shift in median estimate for
the PBPK model population median parameters, sorted in order of decreasing CI width (contuined)
Mouse
Rat
Human
Width of CI on population
Width of CI on
Width of CI on
median
Fold-shift
population median
Fold-shift
population median
Fold-shift
PBPK
in pop.
PBPK
in pop.
PBPK
in pop.
Parameter
Prior
Posterior
median
Parameter
Prior
Posterior
median
Parameter
Prior
Posterior
median
VFatC
3.49
1.83
xl.25
QFatC
3.47
1.66
xl.14
VDCVG
6820
1.34
-12
QP
2.75
1.82
-1.02
VPlasC
2.17
1.55
xl.03
VRespEffC
1.66
1.33
-1.02
VLivC
1.85
1.6
-1.16
PB
2.37
1.51
-1.15
PBodTCOH
3.32
1.32
xl.68
QC
2.1
1.59
xl.2
QC
1.64
1.43
xl.15
VRespLumC
1.65
1.31
-1
PB
2.3
1.54
-1.07
VRespEffC
1.56
1.43
-1
TCAPlas
26.9
1.29
-1.21
QLivC
1.55
1.42
xl.02
VRespLumC
1.56
1.41
Xl
VKidC
1.54
1.28
-1.01
QRapC
1.51
1.41
-1.03
VLivC
1.57
1.4
-1.05
PB
2.04
1.28
-1.04
VGutC
1.38
1.3
-1.01
PLiv
1.67
1.37
xl.05
QRapC
2.22
1.25
xl.34
VBldC
1.34
1.27
-1.02
QLivC
1.53
1.34
xl.04
QGutC
1.59
1.23
-1.19
VRespLumC
1.32
1.26
-1.01
VKidC
1.47
1.33
xl.01
VSlwC
1.66
1.21
xl.07
VRespEffC
1.31
1.26
-1
QKidC
1.39
1.28
Xl
VPlasC
1.39
1.2
xl.01
QGutC
1.52
1.24
xl.15
VGutC
1.38
1.28
-1.01
QKidC
1.36
1.17
-1
VKidC
1.29
1.24
-1
VBldC
1.34
1.25
Xl.01
VBldC
1.34
1.17
xl.02
VRapC
1.3
1.23
-1.01
VRapC
1.34
1.23
Xl
FracLungSys
19.4
1.14
xl.29
VSlwC
1.19
1.11
-1.01
QGutC
1.53
1.22
Xl.14
VRapC
1.22
1.12
Xl
VBodC
1.05
1.03
xl.01
TCAPlas
1.6
1.21
-1.01
kDissoc
1.23
1.12
-1.01
VBodTCOHC
1.04
1.03
xl.01
VSlwC
1.15
1.09
Xl
VGutC
1.22
1.11
Xl.01
VBodC
1.04
1.03
Xl
VBodC
1.04
1.02
-1
VBodTCOHC
1.02
1.01
Xl
VBodTCOHC
1.02
1.01
-1
§•
<3
to
U>
o
o
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
was the fraction of TCE oxidation directly producing TCA, which shifted by fourfold in rats and
fivefold in mice, with prior CIs of 28-fold and 16-fold, respectively. These shifts are still
relatively modest in comparison to the prior CI, and moreover, the posterior CI is quite narrow
(fourfold in rats, twofold in humans), suggesting that the parameter is well identified by the in
vivo data.
In addition, in only a few cases did the interquartile regions of the prior and posterior
distributions not overlap. In most of these cases, including the diffusion rate from respiratory
lumen to tissue, the Kms for renal TCE GSH conjugation and respiratory TCE oxidation, and
several metabolite kinetic parameters, the prior distributions themselves were noninformative.
For a noninformative prior, the lack of overlap would only be an issue if the posterior
distributions were affected by the truncation limit, which was not the case here. The only other
parameter for which there was a lack of interquartile overlap between the prior and posterior
distribution was the Km for hepatic TCE oxidation in mice and in rats, though the prior and
posterior 95% CIs did overlap within each species. As discussed Section 3.3, there is some
uncertainty in the extrapolation of in vitro KM values to in vivo values (within the same species).
In addition, in mice, it has been known for some time that Km values appear to be discordant
among different studies (Abbas and Fisher, 1997; Fisher et al., 1991; Greenberg et al., 1999).
In terms of estimates of population variability, for the vast majority of parameters, the
posterior estimate of the population GSD was either twofold or less, indicating modest
variability. In some cases, while the posterior population GSD was greater than twofold, it was
similar to the prior estimate of the population GSD, indicating limited additional informative
data on variability. This was the case for oral absorption parameters, which are expected to be
highly variable because the current model lumps parameters for different oral dosing vehicles
together, and a relatively wide prior distribution was given. In addition, in some cases this was
due to in vitro data showing a higher degree of variability. Examples of this include TCA
plasma binding parameters in the mouse, and the Vmax for hepatic oxidation and the fraction of
oxidation to TCA in humans. In a few other cases, the in vivo data appeared to indicate greater
than twofold variability between subjects, and these are discussed in more detail below.
In the mouse, the two parameters for which this is the case are the Vmax for respiratory
tract oxidation and the urinary excretion rate for TCOG. In the first case, the variability is driven
by the need for a higher respiratory tract Vmax for males in the Fisher et al. (1991) study as
compared to other studies. In the second case, it is driven by the relatively low estimate of
urinary excretion of TCOG in the Abbas and Fisher (1997), Abbas et al. (1997), and Greenberg
et al. (1999) studies as compared with the relatively high estimate Green and Prout (1985) and
Prout et al. (1985).
This document is a draft for review purposes only and does not constitute Agency policy.
124 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
In the rat, the two parameters for which the in vivo data suggest greater than twofold
variability are the fraction of oxidation to not producting TCA or TCOH, and the Vmax for
respiratory tract oxidation. In the first case, this is driven by three studies that appeared to
require greater (Bernauer et al., 1996; Kimmerle and Eben, 1973a) or lower (Hissink et al., 2002)
estimates for this parameter as compared with the other studies. Nonetheless, the degree of
variability is not much greater than twofold, with a central estimate population GSD of 2.15-fold.
In the case of the Vmax for respiratory tract oxidation, two studies appeared to require higher
(Fisher et al., 1989) or lower (Simmons et al., 2002) values for this parameter as compared with
the other studies.
In humans, as would be expected, more parameters appeared to exhibit greater than
twofold variability. In terms of distribution, the partition coefficients for TCOG had rather large
posterior estimates for the population GSD of eightfold for the body and fivefold for the liver. In
terms of the body, a few of the subjects in Fisher et al. (1998) and all the subjects in Monster
et al. (1976) appeared to require much higher partition coefficients for TCOG. For the liver, the
variability did not have a discernable trend across studies. In addition, almost all the metabolism
and clearance parameters had posterior estimates for population variability of greater than a
twofold GSD. The largest of these was the urinary excretion rate for TCOG, with a GSD of
19-fold. In this case, the variability was driven by individuals in the Chiu et al. (2007) 1 ppm
study, who were predicted to have much lower rate of urinary excretion as compared to that
estimated in the other, higher exposure studies.
In sum, the Bayesian analysis of the updated PBPK model and data exhibited no major
inconsistencies in prior and posterior parameter distributions. The most significant issue in terms
of population central estimates was the KM for hepatic oxidative metabolism, for which the
posterior estimates were low compared to, albeit somewhat uncertain, in vitro estimates, and it
could be argued that a wider prior distribution would have been better. However, the central
estimates were not at or near the truncation boundary, so it is unlikely that wider priors would
change the results substantially. In terms of population variability, in rodents, the estimates of
variability were generally modest, which is consistent with more homogeneous and controlled
experimental subjects and conditions, whereas the estimates of human population variability
were greater—particularly for metabolism and clearance. Overall, there were no indications
based on this evaluation of prior and posterior distributions either that prior distributions were
overly restrictive or that model specification errors led to pathological parameter estimates.
This document is a draft for review purposes only and does not constitute Agency policy.
125 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
3.5.6.1.3. Comparison of Model Predictions With Data
Comparisons of model predictions and data for each species are discussed in the
sub-sections below. First, as an overall summary, for each species and each output
measurement, the data and predictions generated from a random sample of the MCMC chain are
scatter-plotted to show the general degree of consistency between data and predictions. Next, as
with the Hack et al. (2006) model, the sampled subject-specific parameters were used to generate
predictions for comparison to the calibration data (see Figure 3-8). Thus, the predictions for a
particular data set are conditioned on the posterior parameter distributions for same data set.
Because these parameters were "optimized" for each experiment, these subject-specific
predictions should be accurate by design—and, on the whole, were so. In addition, the
"residual-error" estimate for each measurement (see Table 3-41) provides some quantitative
measure of the degree to which there were deviations due to intrastudy variability and model
misspecification, including any difficulties fitting multiple dose levels in the same study using
the same model parameters.
Next, only samples of the population parameters (means and variances) were used, and
new subjects were sampled from appropriate distribution using these population means and
variances (see Figure 3-8). That is, the predictions were only conditioned on the
population-level parameters distributions, representing an "average" over all the data sets, and
not on the specific predictions for that data set. These "new" subjects then represent the
predicted population distribution, incorporating variability in the population as well as
uncertainty in the population means and variances. Because of the limited amount of mouse
data, all available data for that species were utilized for calibration, and there were no data
available for "out-of-sample" evaluation (often referred to as "validation data," but this term is
not used here due to ambiguities as to its definition). In rats, several studies that contained
primarily blood TCE data, which were abundant, were used for out-of-sample evaluation. In
humans, there were substantial individual and aggregated (mean of individuals in a study) data
that was available for out-of-sample evaluation, as computational intensity limited the number of
individuals that could be used in the MCMC-based calibration.
3.5.6.1.4. Mouse model and data
Each panel of Figure 3-9 shows a scatter plot of the calibration data and a random posterior
prediction for each of the measured endpoint. The endpoint abbreviations are listed in
Table 3-41, as are the implied GSDs for the "residual" errors, which include intrastudy
variability, interindividual variability, and measurement and model errors. The residual-error
This document is a draft for review purposes only and does not constitute Agency policy.
126 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
PBPK
modeld
Posterior
Posterior I2
Posterior subject-
specific A
0i / V
Posterior population
prediction;
Yjki
Experiment j
MCMC outputs
Group/
Individual i
Posterior population
0 /\
Posterior group:;
prediction
Yijkl
GSDs are also shown as grey dotted lines in Figure 3-9. Table 3-42 provides an evaluation of
the predictions of the mouse model for each data set, with figures showing individual
time-course data and predictions in Appendix A.
Figure 3-8. Schematic of how posterior predictions were generated for
comparison with experimental data. Two sets of posterior predictions were
generated: population predictions (diagonal hashing) and subject-specific
predictions (vertical hashing). (Same as Figure A-2 in Appendix A)
This document is a draft for review purposes only and does not constitute Agency policy.
127 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-41. Estimates of the residual-error
2
Measurement
abbreviation
Measurement description
GSD for "residual" error
(median estimate)
Mouse
Rat
Human
RetDose
Retained TCE dose (mg)
-
-
1.13
CAlvPPM
TCE concentration in alveolar air (ppm)
-
-
1.44-1.83
CInhPPM
TCE concentration in closed-chamber (ppm)
1.18
1.11-1.12
-
CMixExh
TCE concentration in mixed exhaled air (mg/L)
-
1.5
-
CArt
TCE concentration in arterial blood (mg/L)
-
1.17-1.52
-
CVen
TCE concentration in venous blood (mg/L)
2.68
1.22-4.46
1.62-2.95
CBldMix
TCE concentration in mixed arterial and venous blood
(mg/L)
1.61
1.5
-
CFat
TCE concentration in fat (mg/L)
2.49
1.85-2.66
-
CGut
TCE concentration in gut (mg/L)
-
1.86
-
CKid
TCE concentration in kidney (mg/L)
2.23
1.47
-
CLiv
TCE concentration in liver (mg/L)
1.71
1.67-1.78
-
CMus
TCE concentration in muscle (mg/L)
-
1.65
-
AExhpost
Amount of TCE exhaled postexposure (mg)
1.23
1.12-1.17
-
CPlasTCA
TCA concentration in plasma (mg/L)
1.40
1.13-1.21
1.12-1.17
CBldTCA
TCA concentration in blood (mg/L)
1.49
1.13-1.59
1.12-1.49
CLivTCA
TCA concentration in liver (mg/L)
1.34
1.67
-
AUrnTCA
Cumulative amount of TCA excreted in urine (mg)
1.34
1.18-1.95
1.11-1.54
AUrnTCAcollect
Cumulative amount of TCA collected in urine
(noncontinuous sampling) (mg)
-
-
2-2.79
CTCOH
Free TCOH concentration in blood (mg/L)
1.54
1.14-1.64
1.14-2.1
CLivTCOH
Free TCOH concentration in liver (mg/L)
1.59
-
-
TotCTCOH
Total TCOH concentration in blood (mg/L)
1.85
1.49
1.2-1.69
ABileTCOG
Cumulative amount of bound TCOH excreted in bile
(mg)
-
2.13
-
CTCOG
Bound TCOH concentration in blood
-
2.76
-
CTCOGTCOH
Bound TCOH concentration in blood in free TCOH
equivalents
1.49
-
-
CLivTCOGTCOH
Bound TCOH concentration in liver in free TCOH
equivalents (mg/L)
1.63
-
-
AUrnTCOGTCOH
Cumulative amount of total TCOH excreted in urine
(mg)
1.26
1.12-2.27
1.11-1.13
AUrnTCOGTCOHcol
lect
Cumulative amount of total TCOH collected in urine
(noncontinuous sampling) (mg)
-
-
1.3-1.63
AUrnTCTotMole
Cumulative amount of TCA+total TCOH excreted in
urine (mmol)
-
1.12-1.54
-
CDCVGmol
DCVG concentration in blood (mmol/L)
-
-
1.53
AUrnNDCVC
Cumulative amount of NAcDCVC excreted in urine
(mg)
-
1.17
1.17
3
4 Values higher than twofold are in bold.
This document is a draft for review purposes only and does not constitute Agency policy.
128 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
In terms of total metabolism, closed-chamber data (see Figure 3-9, panel A) were fit
accurately with the updated model, with a small residual-error GSD of 1.18. While the previous
analyses of Hack et al. (2006) allowed for each chamber experiment to be fit with different
parameters, the current analysis made the more restrictive assumption that all experiments in a
single study utilize the same parameters. Furthermore, the accuracy of closed-chamber
predictions did not require the very high values for cardiac output that were used by Fisher et al.
(1991), confirming the suggestion (discussed in Appendix A) that additional respiratory
metabolism would resolve this discrepancy. The accurate model means that uncertainty with
respect to possible wash-in/wash-out, respiratory metabolism, and extrahepatic metabolism could
be well characterized. For instance, the absence of in vivo data on GSH metabolism in mice
means that this pathway remains relatively uncertain; however, the current model should be
reliable for estimating lower and upper bounds on the GSH pathway flux.
In terms of the parent compound TCE (see Figure 3-9, panels B-G), the parent PBPK
model (for TCE) appears to now be robust, with the exception of the remaining over-prediction
of TCE in blood following inhalation exposure. As expected, the venous-blood TCE
concentration had the largest residual-error, with a GSD of 2.7, reflecting largely the difficulty in
fitting TCE blood levels following inhalation exposure. In addition, the fat and kidney TCE
concentrations also are somewhat uncertain, with a GSD for the residual-error of 2.5 and 2.2,
respectively. These tissues were only measured in two studies, Abbas and Fisher (1997) and
Greenberg et al. (1999), and the residual-error reflects the difficulties in simultaneously fitting
the model to the different dose levels with the same parameters. Residual-error GSDs for other
TCE measurements were less than twofold. Thus, most of the problems previously encountered
with the Abbas and Fisher (1997) gavage data were solved by allowing absorption from both the
stomach and duodenal compartments. Notably, the addition of possible wash-in/wash-out,
respiratory metabolism, and extrahepatic metabolism (i.e., kidney GSH conjugation) was
insufficient to remove the long-standing discrepancy of PBPK models over-predicting TCE
blood levels from mouse inhalation exposures, suggesting another source of model or
experimental error is the cause. However, the availability of tissue concentration levels of TCE
somewhat ameliorates this limitation.
In terms of TCA and TCOH, the overall mass balance and metabolic disposition to these
metabolites also appeared to be robust, as urinary excretion following dosing with TCE, TCOH,
as well as TCA could be modeled accurately (see Figure 3-9, panels K and Q). The residual
GSDs for the urinary excretions are small: 1.34 for TCA and 1.26 for total TCOH. In addition,
the blood and tissue concentrations were also accurately predicted (see Figure 3-9, panels H-J,
L-P). All the residual GSDs were less than twofold, with those for TCA measurements less
thanl .5-fold. This improvement over the Hack et al. (2006) model was likely due in part to the
This document is a draft for review purposes only and does not constitute Agency policy.
129 DRAFT—DO NOT CITE OR QUOTE
-------
1 10' icr -icr 10
ClnhPPM Data
10"' 1 10' 10" 10'
CVen Data
10"' 1 10' 10" 10'
CBIdMix Data
CPIasTCA Data
AExhpost Data
CBIdTCA Data
Figure 3-9. Comparison of mouse data and PBPK model predictions from a
random posterior sample. Each panel shows results for a different
measurement. The solid line represents prediction = data, and the grey dotted
lines show prediction = data M GSDcn and data •: GSDerr, where GSDerr is the
median estimate of the residual-error GSD shown in Table 3-41.
This document is a draft for review purposes only and does not constitute Agency policy,
130 DRAFT—DO NOT CITE OR QUOTE
-------
CLivTCOH Data
CTCOGTCOH Data
TotCTCOH Data
CLivTCOGTCOH Data
AUrnTCOGTCOH Data
¦¦¦I i i i i i i i i i i i ¦ ¦ ¦
1 10' 102 1 03 10~2 10~1 1 101 10"2 10~1 1 101
CLivTCA Data AUrnTCA Data CTCOH Data
Figure 3-9 (continued). Comparison of mouse data and PBPK model
predictions from a random posterior sample. Each panel shows results for a
different measurement. The solid line represents prediction = data, and the grey
dotted lines show prediction = data x GSDerr and data ^ GSDerr, where GSDerr is
the median estimate of the residual-error GSD shown in Table 3-41.
This document is a draft for review purposes only and does not constitute Agency policy.
131 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-42. Summary comparison of updated PBPK model predictions and
2 in vivo data in mice
3
Study
Exposure(s)
Discussion
Abbas and Fisher
(1997)
TCE gavage
(corn oil)
Generally, model predictions were quite good, especially with respect
to tissue concentrations of TCE, TCA, and TCOH. There were some
discrepancies in TCA and TCOG urinary excretion and TCA and TCOG
concentrations in blood due to the requirement (unlike in Hack et al., 2006)
that all experiments in the same study utilize the same parameters. Thus,
for instance, TCOG urinary excretion was accurately predicted at
300 mg/kg, underpredicted at 600 mg/kg, over-predicted at 1,200 mg/kg,
and underpredicted again at 2,000 mg/kg, suggesting significant
intraexperimental variability (not addressed in the model).
Population predictions were quite good, with the almost all of the data
within the 95% CI of the predictions, and most within the interquartile
region.
Abbas et al. (1997)
TCOH, TCA
i.v.
Both subject-specific and population predictions were quite good.
Urinary excretion, which was over-predicted by the Hack et al. (2006)
model, was accurately predicted due to the allowance of additional
"untracked" clearance. In the case of population predictions, almost all of
the data were within the 95% CI of the predictions, and most within the
interquartile region.
Fisher and Allen
(1993)
TCE gavage
(corn oil)
Both subject-specific and population predictions were quite good.
Some discrepancies in the time-course of TCE blood concentrations were
evidence across doses in the subject-specific predictions, but not in the
population predictions, suggesting significant intrasubject variability (not
addressed in the model).
Fisher etal. (1991)
TCE inhalation
Blood TCE levels during and following inhalation exposures were still
over-predicted at the higher doses. However, there was the stringent
requirement (absent in Hack et al., 2006) that the model utilize the same
parameters for all doses and in both the closed and open-chamber
experiments. Moreover, the Hack et al. (2006) model required significant
differences in the parameters for the different closed-chamber experiments,
while predictions here were accurate utilizing the same parameters across
different initial concentrations. These conclusions were the same for
subject-specific and population predictions (e.g., TCE blood levels
remained over-predicted in the later case).
Green and Prout
(1985)
TCE gavage
(corn oil)
Both subject-specific and population predictions were adequate,
though the data collection was sparse. In the case of population
predictions, almost all of the data were within the 95% CI of the
predictions, and about half within the interquartile region.
Greenberg et al.
(1999)
TCE inhalation
Model predictions were quite good across a wide variety of measures
that included tissue concentrations of TCE, TCA, and TCOH. However, as
with the Hack et al. (2006) predictions, TCE blood levels were
over-predicted by up to twofold. Population predictions were quite good,
with the exception of TCE blood levels. Almost all of the other data was
within the 95% CI of the predictions, and most within the interquartile
region.
4
This document is a draft for review purposes only and does not constitute Agency policy.
132 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Table 3-42. Summary comparison of updated PBPK model predictions and
in vivo data in mice (continued)
Study
Exposure(s)
Discussion
Larson and Bull
(1992b)
TCE gavage
(aqueous)
Both subject-specific and population predictions were quite good,
though the data collection was somewhat sparse. In the case of population
predictions, all of the data were within the 95% CI of the predictions.
Larson and Bull
(1992a)
TCA gavage
(aqueous)
Both subject-specific and population predictions were quite good. In
the case of population predictions, most of the data were within the
interquartile region.
Merdink et al.
(1998)
TCE i.v.
Both subject-specific and population predictions were quite good,
though the data collection was somewhat sparse. In the case of population
predictions, all of the data were within the 95% CI of the predictions.
Prout et al. (1985)
TCE gavage
(corn oil)
Both subject-specific and population predictions were adequate,
though there was substantial scatter in the data due to the use of single
animals at each data point.
Templin et al.
(1993)
TCE gavage
(aqueous)
Both subject-specific and population predictions were quite good.
With respect to population predictions, almost all of the other data was
within the 95% CI of the predictions, and most within the interquartile
region.
addition of nonurinary clearance ("untracked" metabolism) of TCA and TCOH. Also, the
addition of a liver compartment for TCOH and TCOG, so that first-pass metabolism could be
properly accounted for, was essential for accurate simulation of the metabolite pharmacokinetics
both from intravenous (i.v.) dosing of TCOH and from exposure to TCE.
3.5.6.1.5. Rat model and data
Each panel of Figure 3-10 shows a scatter plot of the calibration data and a random
posterior prediction for each of the measured endpoint. The endpoint abbreviations are listed in
Table 3-41, as are the implied GSDs for the "residual" errors, which include intrastudy
variability, interindividual variability, and measurement and model errors. The residual-error
GSDs are also shown as grey dashed or dotted lines in Figure 3-10. A summary evaluation of
the predictions of the rat model as compared to the data are provided in Tables 3-43 and 3-44,
with figures showing individual time-course data and predictions in Appendix A.
Similar to previous analyses (Hack et al., 2006), the TCE submodel for the rat appears to
be robust, accurately predicting blood and tissue concentrations (see Figure 3-10, panels A-K),
with residual-error GSDs generally less than twofold. The only exceptions are the predictions of
venous blood from Kimmerle and Eben (1973a), which have residual-error GSDs greater than
fourfold, and and predictions of fat concentrations from Simmons et al. (2002), with
This document is a draft for review purposes only and does not constitute Agency policy.
13 3 DRAFT—DO NOT CITE OR QUOTE
-------
1 residual-error GSD of 2.7-fold. For Kimmerle and Eben (1973b), the inaccuracy was primarily
2 at the 3,000-ppm exposure, which might reflect other factors related to the high exposure. For
CBIdMix Data
CKid Data
7 Figure 3-10. Comparison of rat data and PBPK model predictions from a
8 random posterior sample. Each panel shows results for a different
9 measurement. The solid line represents prediction = data, and the grey lines show
10 prediction = data x GSDerr and data ^ GSDerr, where GSDerr is the lowest (dotted)
11 and highest (dashed) median estimate of the residual-error GSD shown in
12 Table 3-41.
13
10_1
1CP
CFat Data
ClnhPPM Data
This document is a draft for review purposes only and does not constitute Agency policy.
134 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
0
1
10" 1 10'
CMus Data
I ""»"l I I I
10~2 10"' 1 101 102
AExhpost Data
1 10' 1(T
CPIasTCA Data
1 101 102
CBIdTCA Data
rmj
1 101 102
CLivTCA Data
TTTT]
10"2 10"1 1
AUrnTCA Data
ABileTCOG Data
10~1 1 10' w
TotCTCOH Data
10"1 1 10' 10'
CTCOH Data
Figure 3-10 (continued). Comparison of rat data and PBPK model
predictions from a random posterior sample. Each panel shows results for a
different measurement. The solid line represents prediction=data, and the grey
lines show prediction = data x GSDerr and data ^ GSDerr, where GSDerr is the
lowest (dotted) and highest (dashed) median estimate of the residual-error GSD
shown in Table 3-41.
This document is a draft for review purposes only and does not constitute Agency policy.
135 DRAFT—DO NOT CITE OR QUOTE
-------
1
o
o
t
o
-4
-3
2 AUmNDCVC Data
3
4 Figure 3-10 (continued). Comparison of rat data and PBPK model
5 predictions from a random posterior sample. Each panel shows results for a
6 different measurement. The solid line represents prediction=data, and the grey
7 lines show prediction = data x GSDerr and data ^ GSDerr, where GSDerr is the
8 lowest (dotted) and highest (dashed) median estimate of the residual-error GSD
9 shown in Table 3-41.
10
10"1 1
CTCOG Data
1 101
AUmTCOGTCOH Data
10"2 10" 1
AUmTCTotMole Data
This document is a draft for review purposes only and does not constitute Agency policy.
136 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
Simmons et al. (2002), the high residual-error appears to reflect scatter due to intra-study
variability. Unlike in the mouse, some data consisting of TCE blood and tissue concentrations
were used for "out-of-sample evaluation" (sometimes loosely termed "validation"). These data
were generally well simulated (see Table 3-44); most of the data within the 95% CI of posterior
predictions. This provides additional confidence in the predictions for the parent compound.
In terms of TCA and TCOH, as with the mouse, the overall mass balance and metabolic
disposition to these metabolites also appeared to be robust: urinary excretion following dosing
with TCE, TCOH, as well as TCA, could be modeled accurately (see Figure 3-10 panels O, T,
and U), with the residual-errors also indicating good predictions in most cases. Residual-error
for these measurements was larger for Green and Prout (1985), Prout et al. (1985), and Stenner
et al. (1997), ranging from GSD of 1.8-2.3, reflecting largely intra-study variability.
Residual-errors for the other studies had GSDs of 1.1-1.5. This improvement over the Hack
et al. (2006) model was likely due in part to the addition of nonurinary clearance ("untracked"
metabolism) of TCA and TCOH. In addition, the addition of a liver compartment for TCOH and
TCOG, so that first-pass metabolism could be properly accounted for, was essential for accurate
simulation of the metabolite pharmacokinetics both from i.v. dosing of TCOH and from TCE
exposure. Blood and plasma concentrations of TCA and free or total TCOH were also fairly
well simulated (see Figure 3-10, panels L, M, P, Q, and S), with GSD for the residual-error of
1.1-1.6. A bit more discrepancy (residual-error GSD of 1.7) was evident with TCA liver
concentrations (see Figure 3-10, panel N). However, TCA liver concentrations were only
available in one study (Yu et al., 2000), and the data show a change in the ratio of liver to blood
concentrations at the last time point, which may be the source of the added residual-error.
Predictions of biliary excretion of TCOG in bile-cannulated rats (see Figure 3-10, panel R), from
Green and Prout (1985), and TCOG in blood (see Figure 3-10, panel S), from Stenner et al.
(1997), were less accurate, with residual-error GSD > 2. However, the biliary excretion data
consisted of a single measurement, and the amount of free TCOH in the same experiment from
Stenner et al. (1997) was accurately predicted.
In terms of total metabolism, as with the mouse, closed-chamber data (see Figure 3-10,
panel A) were fit accurately with the updated model (residual-error GSD of about 1.1). In
addition, the data on NAcDCVC urinary excretion was well predicted (see Figure 3-10, panel V),
with residual-error GSD of 1.18. In particular, the fact that excretion was still ongoing at the end
of the experiment was accurately predicted (see Figure 3-11, panels A and B). Thus, there is
greater confidence in the estimate of the flux through the GSH pathway than there was from the
Hack et al. (2006) model. However, the overall flux is still estimated indirectly, and there
remains some ambiguity as to the relative contributions of respiratory wash-in/wash-out,
respiratory metabolism, extrahepatic metabolism, DCVC bioactivation versus A'-acetylation, and
This document is a draft for review purposes only and does not constitute Agency policy.
13 7 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-43. Summary comparison of updated PBPK model predictions and
2 in vivo data used for "calibration" in rats
3
Study
Exposure(s)
Discussion
Bernauer et al.
(1996)
TCE inhalation
Posterior fits to these data were adequate, especially with the
requirement that all predictions for dose levels utilize the same PBPK model
parameters. Predictions of TCOG and TCA urinary excretion was relatively
accurate, though the time-course of TCA excretion seemed to proceed more
slowly with increasing dose, an aspect not captured in by model.
Importantly, unlike the Hack et al. (2006) results, the time-course of
NAcDCVC excretion was quite well simulated, with the excretion rate
remaining nonnegligible at the last time point (48 hour). It is likely that the
addition of the DCVG submodel between TCE and DCVC, along with prior
distributions that accurately reflected the lack of reliable independent (e.g.,
in vitro) data on bioactivation, allowed for the better fit.
Dallas et al.
(1991)
TCE inhalation
These data, consisting of arterial blood and exhaled breath
concentrations of TCE, were accurately predicted by the model using both
subject-specific and population sampled parameters. In the case of
population predictions, most of the data were within the 95% CI of the
predictions.
Fisher et al.
(1989)
TCE inhalation
These data, consisting of closed-chamber TCE concentrations, were
accurately simulated by the model using both subject-specific and
population sampled parameters. In the case of population predictions, most
of the data were within the 95% CI of the predictions.
Fisher et al.
(1991)
TCE inhalation
These data, consisting of TCE blood, and TCA blood and urine
time-courses, were accurately simulated by the model using both
subject-specific and population sampled parameters. In the case of
population predictions, most of the data were within the 95% CI of the
predictions.
Green and Prout
(1985)
TCE gavage
(corn oil)
TCA i.v.
TCA gavage
(aqueous)
For TCE treatment, these data, consisting of one time point each in
urine for TCA, TCA +TCOG, and TCOG, were accurately simulated by
both subject-specific and population predictions.
For TCA i.v. treatment, the single datum of urinary TCA+TCOG at 24
hour was at the lower 95% CI in the subject-specific simulations, but
accurately predicted with the population sampled parameters, suggesting
intrastudy variability is adequately accounted for by population variability.
For TCA gavage treatment, the single datum of urinary TCA+TCOG at
24 hour was accurately simulated by both subject-specific and population
predictions.
Hissink et al.
(2002)
TCE gavage
(corn oil)
TCE i.v.
These data, consisting of TCE blood, and TCA+TCOG urinary
excretion time-courses, were accurately simulated by the model using
subject-specific parameters. In the case of population predictions,
TCA+TCOH urinary excretion appeared to be somewhat under-predicted.
Kaneko et al.
(1994b)
TCE inhalation
These data, consisting of TCE blood and TCA and TCOG urinary
excretion time-courses, were accurately predicted by the model using both
subject-specific and population sampled parameters. In the case of
population predictions, TCA+TCOH urinary excretion appeared to be
somewhat underpredicted, However, all of the data were within the 95% CI
of the predictions.
4
This document is a draft for review purposes only and does not constitute Agency policy.
13 8 DRAFT—DO NOT CITE OR QUOTE
-------
Table 3-43. Summary comparison of updated PBPK model predictions and
in vivo data used for "calibration" in rats (continued)
Study
Exposure(s)
Discussion
Keys et al.
(2003)
TCE inhalation,
gavage
(aqueous), i.a.
These data, consisting of TCE blood, gut, kidney, liver, muscle and fat
concentration time-courses, were accurately predicted by the model using
both subject-specific and population sampled parameters. In the case of
population predictions, most of the data were within the 95% CI of the
predictions.
Kimmerle and
Eben (1973a)
TCE inhalation
Some inaccuracies were noted in subject-specific predictions,
particularly with TCA and TCOG urinary excretion, TCE exhalation
postexposure, and TCE venous blood concentrations. In the case of TCA
excretion, the rate was underpredicted at the lowest dose (49 mg/kg) and
over-predicted at 330 ppm. In terms of TCOG urinary excretion, the rate
was over-predicted at 175 ppm and underpredicted at 330 ppm. Similarly
for TCE exhaled postexposure, there was some over-prediction at 175 ppm
and some underprediction at 300 ppm. Finally, venous blood concentrations
were over-predicted at 3,000 ppm. However, for population predictions,
most of the data were within with 95% confidence region.
Larson and Bull
(1992b)
TCA gavage
(aqueous)
These data, consisting of TCA plasma time-courses, were accurately
predicted by the model using both subject-specific and population sampled
parameters. In the case of population predictions, all of the data were within
the 95% CI of the predictions.
Larson and Bull
(1992a)
TCE gavage
(aqueous)
These data, consisting of TCE, TCA, and TCOH in blood, were
accurately predicted by the model using both subject-specific and population
sampled parameters. In the case of population predictions, all of the data
were within the 95% CI of the predictions.
Lee et al.
(2000a)
TCE i.v., p.v.
These data, consisting of TCE concentration time course in mixed
arterial and venous blood and liver, were predicted using both the subject
specific and population predictions. In both cases, most of the data were
within the 95% CI of the predictions.
Merdink et al.
(1999)
TCOH i.v.
TCOH blood concentrations were accurately predicted using
subject-specific parameters. However, population-based parameters seemed
to lead to some under-prediction, though most of the data were within the
95% CI of the predictions.
Prout et al.
(1985)
TCE gavage
(corn oil)
Most of these data were accurately predicted using both subject-specific
and population-sampled parameters. However, at the highest two doses
(1,000 and 2,000 mg/kg), there were some discrepancies in the (very
sparsely collected) urinary excretion measurements. In particular, using
subject-specific parameters, TCA+TCOH urinary excretion was
under-predicted at 1,000 mg/kg and over-predicted at 2,000 mg/kg. Using
population-sampled parameters, this excretion was underpredicted in both
cases, though not entirely outside of the 95% CI.
Simmons et al.
(2002)
TCE inhalation
Most of these data were accurately predicted using both subject-specific
and population-sampled parameters. In the open-chamber experiments,
there was some scatter in the data that did not seem to be accounted for in
the model. The closed-chamber data were accurately fit.
This document is a draft for review purposes only and does not constitute Agency policy.
13 9 DRAFT—DO NOT CITE OR QUOTE
-------
Table 3-43. Summary comparison of updated PBPK model predictions and
in vivo data used for "calibration" in rats (continued)
Study
Exposure(s)
Discussion
Stenner et al.
(1997)
TCE
intraduodenal
TCOH i.v.
TCOH i.v.,
bile-cannulated
These data, consisting of TCA and TCOH in blood and TCA and TCOG
in urine, were generally accurately predicted by the model using both
subject-specific and population sampled parameters. However, using
subject-specific parameters, the amount of TCOG in urine was
over-predicted for 100 TCOH mg/kg i.v. dosing, though total TCOH in
blood was accurately simulated. In addition, in bile-cannulated rats, the
TCOG excretions at 5 and 20 mg/kg i.v. were underpredicted, while the
amount at 100 mg/kg was accurately predicted. On the other hand, in the
case of population predictions, all of the data were within the 95% CI of the
predictions, and mostly within the interquartile region, even for TCOG
urinary excretion. This suggests that intrastudy variability may be a source
of the poor fit in using the subject-specific parameters.
Templin et al.
(1995b)
TCE oral
(aqueous)
These data, consisting of TCE, TCA, and TCOH in blood, were
accurately predicted by the model using both subject-specific and population
sampled parameters. In the case of population predictions, all of the data
were within the 95% CI of the predictions.
Yu et al. (2000)
TCA i.v.
These data, consisting of TCA in blood, liver, plasma, and urine, were
generally accurately predicted by the model using both subject-specific and
population sampled parameters. The only notable discrepancy was at the
highest dose of 50 mg/kg, in which the rate of urinary excretion from 0-6
hour appeared to more rapid than the model predicted. However, all of the
data were within the 95% CI of the predictions based on population-sampled
parameters.
2 i.a. = intra-arterial, i.v. = intravenous, p.v. = intraperivenous.
This document is a draft for review purposes only and does not constitute Agency policy.
140 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-44. Summary comparison of updated PBPK model predictions and
2 in vivo data used for "out-of-sample" evaluation in rats
3
Study
Exposure(s)
Discussion
Andersen et al.
(1987a)
TCE inhalation
These closed-chamber data were well within the 95% CI of the
predictions based on population-sampled parameters.
Bruckner et al.
unpublished
TCE inhalation
These data on TCE in blood, liver, kidney, fat, muscle, gut, and
venous blood, were generally accurately predicted based on
population-sampled parameters. The only notable exception was TCE in
the kidney during the exposure period at the 500 ppm level, which were
somewhat under-predicted (though levels postexposure were accurately
predicted).
Fisher et al.
(1991)
TCE inhalation
These data on TCE in blood were well within the 95% CI of the
predictions based on population-sampled parameters.
Jakobson et al.
(1986)
TCE inhalation
These data on TCE in arterial blood were well within the 95% CI of
the predictions based on population-sampled parameters.
Lee et al. (1996)
TCE i.a., i.v., p.v.,
gavage
Except at some very early time-points (<0.5 hour), these data on
TCE in blood were well within the 95% CI of the predictions based on
population-sampled parameters.
Lee et al. (2000a;
2000b)
TCE gavage
These data on TCE in blood were well within the 95% CI of the
predictions based on population-sampled parameters.
4
5 i.a. = intra-arterial, i.v. = intravenous, p.v. = intraperivenous.
6
7
8 oxidation in the liver producing something other than TCOH or TCA. Therefore, there remains a
9 large range of possible values for the flux through the GSH conjugation and other indirectly
10 estimated pathways that are nonetheless consistent with all the available in vivo data. The use of
11 noninformative priors for the metabolism parameters for which there were no in vitro data means
12 that a fuller characterization of the uncertainty in these various metabolic pathways could be
13 achieved. Thus, the model should be reliable for estimating lower and upper bounds on several
14 of these pathways.
15
3.5.6.1.6. Human model and data
16 Each panel of Figure 3-12 shows a scatter plot of the calibration data and a random
17 posterior prediction for each of the measured endpoint. The endpoint abbreviations are listed in
18 Table 3-41, as are the implied GSDs for the "residual" errors, which include intrastudy
19 variability, interindividual variability, and measurement and model errors. The residual-error
20 GSDs are also shown as grey dashed or dotted lines in Figure 3-12. Table 3-45-3-46 provide a
21 summary evaluation of the predictions of the model as compared to the human data, with figures
22 showing individual time-course data and predictions in Appendix A.
This document is a draft for review purposes only and does not constitute Agency policy.
141 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
This document is a draft for review purposes only and does not constitute Agency policy.
142 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Hack et al. (2006) model (raf)
CP
3.
a
I
CM
<
I
m
o
o
o
A
+
+
+
A
i
i
i
! -M<
1
1
1
\
\
1
1
o >;
i
i
i
o
o
1
10
20
Time (h)
r
30
~r
40
CD
"a
0
1_
o
x
(D
C?
>
0
Q
1
«N
O
CM
0
<
1
Updated model (rat)
CM
Time (h)
Hack et al. (2006) model (human)
CP
3.
v
V
X
a
l
CM
<
I
Ui
"O
<0
¦*->
(D
0
X
(D
C?
>
O
a
1
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
With respect to the TCE submodel, retained dose, blood and exhaled air measurements
(see Figure 3-12, panels A-C) appeared more robust than previously found from the Hack et al.
(2006) model. TCE blood concentrations from most studies were well predicted, with
residual-error GSD in most studies < twofold. However, those from Chiu et al. (2007) were
consistently over-predicted (i.e., data <0.1 mg/L in Figure 3-12, panel C), with residual-error
GSD of almost threefold, and a few of those from Fisher et al. (1989) were consistently
underpredicted. Alveolar breath concentrations and retained dose of TCE were well predicted
(residual-error GSD < 1.5-fold) from all studies except Fisher et al. (1998), which had a
residual-error GSD of 1.8-fold. However, the discrepancy in alveolar breath appeared smaller
than that originally reported by Fisher et al. (1998) for their PBPK model. In addition, the
majority of the "out-of-sample" evaluation data consisted of TCE in blood or breath, and were
generally well predicted (see Table 3-46), lending confidence to the model predictions for the
parent compound.
In terms of TCA and TCOH, as with the mouse and rat, the overall mass balance and
metabolic disposition to these metabolites also appeared to be robust, as urinary excretion
following TCE exposure could be modeled accurately (see Figure 3-12, panels F, G, J, and K).
In most cases, the residual-error GSD was less than twofold. However, TCA urinary data from
Chiu et al. (2007) (panel G in Figure 3-12) indicated greater interoccasion variability, reflected in
the residual-error GSD of 2.8. In this study, the same individual exposed to the same
concentration on different occasions sometimes had substantial differences in urinary excretion.
In addition, many TCA urine measurements in this study were saturated, and had to be omitted,
and the fact that the remaining data were sparse and possibly censored may have contributed to
the greater intrastudy variability. Blood and plasma concentrations of TCA and free TCOH (see
Figure 3-12, panels D, E, and H) were fairly well simulated, with GSD for the residual-error of
1.1-1.4, though total TCOH in blood (see Figure 3-12, panel I) had slightly greater residual-error
with GSD of about 1.6. This partially reflects the "sharper" peak concentrations of total TCOH
in the Chiu et al. (2007) data relative to the model predictions. In addition, TCA and TCOH
blood and urine data were available from several studies for "out-of-sample" evaluation and
were generally well predicted by the model (see Table 3-46), lending further confidence to the
model predictions for these metabolites.
In terms of total metabolism, no closed-chamber data exist in humans, but, as discussed
above, alveolar breath concentrations and retained dose (see Figure 3-12, panels A and B) were
generally well simulated, suggesting that total metabolism may be fairly robust. In addition, as
with the rat, the data on NAcDCVC urinary excretion was well predicted (see Figure 3-11,
Figure 3-12 panel M), with residual-error GSD of 1.12). In particular the model accurately
predicted the fact that excretion was still ongoing at the end of the experiment (48 hrs after the
This document is a draft for review purposes only and does not constitute Agency policy.
144 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
CAIvPPM Data
1 T 1 T 1 I . I | '
103
RetDose Data
CVen Data
—|—i 11miij—i i ii»ii|—i 11 mi'i i 11 imij—i i inii^—i 11imi|—i i niiq
icr3 «r2 icr1 1 io1 102 103 io4
CPIasTCA Data
AUrnTCA Data
icr1 1
AUrnTCA collect Data
CBIdTCA Data
CTCOH Data
TotCTCOH Data
Figure 3-12. Comparison of human data and PBPK model predictions from
a random posterior sample. Each panel shows results for a different
measurement. The solid line represents prediction = data, and the grey lines show
prediction = data « GSDerr and data ^ GSDerr, where GSDerr is the lowest (dotted)
and highest (dashed) median estimate of the residual-error GSD shown in
Table 3-41.
This document is a draft for review purposes only and does not constitute Agency policy,
145 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
1 10' 10"
AUrnTCOGTCOH Data
10"" 10"'
AUrnNDCVC Data
o -=
10"° 10"z 10"' 1 10'
AUrnTCOGTCOH_collect Data
10"'
CDCVGmol Data
Figure 3-12 (continued). Comparison of rat data and PBPK model
predictions from a random posterior sample. Each panel shows results for a
different measurement. The solid line represents prediction = data, and the grey
lines show prediction = data x GSDerr and data ^ GSDerr, where GSDerr is the
lowest (dotted) and highest (dashed) median estimate of the residual-error GSD
shown in Table 3-41.
This document is a draft for review purposes only and does not constitute Agency policy.
146 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 3-45. Summary comparison of updated PBPK model predictions and
2 in vivo data used for "calibration" in humans
3
Reference
Exposure(s)
Discussion
Bernauer et al.
(1996)
TCE inhalation
These data, consisting of TCA, TCOG and NAcDCVC excreted in
urine, were accurately predicted by the model using both individual-specific
and population sampled parameters. The posterior NAcDCVC predictions
were an important improvement over the predictions of Hack et al. (2006),
which predicted much more rapid excretion than observed. The fit
improvement is probably a result of the addition of the DCVG submodel
between TCE and DCVC, along with the broader priors on DCVC excretion
and bioactivation. Interestingly, in terms of population predictions, the
NAcDCVC excretion data from this study were on the low end, though still
within the 95% CI.
Chiu et al.
(2007)
TCE inhalation
Overall, posterior predictions were quite accurate across most of the
individuals and exposure occasions. TCE alveolar breath concentrations
were well simulated for both individual-specific and population-generated
simulations, though there was substantial scatter (intraoccasion variability).
However, TCE blood concentrations were consistently over-predicted in most
of the experiments, both using individual-specific and population-generated
parameters. This was not unexpected, as Chiu et al. (2007) noted the TCE
blood measurements to be lower by about twofold relative to previously
published studies. As discussed in Chiu et al. (2007) wash-in/wash-out and
extrahepatic (including respiratory) metabolism were not expected to be able
to account for the difference, and indeed all these processes were added to the
current model without substantially improving the discrepancy.
With respect to metabolite data, TCA and total TCOH in blood were
relatively accurately predicted. There was individual experimental variability
observed for both TCA and TCOH in blood at six hours (end of exposure).
The population-generated simulations over-predicted TCA in blood, while
they were accurate in predicting blood TCOH. Predictions of free TCOH in
blood also showed over-prediction for individual experiments, with
variability at the end of exposure timepoint. However, TCOH fits were
improved for the population-generated simulations. TCA and TCOG urinary
excretion was generally well simulated, with simulations slightly under- or
over-predicting the individual experimental data in some cases.
Fisher et al.
(1998)
TCE inhalation
The majority of the predictions for these data were quite accurate.
Interestingly, in contrast to the predictions for Chiu et al. (2007), TCE blood
levels were somewhat underpredicted in a few cases, both from using
individual-specific and population-generated predictions. These two results
together suggest some unaccounted-for study-to-study variance, though
interindividual variability cannot be discounted as the data from Chiu et al.
(2007) were from individuals in the Netherlands and that from Fisher et al.
(1998) were from individuals in the United States. As reported by Fisher
et al. (1998), TCE in alveolar air was somewhat over-predicted in several
cases, however, the discrepancies seemed smaller than originally reported for
the Fisher et al. model.
4
This document is a draft for review purposes only and does not constitute Agency policy.
147 DRAFT—DO NOT CITE OR QUOTE
-------
Table 3-45. Summary comparison of updated PBPK model predictions and
in vivo data used for "calibration" in humans (continued)
Reference
Exposure(s)
Discussion
Fisher et al.
(1998)
(continued)
TCE inhalation
(continued)
With respect to metabolite data, TCOH and TCA in blood and
TCOG and TCA in urine were generally well predicted, though data for some
individuals appeared to exhibit inter- and/or intraoccasion variability. For
example, in one case in which the same individual (female) was exposed to
both 50 and 100 ppm, the TCOH blood data was over-predicted at the higher
one exposure. In addition, in one individual, initial individual-specific
simulations for TCA in urine were underpredicted but shifted to
over-predictions towards the end of the simulations. The
population-generated results over-predicted TCA in urine for the same
individual. Given the results from Chiu et al. (2007) interoccasion variability
is likely to be the cause, though some dose-related effect cannot be ruled out.
Finally, DCVG data was well predicted in light of the high
variability in the data and availability of only grouped data or data from
multiple individual who cannot be matched to the appropriate TCE and
oxidative metabolite data set. In all cases, the basic shape (plateau and then
sharp decline) and order of magnitude of the time-course were well predicted,
Furthermore, the range of the data was well-captured by the 95% CI of the
population-generated predictions.
Kimmerle and
Eben (1973b)
TCE inhalation
These data were well fit by the model, using either
individual-specific or population-generated parameters.
Monster et al.
(1976)
TCE inhalation
The data simulated in this case were exhaled alveolar TCE, TCE in
venous blood, TCA in blood, TCA in urine, and TCOG in urine. Both using
individual-specific and population-generated simulations, all fits are within
the 95% CI. The one exception was the retained dose for a male exposed to
65 ppm, which was outside the 95% CI for the population-generated results.
Muller et al.
(1974)
TCA,
TCOH oral
The data measured after oral TCA was timecourse TCA measured in
plasma and urine. Individual-specific predictions were accurate, but both
data sets were over-predicted in the population-generated simulations.
The data measured after oral TCOH was timecourse TCOH in
blood, TCOG in urine, TCA in plasma, and TCA in urine.
Individual-specific predictions were accurate, but the population-generated
simulations over-predicted TCOH in blood and TCOG in urine. The
population-based TCA predictions were accurate.
These results indicate that "unusual" parameter values were
necessary in the individual-specific simulations to give accurate predictions.
Paykoc et al.
(1945)
TCA i.v.
These data were well fit by the model, using either
individual-specific or population-generated parameters.
1
2
This document is a draft for review purposes only and does not constitute Agency policy.
148 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Table 3-46. Summary comparison of updated PBPK model predictions and
in vivo data used for "out-of-sample" evaluation in humans
Reference
Exposure(s)
Discussion
Bartonicek (1962)
TCE inhalation
While these data were mostly within the 95% CI of the predictions,
they tended to be at the high end for all the individuals in the study.
Bloemenetal. (2001)
TCE inhalation
These data were all well within the 95% CI of the predictions.
Fernandez et al. (1977)
TCE inhalation
These data were all well within the 95% CI of the predictions.
Lapare et al. (1995)
TCE inhalation
These data were all well within the 95% CI of the predictions.
Monster et al. (1979)
TCE inhalation
These data were all well within the 95% CI of the predictions.
Muller et al. (1974; 1975)
TCE inhalation
Except for TCE in alveolar air, which was over-predicted during
exposure, these data were all well within the 95% CI of the
predictions.
Sato et al. (1977)
TCE inhalation
These data were all well within the 95% CI of the predictions.
Stewart etal. (1970)
TCE inhalation
These data were all well within the 95% CI of the predictions.
Triebig et al. (1976)
TCE inhalation
Except for TCE in alveolar air, these data were all well within the
95% CI of the predictions.
end of exposure). Thus, there is greater confidence in the estimate of the flux through this part of
the GSH pathway than there was from the Hack et al. (2006) model, in which excretion was
completed within the first few hours after exposure (see Figure 3-11, panels C and D).
If only urinary NAcDCVC data were available, as is the case for the rat, the overall GSH
conjugation flux would still be estimated indirectly, and there would remain some ambiguity as
to the relative contributions of respiratory wash-in/wash-out, respiratory metabolism,
extrahepatic metabolism, DCVC bioactivation versus A-acetylation, and oxidation in the liver
producing something other than TCOH or TCA. However, unlike in the rat, the blood DCVG
data, while highly variable, nonetheless provide substantial constraints (at least a strong lower
bound) on the flux of GSH conjugation, and is well fit by the model (see Figure 3-12, panel L,
and Figure 3-13). Importantly, the high residual-error GSD for blood DCVG reflects the fact
that only grouped or unmatched individual data were available, so in this case, the residual-error
includes interindividual variability, which is not included in the other residual-error estimates.
However, as discussed above in Section 3.3.3.2.1, there are uncertainties as to the accuracy of
analytical method used by Lash et al. (1999a) in the measurement of DCVG in blood. Because
these data are so determinative of the overall GSH conjugation flux, these analytical
uncertainties are important to consider in the overall evaluation of the PBPK model predictions
(see below, section 3.5.7).
For the other indirectly estimated pathways, there remain a large range of possible values
that are nonetheless consistent with all the available in vivo data. The use of noninformative
This document is a draft for review purposes only and does not constitute Agency policy.
149 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
priors for the metabolism parameters for which there were no in vitro data means that a fuller
characterization of the uncertainty in these various metabolic pathways could be achieved. Thus,
as with the rat, the model should be reliable for estimating lower and upper bounds on several of
these pathways.
3.5.6.1.7. Sensitivity Analysis With Respect to Calibration Data
To assess the informativeness of the calibration data to the parameters, local sensitivity
analysis is performed with respect to the calibration data points. For each scaling parameter, the
central difference is used to estimate the partial derivatives by centering on the sample mean of
its estimated population mean, and then increasing and decreasing by 5%. The relative change in
the model output f(9) is used to estimate a local sensitivity coefficient (SC) as follows:
SC = 10 x (/(9+) -X0-)}/[>/2 x {/(6+) +X0-)}]. Eq. 3-1
Here, /(0) is one of the model predictions of the calibration data, 0± is the maximum likelihood
estimate (MLE) or baseline value of ± 5%. For log-transformed parameters, 0.05 was added or
subtracted from the baseline value, whereas for untransformed parameters, the baseline value
was multiplied by 1.05 or 0.95. The resulting values of SC are binned into five categories
according to their sensitivity coefficient: negligible (|SC| < 0.01) very low (0.01 < |SC| < 0.1),
low (0.1 < |SC| < 0.5), medium (0.5 < |SC| < 1.0), and high (|SC| > 1.0).
Note that local sensitivity analyses as typically performed in deterministic PBPK
modeling can only inform the "primary" effects of parameter uncertainties—i.e., the direct
change on the quantity of interest due to change in a parameter. They cannot address the
propagation of uncertainties, such as those that can arise due to parameter correlations in the
parameter fitting process. Those can only be addressed in a global sensitivity analysis, which is
left for future research.
The results of local sensitivity analyses are shown in Figures 3-14-3-16. For each
parameter, the number of data points (out of the entire calibration set) which have sensitivity
coefficients in the various categories are shown graphically. As summarized in Table 3-47, most
This document is a draft for review purposes only and does not constitute Agency policy.
150 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Time (h)
of the parameters have at least some calibration data to which they are at least moderately
sensitive (|SC| > 0.5). Across species, the cardiac output (InQCC), ventilation-perfusion ratio
Figure 3-13. Comparison of DCVG concentrations in human blood and
predictions from the updated model. Data are mean concentrations for males
(A) and females (o) reported in Lash et al. (1999a) for humans exposed for
4 hours to 100 ppm TCE in air (thick horizontal line denotes the exposure period).
Data for oxidative metabolites from the same individuals were reported in Fisher
et al. (1998) but could not be matched with the individual DCVG data (Lash
2007, personal communication). The vertical error bars are standard errors of the
mean as reported in Lash et al. (1999a) (n = 8, so standard deviation is 80.5-fold
larger). Lines are PBPK model predictions for individual male (solid) and female
(dashed) subjects. Parameter values used for each prediction are a random sample
from the individual-specific parameters from the human MCMC chains (the last
iteration of the 1st chain was used). See files linked to Appendix A for
comparisons with the full distribution of predictions.
This document is a draft for review purposes only and does not constitute Agency policy.
151 DRAFT—DO NOT CITE OR QUOTE
-------
Number of mouse calibration data points
200 400 600 800 1000
1200
InQCC
InVPRC
QFatC
QGutC
QLivC
QSIw C
InDRespC
QKidC
FracRasC
VFatC
VGutC
VLivC
VRapC
VRespLumC
VRespEffC
VKidC
VBIdC
InPBC
InPFatC
InPGutC
InPLivC
InPRapC
InPRespC
InPKidC
InPSIw C
InPRBCRasTCAC
InPBodTCAC
InPLivTCAC
InkDissocC
InBMaxkDC
InPBodTCOHC
InPLivTCOHC
InPBodTCOGC
InPLivTCOGC
InkTSD
InkAS
InkAD
InkASTCA
InVMaxC
InKMC
InFracOtherC
InFracTCAC
InVMaxDCVGC
InCIDCVGC
InVMaxKidDCVGC
InCIKidDCVGC
InVMaxLungLivC
InKMCIara
InFracLungSysC
InVMaxTCOHC
InKIVTTCOH
InVMaxGlucC
InKMGluc
InkMetTCOHC
InkUrnTCAC
InkMetTCAC
InkBileC
InkEHRC
InkllrnTCOGC
1
|SC|<0.01 0.01<|SC|<0.1 !!!! 0.1<|SC|<0.5 » 0.5<|SC|<1 ¦ 1<|SC|
Figure 3-14. Sensitivity analysis results: Number of mouse calibration data
points with SC in various categories for each scaling parameter.
This document is a draft for review purposes only and does not constitute Agency policy.
152 DRAFT—DO NOT CITE OR QUOTE
-------
200
Number of rat calibration data points
400 600 800 1000
1200
InQCC
InVPRC
QFatC
QGutC
QLivC
QSIw C
InDRespC
QKidC
FracPlasC
VFatC
VGutC
VLivC
VRapC
VRespLumC
VRespEffC
VKidC
VBIdC
InPBC
InPFatC
InPGutC
InPLivC
InPRapC
InPRespC
InPKidC
InPSIw C
InPRBCRasTCAC
InPBodTCAC
InPLivTCAC
InkDissocC
InBMaxkDC
InPBodTCOHC
InPLivTCOHC
InPBodTCOGC
InPLivTCOGC
InkTSD
InkAS
InkAD
InkASTCA
InVMaxC
InKMC
InFracOtherC
InFracTCAC
InVMaxDCVGC
InCIDCVGC
InVMaxKidDCVGC
InCIKidDCVGC
InVMaxLungLivC
InKMCIara
InFracLungSysC
InVMaxTCOHC
InKIVTTCOH
InVMaxGlucC
InKMGluc
InkMetTCOHC
InkUrnTCAC
InkMetTCAC
InkBileC
InkEHRC
InkllrnTCOGC
InkNATC
InkKidBioactC
|SC|<0.01 0.01<|SC|<0.1 !! 0.1<|SC|<0.5 « 0.5<|SC|<1 ¦ 1<|SC|
Figure 3-15. Sensitivity analysis results: Number of rat calibration data
points with SC in various categories for each scaling parameter.
This document is a draft for review purposes only and does not constitute Agency policy.
153 DRAFT—DO NOT CITE OR QUOTE
-------
Number of human calibration data points
500 1000 1500 2000 2500 3000 3500 4000 4500 5000
a
-------
1 Table 3-47. Summary of scaling parameters ordered by fraction of
2 calibration data of moderate or high sensitivity
3
Mouse
Rat
Human
Fraction
Fraction
Fraction
with
with
with
Parameter
|SC|>0.5
Parameter
|SC|>0.5
Parameter
|SC|>0.5
lnVMAXC
0.4405
VLivC
0.4213
InQCC
0.4159
VLivC
0.428
InQCC
0.4182
InVPRC
0.3777
lnPBC
0.3233
InVPRC
0.4158
InClTCOHC
0.2871
InQCC
0.2454
lnVMAXC
0.3984
QGutC
0.2137
InkAD
0.1675
lnPBC
0.2893
InClGlucC
0.186
lnPBodTCAC
0.1642
VFatC
0.1455
lnkUrnTCAC
0.1789
InVPRC
0.1575
QFatC
0.1273
FracPlasC
0.1553
lnFracTCAC
0.1323
lnPBodTCAC
0.1162
lnPBodTCOHC
0.1486
IrtVV!\vGllLcC
0.1147
lnPFatC
0.1154
lnVMAXC
0.1358
lnPFatC
0.093
1ft V viavGIilcC
0.1083
lnPBC
0.1269
lnPLivTCAC
0.0896
QGutC
0.0885
VLivC
0.1225
lnkAS
0.0863
lnkUrnTCAC
0.0696
lnPBodTCAC
0.12
WatC
0.0762
lnPSlwC
0.0664
lnBMaxkDC
0.0897
InKMGluc
0.0762
lnFracTCAC
0.064
VBldC
0.0586
lnkMetTCAC
0.0762
InKMGluc
0.0625
lnkDCVGC
0.0515
lnkUrnTCAC
0.0754
lnkBileC
0.0538
lnPLivTCOGC
0.0446
InKMC
0.0653
lnPLivTCOGC
0.0514
InClDCVGC
0.0435
lnkUrnTCOGC
0.0544
lnPLivC
0.0482
lnkBileC
0.0422
lnVMAXLungLivC
0.0511
InkAD
0.0474
QFatC
0.0401
lnkTSD
0.0469
InKMC
0.0427
lnPSlwC
0.0372
QGutC
0.0452
lnVMAxTCOHC
0.0427
QSlwC
0.0345
QFatC
0.0402
lnPKidC
0.0324
lnKMTCOH
0.0305
lnPLivC
0.0402
lnPGutC
0.03
lnPFatC
0.0292
lnPLivTCOHC
0.0377
lnFracOtherC
0.03
InCIC
0.0288
lnPKidC
0.0352
lnPLivTCAC
0.0292
lnkUrnTCOGC
0.0282
lnPLivTCOGC
0.0352
lnBMaxkDC
0.0285
lnPRBCPlasTCAC
0.0147
lnPRBCPlasTCAC
0.031
lnkMetTCAC
0.0213
lnPLivTCAC
0.0135
lnVMAXTCOHC
0.0235
InV MAXLungLivC
0.0182
lnkMetTCAC
0.013
lnPBodTCOHC
0.0201
lnKMTCOH
0.0182
lnFracTCAC
0.0103
lnPSlwC
0.0134
lnkAS
0.0158
lnPBodTCOGC
0.0095
lnBMaxkDC
0.0134
lnPBodTCOHC
0.015
VRapC
0.0063
lnDRespC
0.0109
FracPlasC
0.0126
VKidC
0.0057
lnkBileC
0.0084
lnkTSD
0.0103
InClKidDCVGC
0.0057
FracPlasC
0.0059
VKidC
0.0095
InkNATC
0.0057
lnPBodTCOGC
0.005
lnVMAXKidDCVGC
0.0095
lnPRapC
0.005
VGutC
0.0025
InkNATC
0.0095
lnPLivTCOHC
0.005
lnPGutC
0.0025
lnDRespC
0.0063
lnkMetTCOHC
0.005
lnKMTCOH
0.0017
QSlwC
0.0055
lnFracOtherC
0.0046
lnkMetTCOHC
0.0017
lnPLivTCOHC
0.0016
VFatC
0.0036
lnkEHRC
0.0017
InkASTCA
0.0016
lnkEHRC
0.0036
QKidC
0.0008
lnkMetTCOHC
0.0016
lnDRespC
0.0011
VKidC
0.0008
VGutC
0.0008
lnKMDCVGC
0.0011
lnPRBCPlasTCAC
0.0008
lnkKidBioactC
0.0002
lnkUrnTCOGC
0.0008
4
5 Parameters not shown have no data with |SC| > 0.5
6
7
This document is a draft for review purposes only and does not constitute Agency policy.
15 5 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
(InVPRC), blood-air partition coefficient (InPBC), Vmax for oxidation (lnVMAxC), and VLivC
are consistently among the most sensitive parameters, with >10% of the calibration data
exhibiting |SC|>0.5 to these parameters. Note that the reason the liver volume is sensitive is that
it is used to scale the capacity or clearance rate for oxidation.
For scaling parameters for which all the calibration data are negligibly sensitive
(|SC| < 0.01), it is important that they either have informative prior data or are unimportant for
dose-metric predictions. For mice, these parameters are the volumes of the respiratory lumen
and tissue (VRespLumC, VRespEffC), the partition coefficient for the respiratory tissue
(InPRespC), and the Vmax values for GSH conjugation in the liver and kidney. For the
respiratory tract parameters, there are prior data to identify the parameters. Moreover, none of
the dose-metric predictions are sensitive to these parameters (see Section 3.5.7.2, below). For
GSH conjugation, it should be noted that for the clearance in the liver and lung (Vmax/Km), some
data are available with sensitivity 0.01 < |SC| <0.1. The data are not at all informative as to the
maximum capacity for GSH conjugation.
For rats, all the scaling parameters have at least one calibration data point with
|SC| > 0.01. However, for the volumes of the respiratory lumen and tissue (VRespLumC,
VRespEffC), the partition coefficient for the respiratory tissue (InPRespC), and the Vmax values
for GSH conjugation in the liver, these consist of only one or two data points. As with
mice,there are prior data to help identify the respiratory tract parameters. Moreover, none of the
dose-metric predictions are sensitive to the respiratory tract parameters (see Section 3.5.7.2,
below). The data are not very informative as to maximum capacity for GSH conjugation in the
liver. However, there are some data that have low or moderate informativeness (0.1 < |SC| < 1)
as to the maximum capacity for GSH conjugation in the kidney, and clearance via GSH
conjugation (Vmax/Km) in the liver and kidney, which have much greater impact on the
dose-metric predictions than the maximum capacity in the liver (see Section 3.5.7.2, below).
For humans, all the scaling parameters have at least one calibration data point with
|SC| > 0.01. However, for the volumes of the respiratory lumen and tissue (VRespLumC,
VRespEffC), the partition coefficient for the respiratory tissue (InPRespC), and the oral
absorption rate for TCA, these consist of only one or two data points. As with mice and rats,
there are prior data to help identify the respiratory tract parameters. Moreover, none of the
dose-metric predictions are sensitive to the respiratory or TCA oral absorption parameters (see
Section 3.5.7.2, below).
This document is a draft for review purposes only and does not constitute Agency policy.
156 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Therefore, the local sensitivity analysis with respect to calibration data confirms that
most of the scaling parameters are informed by at least some of the calibration data. In addition,
the parameters for which the calibration data have very little or negligible sensitivity are either
informed by prior data or have little impact on dose-metric predictions.
3.5.6.1.8. Summary Evaluation of Updated Physiologically Based Pharmacokinetic (PBPK)
Model
Overall, the updated PBPK model, utilizing parameters consistent with the available
physiological and in vitro data from published literature, provides reasonable fits to an extremely
large database of in vivo pharmacokinetic data in mice, rats, and humans. Posterior parameter
distributions were obtained by MCMC sampling using a hierarchical Bayesian population
statistical model and a large fraction of this in vivo database. Convergence of the MCMC
samples for model parameters was good for mice, and adequate for rats and humans. Evaluation
of posterior parameter distributions suggest reasonable results in light of prior expectations and
the nature of the available calibration data. In addition, in rats and humans, the model produced
predictions that are consistent with in vivo data from many studies not used for calibration
(insufficient studies were available in mice for such "out of sample" evaluation). Finally, the
local sensitivity analysis with respect to calibration data confirms that most of the scaling
parameters are informed by at least some of the calibration data, and those that were not either
were informed by prior data or would not have great impact on dose-metric predictions.
3.5.7. Physiologically Based Pharmacokinetic (PBPK) Model Dose-Metric Predictions
3.5.7.1.1. Characterization of Uncertainty and Variability
Since it is desirable to characterize the contributions from both uncertainty in population
parameters and variability within the population, so the following procedure is adopted. First,
500 sets of population parameters (i.e., population mean and variance for each parameter) are
extracted from the posterior MCMC samples—these represent the uncertainty in the population
parameters. To minimize autocorrelation, they were obtained by "thinning" the chains to the
appropriate degree. From each of these sets of population parameters, 100 subject-specific
parameters were generated by Monte Carlo—each of these represents the population variability,
given a particular set of population parameters. Thus a total of 50,000 subjects, representing
100 (variability) each for 500 different populations (uncertainty), were generated.
This document is a draft for review purposes only and does not constitute Agency policy.
157 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Each set was run for a variety of generic exposure scenarios. The combined distribution
of all 50,000 individuals reflects both uncertainty and variability—i.e., the case in which one is
trying to predict the dosimetry for a single random subject. In addition, for each dose-metric, the
mean predicted internal dose was calculated from each of the 500 sets of 100 individuals,
resulting in a distribution for the uncertainty in the population mean. Comparing the combined
uncertainty and variability distribution with the uncertainty distribution in the population mean
gives a sense of how much of the overall variation is due to uncertainty versus variability.
Figures 3-17-3-25 show the results of these simulations for a number of representative
dose-metrics across species continuously exposed via inhalation or orally. For display purposes,
dose-metrics have been scaled by total intake (resulting in a predicted "fraction" metabolized) or
exposure level (resulting in an internal dose per ppm for inhalation or per mg/kg-day for oral
exposures). In these figures, the thin error bars represent the 95% CI for overall uncertainty and
variability, and the thick error bars represent the 95% CI for the uncertainty in the population
mean. The interpretation of these figures is that if the thick error bars are much smaller (or
greater) than the thin error bars, then variability (or uncertainty) contributes the most to overall
uncertainty and variability.
For application to human health risk assessment, the uncertainty in and variability among
rodent internal dose estimates both contribute to uncertainty in human risk estimates. Therefore,
it is appropriate to combine uncertainty and variability when applying rodent dose-metric
predictions to quantitative risk assessment. The median and 95% CI for each dose-metric at
some representative exposures in rodents are given in Tables 3-48-3-49, and the CI in these
tables includes both uncertainty in the population mean and variance as well as variability in the
population. On the other hand, for use in predicting human risk, it is often necessary to separate,
to the extent possible, interindividual variability from uncertainty, and this disaggregation is
summarized in Table 3-50.
3.5.7.1.2. Local Sensitivity Analysis With Respect to Dose-Metric Predictions
To assess the parameter sensitivity of dose-metric predictions, a local sensitivity analysis
is performed. The representative exposure scenarios in Tables 3-48-3-50 are used, but with
metabolic flux dose-metrics converted to "fraction of intake" (i.e., amount metabolized through a
pathway divided by total dose). Each parameter is centered on the sample mean of its estimated
population mean, and then increased and decreased by 5%. The relative change in the model
outputX6) is used to estimate a local SC as follows:
This document is a draft for review purposes only and does not constitute Agency policy.
15 8 DRAFT—DO NOT CITE OR QUOTE
-------
1 sc = 10 x (xe+) -xe)}/11/*x {/(e+) +X0-)}]. Eq. 3-2
2
This document is a draft for review purposes only and does not constitute Agency policy.
159 DRAFT—DO NOT CITE OR QUOTE
-------
Fraction Metabolized
B
Fraction Metabolized
1
2
3
4
5
6
7
~
Mouse
0
Rat
0
Human
CO
o
ID
O
(N
O
O
O
2
CO
o
ID
o
(N
O
O
o
0.1
10
100 1000
Continuous inhalation (ppm )
~
Mouse
a
Rat
0
Human
0.1
10
100 1000
Continuous oral ( mg/kg-d )
Figure 3-17. PBPK model predictions for the fraction of intake that is
metabolized under continuous inhalation (A) and oral (B) exposure
conditions in mice (white), rats (diagonal hashing), and humans (horizontal
hashing). Bars and thin error bars represent the median estimate and 95% CI for
a random subject, and reflect combined uncertainty and variability. Circles and
thick error bars represent the median estimate and 95% CI for the population
mean, and reflect uncertainty only.
This document is a draft for review purposes only and does not constitute Agency policy.
160 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
Fraction Oxidized
B
Fraction Oxidized
~
Mouse
0
Rat
0
Human
~
Mouse
a
Rat
0
Human
Continuous inhalation (ppm )
Continuous oral ( mg/kg-d )
Figure 3-18. PBPK model predictions for the fraction of intake that is
metabolized by oxidation (in the liver and lung) under continuous inhalation
(A) and oral (B) exposure conditions in mice (white), rats (diagonal hashing),
and humans (horizontal hashing). Bars and thin error bars represent the median
estimate and 95% CI for a random subject, and reflect combined uncertainty and
variability. Circles and thick error bars represent the median estimate and 95% CI
for the population mean, and reflect uncertainty only.
This document is a draft for review purposes only and does not constitute Agency policy.
161 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
o -
o
Fraction Conjugated
Fraction Conjugated
;\
\
IVRH IVRH IVRH IVRH
IVRH
T
T
T
1
10 1 10 10 10
Continuous inhalation (ppm)
10 1 10 10 10
Continuous oral (mg/kg-d)
Figure 3-19. PBPK model predictions for the fraction of intake that is
metabolized by GSH conjugation (in the liver and kidney) under continuous
inhalation (A) and oral (B) exposure conditions in mice (dotted line), rats
(dashed line), and humans (solid line). X-values are slightly offset for clarity.
Open circles (connected by lines) and thin error bars represent the median
estimate and 95% CI for a random subject, and reflect combined uncertainty and
variability. Filled circles and thick error bars represent the median estimate and
95% CI for the population mean, and reflect uncertainty only.
This document is a draft for review purposes only and does not constitute Agency policy.
162 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
A Fraction bioactivated in kidney
it
RH
„-i
l
RH
I
RH
I
RH
~~r~
RH
I
10 1 10 10 10
Continuous inhalation (ppm)
3 Fraction bioactivated in kidney
o —
o -=
o —1
"S:—
RH RH RH RH RH
I 1 1 1 1
10~1 1 101 102 103
Continuous oral (mg/kg-d)
Figure 3-20. PBPK model predictions for the fraction of intake that is
bioactivated DCVC in the kidney under continuous inhalation (A) and oral
(B) exposure conditions in rats (dashed line) and humans (solid line).
X-values are slightly offset for clarity. Open circles (connected by lines) and thin
error bars represent the median estimate and 95% CI for a random subject, and
reflect combined uncertainty and variability. Filled circles and thick error bars
represent the median estimate and 95% CI for the population mean, and reflect
uncertainty only.
This document is a draft for review purposes only and does not constitute Agency policy.
163 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
7
o
"?
o -d
Fraction lung oxidation
•-n ~
IVRH IVRH IVRH IVRH IVRH
T
T
T
1
Fraction lung oxidation
IVRH
r~
IVRH IVRH IVRH MRH
T
T
T
1
10"' 1 10' 1(T 10J
Continuous inhalation (ppm)
10"' 1 10' 1(T 1(T
Continuous oral (mg/kg-d)
Figure 3-21. PBPK model predictions for fraction of intake that is oxidized
in the respiratory tract under continuous inhalation (A) and oral (B)
exposure conditions in mice (dotted line), rats (dashed line), and humans
(solid line). X-values are slightly offset for clarity. Open circles (connected by
lines) and thin error bars represent the median estimate and 95% CI for a random
subject, and reflect combined uncertainty and variability. Filled circles and thick
error bars represent the median estimate and 95% CI for the population mean, and
reflect uncertainty only.
This document is a draft for review purposes only and does not constitute Agency policy.
164 DRAFT—DO NOT CITE OR QUOTE
-------
A Fraction 'other1 liver oxidation
o -
o -
o -
\
T
T
T
IVRH IVRH IVRH IVRH IVRH
i
io_l 1 io' icr -icr
Continuous inhalation (ppm)
3 Fraction 'other' liver oxidation
o —
IVRH
I—
IVRH IVRH IVRH MRH
T
T
T
1
10"' 1 10' 1(T 10J
Continuous oral (mg/kg-d)
Figure 3-22. PBPK model predictions for the fraction of intake that is
"untracked" oxidation of TCE in the liver under continuous inhalation (A)
and oral (B) exposure conditions in mice (dotted line), rats (dashed line), and
humans (solid line) X-values are slightly offset for clarity. Open circles
(connected by lines) and thin error bars represent the median estimate and 95% CI
for a random subject, and reflect combined uncertainty and variability. Filled
circles and thick error bars represent the median estimate and 95% CI for the
population mean, and reflect uncertainty only.
This document is a draft for review purposes only and does not constitute Agency policy.
165 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
AUCTCE in blood
per ppm
IVRH IVRH IVRH IVRH IVRH
I 1 1 1 1
1(T1 1 101 102 103
Continuous inhalation (ppm)
o —
I5
E
a3
Q.
O)
E
o —
o —1
„-1
AUCTCE in blood
per mg/kg-d
10 1 10 10 10
Continuous oral (mg/kg-d)
Figure 3-23. PBPK model predictions for the weekly AUC of TCE in venous
blood (mg-hour/L-week) per unit exposure (ppm or mg/kg-day) under
continuous inhalation (A) and oral (B) exposure conditions in mice (dotted
line), rats (dashed line), and humans (solid line). X-values are slightly offset
for clarity. Open circles (connected by lines) and thin error bars represent the
median estimate and 95% CI for a random subject, and reflect combined
uncertainty and variability. Filled circles and thick error bars represent the
median estimate and 95% CI for the population mean, and reflect uncertainty
only.
This document is a draft for review purposes only and does not constitute Agency policy.
166 DRAFT—DO NOT CITE OR QUOTE
-------
AUCTCOH in blood
per ppm
IVRH IVRH IVRH IVRH IVRH
I 1 1 1 1
1(T1 1 101 102 103
Continuous inhalation (ppm)
O -a
J?
O)
E
a>
Q_
D)
E
o -
o -
o —1
AUCTCOH in blood
per mg/kg-d
1
H Si »¦
IVRH
IVRH
—I—
IVRH
—I—
IVRH
—I—
MRH
—I
,-1
10 1 10 10 10
Continuous oral (mg/kg-d)
Figure 3-24. PBPK model predictions for the weekly AUC of TCOH in blood
(mg-hour/L-week) per unit exposure (ppm or mg/kg-day) under continuous
inhalation (A) and oral (B) exposure conditions in mice (dotted line), rats
(dashed line), and humans (solid line). X-values are slightly offset for clarity.
Open circles (connected by lines) and thin error bars represent the median
estimate and 95% CI for a random subject, and reflect combined uncertainty and
variability. Filled circles and thick error bars represent the median estimate and
95% CI for the population mean, and reflect uncertainty only.
This document is a draft for review purposes only and does not constitute Agency policy.
167 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
AUC TCA in liver
per ppm
IVRH IVRH IVRH IVRH IVRH
I 1 1 1 1
1(T1 1 101 102 103
Continuous inhalation (ppm)
O -3
I5
E
a3
Q.
O)
O)
E
o -
o —1
AUC TCA in liver
per mg/kg-d
IVRH
IVRH
—I—
IVRH
—I—
IVRH
—I—
MRH
—I
,-1
10 1 10 10 10
Continuous oral (mg/kg-d)
Figure 3-25. PBPK model predictions for the weekly AUC of TCA in the
liver (mg-hour/L-week) per unit exposure (ppm or mg/kg-day) under
continuous inhalation (A) and oral (B) exposure conditions in mice (dotted
line), rats (dashed line), and humans (solid line). X-values are slightly offset
for clarity. Open circles (connected by lines) and thin error bars represent the
median estimate and 95% CI for a random subject, and reflect combined
uncertainty and variability. Filled circles and thick error bars represent the
median estimate and 95% CI for the population mean, and reflect uncertainty
only.
This document is a draft for review purposes only and does not constitute Agency policy.
168 DRAFT—DO NOT CITE OR QUOTE
-------
Table 3-48. Posterior predictions for representative internal doses: mouse
Dose-metric
Posterior predictions for mouse dose-metrics: median (2.5%, 97.5%)
Units
100 ppm, 7 hour/day,
5 day/wk
600 ppm, 7 hour/day,
5 day/wk
300 mg/kg-day,
5 day/wk
1,000 mg/kg-day,
5 day/wk
ABioactDCVCBW34
0.304 (0.000534, 12.4)
2.35 (0.00603, 37)
0.676 (0.00193, 18.4)
2.81 (0.0086, 42.4)
mg/wk-kg3 4
ABioactDCVCKid
43.7 (0.0774, 1780)
336 (0.801, 5,240)
96.8 (0.281, 2,550)
393 (1.23,6,170)
mg/wk-kg tissue
AMetGSHBW34
0.684 (0.0307, 17.6)
5.15 (0.285,44.9)
1.66 (0.0718, 24.5)
6.37 (0.567, 49.4)
mg/wk-kg3'4
AMetLivlBW34
170 (61.2, 403)
878 (342, 2,030)
400 (125, 610)
874 (233, 1,960)
mg/wk-kg3'4
AMetLivOtherB W3 4
3.81 (0.372, 38.4)
20 (1.86, 192)
8.38 (0.773,80.1)
20 (1.55, 202)
mg/wk-kg3'4
AMetLivOtherLiv
196 (19, 2,070)
1,030 (96.5, 10,100)
437 (39.5, 4,180)
1,020 (82.1, 10,400)
mg/wk-kg tissue
AMetLngBW34
187 (7.75, 692)
263 (10.9, 2,240)
38.5 (3.49, 147)
127 (8.59, 484)
mg/wk-kg3'4
AMetLngResp
638,000
(26,500, 2,510,000)
918,000
(36,800, 7,980,000)
134,000
(12,500,514,000)
433,000
(30,200, 1,690,000)
mg/wk-kg tissue
AUCCBld
96.9 (45,211)
822 (356, 2,040)
110(6.95,411)
592 (56, 1,910)
mg-hour/L-wk
AUCCTCOH
87.9 (9.9, 590)
480 (42.1,4,140)
132 (14.4, 670)
389 (34, 2,600)
mg-hour/L-wk
AUCLivTCA
1,880 (444, 7,190)
5,070(1,310, 18,600)
2,260 (520, 8,750)
4,660 (939, 18,900)
mg-hour/L-wk
TotMetabBW34
377 (140, 917)
1,260 (475, 3,480)
472 (165, 617)
1,110 (303,2,010)
mg/wk-kg3'4
T otOxMetabB W3 4
375 (139, 916)
1,250 (451,3,450)
465 (161,616)
1,100 (294, 2,010)
mg/wk-kg3'4
TotTCAInBW
272 (88.9, 734)
729 (267, 1,950)
334 (106, 875)
694 (185, 1,910)
mg/wk-kg
Note: Mouse body weight is assumed to be 0.03 kg. Predictions are weekly averages over 10 weeks of the specified exposure protocol. CI reflects both
uncertainties in population parameters (mean, variance) as well as population variability.
-------
Table 3-49. Posterior predictions for representative internal doses: rat
Dose-metric
Posterior predictions for rat dose-metrics: median (2.5%,97.5%)
Units
100 ppm, 7 hour/day,
5 day/wk
600 ppm, 7 hour/day,
5 day/wk
300 mg/kg-day, 5 day/wk
1,000 mg/kg-day,
5 day/wk
ABioactDCVCBW34
0.341 (0.0306, 2.71)
2.3 (0.175,22.6)
2.15(0.17, 20.2)
8.89 (0.711,84.1)
mg/wk-kg3 1
ABioactDCVCKid
67.8 (6.03,513)
450 (35.4, 4,350)
420 (31.6,3,890)
1,720 (134, 15,800)
mg/wk-kg tissue
AMetGSHBW34
0.331 (0.0626,2.16)
2.27(0.315, 19.3)
2.13 (0.293, 16)
8.84 (1.35,69.3)
mg/wk-kg3 4
AMetLivlBW34
176 (81.1, 344)
623 (271, 1,270)
539 (176, 1,060)
951 (273,2,780)
mg/wk-kg3'4
AMetLivOtherB W3 4
45.5 (2.52, 203)
160 (7.84, 749)
134 (6.83, 659)
238 (11.3, 1390)
mg/wk-kg3'4
AMetLivOtherLiv
1,870 (92.1,8,670)
6,660 (313,31,200)
5,490 (280, 27,400)
9,900 (492, 59,600)
mg/wk-kg tissue
AMetLngBW34
15 (0.529, 173)
24.5 (0.819, 227)
15.1 (0.527, 115)
32.1 (1.01, 311)
mg/wk-kg3'4
AMetLngResp
41,900 (1,460, 496,000)
67,900 (2,350, 677,000)
40,800 (1,500, 325,000)
85,700 (2,660, 877,000)
mg/wk-kg tissue
AUCCBld
86.7 (39.2, 242)
1,160 (349, 2,450)
670 (47.8, 1,850)
3,340 (828, 8,430)
mg-hour/L-wk
AUCCTCOH
83.6 (1.94, 1,560)
446 (6, 10,900)
304 (4.71, 7,590)
685 (8.14, 32,500)
mg-hour/L-wk
AUCLivTCA
587 (53.7, 4,740)
2,030 (186, 13,400)
1,730(124, 11,800)
3,130 (200,21,000)
mg-hour/L-wk
TotMetabBW34
206 (103, 414)
682 (288, 1,430)
572 (199, 1,080)
1,030 (302, 2,920)
mg/wk-kg3'4
T otOxMetabB W3 4
206 (103, 414)
677 (285, 1,430)
568 (191, 1,080)
1,010 (286, 2,910)
mg/wk-kg3'4
TotTCAInBW
31.7 (3.92, 174)
110 (13.8, 490)
90.1 (10.4,417)
164 (17.3, 800)
mg/wk-kg
Note: Rat body weight is assumed to be 0.3 kg. Predictions are weekly averages over 10 weeks of the specified exposure protocol. CI reflects both
uncertainties in population parameters (mean, variance) as well as population variability.
-------
Table 3-50. Posterior predictions for representative internal doses: human
Posterior predictions for human dose-metrics:
2.5% population: median (2.5%, 97.5%)
50% population: median (2.5%, 97.5%)
97.5% population: median (2.5%, 97.5%)
Dose-metric
Female
0.001 ppm continuous
Male
0.001 ppm continuous
Female
0.001 mg/kg-day continuous
Male
0.001 mg/kg-day continuous
ABioactDCVCBW34
0.000256 (6.97e-5, 0.000872)
0.000254 (6.94e-5, 0.000879)
0.000197 (6.13e-5, 0.000502)
0.0002 (6.24e-5, 0.000505)
0.00203 (0.00087, 0.00408)
0.00202 (0.000859, 0.00413)
0.00262 (0.0012, 0.00539)
0.00271 (0.00125, 0.00559)
0.0119 (0.00713,0.0177)
0.012(0.00699,0.0182)
0.021 (0.0118,0.0266)
0.022 (0.0124, 0.0277)
ABioactDCVCKid
0.02 (0.00549, 0.0709)
0.0207 (0.00558, 0.0743)
0.0152 (0.0048,0.0384)
0.016(0.00493,0.0407)
0.16(0.0671,0.324)
0.163 (0.0679, 0.342)
0.207 (0.0957, 0.43)
0.22 (0.102,0.459)
0.95 (0.56, 1.45)
0.979 (0.563, 1.51)
1.68 (0.956, 2.26)
1.81 (1.03,2.43)
AMetGSHBW34
0.000159 (4.38e-05, 0.000539)
0.000157 (4.37e-05, 0.00054)
0.000121 (3.82e-05, 0.000316)
0.000123 (3.82e-05, 0.000323)
0.00126 (0.000536, 0.00253)
0.00125 (0.000528, 0.00254)
0.00161 (0.000748,0.00331)
0.00167 (0.000777, 0.00343)
0.00736(0.00442,0.011)
0.00736(0.00434,0.0112)
0.013 (0.00725,0.0164)
0.0136 (0.00759, 0.0171)
AMetLivlBW34
0.00161 (0.000619,0.00303)
0.00157 (0.000608, 0.00292)
0.00465 (0.00169, 0.0107)
0.00498 (0.00184, 0.0112)
0.00637(0.00501,0.00799)
0.00619 (0.00484, 0.00779)
0.0172 (0.0153,0.0183)
0.018(0.0161,0.0191)
0.0157 (0.0118,0.0206)
0.0152 (0.0115,0.02)
0.0192 (0.019, 0.0193)
0.02 (0.0198, 0.0201)
AMetLivOtherB W3 4
4.98e-5 (8.59e-6, 0.000222)
4.87e-5 (8.33e-6, 0.000214)
0.000143 (2.35e-5, 0.000681)
0.00015 (2.49e-5, 0.000713)
0.000671 (0.000134, 0.00159)
0.000652 (0.000129, 0.00153)
0.00166 (0.00035, 0.00365)
0.00173 (0.000365, 0.00382)
0.00507 (0.00055, 0.00905)
0.00491 (0.000531,0.00885)
0.00993 (0.00109, 0.0153)
0.0103 (0.00113,0.0159)
AMetLivOtherLiv
0.000748 (0.000138, 0.00335)
0.00065 (0.000119, 0.00288)
0.00214 (0.000354, 0.00979)
0.00197 (0.00033, 0.00907)
0.0104 (0.00225, 0.0237)
0.00898 (0.00193, 0.0203)
0.0253 (0.00564, 0.0543)
0.0234 (0.00526, 0.0503)
0.0805 (0.00871,0.147)
0.0691 (0.00751,0.127)
0.157 (0.0188, 0.251)
0.146 (0.0173,0.232)
AMetLngBW34
6.9e-6(6.13e-7,7.99e-5)
7.25e-6 (6.44e-7, 8.39e-5)
7.54e-8 (6.59e-9, 7.85e-7)
7.05e-8 (6.1e-9, 7.25e-7)
0.00122 (0.000309,0.0032)
0.00127 (0.000325, 0.00329)
1.51e-5 (3.44e-6, 4.6e-5)
1.39e-5 (3.21e-6, 4.24e-5)
0.0123 (0.00563, 0.0197)
0.0124 (0.00582, 0.0199)
0.000396 (0.000104, 0.00097)
0.000366 (9.54e-5, 0.000906)
-------
Table 3-50. Posterior predictions for representative internal doses: human (continued)
Posterior predictions for human dose-metrics:
2.5% population: median (2.5%, 97.5%)
50% population: median (2.5%, 97.5%)
97.5% population: median (2.5%, 97.5%)
Dose-metric
Female
0.001 ppm continuous
Male
0.001 ppm continuous
Female
0.001 mg/kg-day continuous
Male
0.001 mg/kg-day continuous
AMetLngResp
0.0144 (0.00116,0.155)
0.0146 (0.00118,0.157)
0.00015 (1.27e-05, 0.00153)
0.000134 (1.15e-05, 0.00137)
2.44 (0.613,6.71)
2.44 (0.621, 6.65)
0.0313 (0.00725,0.0963)
0.0279 (0.00644, 0.086)
25.8 (12.4, 42.3)
25.3 (12.2,41.2)
0.813 (0.216,2.13)
0.716(0.189, 1.9)
AUCCBld
0.00151 (0.00122, 0.00186)
0.00158 (0.00127, 0.00191)
4.33e-05 (3.3e-05, 6.23e-05)
3.84e-05 (2.89e-05, 5.61e-05)
0.00285 (0.00252, 0.00315)
0.00295 (0.00262, 0.00326)
0.000229 (0.000122, 0.000436)
0.000204 (0.000109, 0.000391)
0.00444 (0.00404, 0.00496)
0.00456 (0.00416, 0.00507)
0.00167 (0.000766, 0.00324)
0.00153 (0.000693, 0.00303)
AUCCTCOH
0.00313 (0.00135,0.00547)
0.00305 (0.00134, 0.00532)
0.00584 (0.00205, 0.0122)
0.00615 (0.00213, 0.0127)
0.0181 (0.0135,0.0241)
0.0179 (0.0133,0.0238)
0.0333 (0.025, 0.0423)
0.035 (0.0264, 0.0445)
0.082 (0.0586, 0.118)
0.0812 (0.0585,0.117)
0.115 (0.0872,0.163)
0.122 (0.0919, 0.172)
AUCLivTCA
0.0152 (0.00668, 0.0284)
0.0137 (0.00598, 0.0258)
0.029 (0.0116, 0.0524)
0.0279 (0.0114,0.0501)
0.126 (0.0784, 0.194)
0.114(0.0704, 0.177)
0.227 (0.138,0.343)
0.219 (0.133,0.33)
0.754 (0.441, 1.38)
0.699 (0.408, 1.3)
1.11 (0.661, 1.87)
1.09 (0.64, 1.88)
TotMetabBW34
0.0049 (0.00383, 0.00595)
0.00482 (0.0038, 0.00585)
0.0163 (0.0136,0.0181)
0.0173 (0.0147,0.019)
0.0107 (0.00893, 0.0129)
0.0105 (0.00877, 0.0127)
0.0191 (0.0188,0.0193)
0.0199 (0.0196, 0.0201)
0.0246 (0.0185,0.0326)
0.0244 (0.0183,0.0324)
0.0194 (0.0194,0.0194)
0.0202 (0.0202, 0.0202)
T otOxMetabB W3 4
0.00273 (0.00143, 0.00422)
0.00269 (0.00143, 0.00415)
0.0049 (0.00183, 0.0108)
0.00516(0.00194,0.0114)
0.00871 (0.0069,0.0111)
0.00857 (0.00675,0.011)
0.0173 (0.0154,0.0183)
0.018(0.0161,0.0191)
0.0224 (0.0158,0.0309)
0.0222 (0.0155,0.0308)
0.0192 (0.019, 0.0193)
0.02 (0.0198, 0.0201)
-------
Table 3-50. Posterior predictions for representative internal doses: human (continued)
Dose-metric
Posterior predictions for human dose-metrics:
2.5% population: median (2.5%, 97.5%)
50% population: median (2.5%, 97.5%)
97.5% population: median (2.5%, 97.5%)
Female
0.001 ppm continuous
Male
0.001 ppm continuous
Female
0.001 mg/kg-day continuous
Male
0.001 mg/kg-day continuous
TotTCAInBW
0.000259 (0.000121, 0.000422)
0.00154 (0.00114, 0.00202)
0.00525 (0.00399, 0.00745)
0.000246 (0.000114, 0.000397)
0.00146 (0.00109, 0.00193)
0.00499 (0.0038, 0.0071)
0.000501 (0.000189, 0.000882)
0.00286 (0.00222, 0.00357)
0.00659 (0.00579, 0.00724)
0.000506 (0.000192, 0.00089)
0.00289 (0.00222, 0.0036)
0.00662 (0.00581, 0.00726)
Note: Human body weight is assumed to be 70 kg for males, 60 kg for females. Predictions are weekly averages over 100 weeks of continuous exposure
(dose-metric units same as previous tables). Each row represents a different population percentile (2.5, 50, and 97.5%), and the CI in each entry reflects
uncertainty in population parameters (mean, variance).
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Here,X6) is one of dose-metric predictions, 0± is the MLE or baseline value of ± 5%. For
log-transformed parameters, 0.05 was added or subtracted from the baseline value, whereas for
untransformed parameters, the baseline value was multiplied by 1.05 or 0.95.
Note that local sensitivity analyses as typically performed in deterministic PBPK
modeling can only inform the "primary" effects of parameter uncertainties— i.e., the direct
change on the quantity of interest due to change in a parameter. They cannot address the
propagation of uncertainties through an analysis, such as those that can arise due to parameter
correlations in the parameter fitting process. Those can only be addressed in a global sensitivity
analysis, which is left for future research.
The results of local sensitivity analyses are shown in Figures 3-26-3-31. As expected,
each dose-metric is sensitive to a only a small fraction of the scaling parameters. Many of these
are well-specified a priori, either due to their being physiological parameters or partition
coefficients that can be measured in vitro. The remaining sensitive parameters are generally
related to metabolism or clearance.
3.5.7.1.3. Implications for the Population Pharmacokinetics of Trichloroethylene (TCE)
3.5.7.1.4. Results
The overall uncertainty and variability in key toxicokinetic predictions, as a function of
dose and species, is shown in Figures 3-17-3-25. As expected, TCE that is inhaled or ingested is
substantially metabolized in all species, predominantly by oxidation (see Figures 3-17-3-18). At
higher exposures, metabolism becomes saturated and the fraction metabolized declines. Mice on
average have a greater capacity to oxidize TCE than rats or humans, and this is reflected in the
predictions at the two highest levels for each route. The uncertainty in the predictions for the
population means for total and oxidative metabolism is relatively modest, therefore, the wide CI
for combined uncertainty and variability largely reflects intersubject variability. Of particular
note is the high variability in oxidative metabolism at low doses in humans, with the 95% CI
spanning from 0.1-0.7 for inhalation and 0.2-1.0 for ingestion.
Predictions of GSH conjugation and renal bioactivation of DCVC are highly uncertain in
rodents, spanning more than 1,000-fold in mice and 100-fold in rats (see Figures 3-19-3-20). In
both mice and rats, the uncertainty in the population mean virtually overlaps with the combined
uncertainty and variability. The uncertainty in mice reflects the lack of GSH-conjugate specific
data in that species, and is, therefore, based on overall mass balance only. The somewhat smaller
uncertainty in rats reflects the fact that, in addition to overall mass balance, urinary NAcDCVC
This document is a draft for review purposes only and does not constitute Agency policy.
174 DRAFT—DO NOT CITE OR QUOTE
-------
1 excretion data are available in that species. However, while the lower bound of GSH
2 conjugation is informed by NAcDCVC excretion data, the upper bound for GSH conjugation and
This document is a draft for review purposes only and does not constitute Agency policy.
175 DRAFT—DO NOT CITE OR QUOTE
-------
a 2
o
Co SS
s <=>
OQ O
^ a
o
t
s
§•
^n
3?
w
%¦
ci
si
>}
FracMetab
FracOxMetab FracMetGSH
^1
Q\
InQCC
InVPRC
QFatC
QGutC
QLivC
QSIwC
InDRespC
QKidC
FracPlasC
VFatC
VGutC
VLivC
VRapC
VRespLumC
VRespEff C
VKidC
VBIdC
InPBC
IriPFatC
InPGutC
InPLivC
InPRapC
InPRespC
InPKidC
InPSIwC
InPRBCPIasTCAC
InPBodTCAC
InPLivTCAC
InkDissocC
IriBMaxkDC
IriPBodTCOHC
InPLivTCOHC
IriPBodTCOGC
InPLivTCOGC
InkTSD
InkAS
InkAD
InkASTCA
InVMaxC
InKMC
InFracOtherC
IriFracTCAC
IriVMaxDCVGC
InCIDCVGC
InVMaxKidDCVGC
InCIKidDCVGC
InVMaxLungLivC
lnKI\fclara
InFracLungSysC
InVMaxTCOHC
InKMTCOH
InVMaxGlucC
InKMGluc
InkMetTCOHC
InkUrnTCAC
IrikMetTCAC
InkBileC
InkEHRC
IrikUrnTCOGC
FracMetLivl FracMetLivOther FracMetLng
ALICCBId
AUCCTCOH
ALICLivTCA
""I 1 1—
-0.5 0.5
SC
—I 1 1—
-0.5 0.5
SC
—i—i—r~
-0.5 0.5
SC
-~i—i—r~
-0.5 0.5
SC
—r~
-0.5
T~
SC
~r~
0.5
-| 1 1—
-0.5 0.5
SC
~i—i—r~
-0.5 0.5
SC
cS
«
—i—r
-0.5
SC
—r~
0.5
b
-~i—i—r~
-0.5 0.5
SC
Figure 3-26. Sensitivity analysis results: SC for mouse scaling parameters with respect to dose-metrics following
100 ppm (light bars) and 600 ppm (dark bars), 7 h/day, 5 day/wk inhalation exposures.
-------
a 2
o
Co SS
s <=>
OQ O
^ a
o
t
s
§•
^n
3?
w
%¦
ci
si
>}
^1
<1
InQCC
InVPRC
QFatC
QGutC
QLivC
QSIwC
InDRespC
QKidC
FracPlasC
VFatC
VGutC
VLivC
VRapC
VRespLumC
VRespEff C
VKidC
VBIdC
InPBC
IriPFatC
InPGutC
InPLivC
InPRapC
InPRespC
InPKidC
InPSIwC
InPRBCPIasTCAC
InPBodTCAC
InPLivTCAC
InkDissocC
InBMaxkDC
IriPBodTCOHC
InPLivTCOHC
IriPBodTCOGC
InPLivTCOGC
InkTSD
InkAS
IrikAD
InkASTCA
InVMaxC
InKMC
InFracOtherC
IriFracTCAC
IriVMaxDCVGC
InCIDCVGC
InVMaxKidDCVGC
InCIKidDCVGC
InVMaxLungLivC
lnKI\fclara
InFracLungSysC
InVMaxTCOHC
InKMTCOH
InVMaxGlucC
InKMGluc
InkMetTCOHC
InkUrnTCAC
IrikMetTCAC
InkBileC
InkEHRC
IrikUrnTCOGC
FracMetab
FracOxMetab
FracMetGSH
FracMetLivl FracMetLivOther FracMetLng
ALICCBId
AUCCTCOH
ALICLivTCA
~i—T
-0.5
SC
—I—
0.5
—i—r
-0.5
SC
—r~
0.5
—i—i—r~
-0.5 0.5
SC
-~i—i—r~
-0.5 0.5
SC
—I 1 1—
-0.5 0.5
SC
cP
9
P
L
I
k
i—r
-0.5
SC
—I—
0.5
—r~
-0.5
T~
SC
0.5
Ei
I.
—i—r
-0.5
SC
—r~
0.5
~i—r
-0.5
SC
—r~
0.5
Figure 3-27. Sensitivity analysis results: SC for mouse scaling parameters with respect to dose-metrics following
300 mg/kg-day (light bars) and 1,000 mg/kg-day (dark bars), 5 day/wk oral gavage exposures.
-------
a 2
o
Co SS
s <=>
OQ O
^ a
o
t
s
§•
^n
3?
w
%¦
ci
si
>}
FracMetab
^1
00
InQCC
InVPRC
QFatC
QGutC
QLivC
QSIwC
InDRespC
QKidC
FracPlasC
VFatC
VGutC
VLivC
VRapC
VRespLumC
VRespEff C
VKidC
VBIdC
InPBC
InPFatC
InPGutC
InPLivC
InPRapC
InPRespC
InPKidC
InPSIwC
InPRBCPIasTCAC
InPBodTCAC
InPLivTCAC
InkDissocC
InBMaxkDC
InPBodTCOHC
IriPLivTCOHC
InPBodTCOGC
InPLivTCOGC
InkTSD
InkAS
InkAD
InkASTCA
InVMaxC
InKMC
InFracOtherC
InFracTCAC
IriVMaxDCVGC
InCIDCVGC
InVMaxKidDCVGC
IriCIKidDCVGC
IriVMaxLungLivC
InKMC lara
InFracLungSysC
InVMaxTCOHC
InKMTCOH
InVMaxGlucC
InKMGluc
InkMetTCOHC
InkUrnTCAC
InkMetTCAC
InkBileC
InkEHRC
IrikUrnTCOGC
InkNATC
InkKidBioactC
FracOxMetab FracMetGSH FracBioactKid FracMetLivl FracMetLivOther FracMetLng
AUCCBId
ALICCTCOH
AUCLivTCA
F
r
F="
r
i
p
r
I
R
i—i—r~
-0.5 0.5
SC
-| 1 1—
-0.5 0.5
SC
-| 1 1—
-0.5 0.5
SC
~i—r
-0.5
SC
—r~
0.5
—i—i—r~
-0.5 0.5
SC
—i—i—r~
-0.5 0.5
SC
-| 1 1—
-0.5 0.5
SC
-| 1 1—
-0.5 0.5
SC
~i—r
-0.5
SC
~l—
0.5
i
eJ
I
I
q
—I 1 1—
-0.5 0.5
SC
Figure 3-28. Sensitivity analysis results: SC for rat scaling parameters with respect to dose-metrics following
100 ppm (light bars) and 600 ppm (dark bars), 7 h/day, 5 day/wk inhalation exposures.
-------
a 2
o
Co SS
s <=>
OQ O
^ a
o
t
s
§•
^n
3?
w
%¦
ci
si
>}
FracMetab
^1
VO
InQCC
InVPRC
QFatC
QGutC
QLivC
QSIwC
InDRespC
QKidC
FracPlasC
VFatC
VGutC
VLivC
VRapC
VRespLumC
VRespEff C
VKidC
VBIdC
InPBC
InPFatC
InPGutC
InPLivC
InPRapC
InPRespC
InPKidC
InPSIwC
InPRBCPIasTCAC
InPBodTCAC
InPLivTCAC
InkDissocC
InBMaxkDC
InPBodTCOHC
IriPLivTCOHC
InPBodTCOGC
InPLivTCOGC
InkTSD
InkAS
InkAD
InkASTCA
InVMaxC
InKMC
InFracOtherC
InFracTCAC
IriVMaxDCVGC
InCIDCVGC
InVMaxKidDCVGC
IriCIKidDCVGC
IriVMaxLungLivC
InKMC lara
InFracLungSysC
InVMaxTCOHC
InKMTCOH
InVMaxGlucC
InKMGluc
InkMetTCOHC
InkUrnTCAC
InkMetTCAC
InkBileC
InkEHRC
IrikUrnTCOGC
InkNATC
InkKidBioactC
FracOxMetab FracMetGSH FracBioactKid FracMetLivl FracMetLivOther FracMetLng
AUCCBId
ALICCTCOH
AUCLivTCA
i—T
-0.5
SC
—r~
0.5
-| 1 1—
-0.5 0.5
SC
-i—r
-0.5
SC
—T"
0.5
~i—r
-0.5
SC
—r~
0.5
~i—r
-0.5
SC
—r~
0.5
—i—i—r~
-0.5 0.5
SC
-i—r
-0.5
SC
T"
0.5
-| 1 1—
-0.5 0.5
SC
~i—r
-0.5
SC
~l—
0.5
—i—i—r~
-0.5 0.5
SC
Figure 3-29. Sensitivity analysis results: SC for rat scaling parameters with respect to dose-metrics following
300 mg/kg-day (light bars) and 1,000 mg/kg-day (dark bars), 5 day/wk oral gavage exposures.
-------
a 2
o
Co SS
s <=>
OQ O
^ a
o
t
s
§•
^n
3?
w
%¦
ci
si
>}
FracMetab
FracOxMetab FracMetGSH FracBioactKid FracMetLivl FracMetLivOther FracMetLng
ALICCBId
ALICCTCOH
AUCLivTCA
00
o
InQCC
InVPRC
QFatC
QGutC
QLivC
QSIwC
InDRespC
QKidC
FracPlasC
VFatC
VGutC
VLivC
VRapC
VRespLumC
VRespEff C
VKidC
VBIdC
InPBC
InPFatC
InPGutC
InPLivC
InPRapC
InPRespC
InPKidC
InPSIwC
InPRBCPIasTCAC
InPBodTCAC
InPLivTCAC
IrikDissocC
InBMaxkDC
InPBodTCOHC
IriPLivTCOHC
InPBodTCOGC
InPLivTCOGC
InPeff DCVG
InkASTCA
inkASTCOH
InVMaxC
InCIC
InFracOtherC
InFracTCAC
InCIDCVGC
InKMDCVGC
IriCIKidDCVGC
InKMKidDCVGC
InVMaxLungLivC
lnKI\fclara
InFracLungSysC
InCITCOHC
InKMTCOH
InCIGIucC
InKMGluc
InkMetTCOHC
InkUrnTCAC
InkMetTCAC
InkBileC
InkEHRC
IrikUrnTCOGC
InkDCVGC
InkNATC
InkKidBioactC
—I 1 1—
-0.5 0.5
SC
-| 1 1—
-0.5 0.5
SC
-i—r
-0.5
SC
—T"
0.5
1—i—r
t—i—r
-|—i—r
-0.5 0.5 -0.5 0.5 -0.5 0.5
SC SC SC
-| 1 1—
-0.5 0.5
SC
-| 1 1—
-0.5 0.5
SC
~i—r
-0.5
SC
~l—
0.5
B
B
I
B
Q
i—i—r~
-0.5 0.5
SC
Figure 3-30. Sensitivity analysis results: SC for female (light bars) and male (dark bars) human scaling
parameters with respect to dose-metrics following 0.001 ppm continuous inhalation exposures.
-------
a 2
o
Co SS
s <=>
OQ O
^ a
o
t
s
§•
^n
3?
w
%¦
ci
si
>}
InQCC
InVPRC
QFatC
QGutC
QLivC
QSIwC
InDRespC
QKidC
FracPlasC
VFatC
VGutC
VLivC
VRapC
VRespLumC
VRespEff C
VKidC
VBIdC
InPBC
InPFatC
InPGutC
InPLivC
InPRapC
InPRespC
InPKidC
InPSIwC
InPRBCPIasTCAC
InPBodTCAC
InPLivTCAC
IrikDissocC
InBMaxkDC
InPBodTCOHC
IriPLivTCOHC
InPBodTCOGC
InPLivTCOGC
InPeff DCVG
InkASTCA
inkASTCOH
InVMaxC
InCIC
InFracOtherC
InFracTCAC
InCIDCVGC
InKMDCVGC
IriCIKidDCVGC
InKMKidDCVGC
InVMaxLungLivC
lnKI\fclara
InFracLungSysC
InCITCOHC
InKMTCOH
InCIGIucC
InKMGluc
InkMetTCOHC
InkUrnTCAC
InkMetTCAC
InkBileC
InkEHRC
IrikUrnTCOGC
InkDCVGC
InkNATC
InkKidBioactC
FracMetab
FracOxMetab FracMetGSH FracBioactKid FracMetLivl FracMetLivOther FracMetLng
ALICCBId
ALICCTCOH
AUCLivTCA
i—i—r~
-0.5 0.5
SC
1—i—r
t—i—r
1—i—r
t—i—r
-|—i—r
-0.5 0.5 -0.5 0.5
SC SC
-0.5 0.5 -0.5 0.5 -0.5 0.5
SC SC SC
-| 1 1—
-0.5 0.5
SC
1—i—r
1—i—r
0
B
1
0
1
I
1—i—r
-0.5 0.5 -0.5 0.5 -0.5 0.5
SC
SC
SC
Figure 3-31. Sensitivity analysis results: SC for female (light bars) and male (dark bars) human scaling
parameters with respect to dose-metrics following 0.001 mg/kg-day continuous oral exposures.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
the amount of DCVC bioactivation are still indirectly estimated from data on other clearance
pathways. In humans, however, overall GSH conjugation is strongly constrained by the
bloodconcentrations of DCVG from Lash et al. (1999a), with 95% CIs on the population mean
spanning only threefold or so. DCVC bioactivation is still indirectly estimated, derived from the
difference between overall GSH conjugation flux and NAcDCVC excretion data from Bernauer
et al. (1996). However, substantial variability is predicted (reflecting variability in the
measurements of Last et al. (1999a), for the error bars for the population mean are substantially
smaller than that for overall uncertainty and variability. Of particular note is the prediction of
one or two orders of magnitude more GSH conjugation and DCVC bioactivation, on average, in
humans than in rats, although importantly, the 95% CIs for the predicted population means do
overlap. However, as discussed above in Section 3.3.3.2.1, there are uncertainties as to the
accuracy of analytical method used by Lash et al. (1999a) in the measurement of DCVG in
blood. Because these data are so influential, the analytical uncertainties contribute substantially
to the overall uncertainty in the estimates of the overall GSH conjugation flux, and may be
greater than the statistical uncertainties calculated using the model.
Predictions for respiratory tract oxidative metabolism were, as expected, greatest in mice,
followed by rats and then humans (see Figure 3-21). In addition, due to the "presystemic" nature
of the respiratory tract metabolism model as well as the hepatic first-pass effect, substantially
more metabolism was predicted from inhalation exposures as compared to oral exposures.
Interestingly, the population means appeared to be fairly well constrained despite the lack of
direct data, suggesting that overall mass balance is an important constraint for the presystemic
respiratory tract metabolism modeled here.
Some constraints were also placed on "other" hepatic oxidation—i.e., through a pathway
that does not result in chloral formation and subsequent formation of TCA and TCOH (see
Figure 3-22). The 95% CI for overall uncertainty and variability spanned about 100-fold, a large
fraction of that due to uncertainty in the population mean. Interestingly, a higher rate per kg
tissue was predicted for rats than for mice or humans, although importantly, the 95% CIs for the
population means overlap among all three species.
The AUC of TCE in blood (see Figure 3-23) showed the expected nonlinear behavior
with increasing dose, with the nonlinearity more pronounced with oral exposure, as would be
expected by hepatic first-pass. Notably, the predicted AUC of TCE in blood from inhalation
exposures corresponds closely with cross-species ppm-equivalence, as is often assumed. For low
oral exposures (<1 mg/kg-day), cross-species mg/kg-day equivalence appears to be fairly
accurate (within twofold), implying the usual assumption of mg/kgy4-day equivalence would be
somewhat less accurate, at least for humans. Interestingly, the AUC of TCOH in blood (see
This document is a draft for review purposes only and does not constitute Agency policy.
182 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Figure 3-24) was relatively constant with dose, reflecting the parallel saturation of both TCE
oxidation and TCOH glucuronidation. In fact, in humans, the mean AUC for TCOH in blood
increases up to 100 ppm or 100 mg/kg-day, due to saturation of TCOH glucuronidation, before
decreasing at 1,000 ppm or 1,000 mg/kg-day, due to saturation of TCE oxidation.
The predictions for the AUC for TCA in the liver showed some interesting features (see
Figure 3-25). The predictions for all three species with within an order of magnitude of each
other, with a relatively modest uncertainty in the population mean (reflecting the substantial
amount of data on TCA). The shape of the curves, however, differs substantially, with humans
showing saturation at much lower doses than rodents, especially for oral exposures. In fact, the
ratio between the liver TCA AUC and the rate of TCA production, although differing between
species, is relatively constant as a function of dose within species (not shown). Therefore, the
shape of the curves largely reflect saturation in the production of TCA from TCOH, not in the
oxidation of TCE itself, for which saturation is predicted at higher doses, particularly via the oral
route (see Figure 3-18). In addition, while for the same exposure (ppm or mg/kg-day TCE) more
TCA (on a mg/kg-day basis) is produced in mice relative to rats and humans, humans and rats
have longer TCA half-lives even though plasma protein binding of TCA is on average greater.
3.5.7.1.5. Discussion
This analysis substantially informs four of the major areas of pharmacokinetic
uncertainty previously identified in numerous reports (reviewed in Chiu et al., 2006b): GSH
conjugation pathway, respiratory tract metabolism, alternative pathways of TCE oxidation
including DCA formation, and the impact of plasma binding on TCA kinetics particularly in the
liver. In addition, the analysis helps identify data that have the potential to further reduce the
uncertainties in TCE toxicokinetics and risk assessment.
With respect to the first, previous estimates of the degree of TCE GSH conjugation and
subsequent bioactivation of DCVC in humans were based on urinary excretion data alone
(Bernauer et al., 1996; Birner et al., 1993). For instance, Bloemen et al. (2001) concluded that
due to the low yield of identified urinary metabolites through this pathway (<0.05% as compared
to 20-30% in urinary metabolites of TCE oxidation), GSH conjugation of TCE is likely of minor
importance. However, as noted by Lash et al. (2000a; 2000b), urinary excretion is a poor
quantitative marker of flux through the GSH pathway because it only accounts for the portion
detoxified, and not the portion bioactivated (a limitation acknowledged by Bloemen et al., 2001).
A reexamination of the available in vitro data on GSH conjugation by Chiu et al.(2006b)
suggested that the difference in flux between TCE oxidation and GSH conjugation may not be as
This document is a draft for review purposes only and does not constitute Agency policy.
183 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
large as suggested by urinary excretion data. For example, the formation rate of DCVG from
TCE in freshly isolated hepatocytes was similar in order of magnitude to the rate measured for
oxidative metabolites (Lash et al., 1999b; Lipscomb et al., 1998a). A closer examination of the
only other available human in vivo data on GSH conjugation, the DCVG blood levels reported in
Lash et al. (Lash et al., 1998a) also suggests a substantially greater flux through this pathway
than inferred from urinary data. In particular, the peak DCVG blood levels reported in this study
were comparable on a molar basis to peak blood levels of TCOH, the major oxidative metabolite,
in the same subjects, as previously reported by Fisher et al. (1998). A lower bound estimate of
the GSH conjugation flux can be derived as follows. The reported mean peak blood DCVG
concentrations of 46 [xM in males exposed to 100 ppm TCE for 4 hours (Lash et al., 1999a),
multiplied by a typical blood volume of 5 L (ICRP, 2003), yields a peak amount of DCVG in
blood of 0.23 mmoles. In comparison, the retained dose from 100 ppm exposure for 4 hours is
4.4 mmol, assuming retention of about 50% (Monster et al., 1976) and minute-volume of
9 L/minute (ICRP, 2003). Thus, in these subjects, about 5% of the retained dose is present in
blood as DCVG at the time of peak blood concentration. This is a strong lower bound on the
total fraction of retained TCE undergoing GSH conjugation because DCVG clearance is ongoing
at the time of peak concentration, and DCVG may be distributed to tissues other than blood. It
should be reiterated that only grouped DCVG blood data were available for PBPK model-based
analysis; however, this should only result in an underestimation of the degree of variation in
GSH conjugation. Finally, this hypothesis of a significant flux through the human GSH
conjugation pathway is consistent with the limited available total recovery data in humans in
which only 60-70% of the TCE dose is recovered as TCE in breath and excreted urinary
metabolites (reviewed in Chiu et al., 2007).
Thus, there is already substantial qualitative and semi-quantitative evidence to suggest a
substantially greater flux through the GSH conjugation pathway than previously estimated based
on urinary excretion data alone. The scientific utility of applying a combination of PBPK
modeling and Bayesian statistical methods to this question comes from being able to
systematically integrate these different types of data—in vitro and in vivo, direct (blood DCVG)
and indirect (total recovery, urinary excretion)—and quantitatively assess their consistency and
implications. For example, the in vitro data discussed above on GSH conjugation were used for
developing prior distributions for GSH conjugation rates, and were not used in previous PBPK
models for TCE. Then, both the direct and indirect in vivo data were used to the extent possible
either in the Bayesian calibration or model evaluation steps.
However, this evidence—both qualitative and quantitative—is highly dependent on the
reliability of the human DCVG measurements, both in vitro and in vivo, from Lash et al. (Lash et
This document is a draft for review purposes only and does not constitute Agency policy.
184 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
al., 1999a; Lash et al., 1999b). In vitro, Green et al. (1997a) reported much lower rates of
DCVG formation in humans using a different analytical method. Similarly, the rates of in vitro
DCVG formation in rats have uneven consistency among studies. In male rat liver cytosol,
Green et al. (1997a) reported a rate of 0.54 pmol/min-mg, consistent with the <2 pmol/min-mg
reported by Dekant et al. (1990), but much less than the 121 pmol/min-mg reported by Lash et al.
(1999b). However, in microsomes, Green et al. (1997a) reported no enzymatic formation,
whereas Dekant et al. (1990) reported a higher rate (i.e., 2 pmol/min-mg) and Lash et al. (1999b)
reported a much higher rate (i.e., 171 pmol/min-mg). Differing results in humans may be
attributable to true interindividual variation (especially since GSTs are known to be
polymorphic). However, this may be less plausible for rats, suggesting that significant
uncertainties remain in the quantitative estimation of GSH conjugation flux.
Several other aspects of the predictions related to GSH conjugation of TCE are worthy of
note. Predictions for rats and mice remain more uncertain due to their having less direct
toxicokinetic data, but are better constrained by total recovery studies. For instance, the total
recovery of 60-70% of dose in exhaled breath and oxidative metabolites in human studies is
substantially less than the >90% reported in rodent studies (also noted by Goeptar et al., 1995).
In addition, it has been suggested that "saturation" of the oxidative pathway for volatiles in
general, and TCE in particular, may lead to marked increases in flux through the GSH
conjugation pathway (Goeptar et al., 1995; 2004a; Slikker et al., 2004b), but the PBPK model
predicts only a modest, at most -twofold, change in flux. This is because there is evidence that
both pathways are saturable in the liver for this substrate at similar exposures and because GSH
conjugation also occurs in the kidney. Therefore, the available data are not consistent with
toxicokinetics alone causing substantially nonlinearites in TCE kidney toxicity or cancer, or in
any other effects associated with GSH conjugation of TCE.
Finally, the present analysis suggests a number of areas where additional data can further
reduce uncertainty in and better characterize the TCE GSH conjugation pathway. The Bayesian
analysis predicts a relatively low distribution volume for DCVG in humans, a hypothesis that
could be tested experimentally. In addition, in vivo measurements of DCVG in blood via a
different, validated analytical method, in humans with known exposures to TCE, would be
highly influential in either corroborating the DCVG blood levels reported in Lash et al. (1999a)
or providing evidence that those reported DCVG blood levels are too high due to analytical
issues. Moreover, it would be useful in such studies to be able to match individuals with respect
to toxicokinetic data on oxidative and GSH conjugation metabolites so as to better characterize
variability. A consistent picture as to which GST isozymes are involved in TCE GSH
conjugation, along with data on variability in isozyme polymorphisms and activity levels, can
This document is a draft for review purposes only and does not constitute Agency policy.
185 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
further inform the extent of human variability. In rodents, more direct data on GSH metabolites,
such as reliably-determined DCVG blood concentrations, preferably coupled with simultaneous
data on oxidative metabolites, would greatly enhance the assessment of GSH conjugation flux in
laboratory animals. Given the large apparent variability in humans, data on interstrain variability
in rodents may also be useful.
With respect to oxidative metabolism, as expected, the liver is the major site of oxidative
metabolism in all three species, especially after oral exposure, where >85% of total metabolism
is oxidation in the liver in all three species. However, after inhalation exposure, the model
predicts a greater proportion of metabolism via the respiratory tract than previous models for
TCE. This is primarily because previous models for TCE respiratory tract metabolism (Clewell
et al., 2000; Hack et al., 2006) were essentially flow-limited—i.e., the amount of respiratory tract
metabolism (particularly in mice) was determined primarily by the (relatively small) blood flow
to the tracheobronchial region. However, the respiratory tract structure used in the present model
is more biologically plausible, is more consistent with that of other volatile organics metabolized
in the respiratory tract (e.g., styrene), and leads to a substantially better fit to closed-chamber
data in mice.
Consistent with the qualitative suggestions from in vitro data, the analysis here predicts
that mice have a greater rate of respiratory tract oxidative metabolism as compared to rats and
humans. However, the predicted difference of 50-fold or so on average between mice and
humans is not as great as the 600-fold suggested by previous reports (Green, 2000; Green et al.,
1997b; NRC, 2006). The suggested factor of 600-fold was based on multiplying the Green et al.
(1997b) data on TCE oxidation in lung microsomes from rats versus mice (23-fold lower) by a
factor for the total CYP content of human lung compared to rat lung (27-fold lower) (incorrectly
cited as being from Raunio et al., 1998; Wheeler and Guenthner, 1990). However, because of
the isozyme-specificity of TCE oxidation, and the differing proportions of different isozymes
across species, total CYP content may not be the best measure of interspecies differences in TCE
respiratory tract oxidative metabolism. Wheeler et al. (1992) reported that CYP2E1 content of
human lung microsomes is about 10-fold lower than that of human liver microsomes. Given that
Green et al. (1997b) report that TCE oxidation by human liver microsomes is about threefold
lower than that in mouse lung microsomes, this suggests that the mouse-to-human comparison
TCE oxidation in lung microsomes would be about 30-fold. Moreover, the predicted amount of
metabolism corresponds to about the detection limit reported by Green et al. (1997b) in their
experiments with human lung microsomes, suggesting overall consistency in the various results.
Therefore, the 50-fold factor predicted by our analysis is biologically plausible given the
This document is a draft for review purposes only and does not constitute Agency policy.
186 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
available in vitro data. More direct in vivo measures of respiratory tract metabolism would be
especially beneficial to reduce its uncertainty as well as better characterize its human variability.
TCA dosimetry is another uncertainty that was addressed in this analysis. In particular,
the predicted interspecies differences in liver TCA AUC are modest, with a range of 10-fold or
so across species, due to the combined effects of interspecies differences in the yield of TCA
from TCE, plasma protein binding, and elimination half-life. This result is in contrast to
previous analyses which did not include TCA protein binding (Clewell et al., 2000; Fisher,
2000), which predicted significantly more than an order of magnitude difference in TCA AUC
across species. In addition, in order to be consistent with available data, the model requires some
metabolism or other clearance of TCA in addition to urinary excretion. That urinary excretion
does not represent 100% of TCA clearance is evident empirically, as urinary recovery after TCA
dosing is not complete even in rodents (Abbas et al., 1997; Yu et al., 2000). Additional
investigation into possible mechanisms, including metabolism to DCA or enterohepatic
recirculation with fecal excretion, would be beneficial to provide a stronger biological basis for
this empirical finding.
With respect to "untracked" oxidative metabolism, this pathway appears to be a relatively
small contribution to total oxidative metabolism. While it is tempting to use this pathway as a
surrogate for DCA production through from the TCE epoxide (Cai and Guengerich, 1999), one
should be reminded that DCA may be formed through multiple pathways (see Section 3.3).
Therefore, this pathway at best represents a lower bound on DCA production. In addition, better
quantitative markers of oxidative metabolism through the TCE epoxide pathway (e.g.,
dichloroacetyl lysine protein adducts, as reported in (e.g., dichloroacetyl lysine protein adducts,
as reported in Forkert et al., 2006) are needed in order to more confidently characterize its flux.
In a situation such as TCE in which there is large database of studies coupled with
complex toxicokinetics, the Bayesian approach provides a systematic method of simultaneously
estimating model parameters and characterizing their uncertainty and variability. While such an
approach is not necessarily needed for all applications, such as route-to-route extrapolation (Chiu
and White, 2006), as discussed in Barton et al. (2007), characterization of uncertainty and
variability is increasingly recognized as important for risk assessment while representing a
continuing challenge for both PBPK modelers and users. If there is sufficient reason to
characterize uncertainty and variability in a highly transparent and objective manner, there is no
reason why our approach could not be applied to other chemicals. However, such an endeavor is
clearly not trivial, though the high level of effort for TCE is partially due to the complexity of its
metabolism and the extent of its toxicokinetic database.
This document is a draft for review purposes only and does not constitute Agency policy.
187 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
It is notable that, with experience, the methodology for the Bayesian approach to PBPK
modeling of TCE has evolved significantly from that of Bois (2000a, b), to Hack et al. (2006), to
the present analysis. Part of this evolution has been a more refined specification of the problem
being addressed, showing the importance of "problem formulation" in risk assessment
applications of PBPK modeling. The particular hierarchical population model for each species
was specified based on the intended use of the model predictions, so that relevant data can be
selected for analysis (e.g., excluding most grouped human data in favor of individual human
data) and data can be appropriately grouped (e.g., in rodent data, grouping by sex and strain
within a particular study). Thus, the predictions from the population model in rodents are the
"average" for a particular "lot" of rodents of a particular species, strain, and sex. This is in
contrast to the Hack et al. (2006) model, in which each dose group was treated as a separate
subject. As discussed above, this previous population model structure led to the unlikely result
that different dose groups within a closed-chamber study had significantly different Vmax values.
In humans, however, interindividual variability is of interest, and furthermore, substantial
individual data are available in humans. Hack et al. (2006) mixed individual- and group-level
data, depending on the availability from the published study, but this approach likely
underestimates population variability due to group means being treated as individuals. In
addition, in some studies, the same individual was exposed more than once, and in Hack et al.
(2006), these were treated as different "individuals." In this case, actual interindividual
variability may be either over- or underestimated, depending on the degree of interoccasion
variability. While it is technically feasible to include interoccasion variability, it would have
added substantially to the computational burden and reduced parameter identifiability. In
addition, a primary interest for this risk assessment is chronic exposure, so the predictions from
the population model in humans are the "average" across different occasions for a particular
individual (adult).
The second aspect of this evolution is the drive towards increased objectivity and
transparency. For instance, available information, or the lack thereof, is formally codified and
explicit either in prior distributions or in the data used to generate posterior distributions, and not
both. Methods at minimizing subjectivity (and hence improving reproducibility) in parameter
estimation include: (1) clear separation between the in vitro or physiologic data used to develop
prior distributions and the in vivo data used to generate posterior distributions; (2) use of
noninformative distributions, first updated using a probabilistic model of interspecies-scaling
that allows for prediction error, for parameters lacking in prior information; and (3) use of a
more comprehensive database of physiologic data, in vitro measurements, and in vivo data for
parameter calibration or for out-of-sample evaluation ("validation"). These measures increase
This document is a draft for review purposes only and does not constitute Agency policy.
188 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
the confidence that the approach employed also provides adequate characterization of the
uncertainty in metabolic pathways for which available data was sparse or relatively indirect, such
as GSH conjugation in rodents and respiratory tract metabolism. Moreover, this approach yields
more confident insights into what additional data can reduce these uncertainties than approaches
that rely on more subjective methods.
3.5.7.1.6. Key Limitations and Potential Implications of Violating Key Assumptions
Like all analyses, this one has a number of limitations and opportunities for refinement,
both biological and statistical. Of course, the modeling results are highly dependent on the
assumed PBPK model structure. However, most of the elements of the model structure are well
established for volatile, lipophilic chemicals such as TCE, and, thus, these assumptions are
unlikely to introduce much bias or inaccuracy. In terms of the statistical model, a key
assumption is the choice of prior and population distributions, —particularly the choice of
unimodal distributions for population variability. While reasonable as a first approximation,
especially without data to suggest otherwise, this assumption may introduce inaccuracies in the
predictions of population variability. For example, if there were an underlying bimodal
distribution, then fitting using a unimodal population distribution would lead to a high estimate
for the variance, and potentially overestimate the degree of population variability. In some cases
in the human model where larger population variance distributions are estimated, this may be the
underlying cause. However, only in the case of GSH conjugation in humans do the larger
estimates of population variability impact the dose-metric predictions used in the dose-response
assessment, so the impact of this assumption is limited for this assessment.
In addition, certain sources of variability, such as between-animal variability in rodents
and between-occasion variability in humans were not included in the hierarchical model, but
were aggregated with other sources of variability in a "residual" error term. Based on the
posterior predictions, it does not appear that this assumption has introduced significant bias in
the estimates because the residuals between predictions and data do not overall appear
systematically high or low. However, this could be verified by addressing between-animal
variability in rodents (requiring a more rigorous treatment of aggregated data, e.g., Chiu and
Bois, 2007) and incorporation of interoccasion variability in humans (e.g., Bernillon and Bois,
2000).
Some key potential refinements are as follows. First would be the inclusion of a CH
submodel, so that pharmacokinetic data, such as that recently published by Merdink et al. (2008),
could be incorporated. In addition, the current analysis is still dependent on a model structure
This document is a draft for review purposes only and does not constitute Agency policy.
189 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
substantially informed by deterministic analyses that test alternative model structures (Evans
et al., submitted), as probabilistic methods for discrimination or selection among complex,
nonlinear models such as that for TCE toxicokinetics have not yet been widely accepted.
Therefore, additional refinement of the respiratory tract model may be possible, though more
direct in vivo data would likely be necessary to strongly discriminating among models. In terms
of validation, application of more sophisticated methods such as cross-validation, may be useful
in further assessing the robustness of the modeling. Finally, additional model changes that may
be of utility to risk assessment, such as development of models for different lifestages (including
childhood and pregnancy), would likely require additional in vivo or in vitro data, particularly as
to metabolism, to ensure model identifiability.
3.5.7.1.7. Overall Evaluation of Physiologically Based Pharmacokinetic (PBPK) Model-
Based Internal Dose Predictions
The utility of the PBPK model developed here for making predictions of internal dose
can be evaluated based on four different components: (1) the degree to which the simulations
have converged to the true posterior distribution; (2) the degree of overall uncertainty and
variability; (3) for humans, the degree of uncertainty in the population; and (4) the degree to
which the model predictions are consistent with in vivo data that are informative to a particular
dose-metric. Table 3-51 summarizes these considerations for each dose-metric prediction. Note
that this evaluation does not consider in any way the extent to which a dose-metric may be the
appropriate choice for a particular toxic endpoint.
Overall, the least uncertain dose-metrics are the fluxes of total metabolism
(TotMetabBW34), total oxidative metabolism (TotOxMetabBW34), and hepatic oxidation
(AMetLivlBW34). These all have excellent posterior convergence (R diagnostic < 1.01),
relatively low uncertainty and variability (GSD < 2), and relatively low uncertainty in human
population variability (GSD for population percentiles <2). In addition, the PBPK model
predictions compare well with the available in vivo pharmacokinetic data.
Predictions for TCE in blood (AUCCBld) are somewhat more uncertain. Although
convergence was excellent across species (R < 1.01), overall uncertainty and variability was
about threefold. In humans, the uncertainty in human population variability was relatively low
(GSD for population percentiles <1.5). TCE blood level predictions were somewhat high in
comparison to the Chiu et al. (2006a) study at 1 ppm, though the predictions were better for most
of the other studies at higher exposure levels. In mice, TCE blood levels were somewhat
over-predicted in open-chamber inhalation studies. In both mice and rats, there were some cases
This document is a draft for review purposes only and does not constitute Agency policy.
190 DRAFT—DO NOT CITE OR QUOTE
-------
1 in which fits were inconsistent across dose groups if the same parameters were used across dose
2 groups, indicating unaccounted-for dose-related effects or intrastudy variability. However, in
3 both rats and humans, TCE blood (humans and rats) and tissue (rats only) concentrations from
This document is a draft for review purposes only and does not constitute Agency policy.
191 DRAFT—DO NOT CITE OR QUOTE
-------
a 2
o S.
a ^3
>! St
££¦ ^
^ c§
OQ o
<* s
a c>
.cj ^
^ O
§•
>1
Table 3-51. Degree of variance in dose-metric predictions due to incomplete convergence (columns 2-4),
combined uncertainty and population variability (columns 5-7), uncertainty in particular human population
percentiles (columns 8-10), model fits to in vivo data (column 11). The GSD is the geometric standard deviation,
which is a "fold-change" from the central tendency.
Dose-metric
Convergence: R for generic
scenarios
GSD for combined
uncertainty and variability
GSD for uncertainty in human
population percentiles
Comments regarding model fits to in
vivo data
abbreviation
Mouse
Rat
Human
Mouse
Rat
Human
1-5%
25-75%
95-99%
ABioactDCVCBW3
4,
ABioactDCVCKid
—
<1.016
<1.015
—
<3.92
<3.77
<2.08
<1.64
<1.30
Good fits to urinary NAcDCVC, and
blood DCVG.
AMetGSHBW34
<1.011
<1.024
<1.015
<9.09
<3.28
<3.73
<2.08
<1.64
<1.29
Good fits to urinary NAcDCVC, and
blood DCVG.
AMetLivlBW34
<1.000
<1.003
<1.004
<2.02
<1.84
<1.97
<1.82
<1.16
<1.16
Good fits to oxidative metabolites.
AMetLivOtherB W 3
4,
AMetLivOtherLiv
<1.004
<1.151
<1.012
<3.65
<3.36
<3.97
<2.63
<1.92
<2.05
No direct in vivo data.
AMetLngB W3 4,
AMetLngResp
<1.001
<1.003
<1.002
<4.65
<4.91
<10.4
<4.02
<2.34
<1.83
No direct in vivo data, but good fits to
closed-chamber.
AUCCBld
<1.001
<1.004
<1.005
<3.04
<3.16
<3.32
<1.20
<1.43
<1.49
Generally good fits, but poor fit to a few
mouse and human studies
AUCCTCOH
<1.001
<1.029
<1.002
<3.35
<8.78
<5.84
<1.73
<1.20
<1.23
Good fits across all three species.
AUCLivTCA
<1.000
<1.005
<1.002
<2.29
<3.18
<2.90
<1.65
<1.30
<1.40
Good fits to rodent data.
TotMetabBW34
<1.001
<1.004
<1.004
<1.92
<1.82
<1.81
<1.13
<1.12
<1.18
Good fits to closed-chamber.
T otOxMetabB W3 4
<1.001
<1.003
<1.004
<1.94
<1.85
<1.96
<1.77
<1.15
<1.20
Good fits to closed-chamber and
oxidative metabolites.
TotTCAInBW
<1.002
<1.002
<1.001
<1.96
<2.69
<2.30
<1.68
<1.19
<1.19
Good fits to TCA data.
<3
vo
to
o
O
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
studies not used for calibration (i.e., saved for "out-of-sample" evaluation/"validation") were
well simulated, adding confidence to the parent compound dose-metric predictions.
For the TCA dose-metric predictions (TotTCAInBW, AUCLivTCA) convergence in all
three species was excellent (R < 1.01). Overall uncertainty and variability was intermediate
between dose-metrics for metabolism and that for TCE in blood, with GSD of about two to
threefold. Uncertainty in human population percentiles was relatively low (GSD of 1.2-1.7).
While liver TCA levels were generally well fit, the data was relatively sparse. Plasma and blood
TCA levels were generally well fit, though in mice, there were again some cases in which fits
were inconsistent across dose groups if the same parameters were used across dose groups,
indicating unaccounted-for dose-related effects or intrastudy variability. In humans, the accurate
predictions for TCA blood and urine concentrations from studies used for "out of sample"
evaluation lends further confidence to dose-metrics involving TCA.
The evaluation of TCOH in blood followed a similar pattern. Convergence in all three
species was good, though the rat model had slightly worse convergence (R ~ 1.03) than the
mouse and humans (R < 1.01). In mice, overall uncertainty and variability was slightly more
than for TCE in blood. There much higher overall uncertainty and variability in the rat
predictions (GSD of almost 9) that likely reflects true interstudy variability. The
population-generated predictions for TCOH and TCOG in blood and urine were quite wide, with
some in vivo data both at the upper and lower ends of the range of predictions. In humans, the
overall uncertainty and variability was intermediate between mice and rats (GSD = 5.8). As with
the rats, this likely reflects true population heterogeneity, as the uncertainty in human population
percentiles was relatively low (GSD of around 1.2-1.7-fold). For all three species, fits to in vivo
data are generally good. In mice, however, there were again some cases in which fits were
inconsistent across dose groups if the same parameters were used across dose groups, indicating
unaccounted-for dose-related effects or intrastudy variability. In humans, the accurate
predictions for TCOH blood and urine concentrations from studies used for "out of sample"
evaluation lends further confidence to those dose-metrics involving TCOH.
GSH metabolism dose-metrics (ABioactDCVCBW34, ABioactDCVCKid,
AMetGSHBW34) had the greatest overall uncertainty in mice but was fairly well characterized
in rats and humans. In mice, there was no in vivo data informing this pathway except for the
indirect constraint of overall mass balance. So although convergence was adequate (R < 1.02),
the uncertainty/variability was very large, with a GSD of ninefold for the overall flux (the
amount of bioactivation was not characterized because there are no data constraining
downstream GSH pathways). For rats, there were additional constraints from (well-fit) urinary
NAcDCVC data, which reduced the overall uncertainty and variability substantially (GSD <
This document is a draft for review purposes only and does not constitute Agency policy.
193 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
fourfold). In humans, in addition to urinary NAcDCVC data, DCVG blood concentration data
was available, though only at the group level. These data, both of which were well fit, in
addition to the greater amount of in vitro metabolism data, allowed for the flux through the GSH
pathway and the rate of DCVC bioactivation to be fairly well constrained, with overall
uncertainty and variability having GSD < fourfold, and uncertainty in population percentiles no
more than about twofold. However, these predictions may need to be interpreted with caution,
given potential analytical issues with quantifying DCVG either in vitro or in vivo (see
Section 3.3.3.2). Thus, the substantial inconsistencies across studies and methods in the
quantification of DCVG following TCE exposure suggest lower confidence in the accuracy of
these predictions.
The final two dose-metrics, respiratory metabolism (AMetLngBW34, AMetLngResp)
and "other" oxidative metabolism (AMetLivOtherBW34, AMetLivOtherLiv), also lacked direct
in vivo data and were predicted largely on the basis of mass balance and physiological
constraints. Respiratory metabolism had good convergence (R < 1.01), helped by the availability
of closed-chamber data in rodents. In rats and mice, overall uncertainty and variability was
rather uncertain (GSD of four-fivefold), but the overall uncertainty and variability was much
greater in humans, with a GSD of about 10-fold. This largely reflects the significant variability
across individuals as well as substantial uncertainty in the low population percentiles (GSD of
fourfold). However, the middle (i.e., "typical" individuals) and upper percentiles (i.e., the
individuals at highest risk) are fairly well constrained with a GSD of around twofold. For the
"other" oxidative metabolism dose-metric, convergence was good in mice and humans
(R < 1.02), but less than ideal in rats (R~ 1.15). In rodents, the overall uncertainty and
variability were moderate, with a GSD around 3.5-fold, slightly higher than that for TCE in
blood. The overall uncertainty and variability in this metric in humans had a GSD of about
fourfold, slightly higher than for GSH conjugation metrics. However, uncertainty in the middle
and upper population percentiles had GSDs of only about twofold, similar to that for respiratory
metabolism.
Overall, as shown in Table 3-51, the updated PBPK model appears to be most reliable for
the fluxes of total, oxidative, and hepatic oxidative metabolism. In addition, dose-metrics related
to blood levels of TCE and oxidative metabolites TCOH and TCA had only modest uncertainty.
In the case of TCE in blood, for some data sets, model predictions over-predicted the in vivo
data, and, in the case of TCOH in rats, substantial interstudy variability was evident. For GSH
metabolism, dose-metric predictions for rats and humans had only slightly greater uncertainty
than the TCE and metabolism metrics. Predictions for mice were much more uncertain,
reflecting the lack of GSD-specific in vivo data. Finally, for "other" oxidative metabolism and
This document is a draft for review purposes only and does not constitute Agency policy.
194 DRAFT—DO NOT CITE OR QUOTE
-------
1 respiratory oxidative metabolism, predictions also had somewhat more uncertainty than the TCE
2 and metabolism metrics, though uncertainty in middle and upper human population percentiles
3 was modest.
4
This document is a draft for review purposes only and does not constitute Agency policy.
195 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
4. HAZARD CHARACTERIZATION
This section presents the hazard characterization of trichloroethylene (TCE) health
effects. Because of the number of studies and their relevance to multiple endpoints, the
evaluation of epidemiologic studies of cancer and TCE is summarized in Section 4.1
(endpoint-specific results are presented in subsequent sections). Genotoxicity data are discussed
in Section 4.2. Due to the large number of endpoints and studies in the toxicity database,
subsequent sections (see Sections 4.3-4.10) are organized by tissue/organ system. Each section
is further organized by noncancer and cancer endpoints, discussing data from human
epidemiologic and laboratory experimental studies. In cases where there is adequate
information, the role of metabolism in toxicity, comparisons of toxicity between TCE and its
metabolites, and carcinogenic mode of action (MOA) are also discussed. Finally, Section 4.11
summarizes the overall hazard characterization and the weight of evidence for noncancer and
carcinogenic effects.
4.1. EPIDEMIOLOGIC STUDIES ON CANCER AND TRICHLOROETHYLENE
(TCE)—METHODOLOGICAL OVERVIEW
This brief overview of the epidemiologic studies on cancer and TCE below provides
background to the discussion contained in Sections 4.4-4.10. Over 50 epidemiologic studies on
cancer and TCE exposure (see Tables 4-1 through 4-3) were examined to assess their ability to
inform weight-of-evidence evaluation, i.e., to inform the cancer hazard from TCE exposure,
according to 15 standards of study design (see Table 4-4), conduct, and analysis. The analysis of
epidemiologic studies on cancer and TCE serves to document essential design features, exposure
assessment approaches, statistical analyses, and potential sources of confounding and bias. This
analysis, furthermore, supports the discussion of site-specific cancer observations in
Sections 4.4-4.9. In those sections, study findings are presented with an assessment and
discussion of their observations according to a study's weight of evidence and the potential for
alternative explanations, including bias and confounding. Tables containing observed findings
for site-specific cancers are also found in Sections 4.4-4.9. Full details of the weight-of
evidence-review to identify a cancer hazard and study selections for meta-analysis may be found
in Appendix B.
This document is a draft for review purposes only and does not constitute Agency policy.
1 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-1. Description of epidemiologic cohort and proportionate mortality ratio (PMR) studies assessing
cancer and TCE exposure
Reference
Description
Study group (N)
comparison group (N)
Exposure assessment and other information
Aircraft and aerospace workers
Radican et al.
(2008); Blair
et al. (1998)
Civilian aircraft-maintenance
workers with at least yr in
1952-1956 at Hill Air Force Base,
UT. VS to 1990 (Blair etal., 1998)
or 2000 (Radican et al., 2008);
cancer incidence 1973-1990 (Blair
etal., 1998)
14,457 (7,204 ever exposed to
TCE).
Incidence (Blair et al., 1998) and
mortality rates (Blair et al., 1998;
Radican et al., 2008) of
nonchemical exposed subjects.
Most subjects (n = 10,718) with potential exposure to 1-25 solvents.
Cumulative TCE assigned to individual subjects using JEM.
Exposure-response patterns assessed using cumulative exposure,
continuous or intermittent exposures, and peak exposure. TCE
replaced in 1968 with 1,1,1-trichloroethane and was discontinued in
1978 in vapor degreasing activities. Median TCE exposures were
about 10 ppm for rag and bucket; 100-200 ppm for vapor
degreasing. Poisson regression analyses controlled for age, calendar
time, sex (Blair et al., 1998) or Cox proportional hazard model for
age and race.
Krishnadasan
et al. (2007)
Nested case-control study within a
cohort of 7,618 workers employed
for between 1950 and 1992, or who
had started employment before
1980 at Boeing/Rockwell/
Rocketdyne [Santa Susana Field
Laboratory, the UCLA cohort of
(Morgenstern et al., 1997)]. Cancer
incidence 1988-1999.
326 prostate cancer cases,
1,805 controls.
Response rate:
Cases, 69%; Controls, 60%.
JEM for TCE, hydrazine, PAHs, benzene, mineral oil constructed
from company records, walk-through, or interviews. Lifestyle factors
obtained from living subjects through mail and telephone surveys.
Conditional logistic regression controlled for cohort, age at diagnosis,
physical activity, SES and other occupational exposure (benzene,
PAHs, mineral oil, hydrazine).
-------
Table 4-1. Description of epidemiologic cohort and proportionate mortality ratio (PMR) studies assessing cancer and
TCE exposure (continued)
Reference
Description
Study group (N)
comparison group (N)
Exposure assessment and other information
Ritz et al.
(1999); Zhao
et al. (2005)
Aerospace workers with >2 yr of
employment at
Rockwell/Rocketdyne (now Boeing)
and who worked at Santa Susana
Field Laboratory, Ventura, CA,
from 1950-1993 [the UCLA cohort
of (Morgenstern et al., 1997)].
Cancer mortality as of December
31,2001. Cancer incidence
1988-2000 for subjects alive as of
1988.
6,044 (2,689 with high cumulative
exposure to TCE). Mortality rates of
subjects in lowest TCE exposure
category.
5,049 (2,227 with high cumulative
exposure to TCE). Incidence rates of
subjects in lowest TCE exposure
category.
JEM for TCE, hydrazine, PAHs, mineral oil, and benzene. IH
ranked each job title ranked for presumptive TCE exposure as high
(3), medium (2), low (1), or no (0) exposure for 3three time periods
(1951-1969, 1970-1979, 1980-1989). Cumulative TCE score: low
(up to 3), medium (over 3 up to 12), high (over 12) assigned to
individual subjects using JEM. Cox proportional hazard, controlled
for time, since 1st employment, SES, age at diagnosis and hydrazine.
Boice
et al.
(2006b)
Aerospace workers with >6 mo
employment at
Rockwell/Rocketdyne (Santa
Susana Field Laboratory and nearby
facilities) from 1948-1999 [IEI
cohort, (IEI, 2005)]. VS to 1999.
41,351, 1,642 male hourly test stand
mechanics (1,111 with potential TCE
exposure).
Mortality rates of U.S. population
and California population. Internal
referent groups including male hourly
nonadministrative Rocketdyne
workers; male hourly,
nonadministrative SSFL workers;
and test stand mechanics with no
potential exposure to TCE.
Potential TCE exposure assigned to test stands workers only whose
tasks included the cleaning or flushing of rocket engines (engine
flush) (n = 639) or for general utility cleaning (n = 472); potential for
exposure to large quantities of TCE was much greater during engine
flush than when TCE used as a utility solvent. JEM for TCE and
hydrazine without semiquantitative intensity estimates. Exposure to
other solvents not evaluated due to low potential for confounding
(few exposed, low exposure intensity, or not carcinogenic).
Exposure metrics included employment duration, employment
decade, yr worked with potential TCE exposure, and yr worked with
potential TCE exposure via engine cleaning, weighted by number of
tests. Lifetable (SMR); Cox proportional hazard controlling for birth
yr, hire yr, and hydrazine exposure.
-------
Table 4-1. Description of epidemiologic cohort and proportionate mortality ratio (PMR) studies assessing cancer and
TCE exposure (continued)
Reference
Description
Study group (N)
comparison group (N)
Exposure assessment and other information
Boice et
al. (1999)
Aircraft-manufacturing workers
with at least 1 yr >1960 at Lockheed
Martin (Burbank, CA). VS to 1996.
77,965 (2,267 with potential routine
TCE exposures and 3,016 with
routine or intermittent TCE
exposure).
Mortality rates of U.S. population
(routine TCE exposed subjects) and
nonexposed internal referents
(routine and intermittent TCE
exposed subjects).
12% with potential routine mixed solvent exposure and 30% with
route or intermittent solvent exposure. JEM for potential TCE
exposure on (1) routine basis or (2) intermittent or routine basis
without semiquantitative intensity estimate. Exposure-response
patterns assessed by any exposure or duration of exposure and
internal control group. Vapor degreasing with TCE before 1966 and
PCE, afterwards. Lifetable analyses; Poisson regression analysis
adjusting for birth date, starting employment date, finishing
employment date, sex and race.
Morgan et al.
(1998)
Aerospace workers with >6 mo
1950-1985 at Hughes (Tucson,
AZ). VS to 1993.
20,508 (4,733 with TCE exposures).
Mortality rates of U.S. population for
overall TCE exposure; mortality rates
of all-other cohort subjects (internal
referents).
TCE exposure intensity assigned using JEM. Exposure-response
patterns assessed using cumulative exposure (low vs. high) and job
with highest TCE exposure rating (peak, medium/high exposure vs.
no/low exposure). "High exposure" job classification defined as
>50 ppm. Vapor degreasing with TCE 1952-1977, but limited IH
data <1975. Limited IH data before 1975 and medium/low rankings
likely misclassified given temporal changes in exposure intensity not
fully considered (NRC, 2006).
Costa et
al. (1989)
Aircraft manufacturing workers
employed 1954-1981at plant in
Italy. VS to 1981.
8,626 subjects
Mortality rates of the Italian
population.
No exposure assessment to TCE and job titles grouped into one of
four categories: blue- and white-collar workers, technical staff, and
administrative clerks. Lifetable (SMR).
Garabrant
et al.
(1988)
Aircraft manufacturing workers >4
yr employment and who had worked
at least 1 d at San Diego, CA, plant
1958-1982. VS to 1982.
14,067
Mortality rates of U.S. population.
TCE exposure assessment for 70 of 14,067 subjects; 14 cases of
esophageal cancer and 56 matched controls. For these 70 subjects,
company work records identified 37% with job title with potential
TCE exposure without quantitative estimates. Lifetable (SMR).
-------
Table 4-1. Description of epidemiologic cohort and proportionate mortality ratio (PMR) studies assessing cancer and
TCE exposure (continued)
Reference
Description
Study group (N)
comparison group (N)
Exposure assessment and other information
Cohorts Identified From Biological Monitoring (U-TCA)
Hansen et al.
(2001)
Workers biological monitored using
U-TCA and air-TCE, 1947-1989.
Cancer incidence from 1964-1996.
803 total.
Cancer incidence rates of the Danish
population.
712 with U-TCA, 89 with air-TCE measurement records, two with
records of both types. U-TCA from 1947-1989; air TCE
measurements from 1974. Historic median exposures estimated
from the U-TCA concentrations were: 9 ppmfor 1947-1964, 5 ppm
for 1965-1973, 4 ppm for 1974-1979, and 0.7 ppmfor 1980-1989.
Air TCE measurements from 1974 onward were 19 ppm (mean) and
5 ppm (median). Overall, median TCE exposure to cohort as
extrapolated from air TCE and U-TCA measurements was
4 ppm (arithmetic mean, 12 ppm). Exposure metrics: yr 1st
employed, employment duration, mean exposure, cumulative
exposure. Exposure metrics: employment duration, average TCE
intensity, cumulative TCE, period 1st employment. Lifetable
analysis (SIR).
Anttila et al.
(1995)
Workers biological monitored using
U-TCA, 1965-1982. VS
1965-1991 and cancer incidence
1967-1992.
3,974 total (3,089 with U-TCA
measurements).
Mortality and cancer incidence rates
of the Finnish population.
Median U-TCA, 63 |imol/L for females and 48 |imol/L for males;
mean U-TCA was 100 |imol/L. Average 2.5 U-TCA measurements
per individual. Using the Ikeda et al. (1972) relationship for TCE
exposure to U-TCA, TCE exposures were roughly 4 ppm
(median) and 6 ppm (mean). Exposure metrics: yr since
1st measurement. Lifetable analysis (SMR, SIR).
Axelson et al.
(1994)
Workers biological monitored using
U-TCA, 1955-1975. VS to 1986
and cancer incidence 1958-1987.
1,421 males.
Mortality and cancer incidence rates
of Swedish male population.
Biological monitoring for U-TCA from 1955 and 1975. Roughly
% of cohort had U-TCA concentrations equivalent to <20 ppm
TCE. Exposure metrics: duration exposure, mean U-TCA.
Lifetable analysis (SMR, SIR).
Other Cohorts
Clapp and
Hoffman
(2008)
Deaths between 1969-2001 among
employees >5 yr employment
duration at an IBM facility
(Endicott, NY).
360 deaths.
Proportion of deaths among New
York residents during 1979-1998.
No exposure assessment to TCE. PMR analysis.
-------
Table 4-1. Description of epidemiologic cohort and proportionate mortality ratio (PMR) studies assessing cancer and
TCE exposure (continued)
Reference
Description
Study group (N)
comparison group (N)
Exposure assessment and other information
Sung et al.
(2007; 2008)
Female workers 1st employed
1973-1997 at an electronics (RCA)
manufacturing factory (Taoyuan,
Taiwan). Cancer incidence
1979-2001 (Sung et al., 2007).
Childhood leukemia 1979-2001
among first born of female subjects
in Sung et al. (2007; 2008)
63,982 females and 40,647 females
with 1st live born offspring.
Cancer incidence rates of Taiwan
population (Sung et al., 2007).
Childhood leukemia incidence rates
of first born live births of Taiwan
population (Sung et al., 2007).
No exposure assessment. Chlorinated solvents including TCE and
PCE found in soil and groundwater at factory site. Company records
indicated TCE not used 1975-1991 and PCE 1975-1991 and PCE
after 1981. No information for other time periods.
Exposure-response using employment duration. Lifetable analysis
(SMR, SIR) (Chang et al., 2003; Chang et al., 2005; Sung et al.,
2007) or Poisson regression adjusting for maternal age, education,
sex, and birth yr (Sung et al., 2008).
Chang et al.
(2005),
Male and female workers employed
1978-1997 at electronics factory as
studied by Sung et al. (2007). VS
from 1985-1997 and cancer
incidence 1979-1997.
86,868 total.
Incidence (Chang et al., 2005)
mortality (Chang et al., 2003) rates
Taiwan population.
ATSDR
(2004)
Workers 1952-1980 at the
View-Master factory (Beaverton,
OR).
616 deaths 1989-2001.
Proportion of deaths between
1989-2001 in Oregon population.
No exposure information on individual subjects. TCE and other
VOCs detected in well water at the time of the plant closure in 1998
were TCE, 1,220-1,670 |ig/L: 1,1-DCE, up to 33 |ig/L; and, PCE up
to 56 ng/L. PMR analysis.
Raaschou-
Nielsen et al.
(2003)
Blue-collar workers employed
>1968 at 347 Danish TCE-using
companies. Cancer incidence
through 1997.
40,049 total (14,360 with presumably
higher level exposure to TCE).
Cancer incidence rates of the Danish
population.
Employers had documented TCE usage. Blue-collar vs. white-collar
workers and companies with <200 workers were variables identified
as increasing the likelihood for TCE exposure. Subjects from iron
and metal, electronics, painting, printing, chemical, and dry cleaning
industries. Median exposures to trichloroethylene were
40-60 ppm for the yr before 1970,10-20 ppm for 1970-1979,
and approximately 4 ppm for 1980-1989. Exposure metrics:
employment duration, yr 1st employed, and # employees in company.
Lifetable (SIR).
-------
Table 4-1. Description of epidemiologic cohort and proportionate mortality ratio (PMR) studies assessing cancer and
TCE exposure (continued)
Reference
Description
Study group (N)
comparison group (N)
Exposure assessment and other information
Ritz (1999a)
Male uranium-processing plant
workers >3 mo employment
1951-1972 at DOE facility
(Fernald, OH). VS 1951-1989,
cancer.
3,814 white males monitored for
radiation (2,971 with potential TCE
exposure).
Mortality rates of the U.S.
population; non-TCE exposed
internal controls for TCE
exposure-response analyses.
JEM for TCE, cutting fluids, kerosene, and radiation generated by
employees and industrial hygienists. Subjects assigned potential
TCE according to intensity: light (2,792 subjects), moderate
(179 subjects), heavy (no subjects). Lifetable (SMR) and conditional
logistic regression adjusted for pay status, date first hire, radiation.
Henschler
etal. (1995)
Male workers >1 yr 1956-1975 at
cardboard factory (Arnsberg region,
Germany). VS to 1992.
169 exposed; 190 unexposed
Mortality rates from German
Democratic Republic (broad
categories) or renal cell carcinoma
incidence rates from Danish
population, German Democratic, or
non-TCE exposed subjects.
Walk-through surveys and employee interviews used to identify
work areas with TCE exposure. TCE exposure assigned to renal
cancer cases using workman's compensation files. Lifetable (SMR,
SIR) or Mantel-Haenszel.
Greenland
etal. (1994)
Cancer deaths, 1969-1984, among
pensioned workers employed <1984
at GE transformer manufacturing
plant (Pittsfield, MA), and who had
job history record; controls were
noncancer deaths among pensioned
workers.
512 cases, 1,202 controls.
Response rate:
Cases, 69%;
Controls, 60%.
Industrial hygienist assessment from interviews and position
descriptions. TCE (no/any exposure) assigned to individual subjects
using JEM. Logistic regression.
Sinks et al.
(1992)
Workers employed 1957-1980 at a
paperboard container manufacturing
and printing plant (Newnan, GA).
VS to 1988. Kidney and bladder
cancer incidence through 1990.
2,050 total.
Mortality rates of the U.S.
population, bladder and kidney
cancer incidence rates from the
Atlanta-SEER registry for the yr
1973-1977.
No exposure assessment to TCE; analyses of all plant employees
including white- and blue-collar employees. Assignment of work
department in case-control study based upon work history; Material
Safety Data Sheets identified chemical usage by department.
Lifetable (SMR, SIR) or conditional logistic regression adjusted for
hire date and age at hire, and using 5- and 10-yr lagged employment
duration.
-------
Table 4-1. Description of epidemiologic cohort and proportionate mortality ratio (PMR) studies assessing cancer and
TCE exposure (continued)
Reference
Description
Study group (N)
comparison group (N)
Exposure assessment and other information
Blair et al.
(1989)
Workers employed 1942-1970 in
U.S. Coast. VS to 1980.
3,781 males of whom 1,767 were
marine inspectors (48%).
Mortality rates of the U.S.
population. Mortality rates of marine
inspectors also compared to that of
noninspectors.
No exposure assessment to TCE. Marine inspectors worked in
confined spaces and had exposure potential to multiple chemicals.
TCE was identified as one of 10 potential chemical exposures.
Lifetable (SMR) and directly adjusted relative risks.
Shannon et al.
(1988)
Workers employed >6 mo at GE
lamp manufacturing plant,
1960-1975. Cancer incidence from
1964-1982.
1,870 males and females, 249 (13%)
in coiling and wire-drawing area.
Cancer incidence rates from Ontario
Cancer Registry.
No exposure assessment to TCE. Workers in CWD had potential
exposure to many chemicals including metals and solvents. A
1955-dated engineering instruction sheet identified trichloroethylene
used as degreasing solvent in CWD. Lifetable (SMR).
Shindell and
Ulrich (1985)
Workers employed >3 mo at a TCE
manufacturing plant 1957-1983.
VS to 1983.
2,646 males and females.
Mortality rates of the U.S.
population.
No exposure assessment to TCE; job titles categorized as either
white- or blue-collar. Lifetable analysis (SMR).
Wilcosky
et al. (1984)
Respiratory, stomach, prostate,
lymphosarcoma, and lymphatic
leukemia cancer deaths 1964-1972
among 6,678 active and retired
production workers at a rubber plant
(Akron, OH); controls were a
20% age-stratified random sample
of the cohort.
183 cases (101 respiratory,
33 prostate, 30 stomach,
9 lymphosarcoma and 10 lymphatic
leukemia cancer deaths).
JEM without quantitative intensity estimates for 20 exposures
including TCE. Exposure metric: ever held job with potential TCE
exposure.
1, 1- DCE = dichloroethylene, CWD = coiling and wire drawing; DOE = U.S. Department of Energy, GE = General Electric, IBM = International Business
Machines Corporation, IEI = International Epidemiology Institute, IH = industrial hygienist, JEM = job-exposure matrix, NRC = National Research Council,
PAH = polycyclic aromatic hydrocarbon, PCE = perchloroethylene, PMR = proportionate mortality ratio, SES = socioeconomic status, SIR = standardized
incidence ratio, SMR = standardized mortality ratio, SSFL = Santa Susanna Field Laboratory, U-TCA = urinary trichloroacetic acid, UCLA = University of
California, Los Angeles, VOCs = volatile organic compounds, VS = vital status.
-------
L/l
K>
Table 4-2. Case-control epidemiologic studies examining cancer and TCE exposure
Reference
Population
Study group (N)
comparison group (N)
response rates
Exposure assessment and other information
Bladder
Pesch et al.
(2000a)
Histologically confirmed urothelial
cancer (bladder, ureter, renal
pelvis) cases from German
hospitals (5 regions) in
1991-1995; controls randomly
selected from residency registries
matched on region, sex, and age.
1,035 cases.
4,298 controls.
Cases, 84%; Controls, 71%.
Occupational history using job title or self-reported exposure. JEM and
JTEM to assign exposure potential to metals and solvents (chlorinated
solvents, TCE, PCE). Lifetime exposure to TCE exposure examined as 30th,
60th, and 90th percentiles (medium, high, and substantial) of exposed control
exposure index. Duration used to examine occupational title and job task
duties and defined as 30th, 60th, and 90th percentiles (medium, long, and very
long) of exposed control durations.
Logistic regression with covariates for age, study center, and smoking.
Siemiatycki
et al. (1994)
Siemiatycki
(1991)
Male bladder cancer cases,
age 35-75 yr, diagnosed in
16 large Montreal-area hospitals in
1979-1985 and histologically
confirmed; controls identified
concurrently at 18 other cancer
sites; age-matched,
population-based controls
identified from electoral lists and
RDD.
484 cases.
533 population controls;
740 other cancer controls.
Cases, 78%; Controls, 72%.
JEM to assign 294 exposures including TCE on semiquantitative scales
categorized as any or substantial exposure. Other exposure metrics included
exposure duration in occupation or job title.
Logistic regression adjusted for age, ethnic origin, socioeconomic status,
smoking, coffee consumption, and respondent status (occupation or job title)
or Mantel-Haenszel stratified on age, income, index for cigarette smoking,
coffee consumption, and respondent status (TCE).
Brain
DeRoos et al.
(2001a);
Olshan et al.
(1999)
Neuroblastoma cases in children of
<19 yr selected from Children's
Cancer Group and Pediatric
Oncology Group with diagnosis in
1992-1994; population controls
(RDD) matched to control on birth
date.
504 cases.
504 controls.
Cases, 73% Controls, 74%.
Telephone interview with parent using questionnaire to assess parental
occupation and self-reported exposure history and judgment-based attribution
of exposure to chemical classes (halogenated solvents) and specific solvents
(TCE). Exposure metric was any potential exposure.
Logistic regression with covariate for child's age and material race, age, and
education.
S?
>!
cs
s;
>1
£
si
o
>!
>!
O
a
a
a
§•
>!
o
o
p
H
o
o
2! . -
<3
Cs
o
a
>j
S"
(3s
H
1—1 s
H 2.
W|
O •
o
c
o
H
W
-------
Table 4-2. Case-control epidemiologic studies examining cancer and TCE exposure (continued)
Reference
Population
Study group (N)
comparison group (N)
response rates
Exposure assessment and other information
Heineman
etal. (1994)
White, male cases, age >30 yr,
identified from death certificates in
1978-1981; controls identified
from death certificates and
matched for age, yr of death and
study area.
300 cases.
386 controls.
Cases, 74%; Controls, 63%.
In-person interview with next-of-kin; questionnaire assessing lifetime
occupational history usins iob title and JEM of Gomez et al. (1994).
Cumulative exposure metric (low, medium or and high) based on weighted
probability and duration.
Logistic regression with covariates for age and study area.
Colon and Rectum
Goldberg et al.
(2001);
Siemiatycki
(1991)
Male colon cancer cases, 35-75 yr,
from 16 large Montreal-area
hospitals in 1979-1985 and
histologically confirmed; controls
identified concurrently at 18 other
cancer sites; age-matched,
population-based controls
identified from electoral lists and
RDD.
497 cases.
533 population controls and
740 cancer controls.
Cases, 82%; Controls, 72%.
In-person interviews (direct or proxy) with segments on work histories (job
titles and self-reported exposures); analyzed and coded by a team of chemists
and industrial hygienists (294 exposures on semiquantitative scales); potential
TCE exposure defined as any or substantial exposure.
Logistic regression adjusted for age, ethnic origin, birthplace, education,
income, parent's occupation, smoking, alcohol consumption, tea consumption,
respondent status, heating source socioeconomic status, smoking, coffee
consumption, and respondent status (occupation, some chemical agents) or
Mantel-Haenszel stratified on age, income, index for cigarette smoking, coffee
consumption, and respondent status (TCE).
Dumas et al.
(2000);
Simeiatycki
(1991)
Male rectal cancer cases, age
35-75 yr, diagnosed in 16 large
Montreal-area hospitals in
1979-1985 and histologically
confirmed; controls identified
concurrently at 18 other cancer
sites; age-matched,
population-based controls
identified from electoral lists and
RDD.
292 cases.
533 population controls and
740 other cancer controls.
Cases, 78%; Controls, 72%.
In-person interviews (direct or proxy) with segments on work histories (job
titles and self-reported exposures); analyzed and coded by a team of chemists
and industrial hygienists (294 exposures on semiquantitative scales); potential
TCE exposure defined as any or substantial exposure.
Logistic regression adjusted for age, education, respondent status, cigarette
smoking, beer consumption and body mass index (TCE) or Mantel-Haenszel
stratified on age, income, index for cigarette smoking, coffee consumption,
ethnic origin, and beer consumption (TCE).
-------
Table 4-2. Case-control epidemiologic studies examining cancer and TCE exposure (continued)
Reference
Population
Study group (N)
comparison group (N)
response rates
Exposure assessment and other information
Fredriksson
etal. (1989)
Colon cancer cases aged 30-75 yr
identified through the Swedish
Cancer Registry among patients
diagnosed in 1980-1983;
population-based controls were
frequency-matched on age and sex
and were randomly selected from a
population register.
329 cases.
658 controls.
Not available.
Mailed questionnaire assessing occupational history with telephone interview
follow-up. Self-reported exposure to TCE defined as any exposure.
Mantel-Haenszel stratified on age, sex, and physical activity.
Esophagus
Parent et al.
(2000b);
Siemiatycki
(1991)
Male esophageal cancer cases,
35-75 yr, diagnosed in 19 large
Montreal-area hospitals in
1979-1985 and histologically
confirmed; controls identified
concurrently at 18 other cancer
sites; age-matched,
population-based controls
identified from electoral lists and
RDD.
292 cases.
533 population controls;
740 subjects with other
cancers.
Cases, 78%; controls, 72%.
In-person interviews (direct or proxy) with segments on work histories (job
titles and self-reported exposures); analyzed and coded by a team of chemists
and industrial hygienists (294 exposures on semiquantitative scales); potential
TCE exposure defined as any or substantial exposure.
Logistic regression adjusted for age, education, respondent status, cigarette
smoking, beer consumption and body mass index (solvents) or
Mantel-Haenszel stratified on age, income, index for cigarette smoking, coffee
consumption, ethnic origin, and beer consumption (TCE).
-------
Table 4-2. Case-control epidemiologic studies examining cancer and TCE exposure (continued)
Reference
Population
Study group (N)
comparison group (N)
response rates
Exposure assessment and other information
Lymphoma
Purdue et al.
(2011)
Cases aged 20-74 with
histologically-confirmed NHL
(B-cell diffuse and follicular,
T-cell, lymphoreticular) without
HIV in 1998-2000 and identified
from four SEER areas (Los
Angelos County and Detroit
metropolitan area, random sample;
SeattlePuget Sound and Iowa, all
consecutive cases); population
controls aged 20-74 with no
previous diagnosis of HIV
infection or NHL, identified
through (1) if >65 yr of age, RDD,
or (2) if >65 yr, identified from
Medicare eligibility files and
stratified on geographic area, age,
and race.
1,321 cases.
1,057 controls.
Cases , 76% ; Controls,
78%.
In-person interview using questionnaire or computer-assisted personal
interview questionnaire specific for jobs held for >1 yr since the age of 16 yr,
hobbies, and medical and family history. For occupational history, 32 job- or
industry-specific interview modules asked for detailed information on
individual jobs and focused on solvents exposure, including TCE, assessment
by expert industrial hygienist blinded to case and control status by levels of
probability, frequency, and intensity. Exposure metric of overall exposure,
average weekly exposure, yr exposed, average exposure intensity, and
cumulative exposure.
Logistic regression adjusted for sex, age, race, education and SEER site.
Gold et al.
Cases aged 35-74 with
histologically-confirmed multiple
myeloma in 2000-2002 and
identified from Seer areas (Detroit,
Seattle-Puget Sound); population
controls.
181 cases.
481 controls.
Cases, 71%; Controls, 52%.
In-person interview using computer-assisted personal interview questionnaire
for jobs held >1 yr since 1941 (cases) or 1946 (controls) and since age 18 yr.
For occupational history, 20 occupations, job- or industry-specific interview
modules asked for detailed information on individual jobs held at least 2 yr and
focused on solvents exposure, including TCE, assessment by expert industrial
hygienist blinded to case and control status by levels of probability, duration
and cumulative exposure.
Logistic regression adjusted for sex, age, race, education and SEER site.
-------
Table 4-2. Case-control epidemiologic studies examining cancer and TCE exposure (continued)
Reference
Population
Study group (N)
comparison group (N)
response rates
Exposure assessment and other information
Cocco et al.
(2010)
Histologically or cytologically
confirmed cases aged >17 yr with
lymphoma (B-cell, T-cell, CLL,
multiple myeloma, Hodgkin) in
1998-2004 and residents of
referral areas from seven European
countries (Czech Republic,
Finland, France, Germany, Ireland,
Italy, and Spain); hospital
(4 participating countries) or
population controls (all others);
controls from (1) Germany and
Italy selected by RDD from
general population and matched
(individually in German and
group-based in Italy) to cases by
sex, age and residence area, and,
(2) for all other countries,
matched hospital controls with
diagnoses other than cancer,
infectious diseases and
immundeficient diseases
(individually in Czech Republic
group-based in all other countries).
2,348 cases.
2,462 controls.
Cases, 88%; controls,
81% hospital and
52% population.
In-person interviews using same structured questionnaire translated to the local
language for information on sociodemographic factors, lifestyle, health history
and all full-time job held >1 yr. Assessment by industrial hygienists in each
participating center to 43 agents, including TCE, by confidence, exposure
intensity, and exposure frequency. Exposure metric of overall TCE exposure
and cumulative TCE exposure for subjects assessed with high degree of
confidence.
Logistic regression adjusted for age, gender, education and study center.
German
centers:
Seidler et al.
(2007); Mester
et al. (2006);
Becker et al.
(2004)
NHL and Hodgkin's disease cases
aged 18-80 yr identified through
all hospitals and ambulatory
physicians in six regions of
Germany between 1998 and 2003;
population controls were identified
from population registers and
matched on age, sex, and region.
710 cases.
710 controls.
Cases, 87%; Controls, 44%.
n-person interview using questionnaire assessing personal characteristics,
ifestyle, medical history, UV light exposure, and occupational history of all
obs held for >1 yr. Exposure of a prior interest were assessed using job
ask-specific supplementary questionnaires. JEM used to assign cumulative
[uantitative TCE exposure metric, categorized according to the distribution
imong the control persons (50th and 90th percentile of the exposed controls).
Conditional logistic regression adjusted for age, sex, region, smoking and
alcohol consumption.
-------
Table 4-2. Case-control epidemiologic studies examining cancer and TCE exposure (continued)
Reference
Population
Study group (N)
comparison group (N)
response rates
Exposure assessment and other information
Wang et al.
(2009)
Cases among females aged 21 and
84 yr with NHL in 1996-2000 and
identified from Connecticut
Cancer Registry; population-based
female controls (1) if <65 yr of
age, having Connecticut address
stratified by 5-yr age groups
identified from random digit
dialing or (2) >65 yr of age, by
random selection from Centers for
Medicare and Medicaid Service
files.
601 cases.
717 controls.
Cases, 72%; Controls, 69%
(<65 yr), 47% (>65 yr)
In-person interview using questionnaire assessment specific for jobs held for
>1 yr. Intensity and probability of exposure to broad category of organic
solvents and to individual solvents, including TCE, estimated using JEM
(Dosemeci et al., 1994; Gomez et al., 1994) and assigned blinded. Exposure
metric of any exposure, exposure intensity (low, medium/high), and exposure
probability (low, medium/high) Logistic regression adjusted for age, family
history of hematopoietic cancer, alcohol consumption and race.
Costantini
et al. (2008);
Miligi et al.
(2006)
Cases aged 20-74 with NHL,
including CLL, all forms of
leukemia, or MM in 1991-1993
and identified through surveys of
hospital and pathology
departments in study areas and in
specialized hematology centers in
eight areas in Italy;
population-based controls
stratified by 5-yr age groups and
by sex selected through random
sampling of demographic or of
National Health Service files.
1,428 NHL + CLL, 586
Leukemia, 263, MM.
1,278 controls (leukemia
analysis).
1,100 controls (MM
analysis).
Cases, 83%;Controls, 73%.
In-person interview primarily at interviewee's home (not blinded) using
questionnaire assessing specific jobs, extra occupational exposure to solvents
and pesticides, residential history, and medical history. Occupational exposure
assessed by job-specific or industry-specific questionnaires. JEM used to
assign TCE exposure and assessed using intensity (2 categories) and exposure
duration (2 categories). All NHL diagnoses and 20% sample of all cases
confirmed by panel of three pathologists.
Logistic regression with covariates for sex, age, region, and education.
Logistic regression for specific NHL included an additional covariate for
smoking.
-------
Table 4-2. Case-control epidemiologic studies examining cancer and TCE exposure (continued)
Reference
Population
Study group (N)
comparison group (N)
response rates
Exposure assessment and other information
Persson and
Fredriksson
(1999);
combined
analysis of
NHL cases in
Persson et al.
(1989; 1993)
Histologically confirmed cases of
B-cell NHL, age 20-79 yr,
identified in two hospitals in
Sweden: Oreboro in 1964-1986
(Persson et al., 1989) and in
Linkoping between 1975-1984
(Persson et al., 1993); controls
were identified from previous
studies and were randomly
selected from population registers.
NHL cases, 199.
479 controls.
Cases, 96% (Oreboro),
90% (Linkoping);
controls, not reported.
Mailed questionnaire to assess self reported occupational exposures to TCE
and other solvents.
Unadjusted Mantel-Haenszel chi-square.
Nordstrom
et al. (1998)
Histologically-confirmed cases in
males of hairy-cell leukemia
reported to Swedish Cancer
Registry in 1987-1992 (includes
one case latter identified with an
incorrect diagnosis date);
population-based controls
identified from the National
Population Registry and matched
(1:4 ratio) to cases for age and
county.
Ill cases.
400 controls.
Cases, 91%; Controls, 83%.
Mailed questionnaire to assess self reported working history, specific exposure,
and leisure time activities.
Univariate analysis for chemical-specific exposures (any TCE exposure).
Fritschi and
Siemiatycki
(1996b);
Siemiatycki,
(1991)
Male NHL cases, age 35-75 yr,
diagnosed in 16 large
Montreal-area hospitals in
1979-1985 and histologically
confirmed; controls identified
concurrently at 18 other cancer
sites; age-matched,
population-based controls
identified from electoral lists and
RDD.
215 cases.
533 population controls
(Group 1) and
1,900 subjects with other
cancers (Group 2).
Cases, 83%; Controls, 71%.
In-person interviews (direct or proxy) with segments on work histories (job
titles and self-reported exposures); analyzed and coded by a team of chemists
and industrial hygienists (294 exposures on semiquantitative scales). Exposure
metric defined as any or substantial exposure.
Logistic regression adjusted for age, proxy status, income, and ethnicity
(solvents) or Mantel-Haenszel stratified by age, body mass index, and cigarette
smoking (TCE).
-------
Table 4-2. Case-control epidemiologic studies examining cancer and TCE exposure (continued)
Reference
Population
Study group (N)
comparison group (N)
response rates
Exposure assessment and other information
Hardell et al.
(1994; 1981)
Histologically-confirmed cases of
NHL in males, age 25-85 yr,
admitted to Swedish (Umea)
hospital between 1974-1978;
living controls (1:2 ratio) from the
National Population Register,
matched to living cases on sex,
age, and place of residence;
deceased controls from the
National Registry for Causes of
Death, matched (1:2 ratio) to dead
cases on sex, age, place of
residence, and yr of death.
105 cases.
335 controls.
Response rate not available.
Self-administered questionnaire assessing self-reported solvent exposure;
phone follow-up with subject, if necessary.
Unadjusted Mantel-Haenszel chi-square.
Persson et al.
(1989; 1993)
Histologically confirmed cases of
Hodgkin's disease, age 20-80 yr,
identified in two hospitals in
Sweden: Oreboro in 1964-1986
(Persson et al., 1989) and in
Linkoping between 1975-1984
(Persson et al., 1993); controls
randomly selected from population
registers.
54 cases (1989 study);
31 cases (1993 study).
275 controls (1989 study);
204 controls (1993 study).
Response rate not available.
Mailed questionnaire to assess self reported occupational exposures to TCE
and other solvents.
Logistic regression with adjustment for age and other exposure; unadjusted
Mantel-Haenszel chi-square.
Childhood Leukemia
Shu et al.
(2004; 1999)
Childhood leukemia cases, <15 yr,
diagnosed between 1989 and 1993
by a Children's Cancer Group
member or affiliated institute;
population controls (random digit
dialing), matched for age, race, and
telephone area code and exchange.
1,842 cases.
1,986 controls.
Cases, 92%; controls, 77%.
Telephone interview with mother, and whenever available, fathers using
questionnaire to assess occupation using job-industry title and self-reported
exposure history. Questionnaire included questions specific for solvent,
degreaser or cleaning agent exposures.
Logistic regression with adjustment for maternal or paternal education, race,
and family income. Analyses of paternal exposure also included age and sex
of the index child.
-------
Table 4-2. Case-control epidemiologic studies examining cancer and TCE exposure (continued)
Reference
Population
Study group (N)
comparison group (N)
response rates
Exposure assessment and other information
Costas et al.
(2002);
MDPH (1997a,
b)
Childhood leukemia (<19 yr age)
diagnosed in 1969-1989 and who
were resident of Woburn, MA;
controls randomly selected from
Woburn public School records,
matched for age.
19 cases.
37 controls.
Cases, 91%; Controls, not
available.
Questionnaire administered to parents separately assessing demographic and
lifestyle characteristics, medical history information, environmental and
occupational exposure and use of public drinking water in the home.
Hydraulic mixing model used to infer delivery of TCE and other solvents
water to residence.
Logistic regression with composite covariate, a weighted variable of individual
covariates.
McKinney
et al. (1991)
Incident childhood leukemia and
non-Hodgkin lymphoma cases,
1974-1988, ages not identified,
from three geographical areas in
England; controls randomly
selected from children of residents
in the three areas and matched for
sex and birth health district.
109 cases.
206 controls.
Cases, 72%; Controls, 77%.
In-person interview with questionnaire with mother to assess maternal
occupational exposure history, and with father and mother, as surrogate, to
assess paternal occupational exposure history. No information provided in
paper whether interviewer was blinded as to case and control status.
Matched pair design using logistic regression for univariate and multivariate
analysis.
Lowengart
et al. (1987)
Childhood leukemia cases aged
<10 yr and identified from the Los
Angeles (CA) Cancer Surveillance
Program in 1980-1984; controls
selected from RDD or from friends
of cases and matched on age, sex,
and race.
123 cases.
123 controls.
Cases, 79%; Controls,
not available.
Telephone interview with questionnaire to assess parental occupational and
self-reported exposure history.
Matched (discordant) pair analysis.
-------
L/l
K>
Table 4-2. Case-control epidemiologic studies examining cancer and TCE exposure (continued)
Reference
Population
Study group (N)
comparison group (N)
response rates
Exposure assessment and other information
Melanoma
Fritschi and
Siemiatycki
(1996b);
Siemiatycki
(1991)
Male melanoma cases, age
35-75 yr, diagnosed in 16 large
Montreal-area hospitals in
1979-1985 and histologically
confirmed; controls identified
concurrently at 18 other cancer
sites; age-matched,
population-based controls
identified from electoral lists and
RDD.
103 cases.
533 population controls and
533 other cancer controls.
Cases, 78%; Controls, 72%.
In-person interviews (direct or proxy) with segments on work histories (job
titles and self-reported exposures); analyzed and coded by a team of chemists
and industrial hygienists (294 exposures on semiquantitative scales); potential
TCE exposure defined as any or substantial exposure.
Logistic regression adjusted for age, education, and ethnic origin (TCE) or
Mantel-Haenszel stratified on age, income, index for cigarette smoking, and
ethnic origin (TCE).
Pancreas
Kernan et al.
(1999)
Pancreatic cancer deaths from
1984-1993 in 24 U.S. states; age-,
sex-, race-, and state-matched
noncancer deaths, excluding other
pancreatic diseases and
pancreatitis, controls.
63,097 cases.
252,386 population controls.
Response rates not
identified.
Exposure surrogate assigned for 111 chlorinated hydrocarbons, including TCE,
and two broad chemical categories using usual occupation on death certificate
and job-exposure-matrix of Gomez et al. (1994).
Race and sex-specific mortality odds ratios from logistic regression analysis
adjusted for age, marital status, metropolitan area, and residential status.
Prostate
Aronson et al.
(1996);
Siemiatycki
(1991)
Male prostate cancer cases, age
35-75 yr, diagnosed in 16 large
Montreal-area hospitals in
1979-1985 and histologically
confirmed; controls identified
concurrently at 18 other cancer
sites; age-matched,
population-based controls
identified from electoral lists and
RDD.
449 cases.
533 population controls
(Group 1) and
other cancer cases from
same study (Group 2).
Cases, 81%; Controls, 72%.
In-person interviews (direct or proxy) with segments on work histories (job
titles and self-reported exposures); analyzed and coded by a team of chemists
and industrial hygienists (294 exposures on semiquantitative scales).
Logistic regression adjusted for age, ethnic origin, socioeconomic status,
Quetlet, and respondent status (occupation) or Mantel-Haenszel stratified on
age, income, index for cigarette smoking, ethnic origin, and respondent status
(TCE).
S?
>!
rs
s;
>1
£
si
1 o
>!
>!
O
a
a
a
§•
s
a
I ^
1 o
a
>1
I
I I
H J
1—1 s
H 2.
W|
O •
p
-------
L/l
K>
Table 4-2. Case-control epidemiologic studies examining cancer and TCE exposure (continued)
Reference
Population
Study group (N)
comparison group (N)
response rates
Exposure assessment and other information
Renal Cell
Moore et al.
(2010)
Cases aged 20-74 yr from
four European countries (Russia,
Romania, Poland, Czech Republic)
with histologically confirmed
kidney cancer in 1999-2003;
hospital controls with diagnoses
unrelated to smoking or
genitourinary disorders in
1998-2003 and frequencymatched
by sex, age and study center.
1,097 cases (825 renal cell
carcinomas).
1,184 controls.
Cases, 90-99%; Controls,
90.3-96%.
In-person interview using questionnaire for information on lifestyle habits,
smoking, antopometric measures, personal and family medical history and
occupational history. Specialized job-specific questionnaire for specific jobs
or industries of interest focused on solvents exposure, including.TCE, with
exposure assignment by expert blinded to case and control status by frequency,
intensity and confidence of TCE exposure. Exposure metric of overall
exposure, duration (total h, yr) and cumulative exposure.
Logistic regression adjusted for sex, age, and study center. BMI, hypertension,
smoking, residence location also included in initial models but did not alter
odds ratios by >10%.
Charbotel et al.
(2006; 2009)
Cases from Arve Valley region in
France identified from local
urologists files and from area
teaching hospitals; age- and
sex-matched controls chosen from
file of same urologist as who
treated case or recruited among the
patients of the case's general
practitioner.
87 cases.
316 controls.
Cases, 74%; controls, 78%.
Telephone interview with case or control, or, if deceased, with next-of-kin
(22% cases, 2% controls). Questionnaire assessing occupational history,
particularly, employment in the screw cutting jobs, and medical history.
Semiquantitative TCE exposure assigned to subjects using a
task/TCE-Exposure Matrix designed using information obtained from
questionnaires and routine atmospheric monitoring of work shops or biological
monitoring (U-TCA) of workers carried out since the 1960s. Cumulative
exposure, cumulative exposure with peaks, and TWA.
Conditional logistic regression with covariates for tobacco smoking and body
mass index.
Briining et al.
(2003)
Histologically-confirmed cases
1992-2000 from German hospitals
(Arnsberg); hospital controls
(urology department) serving area,
and local geriatric department, for
older controls, matched by sex and
age.
134 cases.
401 controls.
Cases, 83%; Controls, not
available.
In-person interviews with case or next-of-kin; questionnaire assessing
occupational history using job title. Exposure metrics included longestjob
held, JEM of Pannett et al. (1985) to assign cumulative exposure to TCE and
PCE, and exposure duration..
Logistic regression with covariates for age, sex, and smoking.
S?
>!
rs
s;
>1
£
si
1 o
>!
>!
O
a
a
a
§•
s
a
I ^
1 o
a
>1
I
I I
H J
1—1 s
H 2.
W|
O •
o
c
o
H
W
-------
Table 4-2. Case-control epidemiologic studies examining cancer and TCE exposure (continued)
Reference
Population
Study group (N)
comparison group (N)
response rates
Exposure assessment and other information
Pesch et al.
(2000b)
Histologically-confirmed cases
from German hospitals (5 regions)
in 1991-1995; controls randomly
selected from residency registries
matched on region, sex, and age.
935 cases.
4,298 controls.
Cases, 88%; Controls, 71%.
In-person interview with case or next-of-kin; questionnaire assessing
occupational history using job title (JEM approach), self-reported exposure, or
job task (JTEM approach) to assign TCE and other exposures.
Logistic regression with covariates for age, study center, and smoking.
Parent et al.
(2000a);
Siemiatycki
(1991)
Male renal cell carcinoma cases,
age 35-75 yr, diagnosed in
16 large Montreal-area hospitals in
1979-1985 and histologically
confirmed; controls identified
concurrently at 18 other cancer
sites; age-matched,
population-based controls
identified from electoral lists and
RDD.
142 cases.
533 population controls
(Group 1) and
other cancer controls
(excluding lung and bladder
cancers) (Group 2).
Cases, 82%; Controls, 71%.
In-person interviews (direct or proxy) with segments on work histories (job
titles and self-reported exposures); analyzed and coded by a team of chemists
and industrial hygienists (about 300 exposures on semiquantitative scales);
TCE defined as any or substantial exposure.
Mantel-Haenszel stratified by age, body mass index, and cigarette smoking
(TCE) or logistic regression adjusted for respondent status, age, smoking, and
body mass index (occupation, job title).
Dosemeci et al.
(1999)
Histologically-confirmed cases,
1988-1990, white males and
females, 20-85 yr, from
Minnesota Cancer Registry;
controls stratified for age and sex
using RDD, 21-64 yr, or from
HCFA records, 64-85 yr.
438 cases.
687 controls.
Cases, 87%; Controls, 86%.
In-person interviews with case or next-of-kin; questionnaire assessing
occupational history of TCE usins iob title and JEM of Gomez et al. (1994).
Exposure metric was any TCE exposure.
Logistic regression with covariates for age, smoking, hypertension, and body
mass index.
Vamvakas
et al. (1998)
Cases who underwent
nephrectomy in 1987-1992 in a
hospital in Arnsberg region of
Germany; controls selected
accident wards from nearby
hospital in 1992.
58 cases.
84 controls.
Cases, 83%; Controls, 75%.
In-person interview with case or next-of-kin; questionnaire assessing
occupational history using job title or self-reported exposure to assign TCE and
PCE exposure.
Logistic regression with covariates for age, smoking, body mass index,
hypertension, and diuretic intake.
-------
Table 4-2. Case-control epidemiologic studies examining cancer and TCE exposure (continued)
Reference
Population
Study group (N)
comparison group (N)
response rates
Exposure assessment and other information
Multiple or Other Sites
Lee et al.,
(2003)
Liver, lung, stomach, colorectal
cancer deaths in males and females
between 1966-1997 from
two villages in Taiwan; controls
were cardiovascular and
cerebral-vascular disease deaths
from same underlying area as
cases.
53 liver,
39 stomach,
26 colorectal,
41 lung cancer cases.
286 controls.
Response rate not reported.
Residence as recorded on death certificate.
Mantel-Haenszel stratified by age, sex, and time period.
Siemiatycki
(1991)
Male cancer cases, 1979-1985,
35-75 yr, diagnosed in
16 Montreal-area hospitals,
histologically confirmed; cancer
controls identified concurrently;
age-matched, population-based
controls identified from electoral
lists and RDD.
857 lung and 117 pancreatic
cancer cases.
533 population controls
(Group 1) and other cancer
cases from same study
(Group 2).
Cases, 79% (lung), 71%
(pancreas); Controls, 72%.
In-person interviews (direct or proxy) with segments on work histories (job
titles and self-reported exposures); analyzed and coded by a team of chemists
and industrial hygienists (294 exposures on semiquantitative scales); TCE
defined as any or substantial exposure.
Mantel-Haenszel stratified on age, income, index for cigarette smoking, ethnic
origin, and respondent status (lung cancer) and age, income, index for cigarette
smoking, and respondent status (pancreatic cancer).
Bolded study(ies) carried forward for consideration in dose-response assessment (see Section 5).
BMI = body mass index, CLL = chronic lymphocytic leukemia, HCFA = Health Care Financing Administration, JEM = job-exposure matrix, JTEM = job-task-
exposure matrix, MM = multiple myeloma, NCI = National Cancer Institute, NHL = non-Hodgkin lymphoma, PCE = perchloroethylene, RDD = random digit
dialing, TWA = time-weighted average, U-TCA = urinary trichloroacetic acid, UV = ultra-violet.
-------
Table 4-3. Geographic-based studies assessing cancer and TCE exposure
Reference
Description
Analysis approach
Exposure assessment
Broome County, NY Studies
ATSDR
(2006a)
(2008)
Total, 22 site-specific, and
childhood cancer incidence from
1980-2001 among residents in
two areas in Endicott, NY.
SIR among all subjects (ATSDR, 2006a) or
among white subjects only (ATSDR, 2008) with
expected numbers of cancers derived using
age-specific cancer incidence rates for New York
State, excluding New York City. Limited
assessment of smoking and occupation using
medical and other records in lung and kidney
cancer subjects (ATSDR, 2008).
Two study areas, Eastern and Western study areas,
identified based on potential for soil vapor intrusion
exposures as defined by the extent of likely soil vapor
contamination. Contour lines of modeled VOC soil
vapor contamination levels based on exposure model
using GIS mapping and soil vapor sampling results
taken in 2003. The study areas were defined by 2000
Census block boundaries to conform to model
predicted areas of soil vapor contamination. TCE was
the most commonly found contaminant in indoor air in
Eastern study area at levels ranging from
0.18-140 |ig/m3, with tetrachloroethylene,
cis-1,2-dichloroethene, 1,1,1 -trichloroethane,
1,1 -dichloroethylene, 1,1 -dichloroethane, and Freon
113 detected at lower levels. PCE was most common
contaminant in indoor air in Western study area with
other VOCs detected at lower levels.
Maricopa County, AZ Studies
Aickin et al.
(1992);
Aickin (2004)
Cancer deaths, including leukemia,
1966-1986, and childhood (<19 yr
old) leukemia incident cases
(1965-1986), Maricopa County,
AZ.
Standardized mortality rate ratio from Poisson
regression modeling. Childhood leukemia
incidence data evaluated using Bayes methods
and Poisson regression modeling.
Location of residency in Maricopa County, AZ, at the
time of death as surrogate for exposure. Some analyses
examined residency in West Central Phoenix and
cancer. Exposure information is limited to TCE
concentration in two drinking water wells in 1982.
Pima County, AZ Studies
ADHS (1990,
1995)
Cancer incidence in children
(<19 yr old) and testicular cancer in
1970-1986 and 1987-1991, Pima
County, AZ.
Standardized incidence RR from Poisson
regression modeling using method of Aickin et al.
(Aickin et al., 1992). Analysis compares
incidence in Tucson Airport Area to rate for rest
of Pima County.
Location of residency in Pima, County, AZ, at the time
of diagnosis or death as surrogate for exposure.
Exposure information is limited to monitoring since
1981 and include VOCs in soil gas samples (TCE,
PCE, 1,1-dichloroethylene, 1,1,1-trichloroacetic acid);
PCBs in soil samples, and TCE in municipal water
supply wells.
-------
Table 4-3. Geographic-based studies assessing cancer and TCE exposure (continued)
Reference
Description
Analysis approach
Exposure assessment
Other
Coyle et al.
(2005)
Incident breast cancer cases among
men and women, 1995-2000,
reported to Texas Cancer Registry.
Correlation study using rank order statistics of
mean average annual breast cancer rate among
women and men and atmospheric release of
12 hazardous air pollutants.
Reporting to EPA Toxic Release Inventory the number
of pounds released for 12 hazardous air pollutants,
(carbon tetrachloride, formaldehyde, methylene
chloride, styrene, tetrachloroethylene,
trichloroethylene, arsenic, cadmium, chromium, cobalt,
copper, and nickel).
Morgan and
Cassady
(2002)
Incident cancer cases, 1988-1989,
among residents of 13 census tracts
in Redlands area, San Bernardino
County, CA.
SIR for all cancer sites and 16 site-specific
cancers; expected numbers using incidence rates
of site-specific cancer of a four-county region
between 1988-1992.
TCE and perchlorate detected in some county wells; no
information on location of wells to residents,
distribution of contaminated water, or TCE exposure
potential to individual residents in studied census
tracts.
Vartiainen
et al. (1993)
Total cancer and site-specific cancer
cases (lymphoma sites and liver)
from 1953-1991 in two Finnish
municipalities.
SIR with expected number of cancers and
site-specific cancers derived from incidence of
the Finnish population.
Monitoring data from 1992 indicated presence of TCE,
tetrachloroethylene and 1,1,1,-triehloroethane in
drinking water supplies in largest towns in
municipalities. Residence in town used to infer
exposure to TCE.
Cohn et al.
(1994b);
Fagliano et al.
(1990)
Incident leukemia and NHL cases,
1979-1987 ,from 75 municipalities
and identified from the New Jersey
State Cancer Registry. Histological
type classified using WHO scheme
and the classification of NIH
Working Formulation Group for
grading NHL.
Logistic regression modeling adjusted for age.
Monitoring data from 1984-1985 on TCE, THM, and
VOCs concentrations in public water supplies, and
historical monitoring data conducted in 1978-1984.
Mallin (1990)
Incident bladder cancer cases and
deaths, 1978-1985, among residents
of nine NW Illinois counties.
SIR and SMR by county of residence and zip
code; expected numbers of bladder cancers using
age-race-sex specific incidence rates from SEER
or bladder cancer mortality rates of the U.S.
population from 1978-1985.
Exposure data are lacking for the study population with
the exception of noting one of two zip code areas with
observed elevated bladder cancer rates also had
groundwater supplies contaminated with TCE, PCE
and other solvents.
-------
Table 4-3. Geographic-based studies assessing cancer and TCE exposure (continued)
Reference
Description
Analysis approach
Exposure assessment
Isacson et al.
(1985)
Incident bladder, breast, prostate,
colon, lung and rectal cancer cases
reported to Iowa cancer registry
between 1969-1981.
Age-adjusted site-specific cancer incidence in
Iowa towns with populations of 1,000-10,000
and who were serviced by a public drinking water
supply.
Monitoring data of drinking water at treatment plant in
each Iowa municipality with populations of
1,000-10,000 used to infer TCE and other volatile
organic compound concentrations in finished drinking
water supplies.
EPA = U.S. Environmental Protection Agency, GIS = geographic information system, NHL = non-Hodgkin lymphoma, NIH = National Institutes of Health,
NW = Northwestern, PCB = polychlorinated biphenyl, PCE = perchloroethylene, -, SEER = Surveillance, Epidemiology, and End Results, SIR = standardized
incidence ratio, SMR = standardized mortality ratio, VOCs = volatile organic compounds, WHO = World Health Organization.
-------
Table 4-4. Standards of epidemiologic study design and analysis use for identifying cancer hazard and TCE
exposure
Category A: Study Design
Clear articulation of study objectives or hypothesis. The ideal is a clearly stated hypothesis or study objectives and the study is designed to achieve the
identified objectives.
Selection and characterization in cohort studies of exposure and control groups and of cases and controls (case-control studies) is adequate. The ideal is
for selection of cohort and referents from the same underlying population and differences between these groups to be due to TCE exposure or level of TCE
exposure and not to physiological, health status, or lifestyle factors. Controls or referents are assumed to lack or to have background exposure to TCE. These
factors may lead to a downward bias including one of which is known as "healthy worker bias," often introduced in analyses when mortality or incidence rates
from a large population such as the U.S. population are used to derive expected numbers of events. The ideal in case-control studies is cases and controls are
derived from the same population and are representative of all cases and controls in that population. Any differences between controls and cases are due to
exposure to TCE itself and not to confounding factors related to both TCE exposure and disease. Additionally, the ideal is for controls to be free of any disease
related to TCE exposure. In this latter case, potential bias is toward the null hypothesis.
Category B: Endpoint Measured
Levels of health outcome assessed. Three levels of health outcomes are considered in assessing the human health risks associated with exposure to TCE:
biomarkers of effects and susceptibility, morbidity, and mortality. Both morbidity as enumerated by incidence and mortality as identified from death certificates
are useful indicators in risk assessment for hazard identification. The ideal is for accurate and predictive indicator of disease. Incidence rates are generally
considered to provide an accurate indication of disease in a population and cancer incidence is generally enumerated with a high degree of accuracy in cancer
registries. Death certifications are readily available and have complete national coverage but diagnostic accuracy is reduced and can vary by specific diagnosis.
Furthermore, diagnostic inaccuracies can contribute to death certificates as a poor surrogate for disease incidence. Incidence, when obtained from
population-based cancer registries, is preferred for identifying cancer hazards.
Changes in diagnostic coding systems for lymphoma, particularly non-Hodgkin lymphoma. Classification of lymphomas today is based on morphologic,
immunophenotypic, genotypic, and clinical features using the World Health Organization (WHO) classification, introduced in 2001, and incorporation of WHO
terminology into International Classification of Disease (ICD)-0-3. ICD Versions 7 and earlier had rubrics for general types of lymphatic and hematopoietic
cancer, but no categories for distinguishing specific types of cancers, such as acute leukemia. Epidemiologic studies based on causes of deaths as coded using
these older ICD classifications typically grouped together lymphatic neoplasms instead of examining individual types of cancer or specific cell types. Before the
use of immunophenotyping, these grouping of ambiguous diseases such as non-Hodgkin lymphoma and Hodgkin's lymphoma may be have misclassified. With
the introduction of ICD-10 in 1990, lymphatic tumors coding, starting in 1994 with the introduction of the Revised European-American Lymphoma
classification, the basis of the current WHO classification, was more similar to that presently used. Misclassification of specific types of cancer, if unrelated to
exposure, would have attenuated estimate of relative risk and reduced statistical power to detect associations. When the outcome was mortality, rather than
incidence, misclassification would be greater because of the errors in the coding of underlying causes of death on death certificates (IOM, 2003). Older studies
that combined all lymphatic and hematopoietic neoplasms must be interpreted with care.
-------
Table 4-4. Standards of epidemiologic study design and analysis use for identifying cancer hazard and TCE
exposure (continued)
Category C: TCE-Exposure Criteria
Adequate characterization of exposure. The ideal is for TCE exposure potential known for each subject and quantitative assessment [job-exposure-matrix
approach] of TCE exposure assessment for each subject as a function of job title, yr exposed, duration, and intensity. The assessment approach is accurate for
assigning TCE intensity [TCE concentration or a time-weighted-average] to individual study subjects and estimates of TCE intensity are validated using
monitoring data from the time period. For the purpose of this report, the objective for cohort and case-controls studies is to differentiate TCE-exposed subjects
from subjects with little or no TCE exposure. A variety of dose-metrics may be used to quantify or classify exposures for an epidemiologic study. They include
precise summaries of quantitative exposure, concentrations of biomarkers, cumulative exposure, and simple qualitative assessments of whether exposure
occurred (yes or no). Each method has implicit assumptions and potential problems that may lead to misclassification. Studies in which it was unclear that the
study population was actually exposed to TCE are excluded from analysis.
Category D: Follow-up (Cohort)
Loss to follow-up. The ideal is complete follow-up of all subjects; however, this is not achievable in practice, but it seems reasonable to expect loss to follow-up
not to exceed 10%. The bias from loss to follow-up is indeterminate. Random loss may have less effect than if subjects who are not followed have some
significant characteristics in common.
Follow-up period allows full latency period for over 50% of the cohort. The ideal to follow all study subjects until death. Short of the ideal, a sufficient
follow-up period to allow for cancer induction period or latency over 15 or 20 yr is desired for a large percentage of cohort subjects.
Category E: Interview Type (Case-control)
Interview approach. The ideal interviewing technique is face-to-face by trained interviewers with more than 90% of interviews with cases and control subjects
conduced face-to-face. The effect on the quality of information from other types of data collection is unclear, but telephone interviews and mail-in
questionnaires probably increase the rate of misclassification of subject information. The bias is toward the null hypothesis if the proportion of interview by type
is the same for case and control, and of indeterminate direction otherwise.
Blinded interviewer. The ideal is for the interviewer to be unaware whether the subject is among the cases or controls and the subject to be unaware of the
purpose and intended use of the information collected. Although desirable for case-control studies, blinding is usually not possible to fully accomplish because
subject responses during the interview provide clues as to subject status. The potential for bias from face-to-face interviews is probably less than with mail-in
interviews. Some studies have assigned exposure status in a blinded manner using a job-exposure matrix and information collected in the unblinded interview.
The potential for bias in this situation is probably less with this approach than for nonblinded assignment of exposure status.
-------
Table 4-4. Standards of epidemiologic study design and analysis use for identifying cancer hazard and TCE
exposure (continued)
Category F: Proxy Respondents
Proxy respondents. The ideal is for data to be supplied by the subject because the subject generally would be expected to be the most reliable source; less than
10% of either total cases or total controls for case-control studies. A subject may be either deceased or too ill to participate, however, making the use of proxy
responses unavoidable if those subjects are to be included in the study. The direction and magnitude of bias from use of proxies is unclear, and may be
inconsistent across studies.
Category G: Sample Size
The ideal is for the sample size is large enough to provide sufficient statistical power to ensure that any elevation of effect in the exposure group, if present,
would be found, and to ensure that the confidence bounds placed on relative risk estimates can be well characterized.
Category H: Analysis Issues
Control for potentially confounding factors of importance in analysis. The ideal in cohort studies is to derive expected numbers of cases based on age-sex-
and time-specific cancer rates in the referent population and in case-control studies by matching on age and sex in the design and then adjusting for age in the
analysis of data. Age and sex are likely correlated with exposure and are also risk factors for cancer development. Similarly, other factors such as cigarette
smoking and alcohol consumption are risk factors for several site-specific cancers reported as associative with TCE exposure. To be a confounder of TCE,
exposure to the other factor must be correlated, and the association of the factor with the site-specific cancer must be causal. The expect effect from controlling
for confounders is to move the estimated relative risk estimate closer to the true value.
Statistical methods are appropriate. The ideal is that conclusions are drawn from the application of statistical methods that are appropriate to the problem and
accurately interpreted.
Evaluation of exposure-response. The ideal is an examination of a linear exposure-response as assessed with a quantitative exposure metric such as cumulative
exposure. Some studies, absent quantitative exposure metrics, examine exposure response relationships using a semiquantitative exposure metric or by duration
of exposure. A positive dose-response relationship is usually more convincing of an association as causal than a simple excess of disease using TCE
dose-metric. However, a number of reasons have been identified for a lack of linear exposure-response finding and the failure to find such a relationship mean
little from an etiological viewpoint.
Documentation of results. The ideal is for analysis observations to be completely and clearly documented and discussed in the published paper, or provided in
supplementary materials accompanying publication.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Twenty-six of the studies identified in a systematic review were selected for inclusion in
the meta-analysis through use of the following meta-analysis inclusion criteria: (1) cohort or
case-control designs; (2) evaluation of incidence or mortality; (3) adequate selection in cohort
studies of exposure and control groups and of cases and controls in case-control studies; (4) TCE
exposure potential inferred to each subject and quantitative assessment of TCE exposure
assessment for each subject by reference to industrial hygiene records indicating a high
probability of TCE use, individual biomarkersjob exposure matrices, water distribution models,
or obtained from subjects using questionnaire (case-control studies); and (5) relative risk
estimates for kidney cancer, liver cancer, or non-Hodgkin lymphoma (NHL) adjusted, at
minimum, for possible confounding of age, sex, and race (see Table 4-5). This evaluation is
summarized below, separately for cohort and case-control studies. Appendix C contains a full
discussion of the meta-analysis, its analytical methodology, including sensitivity analyses, and
findings. The meta-analysis focuses on kidney cancer, liver cancer, and NHL, as most studies
reported relative risks for these sites. Fewer numbers of studies reported relative risks for other
site-specific cancers and TCE exposure and examination of these site-specific cancers and TCE
exposure using meta-analysis was not attempted.
The cohort studies (Anttila et al., 1995; ATSDR, 2004; Axelson et al., 1994; Blair et al.,
1989; Blair et al., 1998; Boice et al., 1999; Boice et al., 2006b; Chang et al., 2003; Chang et al.,
2005; Clapp and Hoffman, 2008; Costa et al., 1989; Garabrant et al., 1988; Greenland et al.,
1994; Hansen et al., 2001; Henschler et al., 1995; Krishnadasan et al., 2007; Morgan et al., 1998;
Raaschou-Nielsen et al., 2003; Radican et al., 2008; Ritz, 1999a; Shannon et al., 1988; Shindell
and Ulrich, 1985; Sinks et al., 1992; Sung et al., 2007; Sung et al., 2008; Wilcosky et al., 1984;
Zhao et al., 2005) (see Table 4-1), with data on the incidence or morality of site-specific cancer
in relation to TCE exposure, range in size 803 (Hansen et al., 2001) to 86,868 (Chang et al.,
2003; Chang et al., 2005), and were conducted in Denmark, Sweden, Finland, Germany, Taiwan,
and the United States (see Table 4-1). Three case-control studies nested within cohorts
(Greenland et al., 1994; Krishnadasan et al., 2007; Wilcosky et al., 1984) are considered as
cohort studies because the summary risk estimate from a nested case-control study, the odds
ratio, was estimated from incidence density sampling. This is considered an unbiased estimate of
the hazard ratio, similar to a relative risk estimate from a cohort study, if, as is the case for these
studies, controls are selected from the same source population as the cases, the sampling rate is
independent of exposure status, and the selection probability is proportional to time-at-risk
(Rothman et al., 2008). Cohort and nested case-control study designs are analytical
epidemiologic studies and are generally relied on for identifying a causal association between
human exposure and adverse health effects (U.S. EPA, 2005c).
This document is a draft for review purposes only and does not constitute Agency policy.
28 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-5. Summary of criteria for meta-analysis study selection
2
Decision
outcome
Studies
Primary reason(s)
Studies recommended for meta-analysis:
Axelson et al. (1994); Greenland et al. (1994);
Hardell etal. (1994); Siemiatycki (1991);
Anttila et al. (1995); Morgan et al. (1998);
Nordstrom et al. (1998); Boice et al. (1999);
Boice et al. (2006b); Dosemeci et al. (1999);
Persson and Fredriksson (1999); Pesch et al.
(2000b); Hansen et al. (2001); Briining et al.
(2003); Raaschou-Nielsen et al. (2003); Zhao
et al. (2005); Miligi et al. (2006); Seidler et al.
(2007); Charbotel et al. (2006; 2009); Blair et
al. (1998); its follow-up Radican et al. (2008);
Wang et al. (2009); Cocco et al. (2010); Moore
et al. (2010); Purdue et al. (2011)
Analytical study designs of cohort or case-control;
evaluation of incidence or mortality;
adequate selection in cohort studies of exposure and
control groups and of cases and controls in
case-control studies;
TCE exposure potential inferred to each subject and
quantitative assessment of TCE exposure assessment
for each subject by reference to industrial hygiene
records indicating a high probability of TCE use,
individual biomarkers, job exposure matrices, water
distribution models, or obtained from subjects using
questionnaire (case-control studies);
relative risk estimates for kidney cancer, liver cancer,
or NHL adjusted, at minimum, for possible
confounding of relevant risk factors, e.g., age, sex, and
race).
Studies not recommended for meta-analysis:
Clapp and Hoffman (2008); ATSDR (2004;
Co hn etal., 1994b)
Weakness with respect to analytical study design (i.e.,
geographic-based, ecological or proportional mortality
ratio design).
Garabrant et al.(1988); Isacson et al.(1985);
Shindell and Ulrich (1985); Wilcosky et al.
(1984); Shannon et al. (1988); Blair et al.
(1989); Costa etal. (1989); (ADHS, 1990,
1995); Mallin (1990); Aickin et al. (1992);
Sinks et al. (1992); Vartiainen et al. (1993);
Morgan and Cassady (2002); Lee et al. (2003);
Aickin (2004); Chang et al. (Chang et al., 2003;
2005); Coyle et al. (2005); ATSDR (2006a);
ATSDR (2008); Sung et al. (2007; 2008)
TCE exposure potential not assigned to individual
subjects using job exposure matrix, individual
biomarkers, water distribution models, or industrial
hygiene data from other process indicating a high
probability of TCE use (cohort studies).
Lowengart et al. (1987); Fredriksson et al.
(1989); McKinney et al. (1991); Heineman
et al. (1994); Siemiatycki et al. (1994);
Aronson et al. (1996); Fritschi and Siemiatycki
(1996b); Dumas et al. (2000); Kernan et al.
(1999); Shu et al. (2004; 1999); Parent et al.
(2000b); Pesch et al. (2000a); DeRoos et al.
(2001a); Goldberg et al. (2001); Costas et al.
(2002); Krishnadasan et al. (2007) Costantini
et al. (2008); Gold et al.
Cancer incidence or mortality reported for cancers
other than kidney, liver, or NHL.
Ritz (1999a)
Subjects monitored for radiation exposure with
likelihood for potential confounding;
cancer mortality and TCE exposure not reported for
kidney cancer and all hemato- and lymphopoietic
cancer reported as broad category.
This document is a draft for review purposes only and does not constitute Agency policy.
29 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Table 4-5. Summary of criteria for meta-analysis study selection (continued)
Decision
outcome
Studies
Primary reason(s)
Studies not recommended for meta-analysis: (continued)
Henschler et al. (1995)
Incomplete identification of cohort and index kidney
cancer cases included in case series.
NHL = non-Hodgkin lymphoma.
While all of these cohort studies are considered in the overall weight of evidence, eleven
of them met all meta-analysis inclusion criteria: the cohorts of Blair et al. (1998) and its
follow-up by Radican et al. (2008); Morgan et al. (1998), Boice et al. (1999; Boice et al., 2006b),
and Zhao et al.(2005), of aerospace workers or aircraft mechanics; and Axelson et al. (1994),
Anttila et al. (1995), Hansen et al. (2001), and Raaschou-Nielsen et al. (2003) of Nordic workers
in multiple industries with TCE exposure; and Greenland et al. (1994) of electrical
manufacturing workers. Subjects or cases and controls in these studies are considered to
sufficiently represent the underlying population, and the bias associated with selection of referent
populations is considered minimal. The exposure-assessment approaches included detailed
job-exposure matrix, biomonitroing data, or use of industrial hygiene data on TCE exposure
pattens and factors that affect such exposure, with high probability of TCE exposure potential to
individual subjects. The statistical analyses methods were appropriate and well documented, the
measured endpoint was an accurate indicator of disease, and the follow-up was sufficient for
cancer latency. These studies are also considered as strong studies for identifying kidney, liver
and NHL cancer hazard. The remaining cohort studies less satisfactorily meet identified criteria
or standards of epidemiologic design and analysis, having deficiencies in multiple criteria
(ATSDR, 2004; Chang et al., 2003; Chang et al., 2005; Clapp and Hoffman, 2008; Costa et al.,
1989; Garabrant et al., 1988; Henschler et al., 1995; Ritz, 1999a; Shindell and Ulrich, 1985;
Sinks et al., 1992; Sung et al., 2007; Sung et al., 2008; Wilcosky et al., 1984). Krishnandansen
et al. (2007), who reported on prostate cancer, met four of the five meta-analysis inclusion
criteria except that for reporting a relative risk estimate cancer of the kidney, liver or NHL, the
site-specific cancers examined using meta-analysis.
The case-control studies on TCE exposure are of several site-specific cancers, including
bladder (Pesch et al., 2000a; Siemiatycki, 1991; Siemiatycki et al., 1994); brain (De Roos et al.,
2001b; Heineman et al., 1994); childhood lymphoma or leukemia (Costas et al., 2002;
Lowengart et al., 1987; McKinney et al., 1991; Shu et al., 2004; Shu et al., 1999); colon cancer
(Goldberg et al., 2001; Siemiatycki, 1991); esophageal cancer (Parent et al., 2000b; Siemiatycki,
This document is a draft for review purposes only and does not constitute Agency policy.
3 0 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
1991); liver cancer (Lee et al., 2003); lung (Siemiatycki, 1991); adult lymphoma or leukemia
(Hardell et al., 1994) [non-Hodgkin lymphoma (NHL), Hodgkin lymphoma]; (Fritschi and
Siemiatycki, 1996a; Siemiatycki, 1991) [NHL]; (Nordstrom etal., 1998) [hairy cell leukemia];
(Persson and Fredrikson, 1999) [NHL]; (Miligi et al., 2006) [NHL and chronic lymphocytic
leukemia (CLL)]; (Seidler et al., 2007) [NHL, Hodgkin lymphoma and subjects included in
(Cocco et al., 2010; Costantini et al., 2008) [leukemia types, CLL included with NHL in (Miligi
et al., 2006; Wang et al., 2009) [NHL]; (Cocco et al., 2010) [B-cell including CLL and multiple
myeloma, T-cell and Hodgkin lymphomas]; (Purdue et al., 2011) [NHL]; Gold et al. [multiple
myeloma]); melanoma (Fritschi and Siemiatycki, 1996b; Siemiatycki, 1991); rectal cancer
(Dumas et al., 2000; Siemiatycki, 1991); renal cell carcinoma, a form of kidney cancer (Briining
et al., 2003; Charbotel et al., 2006; Dosemeci et al., 1999; Moore et al., 2010; Parent et al.,
2000a; Pesch et al., 2000b; Siemiatycki, 1991; Vamvakas et al., 1998); pancreatic cancer
(Siemiatycki, 1991), and prostate cancer (Aronson et al., 1996; Siemiatycki, 1991) (see
Table 4-2). No case-control studies of reproductive cancers (breast or cervix) and TCE exposure
were found in the peer-reviewed literature.
While all of these case-control studies are considered in the overall weight of evidence,
fifteen of them met the meta-analysis inclusion criteria identified in Section B.2.9 (Briining et al.,
2003; Charbotel et al., 2006; Charbotel et al., 2009; Cocco et al., 2010; Dosemeci et al., 1999;
Hardell et al., 1994; Miligi et al., 2006; Moore et al., 2010; Nordstrom et al., 1998; Persson and
Fredrikson, 1999; Pesch et al., 2000b; Purdue et al., 2011; Seidler et al., 2007; Siemiatycki,
1991; Wang et al., 2009). They were of analytical study design, cases and controls were
considered to represent underlying populations and selected with minimal potential for bias;
exposure assessment approaches included assignment of TCE exposure potential to individual
subjects using information obtained from face-to-face, mailed, or telephone interviews; analyses
methods were appropriate, well-documented, included adjustment for potential confounding
exposures, with relative risk estimates and associated confidence intervals reported for kidney
cancer, liver cancer or NHL.
These studies were also considered, to varying degrees, as strong studies for weight-of
evidence characterization of hazard. Both Briining et al. (2003) and Charbotel et al. (2006;
2009) had a priori hypotheses for examining renal cell carcinoma and TCE exposure. Strengths
of both studies are in their examination of populations with potential for high exposure intensity
and in areas with high frequency of TCE usage and their assessment of TCE potential. An
important feature of the exposure assessment approach of Charbotel et al. (2006) is their use of a
large number of studies on biological monitoring of workers in the screw-cutting industry a
predominant industry with documented TCE exposures as support. Other studies were either
This document is a draft for review purposes only and does not constitute Agency policy.
31 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
large multiple-center studies (Cocco et al., 2010; Miligi et al., 2006; Moore et al., 2010; Pesch et
al., 2000a; Pesch et al., 2000b; Purdue et al., 2011; Wang et al., 2009) or reporting from one
location of a larger international study (Dosemeci et al., 1999; Seidler et al., 2007). Cocco et al.
(2010) includes subjects in Seidler et al. (2007) and is preferred because of the larger number of
subjects from four other European countries. In contrast to Briining et al. (2003) and Charbotel
et al. (2006; 2009), two studies conducted in geographical areas with widespread TCE usage and
potential for exposure to higher intensity, in these other studies, a lower exposure prevalence to
TCE is found (any TCE exposure: 15% of cases (Dosemeci et al., 1999); 6% of cases (Miligi et
al., 2006); 13% of cases (Wang et al., 2009); 4% of cases (Cocco et al., 2010) and most subjects
were identified as exposed to TCE probably had minimal contact (3% of cases with
moderate/high TCE exposure (Miligi et al., 2006); 2% of cases with high intensity, but of low
probability TCE exposure (Wang et al., 2009). This pattern of lower exposure prevalence and
intensity is common to community-based population case-control studies (Teschke et al., 2002).
Fourteen case-control studies did not meet specific meta-analysis inclusion criterion
(Aronson et al., 1996; Costas et al., 2002; Dumas et al., 2000; Fritschi and Siemiatycki, 1996b;
Gold et al., In Press; Goldberg et al., 2001; Kernan et al., 1999; Lee et al., 2003; Parent et al.,
2000b; Pesch et al., 2000a; Shu et al., 2004; Shu et al., 1999; Siemiatycki et al., 1994; Vamvakas
et al., 1998). Twelve studies reported relative risk estimates for site-specific cancers other than
kidney, liver, and NHL (Aronson et al., 1996; Costas et al., 2002; Dumas et al., 2000; Fritschi
and Siemiatycki, 1996b; Gold et al., In Press; Goldberg et al., 2001; Kernan et al., 1999; Parent
et al., 2000b; Pesch et al., 2000a; Shu et al., 2004; Shu et al., 1999; Siemiatycki et al., 1994).
Vamvakas et al. (1998) has been subject of considerable controversy (Bloemen and Tomenson,
1995; Cherrie et al., 2001; Green and Lash, 1999; Mandel, 2001; McLaughlin and Blot, 1997;
Swaen, 1995) with questions raised on potential for selection bias related to the study's controls.
This study was deficient in the criterion for adequacy of case and control selection. Briining
et al. (2003), a study from the same region as Vamvakas et al. (1998), is considered a stronger
study for identifying cancer hazard since it addresses many of the deficiencies of Vamvakas et al.
(1998). Lee et al. (2003) in their study of hepatocellular cancer assigns one level of exposure to
all subjects in a geographic area, and inherent measurement error and misclassification bias
because not all subjects are exposed uniformly. Additionally, statistical analyses in this study
did not control for hepatitis viral infection, a known risk factor for hepatocellular cancer and of
high prevalence in the study area.
The geographic-based studies (ADHS, 1990, 1995; Aickin, 2004; Aickin et al., 1992;
ATSDR, 2006a, 2008; Cohn et al., 1994b; Isacson et al., 1985; Mallin, 1990; Morgan and
Cassady, 2002; Vartiainen et al., 1993) with data on cancer incidence are correlation studies to
This document is a draft for review purposes only and does not constitute Agency policy.
32 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
examine cancer outcomes of residents in communities with TCE and other chemicals detected in
groundwater wells or in municipal drinking water supplies (see Table 4-3). These studies did not
meet all five meta-analysis inclusion criteria. The geographic-base studies are not of analytical
designs such as cohort and case-control designs. Another deficiency in all studies is their low
level of detail to individual subjects for TCE. One level of exposure to all subjects in a
geographic area is assigned without consideration of water distribution networks, which may
influence TCE concentrations delivered to a home, or a subject's ingestion rate to estimate TCE
exposure to individual study subjects. Some inherent measurement error and misclassification
bias is likely in these studies because not all subjects are exposed uniformly. Additionally, in
contrast to case-control studies, the geographic-based studies, including the Agency for Toxic
Substances and Disease Registry (ATSDR, 2008), had limited accounting for other potential risk
factors. These studies are of low sensitivity for weight-of evidence characterization of hazard
compared to other cohort and case-control studies.
4.2. GENETIC TOXICITY
This section discusses the genotoxic potential of TCE and its metabolites. A summary is
provided at the end of each section for TCE or its metabolite for their mutagenic potential in
addition to an overall synthesis summary at the end of the genotoxicity section. The liver and
kidney are subjects of study for the genotoxic potential of TCE and its metabolites, and are
discussed more in-depth in sections 4.4.3, 4.4.7, 4.5.6.2.7, 4.5.7, E.2.3, and E.2.4.
The application of genotoxicity data to predict potential carcinogenicity is based on the
principle that genetic alterations are found in all cancers. Genotoxicity is the ability of chemicals
to alter the genetic material in a manner that permits changes to be transmitted during cell
division. Although most tests for mutagenicity detect changes in DNA or chromosomes, some
specific modifications of the epigenome including proteins associated with DNA or RNA, can
also cause transmissible changes. Changes that occur due to the modifications in the epigenome
are discussed in endpoint-specific Sections 4.3-4.9 as well as Sections E.3.1-E.3.4.
Genetic alterations can occur through a variety of mechanisms including gene mutations,
insertions, deletions, translocations, or amplification; evidence of mutagenesis provides
mechanistic support for the inference of potential for carcinogenicity in humans.
Evaluation of genotoxicity data entails a weight-of-evidence approach that includes
consideration of the various types of genetic damage that can occur. In acknowledging that
genotoxicity tests are by design complementary evaluations of different mechanisms of
genotoxicity, a recent International Programme on Chemical Safety (IPCS) publication
This document is a draft for review purposes only and does not constitute Agency policy.
3 3 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
(Eastmond et al., 2009) notes that "multiple negative results may not be sufficient to remove
concern for mutagenicity raised by a clear positive result in a single mutagenicity assay." These
considerations inform the present approach. In addition, consistent with U.S. Environmental
Protection Agency's (EPA) Guidelines on Carcinogenic Risk Assessment and Supplemental
Guidance for Assessing Susceptibility from Early-Life Exposure to Carcinogens) (U.S. EPA,
2005b, d), the approach does not address relative potency (e.g., among TCE metabolites, or of
such metabolites with other known genotoxic carcinogens) per se, nor does it consider
quantitative issues related to the probable production of these metabolites in vivo. Instead, the
analysis of genetic toxicity data presented here focuses on the identification of a genotoxic
hazard of these metabolites; a quantitative analysis of TCE metabolism to reactive intermediates,
via physiologically based pharmacokinetic (PBPK) modeling, is presented in Section 3.5.
TCE and its known metabolites trichloroacetic acid (TCA), dichloroacetic acid (DCA),
chloral hydrate (CH), trichloroethanol (TCOH), S-(l,2-dichlorovinyl)-L-cysteine (DCVC) and
S-dichlorovinyl glutathione (DCVG) have been studied to varying degrees for their genotoxic
potential. The following section summarizes available data on genotoxicity for both TCE and its
metabolites for each potential genotoxic endpoints, when available, in different organisms.
4.2.1. Trichloroethylene (TCE)
4.2.1.1.1. DNA Binding Studies
Covalent binding of TCE to DNA and protein in cell-free systems has been studied by
several investigators. Incubation of [14C]-radio labeled TCE ([14C]TCE) with salmon sperm
DNA in the presence of microsomal preparations from B6C3F1 mice resulted in dose-related
covalent binding of TCE to DNA. The binding was enhanced when the microsomes were taken
from mice pretreated with phenobarbital, which induces cytochrome (CYP) P450 enzymes,
suggesting the binding may be related to an oxidative metabolite, or when
l,2-epoxy-3,3,3-trichloropropane, an inhibitor of epoxide hydrolase, was added to the
incubations (Banerjee and Van Duuren, 1978). In addition, covalent binding of [14C]TCE with
microsomal proteins was detected after incubation with microsomal preparations from mouse
lung, liver, stomach and kidney, and rat liver (Banerjee and Van Duuren, 1978). Furthermore,
incubation of [14C]TCE with calf thymus DNA in the presence of hepatic microsomes from
phenobarbital-pretreated rats yielded significant covalent binding (DiRenzo et al., 1982).
A number of studies have also examined the role TCE metabolism in covalent binding to
DNA and proteins. Miller and Guengerich (1983) used liver microsomes from control,
This document is a draft for review purposes only and does not constitute Agency policy.
34 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
b-naphthoflavone- and phenobarbital-induced B6C3F1 mice, Osborne-Mendel rats, and human
liver microsomes. Significant covalent binding of TCE metabolites to calf thymus DNA and
proteins was observed in all experiments. Phenobarbital treatment increased the formation of
chloral and TCE oxide formation, DNA and protein adducts. In contrast, b-naphthoflavone
treatment did not induce the formation of any microsomal metabolite suggesting that the forms
of CYP induced by phenobarbital are primarily involved in TCE metabolism while the
b-naphthoflavone-inducible forms of CYP have only a minor role in TCE metabolism. TCE
metabolism (based on TCE-epoxide and DNA-adduct formation) was 2.5-3-fold higher in
mouse than in rat microsomes due to differences in rates and clearance of metabolism (discussed
in Section 3.3.3.1). The levels of DNA and protein adducts formed in human liver microsomal
system approximated those observed in liver microsomes prepared from untreated rats. It was
also shown that whole hepatocytes of both untreated mice and phenobarbital-induced rats and
mice could activate TCE into metabolites able to covalently bind extracellular DNA. A study by
Cai and Guengerich (2001b) postulate TCE oxide (an intermediate in the oxidative metabolism
of TCE in rat and mouse liver microsomes) is responsible for the covalent binding of TCE with
protein, and to a lesser extent, DNA. Mass spectrometry was used to analyze the reaction of
TCE oxide (synthesized by m-chloroperbenzoic acid treatment of TCE) with nucleosides,
oligonucleotides and protein to understand the transient nature of the inhibition of enzymes in the
context of adduct formation. Protein amino acid adducts were observed during the reaction of
TCE oxide with the model peptides. The majority of these adducts were unstable under
physiological conditions. Results using other peptides also indicate that adducts formed from the
reaction of TCE oxide with macromolecules and their biological effects are likely to be relatively
short-lived.
Studies have been conducted using in vitro and in vivo systems to understand the DNA
and protein binding capacity of TCE. In a study in male mice, after repeated intraperitoneal (i.p.)
injections of [14C]TCE, radioactivity was detected in the DNA and RNA of all organs studied
(kidney, liver, lung, spleen, pancreas, brain, and testis) (Bergman, 1983). However, in vivo
labeling was shown to be due to metabolic incorporation of CI fragments, particularly in guanine
and adenine, rather than to DNA-adduct formation. In another study (Stott et al., 1982),
following i.p. injection of [14C]TCE in male Sprague-Dawley rats (10-100 mg/kg) and B6C3F1
mice (10-250 mg/kg), high liver protein labeling was observed while very low DNA labeling
was detected. Stott et al. (1982) also observed very low levels of DNA binding (0.62 ± 0.43
alkylation/106 nucleotides) in mice administered 1,200 mg/kg of TCE. In addition, a
dose-dependent binding of TCE to hepatic DNA and protein at low doses in mice was
demonstrated by Kautiainen et al. (1997). In their dose-response study (doses between 2 |ig/kg
This document is a draft for review purposes only and does not constitute Agency policy.
3 5 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
and 200 mg/kg body weight [BW]), the highest level of protein binding (2.4 ng/g protein) was
observed 1 hour after the treatment followed by a rapid decline, indicating pronounced instability
of the adducts and/or rapid turnover of liver proteins. Highest binding of DNA (120 pg/g DNA)
was found between 24 and 72 hours following treatment. Dose-response curves were linear for
both protein and DNA binding. In this study, the data suggest that TCE does bind to DNA and
proteins in a dose-dependent fashion, however, the type and structure of adducts were not
determined.
Mazzullo et al. (1992) reported that TCE was covalently bound in vivo to DNA, RNA
and proteins of rat and mouse organs 22 hours after i.p. injection. Labeling of proteins from
various organs of both species was higher than that of DNA. Bioactivation of TCE to its
intermediates using various microsomal fractions was dependent on CYP enzyme induction and
the capacity of these intermediates to bind to DNA. It appeared that mouse lung microsomes
were more efficient in forming the intermediates than rat lung microsomes, although no other
species specific differences were found (Mazzullo et al., 1992) This also supports the results
described by Miller and Guengerich (1983). The authors suggest some binding ability of TCE to
interact covalently with DNA (Mazzullo et al., 1992).
In summary, studies report that TCE exposure in vivo can lead to binding to nucleic acids
and proteins, and some authors have suggested that such binding is likely due to conversion to
one or more reactive metabolites.
4.2.1.1.2. Bacterial Systems—Gene Mutations
Gene mutation studies (Ames assay) in various Salmonella typhimurium (S. typhimurium)
strains of bacteria exposed to TCE both in the presence and absence of stabilizing agent have
been conducted by different laboratories (Baden et al., 1979; Crebelli et al., 1982; Henschler et
al., 1977; McGregor et al., 1989; Mortelmans et al., 1986; Shimada et al., 1985; Simmon et al.,
1977; Waskell, 1978) (see Table 4-6). It should be noted that these studies have tested TCE
samples of different purities using various experimental protocols. In all in vitro assays,
volatization is a concern when TCE is directly administered.
Waskell (1978) studied the mutagenicity of several anesthetics and their metabolites.
Included in their study was TCE (and its metabolites) using the Ames assay. The study was
conducted both in the presence and absence of a metabolic activation system, S9, and caution
was exercised to perform the experiment under proper conditions (incubation of reaction mixture
in sealed dessicator vials). This study was performed in both TA98 and TA100 S. typhimurium
strains at a dose range of 0.5-10% between 4 and 48 hours. No change in revertant colonies was
This document is a draft for review purposes only and does not constitute Agency policy.
3 6 DRAFT—DO NOT CITE OR QUOTE
-------
1 observed in any of the doses or time courses tested. No information either on the presence or
2 absence of stabilizers in TCE obtained commercially nor its effect on cytotoxicity was provided
3 in the study.
This document is a draft for review purposes only and does not constitute Agency policy.
3 7 DRAFT—DO NOT CITE OR QUOTE
-------
L/l
K>
Table 4-6. TCE genotoxicity: bacterial assays
Test system/endpoint
Doses tested
With activation
Without
activation
Comments
References
S. typhimurium (TA100)
0.1-10 |iL (epoxide-free)
-
-
Plate incorporation assay
Henschler et al. (1977)
S. typhimurium (TA1535, TA100)
1-2.5% (epoxide-free)
+ (TA100)
- (TA1535)
Simmon et al. (1977)
S. typhimurium (TA98, TA100)
0.5-10%
-
-
The study was conducted in
sealed dessicator vials
Waskell (1978)
S. typhimurium (TA100, TA1535)
1-3% (epoxide-free)
+ (TA100)
± (TA1535)
-
Baden etal. (1979)
S. typhimurium (TA100)
5-20% (v/v)
Negative under normal
conditions, but twofold
increase in mutations in a
preincubation assay
Bartschetal. (1979)
0.33-1.33% (epoxide-free)
+
-
Crebelli et al. (1982)
S. typhimurium (TA1535, TA100)
1-5% (higher and lower
purity)
- (higher purity)
+ (lower purity)
-
Extensive cytotoxicity
Shimada et al. (1985)
S. typhimurium (TA98, TA100,
TA1535, TA1537, TA97)
10-1,000 (iL/plate
-
-
Preincubation protocol
Mortelmans et al.
(1986)
S. typhimurium (TA98, TA100,
TA1535)
<10,000 ng/plate
(unstabilized)
-
ND
Vapor assay
McGreeor et al. (1989)
<10,000 ng/plate
(oxirane-stabilized)
+
+
Vapor assay
McGreeor et al. (1989)
S. typhimurium
<10,000 ng/plate
(epoxybutane stabilized)
ND
+
Preincubation assay
McGreeor et al. (1989)
<10,000 ng/plate
(epichlorohydrin stabilized)
ND
+
Vapor assay
McGreeor et al. (1989)
S. typhimurium (YG7108)
1.000-3.000 ng/plate
ND
+
Microcolony assay/revertants
Emmert et al. (2006)
E. coli (K12)
0.9 mM (analytical grade)
+
-
Revertants at arg56 but not
nadl 13 or other loci
Greimetal. (1975)
S?
>!
rs
s;
>1
£
St
oo o
>!
>1
o
a
a
§•
>!
o
0
H
1 5
I ^
81
w (V
2; ^
H J
M s
H 2.
W
V
o
c
o
H
W
ND = not determined.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
In other studies highly purified, epoxide free TCE samples were not mutagenic in
experiments with and without exogenous metabolic activation by S9 in S. typhimurium strain
TA100 using the plate incorporation assay (Henschler et al., 1977). Furthermore, no mutagenic
activity was found in several other strains including TA1535, TA1537, TA97, TA98, and
TAlOOusing the preincubation protocol (Mortelmans et al., 1986). Simmon et al. (1977)
observed a less than twofold but reproducible and dose-related increase in his + revertants in
plates inoculated with S. typhimurium TA100 and exposed to a purified, epoxide-free TCE
sample. The authors observed no mutagenic response in strain TA1535 with S9 mix and in
either TA1535 or TA100 without rat or mouse liver S9. Similar results were obtained by Baden
et al. (1979), Bartsch et al. (1979) and Crebelli et al. (1982). In all these studies purified,
epoxide-free TCE samples induced slight but reproducible and dose-related increases in
his + revertants in S. typhimurium TA100 only in the presence of S9. No mutagenic activity was
detected without exogenous metabolic activation or when liver S9 from naive rats, mice, and
hamsters (Crebelli et al., 1982) was used for activation. Therefore, a number of these studies
showed positive results in TA100 with metabolic activation, but not in other strains or without
metabolic activation.
Shimada et al. (1985) tested a low-stabilized, highly purified TCE sample in an Ames
reversion test, modified to use vapor exposure, in S. typhimurium TA1535 and TA100. No
mutagenic activity was observed—either in the presence or absence of S9 mix. However, at the
same concentrations (1, 2.5, and 5%), a sample of lower purity, containing undefined stabilizers,
was directly mutagenic in TA100 (>fivefold) and TA1535 (>38-fold) at 5% concentration
regardless of the presence of S9. It should be noted that the doses used in this study resulted in
extensive killing of bacterial population, particularly at 5% concentration, more than 95%
toxicity was observed.
A series of studies evaluating TCE (with and without stabilizers) was conducted by
McGregor et al. (1989). The authors tested high purity and oxirane-stabilized TCE samples for
their mutagenic potential in S. typhimurium strains TA1535, TA98, and TA100. Preincubation
protocol was used to test stabilized TCE (up to 10,000 |ig/plate). Mutagenic response was not
observed either in the presence or absence of metabolic activation. When TCE was tested in a
vapor delivery system without the oxirane stabilizers, the authors did not observe any mutagenic
activity. However, TA1535 and TA100 produced a mutagenic response both in the presence and
absence of S9 when exposed to TCE containing 0.5-0.6% 1,2-epoxybutane. Furthermore,
exposure to epichlorohydrin also increased the frequency of mutants.
Emmert et al. (2006) used a CYP2E1-competent bacterial strain (S. typhimurium
containing YG7108pin3ERbs plasmid) in their experiments. TCE was among several other
This document is a draft for review purposes only and does not constitute Agency policy.
3 9 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
compounds investigated and was tested at concentrations of 1,000-3,000 |ig/plate. TCE induced
toxicity and microcolonies at or above 1,000 |ig per plate. A study on Escherichia coli (E. coli)
K12 strain was conducted by Greim et al. (1975) using analytical-grade TCE samples.
Revertants were scored at two loci: argy,, sensitive to base-pair substitution and nadn3, reverted
by frameshift mutagens. In addition, forward mutations to 5-methyltryptophan resistance and
galactose fermentation were selected. Approximately twofold increase in arg + colonies was
observed. No change in other sites was observed. No definitive conclusion can be drawn from
this study due to lack of information on reproducibility and dose-response. .
In addition to the above studies, the ability of TCE to induce gene mutations in bacterial
strains has been reviewed and summarized by several authors (Clewell and Andersen, 2004;
Crebelli and Carere, 1989; Douglas et al., 1999; Fahrig et al., 1995; Moore and Harrington-
Brock, 2000). In summary, TCE, in its pure form as a parent compound is unlikely to induce
point mutations in most bacterial strains. It is possible that some mutations observed in response
to exposure to technical grade TCE may be contributed by the contaminants/impurities such as
1,2-epoxybutane and epichlorohydrin, which are known bacterial mutagens. However, several
studies of TCE reported low, but positive responses in the TA100 strain in the presence of S9
metabolic activation, even when genotoxic stabilizers were not present.
4.2.1.1.3. Fungal and Yeast Systems—Gene Mutations, Conversions and Recombination
Gene mutations, conversions, and recombinations have been studied to identify the effect
of TCE in fungi and yeast systems (see Table 4-7).
Crebelli et al. (1985) studied the mutagenicity of TCE in Aspergillus nidulans
(A. nidulans) both for gene mutations and mitotic segregation. No increase in mutation
frequency was observed when A. nidulans was plated on selective medium and then exposed to
TCE vapors. A small but statistically significant increase in mutations was observed when
conidia of cultures were grown in the presence of TCE vapors and then plated on selective
media. Since TCE required actively growing cells to exerts its genotoxic activity and previous
studies (Bignami et al., 1980) have shown activity in the induction of methGl suppressors by
trichloroethanol and chloral hydrate, it is possible that endogenous metabolic conversion of TCE
into trichloroethanol or chloral hydrate may have been responsible for the positive response.
To understand the cytochrome P450 mediated genotoxic activity of TCE, Callen et al.
(1980) conducted a study in two yeast strains (D7 and D4) CYP. The D7 strain in it log-phase
had a CYP concentration up to five times higher than a similar cell suspension of D4 strain. Two
This document is a draft for review purposes only and does not constitute Agency policy.
40 DRAFT—DO NOT CITE OR QUOTE
-------
1 different concentrations (15 and 22 mM) at two different time points (1 and 4 hours) were
2 studied. A significant increase in frequencies of mitotic gene conversion and recombination was
This document is a draft for review purposes only and does not constitute Agency policy.
41 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-7. TCE genotoxicity: fungal and yeast systems
Test system/endpoint
Doses tested
With
activation
Without
activation
Comments
References
Gene conversions
S. cerevisiae D7 and D4
15 and 22 mM; 1
and 4 h
ND
+ at 1 h, D7
strain;
- at 4 h, both
D7 and D4
gene conversion;
CYP content fivefold greater
in D7 strain;
high cytotoxicity at 22 mM
Callenetal. (1980)
S. cerevisiae D7
11.1, 16.6, and 22.2
mM
~
~
both stationary and log
phase/production of
phototropic colonies
Kochetal. (1988)
S. pombe
0.2-200 mM
("pure" and
technical grade)
~
~
forward mutation, different
experiments with different
doses and time
Rossi et al. (1983)
S. cerevisiae D7
+
-
Bronzetti et al. (1980)
A. nidulans
no data
+
forward mutation
Crebelli et al. (1985)
Recombination
S. cerevisiae
+
-
gene conversion
Bronzetti et al. (1980)
S. cerevisiae D7 and D4
15 and 22 mM;
1 and 4 h
ND
+
Callenetal. (1980)
A. nidulans
ND
+
gene cross over
Crebelli et al. (1985)
Mitotic aneuploidy
S. cerevisiae D61.M
5.5, 11.1, and 16.6
mM
+
+
loss of dominant color
homolog
Kochetal. (1988)
ND = not determined .
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
observed at 15 mM concentrations at 1-hour exposure period in the D7 strain, however, the
22 mM concentration was highly cytotoxic (only 0.3% of the total number of colonies survived).
No changes were seen in D4 strain, suggesting that metabolic activation via CYP played an
important role in both genotoxicity and cytotoxicity. However, marginal or no genotoxic activity
was observed when incubation of cells and test compounds were continued for 4 hours in either
strain, possibly because of increased cytotoxicity, or a destruction of the metabolic system.
Koch et al. (1988) studied the genotoxic effects of chlorinated ethylenes including TCE
in various yeast Saccharomyces cerevisiae strains. Strain D7 was tested (11.1, 16.6, and 22.2
mM TCE) both in stationary-phase cells without S9, stationary-phase cells with S9 and
logarithmic-phase cells using different concentrations. No significant change in mitotic gene
conversion or reverse mutation was observed in either absence or presence of S9. In addition,
there was a considerable increase in the induction of mitotic aneuploidy in Strain D61.M, though
no statistical analysis was performed.
Rossi et al. (1983) studied the effect of TCE on yeast species S. pombe both using in vitro
and host mediated mutagenicity studies and the effect of two stabilizers, epichlorohydrin and
1,2-epoxybutane that are contained in the technical grade of TCE. The main goal of this study
was to evaluate genotoxic activity of TCE samples of different purity and if the effect is due to
the additives present in the TCE or TCE itself. Forward mutations at five loci (ade 1, 3, 4, 5, 9)
of the adenine pathway in the yeast, strain PI was evaluated. The stationary-phase cells were
exposed to 25 mM concentration of TCE for 2, 4, and 8 hours in the presence and absence of S9.
No change in mutation frequency was observed either in pure-grade samples or technical-grade
samples either in the presence or absence of S9 at any of the time-points tested. Interestingly,
this suggests that the stabilizers used in technical-grade TCE are not genotoxic in yeast. In a
follow-up experiment, the same authors studied the effect of different concentrations (0.22, 2.2
and 22.0 mM) in a host mediated assay using liver microsome preparations obtained from
untreated mice, from phenobarbital-pretreated and naphthoflavone-pretreated mice and rats,
which also suggested that stabilizers were not genotoxic in yeast. This experiment is described
in more detail in Section 4.2.1.4.1.
Furthermore, TCE was tested for its ability to induce both point mutation and mitotic
gene conversion in diploid strain of yeast S. cerevisiae (strain D7) both with and without a
mammalian microsomal activation system. In a suspension test with D7, TCE was active only
with microsomal activation (Bronzetti et al., 1980).
These studies are consistent with those of bacterial systems in indicating that pure TCE as
a parent compound is not likely to cause mutations, gene conversions, or recombinations in
fungal or yeast systems. In addition, the data suggest that contaminants used as stabilizers in
This document is a draft for review purposes only and does not constitute Agency policy.
43 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
technical grade TCE are not genotoxic in these systems, and that the observed genotoxic activity
in these systems is predominantly mediated by TCE metabolites.
4.2.1.1.4. Mammalian Systems Including Human Studies
4.2.1.1.5. Gene mutations (bacterial, fungal, or yeast with a mammalian host)
Very few studies have been conducted to identify the effect of TCE, particularly on gene
(point) mutations using mammalian systems (see Table 4-8). An overall summary of different
endpoints using mammalian systems will be provided at the end of this section. In order to
assess the potential mutagenicity of TCE and its possible contaminants, Rossi et al. (1983)
performed genotoxicity tests using two different host mediated assays with pure- and
technical-grade TCE. Male mice were administered with one dose of 2 g/kg of pure or technical
grade TCE by gavage. Following the dosing, for the intraperitoneal host-mediated assay, yeast
cell suspensions (2 x 109 cells/mL) were inoculated into the peritoneal cavity of the animals.
Following 16 hours, animals were sacrificed and yeast cells were recovered to detect the
induction of forward mutations at five loci (ade 1, 2, 4, 5, 9) of the adenine pathway. A second
host-mediated assay was performed by exposing the animals to 2 g/kg of pure or technical grade
TCE and inoculating the cells into the blood system. Yeast cells were recovered from livers
following 4h of exposure. Forward mutations in the five loci {ade 1,2,4,5,9) were not observed
in host-mediated assay either with pure or technical-grade TCE. Genotoxic activity was not
detected when the mutagenic epoxide stabilizers were tested for mutagenicity independently or
in combination. To confirm the sensitivity of the assay, the authors tested a positive
control—A-nitroso-dimethyl-nitrosamine (1 mg/kg) and found a mutation frequency of more
than 20 times the spontaneous level. The authors suggest that the negative result could have
been due to an inadequate incubation time of the sample with the yeast cells.
Male and female transgenic lac Z mice were exposed by inhalation to an actual
concentrations of 0, 203, 1,153, and 3,141 ppm TCE, 6 hours/day for 12 days (Douglas et al.,
1999). Following 14 and 60 days of last exposure, animals were sacrificed and the mutation
frequencies were determined in various organs such as bone marrow, kidney, spleen, liver, lung,
and testicular germ cells. No statistically significant increases in base-changes or small-deletions
were observed at any of the doses tested in male or female lung, liver, bone marrow, spleen, and
kidney, or in male testicular germ cells when the animals were sampled 60 days after exposure.
In addition, statistically significantly increased gene mutations were not observed in the lungs at
14 days after the end of exposure (Douglas et al., 1999). The authors acknowledge that lacZ
This document is a draft for review purposes only and does not constitute Agency policy.
44 DRAFT—DO NOT CITE OR QUOTE
-------
1 bacteriophage transgenic assay does not detect large deletions. The authors also acknowledge
2 that their hypothesis does not readily explain the increases in small deletions and base-change
This document is a draft for review purposes only and does not constitute Agency policy.
45 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-8. TCE genotoxicity: mammalian systems—gene mutations and chromosome aberrations
Test system/endpoint
Doses tested
With
activation
Without
activation
Comments
References
Gene mutations (forward mutations)
Schizosaccharomyces pombe
2 g/kg, 4 and 16 h
ND
Host-mediated: intravenous
and intraperitoneal injections
of yeast cells
Rossi et al. (1983)
Gene mutations (mutations frequency)
lac Z transgenic mice
0, 203, 1,153, or
3,141 ppm
No base
changes or
small deletions
No base
changes or
small deletions
Lung, liver, bone marrow,
spleen, kidney, testicular
germ cells used
Douglas et al. (1999)
Chromosomal aberrationsa
CHO
745-14,900 (ig/mL
ND
-
8-14 h
Galloway et al. (1987)
499-14,900 (ig/mL
-
ND
2 h exposure
Galloway et al. (1987)
C57BL/6J mice
5, 50, 500, or 5,000
ppm (6 h)
-
NA
Splenocytes
Kligerman et al. (1994)
S-D rats
5, 50, 500, or 5,000
ppm (6 h, single and
4-d exposure)
-
NA
Peripheral blood lymphocytes
Kligerman et al. (1994)
a It should be noted that results of most chromosomal aberration assays report the combined incidence of multiple effects, including chromatid breaks,
isochromatid or chromosome breaks, chromatid exchanges, dicentric chromosomes, ring chromosomes, and other aberrations.
ND = not determined, NA = not applicable.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
mutations found in the von Hippel-Lindau tumor suppressor gene in renal cell carcinomas of the
TCE-exposed population. DC A, a TCE metabolite has been shown to increase lacl mutations in
transgenic mouse liver, however, only after 60-weeks-of-exposure to high concentration
(>1,000 ppm) in drinking water (Leavitt et al., 1997). DCA induced relatively small increase in
lac / mutations when the animals were exposed for 60 weeks, a significantly longer duration than
the TCE exposure in the Douglas et al. (1999) study (<2 weeks). Because a relatively small
fraction of TCE is metabolized to DCA (see Section 3.3), the mutagenic effect of DCA is
unlikely to have been detected in the experiments in Douglas et al. (1999). Glutathione (GSH)
conjugation, which leads to the production of genotoxic metabolites (see Section 4.2.5),
constitutes a relatively small (and relatively uncertain) portion of TCE metabolism in mice, with
little data on the extent of renal DCVC bioactivation versus detoxification in mice (see
Sections 3.3 and 3.5). In addition, statistically significantly increased kidney tumors have not
been reported in mice with TCE treatment, and the increased incidence of kidney tumors in rats,
while considered biologically significant, are quite low and not always statistically significant
(see Section 4.4). Therefore, although Douglas et al. (1999) did not detect increased mutations
in the kidney, these results are not highly informative as to the role of mutagenicity in
TCE-induced kidney tumors, given the uncertainties in the production in genotoxic GSH
conjugation metabolites in mice and the low carcinogenic potency of TCE for kidney tumors in
rodents relative to what is detectable in experimental bioassays.
4.2.1.1.6. von Hippel-Lindau (VHL) gene mutations
Studies have been conducted to determine the role of VHL gene mutations in renal cell
carcinoma, with and without TCE exposure, and are summarized here. Most of these studies are
epidemiologic, comparing VHL mutation frequencies of TCE-exposed to nonexposed cases from
renal cell carcinoma case-control studies, or to background mutation rates among other renal cell
carcinoma case series (described in Section 4.4.3). Inactivation of the VHL gene through
mutations, loss of heterozygosity and imprinting has been observed in about 70% of renal clear
cell carcinomas (Alimov et al., 2000; Kenck et al., 1996). Recent studies have also examined the
role of other genes or pathways in renal cell carcinoma subtypes, including c-myc activation and
vascular endothelial growth factor (VEGF) (Furge et al., 2007; Toma et al., 2008).
Several studies have examined the role of VHL gene inactivation in renal cell carcinoma,
including a recent study that measured not only mutations but also promoter hypermethylation
(Nickerson et al., 2008). This study focused on kidney cancer regardless of cause, and found that
91% of cc-renal cell carcinoma (RCC) exhibited alterations of the VHL gene, suggesting a role
This document is a draft for review purposes only and does not constitute Agency policy.
47 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
for VHL mutations as an early event in cc-RCC. A recent analysis of current epidemiological
studies of renal cell cancer suggests VHL gene alterations as a marker of cc-RCC, but that
limitations of previous studies may make the results difficult to interpret (Chow and Devesa,
2008). Conflicting results have been reported in epidemiological studies of VHL mutations in
TCE-exposed cases and are described in detail in Section 4.5.2. Both Briining et al. (1997b) and
Brauch et al. (1999; 2004) associated increased VHL mutation frequency in TCE-exposed renal
cell carcinoma cases. The two other available studies of Schraml et al. (1999) and
Charbotel et al. (2007) because of their limitations and lower mutation detection rate in the case
of Charbotel et al. (2007)neither add nor detract to the conclusions from the earlier studies.
Additional discussion of these data are in Section 4.4.3.
Limited animal studies have examined the role of TCE and VHL mutations, although
Mally et al. (2006) have recently conducted both in vitro and in vivo studies using the Eker rat
model (see Section 4.4.6.1.1). The Eker rat model (Tsc-2"b) is at increased risk for the
development of spontaneous renal cell carcinoma and as such has been used to understand the
mechanisms of renal carcinogenesis (Stemmer et al., 2007; Wolf et al., 2000). One study has
demonstrated similar pathway activation in Eker rats as that seen in humans with VHL mutations
leading to renal cell carcinoma, suggesting Tsc-2 inactivation is analogous to inactivation of VHL
in human renal cell carcinoma (Liu et al., 2003). In Mally et al. (2006), male rats carrying the
Eker mutation were exposed to TCE (0, 100, 250, 500, or 1,000 mg/kg BW by gavage, 5 days a
week) for 13 weeks to determine the renal effects (additional data from this study on in vitro
DCVC exposure are discussed below, Section 4.2.5). A significant increase in labeling index in
kidney tubule cells was observed, however, no enhancement of preneoplastic lesions or tumor
incidence was found in Eker rat kidneys compared to controls. In addition, no VHL gene
mutations in exons 1-3 were detected in tumors obtained from either control or TCE-exposed
Eker rats. Although no other published studies have directly examined VHL mutations following
exposure to TCE, two studies performed mutational analysis of archived formalin-fixed paraffin
embedded tissues from renal carcinomas from previous rat studies. These carcinomas were
induced by the genotoxic carcinogens potassium bromate (Shiao et al., 2002) or
A'-nitrosodi methyl amine (Shiao et al., 1998). Limited mutations in the VHL gene were observed
in all samples, but, in both studies, these were found only in the clear cell renal carcinomas.
Limitations of these two studies include the small number of total samples analyzed, as well as
potential technical issues with DNA extraction from archival samples (see Section 4.4.3).
However, analyses of VHL mutations in rats may not be informative as to the potential
genotoxicity of TCE in humans because the VHL gene may not be the target for
nephrocarcinogenesis in rats to the extent that it appears to be in humans.
This document is a draft for review purposes only and does not constitute Agency policy.
48 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
4.2.1.1.7. Chromosomal aberrations
A few studies were conducted to investigate the ability of TCE to induce chromosomal
aberrations in mammalian systems (see Table 4-8). Galloway et al. (1987) studied the effect of
TCE on chromosome aberrations in Chinese hamster ovary cells. When the cells were exposed
to TCE (499-14,900 |ig/mL) for 2 hours with metabolic activation, S9, no chromosomal
aberrations were observed. Furthermore, without metabolic activation, no changes in
chromosomal aberrations were found when the cells were exposed to TCE concentrations of
745-14,900 |ig/mL for 8-14 hours. It should be noted that in this study, liquid incubation
method was used and the experiment was part of a larger study to understand the genotoxic
potential of 108 chemicals.
Three inhalation studies in mice and rats examined if TCE could induce cytogenetic
damage (Kligerman et al., 1994). In the first two studies, CD rats or C57B1/6 mice, were
exposed to 0-, 5-, 500-, or 5,000-ppm TCE for 6 hours. Peripheral blood lymphocytes in rats and
splenocytes in mice were analyzed for induction of chromosomal aberrations, sister chromatid
exchanges, and micronucleus formation. The results of micronucleus and sister chromatid
exchanges will be discussed in the next sections (see Sections 4.2.1.4.4 and 4.2.1.4.5). No
significant increase in chromosomal aberrations was observed in binucleated peripheral blood
lymphocytes. In the third study, the authors exposed the same strain of rats for 6 hours/day over
4 consecutive days. No statistically significant concentration-related increases in chromosomal
aberrations were observed. The limited results of the above studies have not reported TCE to
cause chromosomal aberrations either in in vitro or in vivo mammalian systems.
4.2.1.1.8. Micronucleus induction
The appearance of micronuclei is another endpoint that can demonstrate the genotoxic
effect of a chemical. Several studies have been conducted to identify if TCE can cause
micronucleus formation (see Table 4-9).
Wang et al. (2001) investigated micronucleus formation by TCE administered as a vapor
in CHO-K1 cells in vitro. Cells were grown in culture media with an inner Petri dish containing
TCE that would evaporate into the media containing cells. The concentration of TCE in cultured
medium was determined by gas chromatography. The actual concentration of TCE ranged from
0.8 and 1.4 ppm after a 24-hour treatment. A significant dose-dependent increase in micronuclei
formation was observed. A dose-dependent decrease in cell growth and cell number was also
observed. The authors did not test if the micronuclei formed was due to direct damage to the
DNA or spindle formation.
This document is a draft for review purposes only and does not constitute Agency policy.
49 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-9. TCE genotoxicity: mammalian systems—micronucleus, sister chromatic exchanges
Test system/endpoint
Doses tested
With
activation
Without
activation
Comments
References
Micronucleus
Human hepatoma HepG2 cells
0.5-4 mM, 24 h
NA
+
Hu et al. (2008)
Primary cultures of human and rat kidney
cells
1.0, 2.0, or 4.0 mM
NA
+
Dose-dependent significant
increase
Robbiano et al. (2004)
Sprague-Dawley rats
3,591 mg/kg
+
-
Robbiano et al. (2004)
CHO-K1 cells
0.8-1.4 ppm
+
Dose-dependent significant
increase
Wang et al. (2001)
Male CD-I mice
457 mg/kg
+
NA
Bone marrow, correlated with
TCOH in urine
Hrelia et al. (1994)
C56BL/6J mice
5, 50, 500, or
5,000 ppm
-
NA
Splenocytes
Kligerman et al. (1994)
S-D rats
5, 50, 500, or
5,000 ppm
+
NA
Dose dependent; peripheral
blood lymphocytes
Kligerman et al. (1994)
Sister chromatid exchanges
CHO
0.17%
-
ND
1 h (vapor)
White etal. (1979)
17.9-700 (ig/mL
ND
+
25 h (liquid)
Galloway et al. (1987)
49.7-14,900 ng/mL
+
ND
2 h
Galloway et al. (1987)
Human lymphocytes
178 (ig/mL
ND
+
Gu et al. (1981a; 1981b)
S-D rats
5, 50, 500, or
5,000 ppm
-
NA
Peripheral blood lymphocytes
Kligerman et al. (1994)
Peripheral blood lymphocytes from humans
occupationally exposed
Occupational
exposure
-
NA
Nagayaetal., 1989a
C57BL/6J mice
5, 50, 500, or
5,000 ppm
—
NA
Splenocytes
Kligerman et al. (1994)
ND = not determined, NA = not applicable.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Robbiano et al. (2004) conducted an in vitro study on DNA damage and micronuclei
formation in rat and human kidney cells exposed to six carcinogenic chemicals including TCE.
The authors examined for the ability of TCE to induce DNA fragmentation and formation of
micronuclei in primary cultures of rat and human kidney cells derived from kidney cancer
patients with 1-4 mM TCE concentrations. A significant dose-dependent increase in the
frequency of micronuclei was obtained in primary kidney cells from both male rats and human of
both genders. The authors acknowledge that the significance of the results should be considered
in light of the limitations including (1) examination of TCE on cells from only three rats,
(2) considerable variation in the frequency of DNA lesions induced in the cells, and (3) the
possibility that kidney cells derived from kidney cancer patients may be more sensitive to
DNA-damaging activity due to a more marked expression of enzymes involved in the metabolic
activation of kidney procarcinogens and suppression of DNA repair processes. Never the less,
this study is important and provides information of the possible genotoxic effects of TCE.
In the same study, Robbiano et al. (2004) administered rats a single oral dose of TCE
(3,591 mg/kg) corresponding to '/2 LD50 which had been pre-exposed to folic acid for 48 hours
and the rats were euthanized 48 hours later following exposure to TCE. The frequency of
binucleated cells was taken as an index of kidney cell proliferation. A statistically significant
increase in the average frequency of micronucleus was observed.
Hu et al. (2008) studied the effect of TCE on micronuclei frequencies using human
hepatoma HepG2 cells. The cells were exposed to 0.5, 1, 2, and 4 mM TCE for 24 hours. TCE
caused a significant increase in micronuclei frequencies at all concentrations tested. It is
important to note that similar concentrations were used in Robbiano et al. (2004).
As described in the chromosomal aberration section (see Section 4.2.1.4.3), inhalation
studies were performed using male C57BL/6 mice and CD rats (Kligerman et al., 1994) to
determine if TCE could induce micronuclei. In the first and second study, rats or mice
respectively, were exposed to 0-, 5-, 500-, or 5,000-ppm TCE for 6 hours. Peripheral blood
lymphocytes in rats and splenocytes in mice were cultured and analyzed for induction of
micronuclei formation. Bone marrow polychromatic erythrocytes (PCEs) were also analyzed for
micronuclei. TCE caused a statistically significant increase in micronuclei formation at all
concentrations in rat bone marrow PCEs but not in mice. The authors note that TCE was
significantly cytotoxic at the highest concentration tested as determined by significant
concentration-related decrease in the ratio of PCEs/normochromatic erythrocytes. In the
third study, to confirm the results of the first study, the authors exposed rats to one dose of
5,000 ppm for 6 hours. A statistical increase in bone marrow micronuclei-PCEs was observed
confirming the results of the first study.
This document is a draft for review purposes only and does not constitute Agency policy.
51 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Hrelia et al. (1994) treated male CD-I mice with TCE (457 mg/kg BW; i.p.) for 30 hours.
Bone marrow cells were harvested for determination of micronuclei frequencies in PCEs. An
increase in micronuclei frequency at 30 hours after treatment was observed. Linear regression
analysis showed that micronuclei frequency induced by TCE correlated with trichloroethanol
concentrations in urine, a marker of TCE oxidative metabolism (Hrelia et al., 1994).
In summary, based on the results of the above studies, TCE is capable of inducing
micronuclei in different in vitro and in vivo systems tested. Specific methods were not used that
could definitively identify the mechanism of micronuclei formation. These are important
findings that indicate TCE has genotoxic potential as measured by the micronucleus formation.
4.2.1.1.9. Sister chromatid exchanges (SCEs)
Studies have been conducted to understand the ability of TCE to induce sister chromatid
exchanges (SCEs) both in vitro and in vivo systems (see Table 4-9). White et al. (1979)
evaluated the possible induction of SCE in CHO using a vapor exposure procedure by exposing
the cells to TCE (0.17%) for 1 hour in the presence of S9 metabolic activation. No change in
SCE frequencies were observed between the control and the treatment group. However, in
another study by Galloway et al. (1987) a dose-related increase in SCE frequency in repeated
experiments both with and without metabolic activation was observed. It should be noted that in
this study, liquid incubation was used, and the exposure times were 25 hours without metabolic
activation at a concentration between 17.9-700 |ig/mL and 2 hours in the presence of S9 at a
concentration of 49.7-14,900 |ig/mL. Due to the difference in the dose, length of exposure and
treatment protocol (vapor exposure vs. liquid incubation), no direct comparison can be made. It
should also be noted that inadequacy of dose selection and the absence of positive control in the
White et al. (1979) makes it difficult to interpret the study. In another study (Gu et al.,
1981a)sss, a small but positive response was observed in assays with peripheral lymphocytes.
No statistically significant increase in SCEs was found when male C57B1/6 mice or CD
rats were exposed to TCE at concentrations of 5,500, or 5,000 ppm for 6 hours (Kligerman et al.,
1994). Furthermore, in another study by Nagaya et al. (1989a), lymphocytes of TCE-exposed
workers (n = 22) and matched controls (// = 22) were analyzed for SCEs. The workers had
constantly used TCE in their jobs although the exact exposure was not provided. The duration of
their employment ranged from 0.7-34 years, averaging about 10 years. It should be noted that
there were both smokers and nonsmokers among the exposed population. If a subject had not
smoked for at least 2 years before the samples were taken, then they were considered as
nonsmokers. There were eight nonsmokers in the group. If they were classified as smokers,
This document is a draft for review purposes only and does not constitute Agency policy.
52 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
then they smoked between 10-50 cigarettes per day. No significant increase in mean SCE
frequencies were found in exposed population compared to controls, though the study is
relatively small.
In summary, induction of SCEs have been reported in several, though not all, paradigms
of TCE exposure, consistent with the structural damage to DNA/chromosomes indicated by
excess micronuclei formation.
4.2.1.1.10. Unscheduled DNA synthesis
In vitro studies are briefly described here, with additional discussion of effects related to
TCE-induced unscheduled DNA synthesis in the context of the liver in Section E.2.4.1. Perocco
and Prodi (1981) studied unscheduled DNA synthesis in human lymphocytes cultured in vitro
(see Table 4-10). Three doses of TCE (2.5, 5.0, and 10 |iL/mL) were used as final
concentrations with and without S9. The results indicate that there was an increase in
unscheduled DNA synthesis (UDS) only in the presence of S9, and in addition, the increase was
maximal at the TCE concentration of 5 |iL/mL. Three chlorinated ethane and ethylene solvent
products were examined for their genotoxicity in hepatocyte primary culture DNA repair assays
using vapor phase exposures. Rat hepatocytes primary cultures were initiated and exposed to
low-stabilized or standard stabilized TCE (0.1-2.5%) for 3 or 18 hours. Unscheduled DNA
synthesis or DNA repair was not observed using either low or standard stabilized TCE, even at
vapor phase doses up to those that produced extensive cell killing after 3 or 18 hour exposure
(Shimada et al., 1985). Costa and Ivanetich (1984) examined the ability of TCE to induce
unscheduled DNA synthesis hepatocytes isolated from phenobarbital treated rats. The UDS was
assessed only at the highest concentration that is tolerated by the hepatocytes (2.8 mM TCE).
These results indicate that TCE stimulated unscheduled DNA synthesis in isolated rodent
hepatocytes, and, importantly, in human lymphocytes in vitro.
4.2.1.1.11. DNA strand breaks
DNA damage in response to TCE exposure was studied using comet assay in human
hepatoma HepG2 cells (Hu et al., 2008; see Table 4-10). The cells were exposed to 0.5, 1, 2, and
4 mM for 24 hours. TCE increased the DNA migration in a significant dose-dependent manner
at all tested concentrations suggesting TCE caused DNA strand breaks and chromosome damage.
TCE (4-10 mmol/kg BW) were given to male mice by i.p. injection. The induction of
single-strand breaks (SSB) in DNA of liver, kidney, and lung was studied by the DNA
This document is a draft for review purposes only and does not constitute Agency policy.
53 DRAFT—DO NOT CITE OR QUOTE
-------
1 unwinding technique. There was a linear increase in the level of single strand breaks in kidney
2 and liver DNA but not in lung DNA 1 hour after administration (Walles, 1986).
This document is a draft for review purposes only and does not constitute Agency policy.
54 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-10. TCE genotoxicity: mammalian systems—unscheduled DNA synthesis, DNA strand breaks/protein
crosslinks, cell transformation
Test system/endpoint
Doses tested
With
activation
Without
activation
Comments
References
Unscheduled DNA synthesis
Rat primary hepatocytes
ND
-
Shimada et al. (1985)
Human lymphocytes
2.5, 5, 10 (iL/mL
±
~
Increase was only in certain
doses and maximum at
5 |iL/mL conc.
Perocco and Prodi (1981)
Phenobarbital induced rat hepatocytes
2.8 mM
ND
+
Costa and Ivanetich (1984)
DNA strand breaks/protein crosslinks
Primary rat kidney cells
0.5, 1.0,2.0,
4.0 mM
NA
+
Dose-dependent significant
increase
Robbiano et al. (2004)
Primary cultures of human kidney cells
1.0, 2.0, 4.0 mM
ND
+
Dose-dependent significant
increase
Robbiano et al. (2004)
Sprague-Dawley rats
3,591 mg/kg
+
NA
Single oral administration
Robbiano et al. (2004)
Sprague-Dawley rats
500, 1,000, and
2,000 ppm
-
NA
Comet assay
Clay (2008)
Cell transformation
BALB/c 3T3 mouse cells
4, 20, 100,
250 (ig/mL
NA
+
Weakly positive compared to
other halogenated compounds
tested in the same experiment
Tu et al. (1985)
Rat embryo cells
NA
+
Price et al. (1978)
Syrian hamster embryo cells
5, 10, 25 (ig/mL
NA
-
Amacher and Zelljadt
(1983)
ND = not determined, NA = not applicable.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Robbiano et al. (2004) conducted an in vitro study on DNA damage in rat and human
kidney cells exposed to six carcinogenic chemicals including TCE in the comet assay. The
authors examined the ability of TCE to induce DNA fragmentation in primary cultures of rat and
human kidney cells with 1-4 mM TCE concentrations. TCE was dissolved in ethanol with a
maximum concentration of 0.3% and the rat cultures were exposed to 20 hours. Primary human
kidney cells were isolated from fragments of kidney discarded during the course of surgery for
carcinoma of both male and female donors with an average age of 64.2 years and were also
exposed to 20 hours. Significant dose-dependent increases in the ratio of treated/control tail
length (average 4-7 |iM compared to control) was observed as measured by comet assay in
primary kidney cells from both male rats and human of both genders.
Clay et al. (2008) studied the DNA damage inducing capacity of TCE using the comet
assay in rat kidney proximal tubules. Rats were exposed by inhalation to a range of TCE
concentrations (500, 1,000, or 2,000 ppm) for 6 hours/day for 5 days. TCE did not induce DNA
damage (as measured by tail length and percentage tail DNA and tail movement) in rat kidney
proximal tubules in any of the doses tested possibly due to study limitations (small number of
animals tested [n = 5] and limited exposure time [6 hours/day for only 5 days]). These results
are in contrast to the findings of Robbiano et al. (2004) which showed DNA damage and
increased micronuclei in the rat kidney 20 hours following a single dose (3,591 mg/kg BW) of
TCE. The DNA damage reported by comet assay is consistent with results for other markers
of chromosomal damage or DNA structural damage such as excess micronuclei formation and
SCE induced by TCE exposure.
4.2.1.1.12. DNA damage related to oxidative stress, polymorphisms
A detailed description of studies related to lipid peroxidation of TCE is presented in
conjunction with discussion of liver toxicity (see Section 4.5, E.2.4.3, and E.3). A recent study
reported on genetic polymorphism in solvent exposed population (Kumar et al., 2009). Normal
(n = 220) and solvent-exposed (n = 97) population was genotyped for CYP1A1, GSTM1,
GSTT1 and GSTP1 polymorphisms. No exposure related differences were observed. In
addition, the authors also examined TCE-exposed lymphocytes for the presence of chromosomal
aberrations and micronucleus at concentration of 2, 4 or 6 mM TCE. No significant changes in
any of the parameters were observed.
This document is a draft for review purposes only and does not constitute Agency policy.
56 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
4.2.1.1.13. Cell transformation
In vitro cell transformation using BALB/c-3T3 cells was conducted using TCE with
concentrations varying from 0-250 |ig/mL in liquid phase exposed for 72 hours (see Table 4-10).
The cytotoxicity of TCE at the concentration tested in the transformation assay was determined
by counting cells from duplicate plates of each test conditions at the end of the treatment period.
A dose-dependent increase in Type III foci was observed although no statistical analysis was
conducted (Tu et al., 1985). In another study by Amacher and Zelljadt (1983), Syrian hamster
embryo cells were exposed to 5, 10, or 25 |ag/m L of TCE. In this experiment, two different
serums (horse serum and fetal bovine serum) were also tested to understand the importance of
serum quality in the transformation assay. Preliminary toxicity assay was performed to select
dose levels which had 50-90% cell survival. One week after dosing, the cell colonies were fixed
and counted for variability determination and examination of individual colonies for the evidence
of morphological transformation. No significant change in morphological cell transformation
was obtained. Furthermore, no significant changes were seen in transformed colonies when
tested in different serum. However, these studies are of limited use for determining the
genotoxic potential of TCE because they did not examine the foci for mutations, for instance in
oncogenes or tumor suppressor genes.
4.2.1.1.14. Summary
Evidence from a number of different analyses and a number of different laboratories
using a fairly complete array of endpoints suggests that TCE, following metabolism, has the
potential to be genotoxic. A series of carefully controlled studies evaluating TCE itself (without
mutagenic stabilizers and without metabolic activation) found it to be incapable of inducing gene
mutations in most standard mutation bacterial assays (Baden et al., 1979; Bartsch et al., 1979;
Crebelli et al., 1982; Henschler et al., 1977; Mortelmans et al., 1986; Shimada et al., 1985;
Simmon et al., 1977; Waskell, 1978). Therefore, it appears that it is unlikely that TCE is a
direct-acting mutagen, though TCE has shown potential to affect DNA and chromosomal
structure. Low, but positive responses were observed in the TA100 strain in the presence of S9
metabolic activation, even when genotoxic stabilizers were not present, suggesting metabolites
of TCE are genotoxic. TCE is also positive in some but not all fungal and yeast systems (Callen
et al., 1980; Crebelli et al., 1985; Koch et al., 1988; Rossi et al., 1983). Data from human
epidemiological studies support the possible mutagenic effect of TCE leading to VHL gene
damage and subsequent occurrence of renal cell carcinoma. Association of increased VHL
This document is a draft for review purposes only and does not constitute Agency policy.
57 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
mutation frequency in TCE-exposed renal cell carcinoma cases has been observed (Brauch et al.,
1999; Brauch et al., 2004; Briining et al., 1997b).
TCE can lead to binding to nucleic acids and proteins (Bergman, 1983; DiRenzo et al.,
1982; Kautiainen et al., 1997; Mazzullo et al., 1992; Miller and Guengerich, 1983), and such
binding appears to be due to conversion to one or more reactive metabolites. For instance,
increased binding was observed in samples bioactivated with mouse and rat microsomal fractions
(Banerjee and Van Duuren, 1978; DiRenzo et al., 1982; Mazzullo et al., 1992; Miller and
Guengerich, 1983). DNA binding is consistent with the ability to induce DNA and chromosomal
perturbations. Several studies report the induction of micronuclei in vitro and in vivo from TCE
exposure (Hrelia et al., 1994; Hu et al., 2008; Kligerman et al., 1994; Robbiano et al., 2004;
Wang et al., 2001). Reports of SCE induction in some studies are consistent with DNA effects,
but require further study (Gu et al., 1981a; Gu et al., 1981b; Kligerman et al., 1994; Nagaya et
al., 1989a; White etal., 1979).
Overall, evidence from a number of different analyses and a number of different
laboratories using various genetic endpoints indicates that TCE has a potential to induce damage
to the structure of the chromosome in a number of targets but has a more limited ability to induce
mutation in bacterial systems.
Below, the genotoxicity data for TCE metabolites TCA, DCA, TCOH, chloral hydrate,
DCVC, and DCVG are briefly reviewed. The contributions of these data are twofold. First, to
the extent that these metabolites may be formed in the in vitro and in vivo test systems for TCE,
they provide insight into what agent or agents may contribute to the limited activity observed
with TCE in these genotoxicity assays. Second, because the in vitro systems do not necessarily
fully recapitulate in vivo metabolism, the genotoxicity of the known in vivo metabolites
themselves provide data as to whether one may expect genotoxicity to contribute to the toxicity
of TCE following in vivo exposure.
4.2.2. Trichloroacetic Acid (TCA)
The TCE metabolite, TCA, has been studied using a variety of genotoxicity assay for its
genotoxic potential (see International Agency for Research on Cancer [IARC, 2004] for
additional information). Evaluation of in vitro studies of TCA must consider toxicity and
acidification of medium resulting in precipitation of proteins, as TCA is commonly used as a
reagent to precipitate proteins.
This document is a draft for review purposes only and does not constitute Agency policy.
5 8 DRAFT—DO NOT CITE OR QUOTE
-------
4.2.2.1.1. Bacterial Systems—Gene Mutations
1 TCA has been evaluated in a number of in vitro test systems including the bacterial
2 assays (Ames) using different S. typhimurium strains such as TA98, TA100, TA104, TA1535,
3 and RSJ100 (see Table 4-11). The majority of these studies did not report positive findings for
4 genotoxicity Shirasu et al., (1976); Nestmann et al., (1980) Moriya et al., (1983) (DeMarini et
5 al., 1994; Kargalioglu et al., 2002; Nelson et al., 2001; Rapson et al., 1980; Waskell, 1978).
6 Waskell (1978) studied the effect of TCA (0.45 mg/plate) on bacterial strains TA98 and TA100
7 both in
8 Table 4-11. Genotoxicity of trichloroacetic acid—bacterial systems
Test system/endpoint
Doses
(LED or HID)a
Resultsb
Reference
With
activation
Without
activation
1 Prophage induction, E. coli WP2s
10,000
-
-
DeMarini et al. (1994)
SOS chromotest, Escherichia coli PQ37
10,000
-
-
Giller et al. (1997)
S. typhimurium TA1535, 1536, 1537,
1538, reverse mutation
20 ng/plate
NT
-
Shirasu et al., (1976)
S. typhimurium TA100, 98, reverse
mutation
450 ng/plate
-
-
Waskell (1978)
S. typhimurium TA100, 1535, reverse
mutation
4,000 ng/plate
-
-
Nestmann et al., (1980)
S. typhimurium TA1537, 1538, 98, reverse
mutation
2,000 ng/plate
-
-
Nestmann et al., (1980)
S. typhimurium TA100, reverse mutation
520 ng/plate
NT
-
Rapson etal. (1980)
S. typhimurium TA100, 98, reverse
mutation
5,000 ng/plate
-
-
Moriya et al., (1983)
S. typhimurium TA100, reverse mutation
600 ppm
-
-
DeMarini et al. (1994)
S. typhimurium TA100, reverse mutation,
liquid medium
1,750
+
+
Giller etal. (1997)
S. typhimurium TA104, reverse mutation,
microsuspension
250 ng/plate
-
-
Nelson etal. (2001)
S. typhimurium TA100, RSJ100, reverse
mutation
16,300
-
-
Kargalioglu et al.
(2002)
S. typhimurium TA98, reverse mutation
13,100
-
-
Kargalioglu et al.
(2002)
S. typhimurium TA1535, SOS DNA repair
+
-
Ono et al. (1991)
10
11 1 LED = lowest effective dose; HID = highest ineffective dose; doses are in ng/mL for in vitro tests unless specified.
12 b Results: +, positive; negative; NT, not tested.
13
14 Table adapted from IARC monograph (2004a) and modified/updated for newer references.
This document is a draft for review purposes only and does not constitute Agency policy.
59 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
the presence and absence of S9. The author did not find any revertants at the maximum nontoxic
dose tested. Following exposure to TCA, Rapson et al. (1980) reported no change in mutagenic
activity in strain TA100 in the absence of S9. DeMarini et al. (1994) performed different studies
to evaluate the genotoxicity of TCA, including the Microscreen prophage-induction assay (TCA
concentrations 0-10 mg/mL) and use of the S. typhimurium TA100 strain using bag vaporization
technique (TCA concentrations 0-100 ppm), neither of which yielded positive results.
Nelson et al. (2001) reported no positive findings with TCA using a S. typhimurium
microsuspension bioassay (S. typhimurium strain TA104) following incubation of TCA for
various lengths of time, with or without rat cecal microbiota. Similarly, no activity was observed
in a study conducted by Kargalioglu et al. (2002) where S. typhimurium strains TA98, TA100,
and RSJ100 were exposed to TCA (0.1-100 mM) either in the presence or absence of S9
(Kargalioglu et al., 2002).
TCA was also negative in other bacterial systems. The SOS chromotest (which measures
DNA damage and induction of the SOS repair system) in E. coli PQ37, ± S9 (Giller et al., 1997)
evaluated the genotoxic activity of TCA ranging from 10-10,000 |ag/m L and did not find any
response. Similarly, TCA was not genotoxic in the Microscreen prophage-induction assay in E.
coli with TCA concentrations ranging from 0-10,000 |ig/mL, with and without S9 activation
(DeMarini etal., 1994).
However, TCA induced a small increase in SOS DNA repair (an inducible error-prone
repair system) in S. typhimurium strain TA1535 in the presence of S9 (Ono et al., 1991).
Furthermore, Giller et al. (1997) reported that TCA demonstrated genotoxic activity in an Ames
fluctuation test in S. typhimurium TA100 in the absence of S9 at noncytotoxic concentrations
ranging from 1,750-2,250 |ig/mL. The addition of S9 decreased the genotoxic response, with
effects observed at 3,000-7,500 |ig/mL. Cytotoxic concentrations in the Ames fluctuation assay
were 2,500 and 10,000 |ag/m L without and with microsomal activation, respectively.
4.2.2.1.2. Mammalian Systems
4.2.2.1.3. Gene mutations
The mutagenicity of TCA has also been tested in cultured mammalian cells (see
Table 4-12). Harrington-Brock et al. (1998) examined the potential of TCA to induce mutations
in L5178Y/TK ± -3.7.2C mouse lymphoma cells. In this study, mouse lymphoma cells were
incubated in culture medium treated with TCA concentrations up to 2,150 |ag/m L in the presence
This document is a draft for review purposes only and does not constitute Agency policy.
60 DRAFT—DO NOT CITE OR QUOTE
-------
1 of S9 metabolic activation and up to 3,400 |ag/mL in the absence of S9 mixture. In the presence
2 of S9, a doubling of mutant frequency was seen at concentrations of 2,250 |ag/m L and higher,
3 including several concentrations with survival >10%. In the absence of S9, TCA increased the
4 mutant frequency by twofold or greater only at concentrations of 2,000 |ag/m L or higher. These
5 results were obtained at <11% survival rates. The authors noted that the mutants included both
6 large-colony and small-colony mutants. The small-colony mutants are indicative of
7 chromosomal damage. It should be noted that no rigorous statistical evaluation was conducted
8 on these data. Cytotoxic and genotoxic effects of TCA were tested in a microplate-based
9 cytotoxicity test and a HGPRT gene mutation assay using Chinese hamster ovary K1 cells,
10 respectively (Zhang et al., 2010). TCA was the least cytotoxic when compared to six other
This document is a draft for review purposes only and does not constitute Agency policy.
61 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-12. TCA Genotoxicity—mammalian systems (both in vitro and in vivo)
Test system/endpoint
Doses
(LED or HID)a
Resultsb
Reference
With
activation
Without
activation
Gene mutation, mouse lymphoma L5178Y/TK ± cells, in vitro
3,000
(+)
?
Harrington-Brock et al.
(1998")
Gene mutation, Chinese hamster ovary cells in vitro, HGPRT gene mutation assay
10,000 nM
NT
-
Zhang et al. (2010)
DNA strand breaks, B6C3F1 mouse and Fischer 344 rat hepatocytes, in vitro
1,630
NT
-
Chang et al. (1992)
DNA strand breaks, human CCRF-CEM lymphoblastic cells, in vitro
1,630
NT
-
Chang et al. (1992)
DNA damage, Chinese hamster ovary cells, in vitro, comet assay
3 mM
NT
-
Plewa et al. (2002)
DNA strand breaks, B6C3F1 mouse liver, in vivo
1.0, oral, x 1
+
Nelson and Bull (1988)
DNA strand breaks, B6C3F1 mouse liver, in vivo
500, oral, x 1
+
Nelson etal. (1989)
DNA strand breaks, B6C3F1 mouse liver, in vivo
500, oral,
10 repeats
-
Nelson etal. (1989)
DNA strand breaks, B6C3F1 mouse liver and epithelial cells from stomach and
duodenum, in vivo
1,630, oral, x 1
-
Chang et al. (1992)
DNA strand breaks, male B6C3F1 mice, in vivo
500 (neutralized)
-
Styles et al. (1991)
Micronucleus formation, Swiss mice, in vivo
125, i.p., x 2
+
Bhunya and Behera (1987)
Micronucleus formation, female C57BL/6JfBL10/Alpk mouse bone-marrow
erythrocytes, in vivo
1,300, i.p., x 2
-
Mackay et al. (1995)
Micronucleus formation, male C57BL/6JfBL10/Alpk mouse bone-marrow
erythrocytes, in vivo
1,080, i.p., x 2
-
Mackay et al. (1995)
Micronucleus formation, Pleurodeles waltl newt larvae peripheral erythrocytes, in
vivo
80
+
Giller et al. (1997)
Chromosomal aberrations, Swiss mouse bone-marrow cells in vivo
125, i.p., x 1
+
Bhunya and Behera (1987)
100, i.p., x 5
+
Bhunya and Behera (1987)
500, oral, x 1
+
Bhunya and Behera (1987)
Chromosomal aberrations, chicken Gallus domesticus bone marrow, in vivo
200, i.p., x l
+
Bhunya and Jena (1996)
-------
This document is a draft for review purposes only and does not constitute Agency policy.
5/12/11 63 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-12. TCA Genotoxicity—mammalian systems (both in vitro and in vivo) (continued)
Test system/endpoint
Doses
(LED or HID)a
Resultsb
Reference
With Without
activation activation
Chromosomal aberrations, human lymphocytes, in vitro
5,000,
(neutralized)
-
Mackay et al. (1995)
Sperm morphology, Swiss mice, in vivo
125, i.p., x 5
+
Bhunya and Behera (1987)
a LED = lowest effective dose; HID = highest ineffective dose; doses are in ng/mL for in vitro tests; mg/kg for in vivo tests unless specified.
b Results: + = positive; (+) = weakly positive; - = negative; NT = not tested; ? = inconclusive.
Table adapted from IARC monograph (2004a) and modified/updated for newer references.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
haloacetic acids. TCA at concentrations 0, 200, 1,000, 5,000 and 10,000 |iM induced a visible
increase in mutant frequency but did not show any statistically significant increase at any of the
doses tested.
4.2.2.1.4. Chromosomal aberrations
Mackay et al. (1995) investigated the ability of TCA to induce chromosomal damage in
an in vitro chromosomal aberration assay using cultured human cells. The authors treated the
cells with TCA as free acid, both in the presence and absence of metabolic activation. TCA
induced chromosomal damage in cultured human peripheral lymphocytes at concentrations
(2,000 and 3,500 |ig/mL) that significantly reduced the pH of the medium. However, exposure
of cells to neutralized TCA did not have any effect even at a cytotoxic concentration of
5,000 |ig/mL. It is possible that the reduced pH was responsible for the TCA-induced
clastogenicity in this study. To further evaluate the role of pH changes in the induction of
chromosome damage, the authors isolated liver-cell nuclei from B6C3F1 mice and suspended in
a buffer at various pH levels. The cells were stained with chromatin-reactive (fluorescein
isothiocyanate) and DNA-reactive (propidium iodide) fluorescent dyes. A decrease in chromatin
staining intensity was observed with the decrease in pH, suggesting that pH changes,
independent of TCA exposure, can alter chromatin conformation. It was concluded by the
authors that TCA-induced pH changes are likely to be responsible for the chromosomal damage
induced by un-neutralized TCA. In another in vitro study, Plewa et al. (2002) evaluated the
induction of DNA strand breaks induced by TCA (1-25 mM) in CHO cells and did not observe
any genotoxicity.
4.2.2.1.5. Micronucleus
Relative genotoxicity of TCA was tested in a mouse in vivo system (see Table 4-12)
using three different cytogenetic assay (bone marrow chromosomal aberrations, micronucleus
and sperm-head abnormalities) (Bhunya and Behera, 1987) and for chromosomal aberrations in
chicken (Bhunya and Jena, 1996). TCA induced a variety of anomalies including micronucleus
in the bone marrow of mice and chicken. A small increase in the frequency of micronucleated
erythrocytes at 80 |ag/m L in a newt (Pleurodeles waltl larvae) micronucleus test was observed in
response to TCA exposure (Giller et al., 1997). Mackay et al. (1995) investigated the ability of
TCA to induce chromosomal DNA damage in the in vivo bone-marrow micronucleus assay in
mice. C57BL mice were given TCA intraperitoneally at doses of 0, 337, 675, or
This document is a draft for review purposes only and does not constitute Agency policy.
65 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
1,080 mg/kg-day for males and 0, 405, 810, or 1,300 mg/kg-day for females for two consecutive
days, and bone-marrow samples were collected 6 and 24 hours after the last dose. The
administered doses represented 25, 50, and 80% of the median lethal dose, respectively. No
treatment-related increase in micronucleated polychromatic erythrocytes was observed.
4.2.2.1.6. Other DNA damage studies
DNA unwinding assays have been used as indicators of single strand breaks and are
discussed in detail in Section E.2.3. Studies were conducted on the ability of TCA to induce
single-strand breaks (Chang et al., 1992; Nelson and Bull, 1988; Nelson et al., 1989; Styles et al.,
1991; see Table 4-12). Nelson and Bull (1988) evaluated the ability of TCA and other
compounds to induce single-strand DNA breaks in vivo in Sprague-Dawley rats and B6C3F1
mice. Single oral doses were administered to three groups of three animals, with an additional
group as a vehicle control. Animals were sacrificed after 4 hours, and 10% liver suspensions
were analyzed for single-strand DNA breaks by the alkaline unwinding assay. Dose-dependent
increases in single-strand DNA breaks were induced in both rats and mice, with mice being more
susceptible than rats. The lowest dose of TCA that produced significant SSBs was 0.6 mmol/kg
(98 mg/kg) in rats but 0.006 mmol/kg (0.98 mg/kg) in mice.
However, in a follow-up study, Nelson et al. (1989) male B6C3F1 mice were treated with
500 mg/kg TCA, and single strand breaks in whole liver homogenate were examined, and no
significant differences from controls were reported. Moreover, in the experiments in the same
study with DC A, increased single strand breaks were reported, but with no dose-response
between 10 and 500 mg/kg, raising concerns about the reliability of the DNA unwinding assay
used in these studies. For further details, see Section E.2.3. In an additional follow-up
experiment with a similar experimental paradigm, Styles et al. (1991) tested TCA for its ability
to induce strand breaks in male B6C3F1 mice in the presence and absence of liver growth
induction. The test animals were given one, two, or three daily doses of neutralized TCA
(500 mg/kg) by gavage and killed 1 hour after the final dose. Additional mice were given a
single 500-mg/kg gavage dose and sacrificed 24 hours after treatment. Liver nuclei DNA were
isolated, and the induction of single strand breaks was evaluated using the alkaline unwinding
assay. Exposure to TCA did not induce strand breaks under the conditions tested in this assay.
In a study by Chang et al. (1992), administration of single oral doses of TCA (1-10 mmol/kg) to
B6C3F1 mice did not induce DNA strand breaks in a dose-related manner as determined by the
alkaline unwinding assay. No genotoxic activity (evidence for strand breakage) was detected in
F344 rats administered by gavage up to 5 mmol/kg (817 mg/kg).
This document is a draft for review purposes only and does not constitute Agency policy.
66 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
In summary, although Nelson and Bull (1988) report effects on DNA unwinding for TCE
and its metabolites with DCA having the highest activity and TCA the lowest, Nelson et al.
(1989), using the same assay, reported no effect for TCA and the same effect at 10 and
500 mg/kg for DCA in mice. Moreover, Styles et al. (1991) did not find a positive result for
TCA using the same paradigm as Nelson and Bull (1988) and Nelson et al. (1989). Furthermore,
Chang et al. (1992) also did not find increased single strand breaks for TCA exposure in rats,
(see Section E.2.4.3).
4.2.2.1.7. Summary
In summary, TCA has been studied using a variety of genotoxicity assays, including the
recommended battery. No mutagenicity was reported in S. typhimurium strains in the presence
or absence of metabolic activation or in an alternative protocol using a closed system, except in
one study on strain TA100 using a modified protocol in liquid medium. This is largely
consistent with the results from TCE, which was negative in most bacterial systems except some
studies with the TA100 strain. Mutagenicity in mouse lymphoma cells was only induced at
cytotoxic concentrations. Measures of DNA-repair responses in bacterial systems have been
inconclusive, with induction of DNA repair reported in S. typhimurium but not in E. coli.
TCA-induced clastogenicity may be secondary to pH changes and not a direct effect of TCA.
4.2.3. Dichloroacetic Acid (DCA)
DCA is another metabolite of TCE that has been studied using a variety of genotoxicity
assay for its genotoxic potential (see Tables 4-13 and 4-14; see IARC (2004a) for additional
information).
4.2.3.1.1. Bacterial and Fungal Systems—Gene Mutations
Studies were conducted to evaluate mutagenicity of DCA in different S. typhimurium and
E. coli strains (DeMarini et al., 1994; Fox et al., 1996a; Fox et al., 1996b; Giller et al., 1997;
Herbert et al., 1980; Kargalioglu et al., 2002; Nelson et al., 2001; Waskell, 1978). DCA was
mutagenic in three strains of S. typhimurium: strain TA100 in three of five studies, strain RSJ100
in a single study, and strain TA98 in two of three studies. DCA failed to induce point mutations
in other strains of S. typhimurium (TA104, TA1535, TA1537, and TA1538) or in E. coli strain
This document is a draft for review purposes only and does not constitute Agency policy.
67 DRAFT—DO NOT CITE OR QUOTE
-------
1 WP2uvrA. In one study, DCA caused a weak induction of SOS repair in E. coli strain PQ37
2 (Gilleretal., 1997).
3 DeMarini et al. (1994), in the same study as described in the TCA section of this section,
4 also studied DCA as one of their compounds for analysis. In the prophage-induction assay using
5 E. coli, DCA, in the presence of S9, was genotoxic producing 6.6-7.2 plaque-forming units
6 (PFU)/mM and slightly less than threefold increase in PFU/plate in the absence of S9. In the
7 second set of studies, which involved the evaluation of DCA at concentrations of 0-600 ppm for
This document is a draft for review purposes only and does not constitute Agency policy.
68 DRAFT—DO NOT CITE OR QUOTE
-------
L/l
K>
Table 4-13. Genotoxicity of dichloroacetic acid (bacterial systems)
Test system/endpoint
Doses
(LED or HID)a
Resultsb
Reference
With
activation
Without
activation
1 Prophage induction, E. coli WP2s
2,500
+
-
DeMarini et al. (1994)
SOS chromotest, E. coli PQ37
500
-
(+)
Gilleretal. (1997)
S. typhimurium, DNA repair-deficient strains TS24, TA2322, TA1950
31,000
-
-
Waskell (1978)
S. typhimurium TA100, TA1535, TA1537, TA1538, reverse mutation
-
-
Herbert et al. (1980)
S. typhimurium TA100, reverse mutation
50
+
+
DeMarini et al. (1994)
S. typhimurium TA100,TA1535, TA1537, TA98, reverse mutation
5,000
-
-
Fox et al., (1996b)
S. typhimurium TA100, reverse mutation, liquid medium
100
+
+
Gilleretal. (1997)
S. typhimurium RSJ100, reverse mutation
1,935
-
+
Kargalioglu et al. (2002)
S. typhimurium TA104, reverse mutation, microsuspension
150 ng/plate
-
-
Nelson etal. (2001)
S. typhimurium TA98, reverse mutation
10 ng/plate
(+)
-
Herbert et al. (1980)
5,160
-
+
Kargalioglu et al. (2002)
S. typhimurium TA100, reverse mutation
1,935
+
+
Kargalioglu et al. (2002)
E. coli WP2uvrA, reverse mutation
5,000
-
-
Fox etal., (1996b)
S?
>!
rs
s;
>1
£
st
G\ -3
§
>!
>1
o
a
a
HH
0 °
HH ^
'/O >!
fel
H §
1 >!
81
2; ^
1—1 s
H 2.
«£
o
c
o
H
W
aLED = lowest effective dose; HID = highest ineffective dose; doses are in ng/mL for in vitro tests unless specified.
b Results: + = positive; (+) = weakly positive; - = negative.
Table adapted from IARC monograph (2004a) and modified/updated for newer references.
-------
Table 4-14. Genotoxicity of dichloroacetic acid—mammalian systems
Test system/endpoint
Doses
(LED or HID)a
Resultsb
Reference
With
activation
Without
activation
Gene mutation, mouse lymphoma cell line L5178Y/TK ± in vitro
5,000
-
-
Fox etal., (1996b)
Gene mutation, mouse lymphoma cell line L5178Y/TK ± -3.7.2C in vitro
400
NT
+
Harrington-Brock et al.
(1998)
Gene mutation, Chinese hamster ovary cells in vitro, HGPRT gene mutation
assay
1,000 nM
NT
+
Zhang et al. (2010)
DNA strand breaks and alkali-labile damage, Chinese hamster ovary cells in
vitro (single-cell gel electrophoresis assay)
3,225 ng/mL
NT
-
Plewa et al. (2002)
DNA strand breaks, B6C3F1 mouse hepatocytes in vitro
2,580
NT
-
Chang et al. (1992)
DNA strand breaks, Fischer 344 rat hepatocytes invito
1,290
NT
-
Chang et al. (1992)
Micronucleus formation, mouse lymphoma L5178Y/TK ± -3.7.2C cell line in
vitro
800
NT
-
Harrington-Brock et al.
(1998)
Chromosomal aberrations, Chinese hamster ovary in vitro
5,000
-
-
Fox etal., (1996b)
Chromosomal aberrations, mouse lymphoma L5178Y/Tk ± -3.7.2C cell line
in vitro
600
NT
+
Harrington-Brock et al.
(1998)
Aneuploidy, mouse lymphomaL5178Y/Tk± -3.7.2C cell line invito
800
NT
-
Harrington-Brock et al.
(1998)
DNA strand breaks, human CCRF-CEM lymphoblastoid cells invito
1,290
NT
-
Chang et al. (1992)
DNA strand breaks, male B6C3F1 mouse liver in vivo
13, oral, x 1
+
Nelson and Bull (1988)
DNA strand breaks, male B6C3F1 mouse liver in vivo
10, oral, x 1
+
Nelson etal. (1989)
DNA strand breaks, male B6C3F1 mouse liver in vivo
1,290, oral, x 1
-
Chang et al. (1992)
DNA strand breaks, male B6C3F1 mouse splenocytes in vivo
1,290, oral, x 1
-
Chang et al. (1992)
DNA strand breaks, male B6C3F1 mouse epithelial cells from stomach and
duodenum in vivo
1,290, oral, x 1
-
Chang et al. (1992)
DNA strand breaks, male B6C3F1 mouse liver in vivo
5,000, dw, x 7-14 d
-
Chang et al. (1992)
-------
DNA strand breaks, alkali-labile sites, cross linking, male B6C3F1 mouse
blood leukocytes in vivo (single-cell gel electrophoresis assay)
3,500, dw, x 28 d
+
Fuscoe et al. (1996)
Table 4-14. Genotoxicity of Dichloroacetic Acid—mammalian systems (continued)
Test system/endpoint
Doses
(LED or HID)a
Resultsb
Reference
With Without
activation activation
DNA strand breaks, male Sprague-Dawley rat liver in vivo
30, oral, x 1
+
Nelson and Bull (1988)
DNA strand breaks, male Fischer 344 rat liver in vivo
645, oral, x 1
-
Chang et al. (1992)
DNA strand breaks, male Fischer 344 rat liver in vivo
2,000, dw, x 30 wk
-
Chang et al. (1992)
Gene mutation, lacl transgenic male B6C3F1 mouse liver assay in vivo
1,000, dw, x 60 wk
+
Leavitt et al. (1997)
Micronucleus formation, male B6C3F1 mouse peripheral erythrocytes in vivo
3,500, dw, x 9 d
+
Fuscoe et al. (1996)
Micronucleus formation, male B6C3F1 mouse peripheral erythrocytes in vivo
3,500, dw, x 28 d
-
Fuscoe et al. (1996)
Micronucleus formation, male B6C3F1 mouse peripheral erythrocytes in vivo
3,500, dw, x 10 wk
+
Fuscoe et al. (1996)
Micronucleus formation, male and female Crl:CD (S-D) BR rat bone-marrow
erythrocytes in vivo
1,100, i.v., x 3
-
Fox et al., (1996b)
Micronucleus formation, Pleurodeles waltl newt larvae peripheral
erythrocytes in vivo
80 d
—
Giller et al. (1997)
a LED = lowest effective dose; HID = highest ineffective dose; doses are in ng/mL for in vitro tests; mg/kg for in vivo tests unless specified; dw = drinking-water
(in mg/L); d = day; wk = week; i.v. = intravenous.
b Results: + = positive; - = negative; NT = not tested.
Table adapted from IARC monograph (2004a) and modified/updated for newer references.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
mutagenicity in S. typhimurium TA100 strain, DC A was mutagenic both in the presence and
absence of S9, producing three-five times increases in the revertants/plate compared to the
background. The lowest effective concentration for DC A without S9 was 100 ppm and 50 ppm
in the presence of S9. In the third and most important study, mutation spectra of DC A were
determined at the base-substitution allele hisG46 of S. typhimurium TA100. DCA-induced
revertants were chosen for further molecular analysis at concentrations that produced mutant
yields that were two-fivefold greater than the background. The mutation spectra of DCA were
significantly different from the background mutation spectrum. Thus, despite the modest
increase in the mutant yields (three-five times) produced by DCA, the mutation spectra confirm
that DCA is mutagenic. DCA primarily induced GC-AT transitions.
Kargalioglu et al. (2002) analyzed the cytotoxicity and mutagenicity of the drinking
water disinfection by-products including DCA in S. typhimurium strains TA98, TA100, and
RSJ100 ± S9. DCA was mutagenic in this test although the response was low when compared to
other disinfection by-products tested in strain TA100. This study was also summarized in a
review by Plewa et al. (2002). Nelson et al. (2001) investigated the mutagenicity of DCA using
a S. typhimurium microsuspension bioassay following incubation of DCA for various lengths of
time, with or without rat cecal microbiota. No mutagenic activity was detected for DCA with
S. typhimurium strain TA104.
Although limited data, it appears that DCA has mutagenic activity in the S. typhimurium
strains, particularly TA100.
4.2.3.1.2. Mammalian Systems
4.2.3.1.3. Gene mutations
The mutagenicity of DCA has been tested in mammalian systems, particularly, mouse
lymphoma cell lines in vitro (Fox et al., 1996b; Harrington-Brock et al., 1998); and lacl
transgenic mice in vivo (Leavitt et al., 1997). Harrington-Brock et al. (1998) evaluated DCA for
it mutagenic activity in L5178Y/TK ± (-) 3.7.2C mouse lymphoma cells. A dose-related
increase in mutation (and cytotoxic) frequency was observed at concentrations between 100 and
800 |ig/mL, Most mutagenic activity of DCA at the Tk locus was due to the production of
small-colony Tk mutants (indicating chromosomal mutations). Different pH levels were tested
in induction of mutant frequencies and it was determined that the mutagenic effect observed was
due to the chemical and not pH effects.
This document is a draft for review purposes only and does not constitute Agency policy.
72 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Mutation frequencies were studied in male transgenic B6C3F1 mice harboring the
bacterial laclgene administered DCA at either 1.0 or 3.5 g/L in drinking water (Leavitt et al.,
1997). No significant difference in mutant frequency was observed after 4 or 10 weeks of
treatment in both the doses tested as compared to control. However, at 60 weeks, mice treated
with 1.0 g/L DCA showed a slight increase (1.3-fold) in the mutant frequency over the control,
but mice treated with 3.5 g/L DCA had a 2.3-fold increase in the mutant frequency. Mutational
spectra analysis revealed that -33% had G:C-A:T transitions and 21% had G:C-T:A
transversions and this mutation spectra was different than that was seen in the untreated animals,
indicating that the mutations were likely induced by the DCA treatment. The authors conclude
that these results are consistent with the previous observation that the proportion of mutations at
T:A sites in codon 61 of the H-ras gene was increased in DCA-induced liver tumors in B6C3F1
mice (Leavitt et al., 1997).
Zhang et al. (2010) tested the cytotoxic and genotoxic effects of DCA in a
microplate-based cytotoxicity test and HGPRT gene mutation assay using Chinese hamster ovary
K1 cells, respectively. The concentrations at which these tests were conducted was 0, 200,
1,000, 5,000 and 10,000 |iM. Two parameters were used to indicate chronic cytotoxicity: the
lowest cytotoxic concentration and the percent Cl/2 value. The lowest cytotoxic concentration
for DCA was 2.87 x 10 M. Statistically significant increase in HGPRT mutant frequency was
observed at concentration 1,000 |iM and above.
4.2.3.1.4. Chromosomal aberrations and micronucleus
Harrington-Brock et al. (1998) evaluated DCA for its potential to induce chromosomal
aberrations in DCA-treated (0, 600, and 800 |ig/mL) mouse lymphoma cells. A clearly positive
induction of aberrations was observed at both concentrations tested. No significant increase in
micronucleus was observed in DCA-treated (0, 600, and 800 |ig/mL) mouse lymphoma cells
(Harrington-Brock et al., 1998). However, no chromosomal aberrations were found in Chinese
hamster ovary cells exposed to DCA (Fox et al., 1996b)
Fuscoe et al. (1996) investigated in vivo genotoxic potential of DCA in bone marrow and
blood leukocytes using the peripheral-blood-erythrocyte micronucleus assay (to detect
chromosome breakage and/or malsegregation) and the alkaline single cell gel electrophoresis
(comet) assay, respectively. Mice were exposed to DCA in drinking water, available ad libitum,
for up to 31 weeks. A statistically significant dose-related increase in the frequency of
micronucleated PCEs was observed following subchronic exposure to DCA for 9 days.
Similarly, a significant increased was also observed when exposed for >10 weeks particularly at
This document is a draft for review purposes only and does not constitute Agency policy.
73 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
the highest dose of DCA tested (3.5 g/L). DNA cross-linking was observed in blood leukocytes
in mice exposed to 3.5 g/L DCA for 28 days. These data provide evidence that DCA may have
some potential to induce chromosome damage when animals were exposed to concentrations
similar to those used in the rodent bioassay.
4.2.3.1.5. Other DNA damage studies
Nelson and Bull (1988) and Nelson et al. (1989) have been described above in
Section 4.2.2.4 and E.2.3, with positive results for DNA unwinding for DCA, though Nelson
et al. (1989) reported the same response at 10 and 500 mg/kg in mice, raising concerns about the
reliability of the assay in these studies. Chang et al. (1992) conducted both in vitro and in vivo
studies to determine the ability of DCA to cause DNA damage. Primary rat (Fischer 344)
hepatocytes and primary mouse hepatocytes treated with DCA for 4 hours did not in induce
DNA single strand breaks as detected by alkaline DNA unwinding assay. No DNA strand breaks
were observed in human CCRF-CEM lymphoblastoid cells in vitro exposed to DCA. Similarly,
analysis of the DNA single strand breaks in mice killed 1 hour after a single dose of 1, 5 or
10 mM/kg DCA did not cause DNA damage. None of the Fischer 344 rats killed 4 hours after a
single gavage treatment (1-10 mM/kg) produced any detectable DNA damage.
4.2.3.1.6. Summary
In summary, DCA has been studied using a variety but limited number of genotoxicity
assays. Within the available data, DCA has been demonstrated to be mutagenic in the
S. typhimurium assay, particularly in strain TA100, the in vitro mouse lymphoma assay and in
vivo cytogenetic and gene mutation assays. DCA can cause DNA strand breaks in mouse and rat
liver cells following in vivo administration by gavage.
4.2.4. Chloral Hydrate
Chloral hydrate has been evaluated for its genotoxic potential using a variety of
genotoxicity assays (see Tables 4-15, 4-16, and 4-17). These data are particularly important
because it is known that a large flux of TCE metabolism leads to chloral hydrate as an
intermediate, so a comparison of their genotoxicity profiles is likely to be highly informative.
This document is a draft for review purposes only and does not constitute Agency policy.
74 DRAFT—DO NOT CITE OR QUOTE
-------
4.2.4.1.1. DNA Binding Studies
1 Limited analysis has been performed examining DNA binding potential of chloral
2 hydrate (Keller and Hd'A, 1988; Ni et al., 1995; Von Tungeln et al., 2002). Keller and Heck
3 (1988) conducted both in vitro and in vivo experiments using B6C3F1 mouse strain. The mice
4 were pretreated with 1,500 mg/kg TCE for 10 days and then given 800 mg/kg [14C] chloral. No
5 detectable covalent binding of [14C] to DNA in the liver was observed. Another study with in
6 vivo exposures to nonradioactive chloral hydrate at a concentration of 1,000 and 2,000 nmol in
7 mice B6C3F1 demonstrated an increase in malondialdehyde-derived and 8-oxo-2'-
8 deoxyguanosine adducts in liver DNA (Von Tungeln et al., 2002). Ni et al. (1995) observed
This document is a draft for review purposes only and does not constitute Agency policy.
75 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-15. Chloral hydrate genotoxicity: bacterial, yeast and fungal systems
Test system/endpoint
Doses
(LED or HID)a
Resultsb
Reference
With
activation
Without
activation
SOS chromotest, Escherichia coli PQ37
10,000
-
-
Gilleretal. (1995)
S. typhimurium TA100, TA1535, TA98, reverse mutation
10,000
-
-
Waskell (1978)
S. typhimurium TA100, TA1537, TA1538, TA98, reverse mutation
1,000
+
+
Haworth et al. (1983)
S. typhimurium TA100, reverse mutation
5,000 ng/plate
-
-
Leuschner and Leuschner (1991)
S. typhimurium TA100, reverse mutation
2,000 ng/plate
+
+
Ni et al. (1994)
S. typhimurium TA100, reverse mutation, liquid medium
300
+
-
Gilleretal. (1995)
S. typhimurium TA100, TA104, reverse mutation
1,000 ng/plate
+
+
Beland (1999)
S. typhimurium TA104, reverse mutation
1,000 ng/plate
+
+
Ni et al. (1994)
S. typhimurium TA1535, reverse mutation
1,850
-
-
Leuschner and Leuschner (1991)
S. typhimurium TA1535, TA1537 reverse mutation
6,667
-
-
Haworth etal. (1983)
S. typhimurium TA1535, reverse mutation
10,000
-
-
Beland (1999)
S. typhimurium TA98, reverse mutation
7,500
-
-
Haworth etal. (1983)
S. typhimurium TA98, reverse mutation
10,000 ng/plate
-
+
Beland (1999)
A.nidulans, diploid strain 35X17, mitotic cross-overs
1,650
NT
-
Crebelli et al. (1985)
A. nidulans, diploid strain 30, mitotic cross-overs
6,600
NT
-
Kafer (1986)
A. nidulans, diploid strain NH, mitotic cross-overs
1,000
NT
-
Kappas (1989)
A. nidulans, diploid strain PI, mitotic cross-overs
990
NT
-
Crebelli et al. (1991)
A. nidulans, diploid strain 35X17, nondisjunctions
825
NT
+
Crebelli et al. (1985)
A. nidulans, diploid strain 30, aneuploidy
825
NT
+
Kafer (1986)
A. nidulans, haploid conidia, aneuploidy, polyploidy
1,650
NT
+
Kafer (1986)
A. nidulans, diploid strain NH, nondisjunctions
450
NT
+
Kappas (1989)
-------
Table 4-15. Chloral hydrate genotoxicity: bacterial, yeast and fungal systems (continued)
Test system/endpoint
Doses
(LED or HID)a
Resultsb
Reference
With
activation
Without
activation
A. nidulans, diploid strain PI, nondisjunctions
660
NT
+
Crebelli et al. (1991)
A. nidulans, haploid strain 35, hyperploidy
2,640
NT
+
Crebelli et al. (1991)
S. cerevisiae, meiotic recombination
3,300
NT
?
Sora and Agostini Carbone
(1987)
S. cerevisiae, disomy in meiosis
2,500
NT
+
Sora and Agostini Carbone
(1987)
S. cerevisiae, disomy in meiosis
3,300
NT
+
Sora and Agostini Carbone
(1987)
S. cerevisiae, D61.M, mitotic chr. malsegregation
1,000
NT
+
Albertini, (1990)
Drosophila melanogaster, somatic mutation wing spot test
825
+
Zordanetal. (1994)
Drosophila melanogaster, induction of sex-linked lethal mutation
37.2 feed
?
Beland (1999)
Drosophila melanogaster, induction of sex-linked lethal mutation
67.5 inj
-
Beland (1999)
aLED = lowest effective dose; HID = highest ineffective dose; doses are in |ig/mL for in vitro tests; inj = injection.
b Results: + = positive; - = negative; NT = not tested; ? = inconclusive.
Table adapted from IARC monograph (2004a) and modified/updated for newer references.
-------
Table 4-16.. Chloral hydrate genotoxicity: mammalian systems—all genetic endpoints, in vitro
Test system/endpoint
Doses
(LED or
HID)a
Resultsb
Reference
With
activation
Without
activation
DNA-protein cross-links, rat nuclei in vitro
41,250
NT
-
Keller and Heck (1988)
DNA single-strand breaks, rat primary hepatocytes in vitro
1,650
NT
-
Chang et al. (1992)
Gene mutation, mouse lymphoma L5178Y/TK ±, in vitro
1,000
(+)
Harrington-Brock et al. (1998)
Sister chromatid exchange, CHO cells, in vitro
100
+
+
Beland (1999)
Micronucleus formation, (kinetochore-positive), Chinese hamster CI cells, in vitro
165
NT
+
Degrassi and Tanzarella (1988)
Micronucleus formation, (kinetochore-negative), Chinese hamster CI cells, in vitro
250
NT
-
Degrassi and Tanzarella (1988)
Micronucleus formation, (kinetochore-positive), Chinese hamster LUC2 cells, in vitro
400
NT
+
Parry etal., (1990)
Micronucleus formation, (kinetochore-positive), Chinese hamster LUC2 cells, in vitro
400
NT
+
Lynch and Parry (1993)
Micronucleus formation, Chinese hamster V79 cells, in vitro
316
NT
+
Seelbach et al. (1993)
Micronucleus formation, mouse lymphoma L5178Y/TK ±, in vitro
1,300
NT
-
Harrington-Brock et al. (1998)
Micronucleus formation, mouse lymphoma L5178Y/TK ±„ in vitro
500
NT
+
Nesslany and Marzin (1999)
Chromosomal aberrations, Chinese Hamster CHED cells, in vitro
20
NT
+
Furnus et al. (1990)
Chromosomal aberrations, Chinese Hamster ovary cells, in vitro
1,000
+
+
Beland (1999)
Chromosomal aberrations, mouse lymphoma L5178Y/TK ± cells line, in vitro
1,250
NT
(+)
Harrington-Brock et al. (1998)
Aneuploidy, Chinese hamster CHED cells, in vitro
10
NT
+
Furnus et al. (1990)
Aneuploidy, primary Chinese hamster embryonic cells, in vitro
250
NT
+
Natarajan et al. (1993)
Aneuploidy, Chinese hamster LUC2p4 cells, in vitro
250
NT
+
Warretal. (1993)
Aneuploidy, mouse lymphoma L5178Y/TK . invito
1,300
NT
-
Harrington-Brock et al. (1998)
Tetraploidy and endoredupliation, Chinese hamster LUC2p4cells, in vitro
500
NT
+
Warretal. (1993)
Cell transformation, Syrian hamster embryo cells (24-h treatment)
350
NT
+
Gibson etal. (1995)
Cell transformation, Syrian hamster dermal cell line (24-h treatment)
50
NT
+
Parry et al. (1996)
DNA single-strand breaks, human lymphoblastoid cells, in vitro
1,650
NT
-
Chang et al. (1992)
-------
Table 4-16. Chloral hydrate genotoxicity: mammalian systems—all genetic endpoints, in vitro (continued)
Test system/endpoint
Doses
(LED or
HID)a
Resultsb
Reference
With
activation
Without
activation
Gene mutation, tk and hprt locus, human lymphoblastoid
1,000
NT
+
Beland (1999)
Sister chromatid exchanges, human lymphocytes, in vitro
54
NT
(+)
Guetal. (1981a)
Micronucleus formation, human lymphocytes, in vitro
100
-
+
Van Hummelen and Kirsch-
Volders (1992)
Micronucleus formation, human lymphoblastoid AHH-1 cell line, in vitro
100
NT
+
Parry et al. (1996)
Micronucleus formation, human lymphoblastoid MCL-5 cell line, in vitro
500
NT
-
Parry et al. (1996)
Micronucleus formation (kinetochore-positive), human diploid LEO fibroblasts, in vitro
120
NT
+
Bonatti et al. (1992)
Aneuploidy (double Y induction), human lymphocytes, in vitro
250
NT
+
Vagnarelli et al. (1990)
Aneuploidy (hyperdiploidy and hypodiploidy), human lymphocytes in vitro
50
NT
+
Sbrana et al. (1993)
Polyploidy, human lymphocytes, in vitro
137
NT
+
Sbrana et al. (1993)
C-Mitosis, human lymphocytes, in vitro
75
NT
+
Sbrana et al. (1993)
1 LED = lowest effective dose; HID = highest ineffective dose; doses are in ng/mL for in vitro tests.
b Results: + = positive; (+) = weakly positive in an inadequate study; - = negative; NT = not tested.
Table adapted from IARC monograph (2004a) and modified/updated for newer references.
-------
Table 4-17.. Chloral hydrate genotoxicity: mammalian systems—all genetic damage, in vivo
Test system/endpoint
Doses
(LED or HID)a
Resultsb
Reference
DNA single-strand breaks, male Sprague-Dawley rat liver
300, oral
+
Nelson and Bull (1988)
DNA single-strand breaks, male Fischer 344 rat liver
1,650, oral
-
Chang et al. (1992)
DNA single-strand breaks, male B6C3F1 mouse liver
100, oral
+
Nelson and Bull (1988)
DNA single-strand breaks, male B6C3F1 mouse liver
825, oral
-
Chang et al. (1992)
Micronucleus formation, male and female NMRI mice, bone-marrow erythrocytes
500, i.p.
-
Leuschner and Leuschner (1991)
Micronucleus formation, BALB/c mouse spermatids
83, i.p.
-
Russo and Levis, 1992
Micronucleus formation, male BALB/c mouse bone-marrow erythrocytes and early
spermatids
83, i.p.
+
Russo and Levis, 1992
Micronucleus formation, male BALB/c mouse bone-marrow erythrocytes
200, i.p.
+
Russo et al. (1992)
Micronucleus formation, male F1 mouse bone-marrow erythrocytes
400, i.p.
-
Leopardi et al. (1993)
Micronucleus formation, C57B1 mouse spermatids
41, i.p.
+
Allen etal., 1994
Micronucleus formation, male Swiss CD-I mouse bone-marrow erythrocytes
200, i.p.
+
Marrazzini et al., (1994)
Micronucleus formation, B6C3F1 mouse spermatids after spermatogonial stem-cell treatment
165, i.p.
+
Nutley et al. (1996)
Micronucleus formation, B6C3F1 mouse spermatids after meiotic cell treatment
413, i.p.
-
Nutley et al. (1996)
Micronucleus formation, male Fl, BALB/c mouse peripheral-blood erythrocytes
200, i.p.
Grawe et al. (1997)
Micronucleus formation, male B6C3F1 mouse bone-marrow erythrocytes
500, i.p., x 3
+
Beland (1999)
Micronucleus formation, infants, peripheral lymphocytes
50, oral
+
Ikbal et al. (2004)
Chromosomal aberrations, male and female Fl mouse bone marrow cells
600, i.p.
-
Xu and Alder (1990)
Chromosomal aberrations, male and female Sprague-Dawley rat bone-marrow cells
1,000, oral
Leuschner and Leuschner (1991)
Chromosomal aberrations, BALB/c mouse spermatogonia treated
83, i.p.
-
Russo and Levis, (1992b)
Chromosomal aberrations, Fl mouse secondary spermatocytes
82.7, i.p.
+
Russo et al. (1984)
Chromosomal aberrations, male Swiss CD-I mouse bone-marrow erythrocytes
400, i.p.
Marrazzini et al. (1994)
Chromosomal aberrations, ICR mouse oocytes
600, i.p.
-
Mailhes et al. (1993)
-------
Table 4-17. Chloral hydrate genotoxicity: mammalian systems—all genetic damage, in vivo (continued)
Test system/endpoint
Doses
(LED or
HID)a
Resultsb
Reference
Micronucleus formation, infants, peripheral lymphocytes
50, oral
+
Ikbal et al. (2004)
Polyploidy, male and female Fl, mouse bone-marrow cells
600, i.p.
-
Xu and Alder (1990)
Aneuploidy Fl mouse secondary spermatocytes
200, i.p.
+
Miller and Adler (1992)
Aneuploidy, male Fl mouse secondary spermatocytes
400, i.p.
Leopardi et al. (1993)
Hyperploidy, male Swiss CD-I mouse bone-marrow erythrocytes
200, i.p.
+
Marrazzini et al., 1994
aLED = lowest effective dose; HID = highest ineffective dose; doses are in mg/kg BW for in vivo tests, i.p. = intraperitoneally.
b Results: + = positive; - = negative.
Table adapted from IARC monograph (2004a) and modified/updated for newer references.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
malondialdehyde adducts in calf thymus DNA when exposed to chloral hydrate and microsomes
from male B6C3F1 mouse liver.
Keller and Heck (1988) investigated the potential of chloral to form DNA-protein
cross-links in rat liver nuclei using concentrations 25, 100, or 250 mM. No statistically
significant increase in DNA-protein cross-links was observed. DNA and RNA isolated from the
[14C] chloral-treated nuclei did not have any detectable [14C] bound. However, the proteins from
choral-treated nuclei did have a concentration-related binding of [14C],
4.2.4.1.2. Bacterial and Fungal Systems—Gene Mutations
Chloral hydrate induced gene mutations in S. typhimurium TA100 and TA104 strains, but
not in most other strains assayed. Four of six studies of chloral hydrate exposure in
S. typhimurium TA100 and two of two studies in S. typhimurium TA104 were positive for
revertants (Beland, 1999; Giller et al., 1995; Haworth et al., 1983; Ni et al., 1994). Waskell
(1978) studied the effect of chloral hydrate along with TCE and its other metabolites. Chloral
hydrate was tested at different doses (1.0-13 mg/plate) in different S. typhimurium strains
(TA98, TA100, TA1535) for gene mutations using Ames assay. No revertant colonies were
observed in strains TA98 or TA1535 both in the presence and absence of S9 mix. Similar results
were obtained by Leuschner and Leuschner (1991). However, in TA100, a dose-dependent
statistically significant increase in revertant colonies was obtained both in the presence and
absence of S9. It should be noted that chloral hydrate that was purchased from Sigma was
recrystallized from one to six times from chloroform and the authors describe this as crude
chloral hydrate. However, this positive result is consistent with other studies in this strain as
noted above. Furthermore, Giller et al. (1995) studied chloral hydrate genotoxicity in
three short-term tests. Chloral-induced mutations in strain TA100 of S. typhimurium (fluctuation
test). Similar results were obtained by Haworth et al. (1983). These are consistent with several
studies of TCE, in which low, but positive responses were observed in the TA100 strain in the
presence of S9 metabolic activation, even when genotoxic stabilizers were not present.
A significant increase in mitotic segregation was observed in Aspergillus nidulans when
exposed to 5 and 10 mM chloral hydrate (Crebelli et al., 1985). Studies of mitotic crossing-over
in Aspergillus nidulans have been negative while these same studies were positive for
aneuploidy (Crebelli etal., 1985, 1991; Kafer, 1986; Kappas, 1989).
Two studies were conducted in Saccharomyces cerevisiae to understand the
chromosomal malsegregation as a result of exposure to chloral hydrate (Albertini, 1990; Sora
and Agostini Carbone, 1987). Chloral hydrate (1-25 mM) was dissolved in sporulation medium
This document is a draft for review purposes only and does not constitute Agency policy.
82 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
and the frequencies of various meiotic events such as recombination, disomy were analyzed.
Chloral hydrate inhibited sporulation as a function of dose and increased diploid and disomic
clones . Chloral hydrate was also tested for mitotic chromosome malsegregation using
Saccharomyces cerevisiae D61.M (Albertini, 1990). The tester strain was exposed to a dose
range of 1-8 mg/mL. An increase in the frequency of chromosomal malsegregation was
observed as a result of exposure to chloral hydrate.
Limited analysis of chloral hydrate mutagenicity has been performed in Drosophila
(Beland, 1999; Zordan et al., 1994). Of these two studies, chloral hydrate was positive in the
somatic mutation wing spot test (Zordan et al., 1994), equivocal in the induction of sex-linked
lethal mutation when in feed but negative when exposed via injection (Beland, 1999).
4.2.4.1.3. Mammalian Systems
4.2.4.1.4. Gene mutations
Harrington-Brock et al. (1998) noted that chloral hydrate-induced concentration related
cytotoxicity in TK± mouse lymphoma cell lines without S9 activation. A nonstatistical increase
in mutant frequency was observed in cells treated with chloral hydrate. The mutants were
primarily small colony TK mutants, indicating that most chloral hydrate-induced mutants
resulted from chromosomal mutations rather than point mutations. It should be noted that in
most concentrations tested (350-1,600 ng/mL), cytotoxicity was observed. Percentage cell
survival ranged from 96-4%.
4.2.4.1.5. Micronucleus
Micronuclei induction following exposure to chloral hydrate is positive in most test
systems in both in vitro and in vivo assays, although some negative tests do also exist (Allen et
al., 1994; Harrington-Brock et al., 1998; Marrazzini et al., 1994) (Beland, 1999; Bonatti et al.,
1992; Degrassi and Tanzarella, 1988; Giller et al., 1995; Grawe et al., 1997; Ikbal et al., 2004;
Leopardi et al., 1993; Leuschner and Leuschner, 1991; Lynch and Parry, 1993; Nesslany and
Marzin, 1999; Nutley et al., 1996; Parry et al., 1996; Russo and Levis, 1992a; Russo and Levis,
1992b; Russo et al., 1992; Seelbach et al., 1993; Van Hummelen and Kirsch-Volders, 1992).
Some studies have attempted to make inferences regarding aneuploidy induction or
clastogenicity as an effect of chloral hydrate. Aneuploidy results from defects in chromosome
This document is a draft for review purposes only and does not constitute Agency policy.
83 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
segregation during mitosis and is a common cytogenetic feature of cancer cells (see
Section E.3.1.5).
Giller et al. (1995) studied chloral hydrate genotoxicity in three short-term tests. Chloral
hydrate caused a significant increase in the frequency of micronucleated erythrocytes following
in vivo exposure of the amphibian Pleurodeles waltl newt larvae.
Chloral hydrate induced aneuploidy in vitro in multiple Chinese hamster cell lines
(Furnus et al., 1990; Natarajan et al., 1993; Warr et al., 1993) and human lymphocytes (Sbrana et
al., 1993; Vagnarelli et al., 1990) but not mouse lymphoma cells Harrington-Brock et al. (1998).
In vivo studies performed in various mouse strains led to increased aneuploidy in spermatocytes
(Liang and Pacchierotti, 1988; Miller and Adler, 1992; Russo et al., 1984)but not oocytes
(Mailhes et al., (1993)) or bone marrow cells (Leopardi et al., 1993; Xu and Adler, 1990).
The potential of chloral hydrate to induce aneuploidy in mammalian germ cells has been
of particular interest since Russo et al. (1984) first demonstrated that chloral hydrate treatment of
male mice results in significant increase in frequencies of hyperploidy in metaphase II cells.
This hyperploidy was thought to have arisen from chromosomal nondisjunction in
premeiotic/meiotic cell division and may be a consequence of chloral hydrate interfering with
spindle formation [reviewed by Russo et al. (1984) and Liang and Brinkley (1985)]. Chloral
hydrate also causes meiotic delay, which may be associated with aneuploidy (Miller and Adler,
1992). Chloral hydrate has been shown to induce micronuclei but not structural chromosomal
aberrations in mouse bone-marrow cells. Micronuclei induced by nonclastogenic agents are
generally believed to represent intact chromosomes that failed to segregate into either
daughter-cell nucleus at cell division (Russo et al., 1992; Xu and Adler, 1990). Furthermore,
chloral hydrate-induced micronuclei in mouse bone-marrow cells (Russo et al., 1992) and in
cultured mammalian cells (Bonatti et al., 1992; Degrassi and Tanzarella, 1988) have shown to be
predominantly kinetochore-positive in composition upon analysis with immunofluorescent
methods. The presence of a kinetochore in a micronucleus is considered evidence that the
micronucleus contains a whole chromosome lost at cell division (Degrassi and Tanzarella, 1988;
Eastmond and Tucker, 1989; Hennig et al., 1988). Therefore, both TCE and chloral hydrate
appear to increase the frequency of micronuclei.
Allen et al. (1994) treated male C57B1/6J mice were given a single intraperitoneal
injection of 0, 41, 83, or 165 mg/kg chloral hydrate. Spermatids were harvested at 22 hours, 11,
13.5, and 49 days following exposure (Allen et al., 1994). Harvested spermatids were processed
to identify both kinetochore-positive micronucleus (aneugen) and kinetochore-negative
micronucleus (clastogen). All chloral hydrate doses administered 49 days prior to cell harvest
were associated with significantly increased frequencies of kinetochore-negative micronuclei in
This document is a draft for review purposes only and does not constitute Agency policy.
84 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
spermatids, however, dose dependence was not observed. This study is in contrast with other
studies (Bonatti et al., 1992; Degrassi and Tanzarella, 1988) who demonstrated predominantly
kinetochore-positive micronucleus.
The ability of chloral hydrate to induce aneuploidy and polyploidy was tested in human
lymphocyte cultures established from blood samples obtained from two healthy nonsmoking
donors (Sbrana et al., 1993). Cells were exposed for 72 and 96 hours at doses between 50 and
250 |ig/mL, No increase in percentage hyperdiploid, tetraploid, or endoreduplicated cells were
observed when cells were exposed to 72 hours at any doses tested. However, at 96 hours of
exposure, significant increase in hyperdiploid was observed at one dose (150 ng/mL) and was
not dose dependent. Significant increase in tetraploid was observed at dose 137 mg/mL, again,
no dose dependence was observed.
Ikbal et al. (2004) assessed the genotoxic effects in cultured peripheral blood
lymphocytes of 18 infants (age range of 31-55 days) before and after administration of a single
dose of chloral hydrate (50 mg/kg of body weight) for sedation before a hearing test for
micronucleus frequency. A significant increase in micronuclei frequency was observed after
administration of chloral hydrate.
4.2.4.1.6. Chromosomal aberrations
Several studies have included chromosomal aberration analysis in both in vitro and in
vivo systems exposed to chloral hydrate and have resulted in positive in in vitro
studies—although not all studies had statistically significant increase Harrington-Brock et al.
(1998) (Beland, 1999; Furnus et al., 1990).
Analysis of chloral hydrate treated mouse lymphoma cell lines for chromosomal
aberrations resulted in a nonsignificant increase in chromosomal aberrations Harrington-Brock
et al. (1998). However, it should be noted that the concentrations tested (1,250 and
1,300 |ig/mL) were cytotoxic (with a cell survival of 11 and 7%, respectively). Chinese hamster
embryo cells were also exposed to 0.001, 0.002, and 0.003% chloral hydrate for 1.5 hours
(Furnus et al., 1990). A nonstatistically significant increase in frequency of chromosomal
aberrations was observed only 0.002 and 0.003% concentrations, with the increase not
dose-dependent. In this study, it should be noted that the cells were only exposed for 1.5 hours
to chloral hydrate and cells were allowed to grow for 48 hours (two cell cycles) to obtain similar
mitotic index before analyzing for chromosomal aberrations. No information on cytotoxicity
was provided except that higher doses decreased the frequency of mitotic cells at the time of
fixation.
This document is a draft for review purposes only and does not constitute Agency policy.
85 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
In vivo chromosome aberration studies have mostly reported negative or null results
(Leuschner and Leuschner, 1991; Liang and Pacchierotti, 1988; Mailhes et al., 1993; Russo and
Levis, 1992a; Russo and Levis, 1992b; Xu and Adler, 1990) with the exception of one study
(Russo et al., 1984) in an F1 cross of mouse strain between C57B1/Cne x C3H/Cne.
4.2.4.1.7. Sister chromatid exchanges (SCEs)
SCEs were assessed by Ikbal et al. (2004) in cultured peripheral blood lymphocytes of
18 infants (age range of 31-55 days) before and after administration of a single dose of chloral
hydrate (50 mg/kg of body weight) for sedation before a hearing test. The authors report a
significant increase in the mean number of SCEs, from before administration (7.03 ± 0.18
SCEs/cell) and after administration (7.90 ± 0.19 SCEs/cell), with each of the 18 individuals
showing an increase with treatment. Micronuclei were also significantly increased. SCEs were
also assessed by Gu et al. (1981a) in human lymphocytes exposed in vitro with inconclusive
results, although positive results were observed by Beland (1999) in Chinese hamster ovary cells
exposed in vitro with and without an exogenous metabolic system.
4.2.4.1.8. Cell transformation
Chloral hydrate was positive in the two studies designed to measure cellular
transformation (Gibson et al., 1995; Parry et al., 1996). Both studies exposed Syrian hamster
cells (embryo and dermal) to chloral hydrate and induced cellular transformation.
4.2.4.1.9. Summary
Chloral hydrate has been reported to induce micronuclei formation, aneuploidy, and
mutations in multiple in vitro systems and in vivo. In vivo studies have limited results to an
increased micronuclei formation mainly in mouse spermatocytes. CH is positive to in some
studies in in vitro genotoxicity assays that detect point mutations, micronuclei induction,
chromosomal aberrations, and/or aneuploidy. The in vivo data exhibit mixed results (Allen et
al., 1994) (Adler, 1993; Leuschner and Beuscher, 1998; Mailhes et al., 1993; Nutley et al., 1996;
Russo et al., 1992; Xu and Adler, 1990). Most of the positive studies show that chloral hydrate
induces aneuploidy. Based on the existing array of data, CH has the potential to be genotoxic,
particularly when aneuploidy is considered in the weight of evidence for genotoxic potential.
Some have suggested that chloral hydrate may act through a mechanism of spindle poisoning and
This document is a draft for review purposes only and does not constitute Agency policy.
86 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
resulting in numerical changes in the chromosomes, but some data also suggest induction of
chromosomal aberrations. These results are consistent with TCE, albeit there are more limited
data on TCE for these genotoxic endpoints.
4.2.5. Dichlorovinyl Cysteine (DCVC) and S-Dichlorovinyl Glutathione (DCVG)
DCVC and DCVG have been studied for their genotoxic potential; however, since there
is limited number of studies to evaluate them based on each endpoint, particularly in mammalian
systems, the following section has been combined to include all the available studies for different
endpoints of genotoxicity. Study details can be found in Table 4-18.
DCVC and DCVG, cysteine intermediates of TCE formed by the glutathione-
»Y-trnasferase (GST) pathway, are capable of inducing point mutations as evidenced by the fact
that they are positive in the Ames assay. Dekant et al. (1986c) demonstrated mutagenicity of
DCVC in S. typhimurium strains (TA100, TA2638, and TA98) using the Ames assay in the
absence of S9. The effects were decreased with the addition of a beta-lyase inhibitor
aminooxyacetic acid, suggesting that bioactivation by this enzyme plays a role in genotoxicity.
Vamvakas et al. (1987) tested iV-acetyl-S-(l,2-dichlorovinyl)-L-cysteine) (NAcDCVC) for
mutagenicity following addition of rat kidney cytosol and found genotoxic activity.
Furthermore, Vamvakas (1988b), in another experiment, investigated the mutagenicity of DCVG
and DCVC in S. typhimurium strain TA2638, using kidney subcellular fractions for metabolic
activation and AOAA (a beta-lyase inhibitor) to inhibit genotoxicity. DCVG and DCVC both
exhibited direct-acting mutagenicity, with kidney mitochondria, cytosol, or microsomes
enhancing the effects for both compounds and AOAA diminishing, but not abolishing the
effects. Importantly, addition of liver subcellular fractions did not enhance the mutagenicity of
DCVG, consistent with in situ metabolism playing a significant role in the genotoxicity of these
compounds in the kidney.
While additional data are not available on DCVG or NAcDCVC, the genotoxicity of
DCVC is further supported by the predominantly positive results in other available in vitro and
in vivo assays. Jaffe et al. (1985) reported DNA strand breaks due to DCVC administered in
vivo, in isolated perfused kidneys, and in isolated proximal tubules of albino male rabbits.
Vamvakas et al. (1989) reported dose-dependent increases in unscheduled DNA synthesis in
LLC-PK1 cell clones at concentrations without evidence of cytotoxicity. In addition,
Vamvakas et al. (1996) reported that 7-week DCVC exposure to LLC-PK1 cell clones at
noncytotoxic concentrations induces morphological and biochemical de-differentiation that
persists for at least 30 passages after removal of the compound. This study also reported
This document is a draft for review purposes only and does not constitute Agency policy.
87 DRAFT—DO NOT CITE OR QUOTE
-------
1 increased expression of the proto-oncogene c-fos in the cells in this system. In a Syrian hamster
2 embryo fibroblast system, DCVC did not induce micronuclei, but demonstrated an unscheduled
3 DNA synthesis response (Vamvakas et al., 1988a).
4 Two more recent studies are discussed in more detail. Mally et al. (2006) isolated
5 primary rat kidney epithelial cells from Tsc-2Ek/+ (Eker) rats, and reported increased
6 transformation when exposed to 10 |iM DCVC, similar to that of the genotoxic renal carcinogens
7 jV-methyl-TV-nitro-jV-nitrosoguanidine (Horesovsky et al., 1994). The frequency was variable
8 but consistently higher than background. No loss-of-heterozygosity (LOH) of the Tsc-2 gene
This document is a draft for review purposes only and does not constitute Agency policy.
88 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-18. TCE GSH conjugation metabolites genotoxicity
Test system/endpoint
Doses tested
With
activation
Without
activation
Comments
References
Gene mutations (Ames test)
S. typhimurium, TA100, 2638, 98
0.1-0.5 nmol
ND
+
DCVC was mutagenic in all three strains of
S. typhimurium without the addition of
mammalian subcellular fractions.
Dekant et al.,
(1986c)
S. typhimurium, TA2638
50-300 nmol
+
+
Increase in number of revertants in DCVC
alone at low doses; further increase in
revertants was observed in the presence of
microsomal fractions. Toxicity as indicated
by decreased revertants per plate were seen at
higher doses.
Vamvakas et al.
(1988b)
Mutation analysis
In vitro—rat kidney epithelial cells, LOH
in Tsc gene
10 |iM
NA
-
Only 1/9 transformed cells showed LOH.
Mally et al. (2006)
In vitro—rat kidney epithelial cells, VIII.
gene (exons 1-3)
10 nM
NA
~
No mutations in VIII. eene. Note: VHL is not
a target gene in rodent models of chemical-
induced or spontaneous renal carcinogenesis.
Mally et al. (2006)
Unscheduled DNA synthesis
Porcine kidney tubular epithelial cell line
(LLC-PK1)
2.5 (iM-5, 10, 15,
24 h; 2.5-100 \iM
NA
+
Dose-dependent in UDS up to 24 h tested at
2.5 |iM. Also, there was a dose dependent
increase at lower conc. Higher
concentrations were cytotoxic as determined
by LDH release from the cells.
Vamvakas et al.
(1989)
Syrian hamster embryo fibroblasts
NA
+
Increase in UDS in treatment groups.
Vamvakas et al.
(1988a)
-------
Table 4-18. TCE GSH conjugation metabolites genotoxicity (continued)
Test system/endpoint
Doses tested
With
activation
Without
activation
Comments
References
DNA strand breaks
Male rabbit renal tissue (perfused kidneys
and proximal tubules)
0-100 mg/kg or
10 ^M to 10 mM
ND
+
Dose dependent increase in strand breaks in
both i.v. and i.p. injections (i.v. injections
were done only for 10 and 20 mg/kg) were
observed. Perfusion of rabbit kidney (45 min
exposure) and proximal tubules (30 min
exposure) experimentresulted in a dose
dependent difference in the amount of single
strand breaks.
Jaffe et.al. (1985)
Primary kidney cells from both male rats
and human
1-4 mM; 20 h
exposure
NA
+
Statistically significant increase in all doses
(1, 2, or 4 mM) both in rats and human cells.
Robbiano et al.
(2004)
In vivo—male Sprague-Dawley rats
exposed to TCE or DCVC—comet assay
TCE: 500-2,000
ppm, inhalation,
6 h/d, 5 d
DCVC: 1 or
10 mg/kg, single
oral dose for 16 h
+ (DCVC)
- (TCE)
NA
No significant increase in tail length in any of
the TCE exposed groups. InExpt. 1. 2 h
exposure—1 or 10 mg to DCVC resulted in
significant increase with no dose-response,
but not at 16 h. In Expt. 2 ND for 1 mg,
significant increase at 10 mg.
Clay (2008)
Micronucleus
Syrian hamster embryo fibroblasts
NA
-
No micronucleus formation.
Vamvakas et al.
(1988a)
Primary kidney cells from both male rats
and human
1-4 mM; 20 h
exposure
NA
+
Statistically significant increase in all doses
(1,2, and 4 mM) both in rats and human
cells.
Robbiano et al.
(2004)
Male Sprague-Dawley rats; proximal
tubule cells (in vivo)
4 mM/kg TCE
exposure, single
dose
NA
+
Statistically significant increase in the
average frequency of micronucleated kidney
cells was observed.
Robbiano et al.
(1998)
-------
Table 4-18. TCE GSH conjugation metabolites genotoxicity (continued)
Test system/endpoint
Doses tested
With
activation
Without
activation
Comments
References
Cell transformation
Kidney tubular epithelial cell line
(LLC-PK1)
lor 5 |iM; 7 wk
NA
+
Induced morphological cell transformation at
both concentrations tested. Furthermore,
cells maintained both biochemical and
morphological alterations remained stable for
30 passages.
Vamvakas et al.
(1996)
Rat kidney epithelial cells (in vitro)
10 \M- 24 h
exposure, 7 wk post
incubation
NA
+
Cell transformation was higher than control,
however, cell survival percentage ranged
from 39-64% indicating cytotoxicity.
Mally et al. (2006)
Gene expression
Kidney tubular epithelial cell line
(LLC-PK1)
lor 5 |iM clones, 30,
60, 90 min
NA
+
Increased c-fos expression in 1 and 5 |iM
exposed clones at three different times tested.
Vamvakas et al.
(1996)
Kidney tubular epithelial cell line
(LLC-PK1)
NA
+
Expression of c-fos and c-myc increased in a
time-dependent manner.
Vamvakas et al.
(1993)
i.v. = intravenous, LDH = lactate dehydrogenase, ND = not determined, NA = not applicable.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
was reported either in these DCVC transformants or in renal tumors (which were not increased in
incidence) from TCE-treated Eker rats, which Mally et al. (2006) suggested support a
nongenotoxic mechanism because a substantial fraction of spontaneous renal tumors in Eker rats
showed LOH at this locus (Kubo et al., 1994; Yeung et al., 1995) and because LOH was
exhibited both in vitro and in vivo with 2,3,4-tris(glutathion-S-yl)-hydroquinone treatment in
Eker rats (Yoon et al., 2001). However, 2,3,4-tris(glutathion-S-yl)-hydroquinone is not
genotoxic in standard mutagenicity assays (Yoon et al., 2001), and Kubo et al. (1994) also
reported that none of renal tumors induced by the genotoxic carcinogen A-ethyl-A'-nitrosourea
showed LOH. Therefore, the lack of LOH at the Tsc-2 locus induced by DCVC in vitro, or TCE
in vivo, reported by Mally et al. (2006) is actually more similar to the response from the
genotoxic carcinogen A'-ethyl-A-nitrosourea than the nongenotoxic carcinogen
2,3,4-tris(glutathion-S-yl)-hydroquinone. Therefore, these data do not substantially contradict
the body of evidence on DCVC genotoxicity.
Finally, Clay (2008) evaluated the genotoxicity of DCVC in vivo using the comet assay
to assess DNA breakage in the proximal tubules of rat kidneys. Rats were exposed orally to a
single dose of DCVC (1 or 10 mg/kg). The animals were sacrificed either 2 or 16 hours after
dosing and samples prepared for detecting the DNA damage. DCVC (1 and 10 mg/kg) induced
no significant DNA damage in rat kidney proximal tubules at the 16-hour sampling time or after
1 mg/kg DCVC at the 2-hour sampling time. While Clay et al. (2008) concluded that these data
were insufficient to indicate a positive response in this assay, the study did report a statistically
significant increase in percentage tail DNA 2 hours after treatment with 10 mg/kg DCVC,
despite the small number of animals at each dose (n = 5) and sampling time. Therefore, these
data do not substantially contradict the body of evidence on DCVC genotoxicity.
Overall, DCVC, and to a lesser degree DCVG and NAcDCVC, have demonstrated
genotoxicity based on consistent results in a number of available studies. While some recent
studies (Clay, 2008; Mally et al., 2006) have reported a lack of positive responses in some in
vivo measures of genotoxicity with DCVC treatment, due to a number of limitations discussed
above, these studies do not substantially contradict the body of evidence on DCVC genotoxicity.
It is known that these metabolites are formed in vivo following TCE exposure, specifically in the
kidney, so they have the potential to contribute to the genotoxicity of TCE, especially in that
tissue. Moreover, DCVC and DCVG genotoxic responses were enhanced when metabolic
activation using kidney subcellular fractions was used (Vamvakas et al., 1988b). Finally, the
lack of similar responses in in vitro genotoxicity assays with TCE, even with metabolic
activation, is likely the result of the small yield (if any) of DCVC under in vitro conditions, since
in vivo, DCVC is likely formed predominantly in situ in the kidney while S9 fractions are
This document is a draft for review purposes only and does not constitute Agency policy.
92 DRAFT—DO NOT CITE OR QUOTE
-------
1 typically derived from the liver. This hypothesis could be tested in experiments in which TCE is
2 incubated with subcellular fractions from the kidney, or from both the kidney and the liver (for
3 enhanced GSH conjugation).
4
4.2.6. Trichloroethanol (TCOH)
5 Limited studies are available on the effect of TCOH on genotoxicity (see Table 4-19).
6 TCOH is negative in the S. typhimurium assay using the TA100 strain (Bignami et al., 1980;
7 DeMarini et al., 1994; Waskell, 1978). A study by Beland (1999) using S. typhimurium strain
8 TA104 did not induce reverse mutations without exogenous metabolic activation, however did
9 increase mutant frequency in the presence of exogenous metabolic activation at a dose above
10 2,500 |ig/plate. TCOH has not been evaluated in the other recommended screening assays.
11 Therefore, the database is limited for the determination of TCOH genotoxicity.
12
13
14 Table 4-19. Genotoxicity of trichloroethanol
15
Test system/endpoint
Doses
(LED or HID)a
Resultsb
Reference
With
activation
Without
activation
S. typhimurium TA100, 98, reverse
mutation
7,500 ng/plate
-
-
Waskell (1978)
S. typhimurium TA100, reverse
mutation
0.5 |ig/cm3 vapor
-
-
DeMarini et al. (1994)
S. typhimurium TA104, reverse
mutation
2,500 ng/plate
+
-
Beland (1999)
S. typhimurium TA100, 1535
reverse mutation
NA
-
-
Bignami et al. (1980)
Sister chromatid exchanges
NA
NA
+
Gu et al., 1981a
16
17 aLED = lowest effective dose; HID = highest ineffective dose.
18 b Results: + = positive; - = negative; NA = doses not available, results based on the abstract.
19
20
This document is a draft for review purposes only and does not constitute Agency policy.
93 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
4.2.7. Synthesis and Overall Summary
Trichloroethylene and its metabolites (TCA, DC A, CH, DCVC, DCVG, and TCOH) have
been evaluated to varying degrees for their genotoxic activity in several of in vitro systems such
as bacteria, yeast, and mammalian cells and, also, in in vivo systems.
There are several challenges in interpreting the genotoxicity results obtained from TCE
exposure. For example, some studies in bacteria should be interpreted with caution if conducted
using technical grade TCE since it may contain known bacterial mutagens in trace amounts as
stabilizers (e.g., 1,2-epoxybutane and epichlorohydrin). Because of the volatile nature of TCE,
there could be false negative results if proper precautions are not taken to limit evaporation, such
as the use of a closed sealed system. The adequacy of the enzyme-mediated activation of TCE in
vitro tests is another consideration. For example, it is not clear if standard S9 fractions can
adequately recapitulate the complex in vivo metabolism of TCE to reactive intermediates, which
in some cases entails multiple sequential steps involving multiple enzyme systems (e.g., CYP,
GST, etc.) and interorgan processing (as is described in more detail in Section 3.3). In addition,
the relative potency of the metabolites in vitro may not necessarily inform their relative
contribution to the overall mechanistic effects of the parent compound, TCE. Furthermore,
although different assays provided data relevant to different types of genotoxic endpoints, not all
effects that are relevant for carcinogenesis are encompassed. The standard battery of prokaryotic
as well as mammalian genotoxicity test protocols typically specify the inclusion of significantly
cytotoxic concentrations of the test compound.
With respect to potency, several TCE studies have been conducted along with numerous
other chlorinated compounds and the results interpreted as a comparison of the group of
compounds tested (relative potency). However, for the purposes of hazard characterization, such
comparisons are not informative—particularly if they are not necessarily correlated with in vivo
carcinogenic potency. Also, differentiating the effects of TCE with respect to its potency can be
influenced by many factors such as the type of cells, their differing metabolic capacities,
sensitivity of the assay, need for greater concentration to show any effect, interpretation of data
when the effects are marginal, and gradation of severity of effects.
Also, type of samples used, methodology used for the isolation of genetic material, and
duration of exposure can particularly influence the results of several studies. This is particularly
true for human epidemiological studies. For example, while some studies use tissues obtained
directly from the patients others use formalin fixed tissues sections to isolate DNA for mutation
detection. Type of fixing solution, fixation time, and period of storage of the tissue blocks often
affect the quality of DNA. Formic acid contained in the formalin solution or picric acid
contained in Bouin's solution is known to degrade nucleic acids resulting in either low yield or
This document is a draft for review purposes only and does not constitute Agency policy.
94 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
poor quality of DNA. In addition, during collection of tumor tissues, contamination of
neighboring normal tissue can easily occur if proper care is not exercised. This could lead to the
'dilution effect' of the results, i.e., because of the presence of some normal tissue; frequency of
mutations detected in the tumor tissue can be lower than expected. Due to some of these
technical difficulties in obtaining proper material (DNA) for the detection of mutation, the results
of these studies should be interpreted cautiously.
The following synthesis, summary, and conclusions focus on the available studies that
may provide some insight into the potential genotoxicity of TCE considering the above
challenges when interpreting the mutagenicity data for TCE.
Overall, evidence from a number of different analyses and a number of different
laboratories using a fairly complete array of endpoints suggests that TCE, following metabolism,
has the potential to be genotoxic. TCE has a limited ability to induce mutation in bacterial
systems, but greater evidence of potential to bind or to induce damage in the structure of DNA or
the chromosome in a number of targets. A series of carefully controlled studies evaluating TCE
itself (without mutagenic stabilizers and without metabolic activation) found it to be incapable of
inducing gene mutations in most standard mutation bacterial assays (Baden et al., 1979; Bartsch
et al., 1979; Crebelli et al., 1982; Henschler et al., 1977; Mortelmans et al., 1986; Shimada et al.,
1985; Simmon et al., 1977; Waskell, 1978). Therefore, it appears that it is unlikely that TCE is a
direct-acting mutagen, though TCE has shown potential to affect DNA and chromosomal
structure. TCE is also positive in some but not all fungal and yeast systems (Callen et al., 1980;
Crebelli et al., 1985; Koch et al., 1988; Rossi et al., 1983). Data from human epidemiological
studies support the possible mutagenic effect of TCE leading to VHL gene damage and
subsequent occurrence of renal cell carcinoma. Association of increased VHL mutation
frequency in TCE-exposed renal cell carcinoma cases has been observed (Brauch et al., 1999;
Brauch et al., 2004; Briining et al., 1997b).
TCE can lead to binding to nucleic acids and proteins (Bergman, 1983; DiRenzo et al.,
1982; Kautiainen et al., 1997; Mazzullo et al., 1992; Miller and Guengerich, 1983), and such
binding appears to be due to conversion to one or more reactive metabolites. For instance,
increased binding was observed in samples bioactivated with mouse and rat microsomal fractions
(Banerjee and Van Duuren, 1978; DiRenzo et al., 1982; Mazzullo et al., 1992; Miller and
Guengerich, 1983). DNA binding is consistent with the ability to induce DNA and chromosomal
perturbations. Several studies report the induction of micronuclei in vitro and in vivo from TCE
exposure (Hrelia et al., 1994; Hu et al., 2008; Kligerman et al., 1994; Robbiano et al., 2004;
Wang et al., 2001). Reports of SCE induction in some studies are consistent with DNA effects,
This document is a draft for review purposes only and does not constitute Agency policy.
95 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
but require further study (Gu et al., 1981a; Gu et al., 1981b; Kligerman et al., 1994; Nagaya et
al., 1989a; White etal., 1979).
TCA, an oxidative metabolite of TCE, exhibits little, if any genotoxic activity in vitro.
TCA did not induce mutations in S. typhimurium strains in the absence of metabolic activation or
in an alternative protocol using a closed system (DeMarini et al., 1994; Giller et al., 1997;
Kargalioglu et al., 2002; Nelson et al., 2001; Rapson et al., 1980; Waskell, 1978) but a
mutagenic response was induced in TA100 in the Ames fluctuation test (Giller et al., 1997).
However, in vitro experiments with TCA should be interpreted with caution if steps have not
been taken to neutralize pH changes caused by the compound (Mackay et al., 1995). Measures
of DNA-repair responses in bacterial systems have shown induction of DNA repair reported in S.
typhimurium but not in E. coli. Mutagenicity in mouse lymphoma cells was only induced at
cytotoxic concentrations (Harrington-Brock et al., 1998). TCA was positive in some
genotoxicity studies in vivo mouse, newt, and chick test systems (Bhunya and Behera, 1987;
Bhunya and Jena, 1996; Birner et al., 1994; Giller et al., 1997). DNA unwinding assays have
either shown TCA to be much less potent than DC A (Nelson and Bull, 1988) or negative (Nelson
et al., 1989; Styles et al., 1991). Due to limitations in the genotoxicity database, the possible
contribution of TCA to TCE genotoxicity is unclear.
DC A, a chloroacid metabolite of TCE, has also been studied using different types of
genotoxicity assays. Although limited studies are conducted for different genetic endpoints,
DCA has been demonstrated to be mutagenic in the S. typhimurium assays, in vitro (DeMarini et
al., 1994; Kargalioglu et al., 2002; Plewa et al., 2002) in some strains, mouse lymphoma assay,
(Harrington-Brock et al., 1998) in vivo cytogenetic tests (Fuscoe et al., 1996; Leavitt et al.,
1997), the micronucleus induction test, the Big Blue mouse system, and other tests (Bignami et
al., 1980; Chang et al., 1992; DeMarini et al., 1994; Fuscoe et al., 1996; Harrington-Brock et al.,
1998; Leavitt et al., 1997; Nelson and Bull, 1988; Nelson et al., 1989). DCA can cause DNA
strand breaks in mouse and rat liver cells following in vivo mice and rats (Fuscoe et al., 1996).
Because of uncertainties as to the extent of DCA formed from TCE exposure, inferences as to
the possible contribution from DCA genotoxicity to TCE toxicity are difficult to make.
Chloral hydrate is mutagenic in the standard battery of screening assays. Effects include
positive results in bacterial mutation tests for point mutations and in the mouse lymphoma assay
for mutagenicity at the Tk locus (Haworth et al., 1983). In vitro tests showed that CH also
induced micronuclei and aneuploidy in human peripheral blood lymphocytes and Chinese
hamster pulmonary cell lines. Micronuclei were also induced in Chinese hamster embryonic
fibroblasts. Several studies demonstrate that chloral hydrate induces aneuploidy (loss or gain of
whole chromosomes) in both mitotic and meiotic cells, including yeast (Gualandi, 1987; Kafer,
This document is a draft for review purposes only and does not constitute Agency policy.
96 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
1986; Singh and Sinha, 1976, 1979; Sora and Agostini Carbone, 1987), cultured mammalian
somatic cells (Degrassi and Tanzarella, 1988), and spermatocytes of mice (Liang and
Pacchierotti, 1988; Russo et al., 1984). Chloral hydrate was negative for sex-linked recessive
lethal mutations in Drosophila (Yoon et al., 1985). It induces SSB in hepatic DNA of mice and
rats (Nelson and Bull, 1988) and mitotic gene conversion in yeast (Bronzetti et al., 1984).
Schatten and Chakrabarti (1998) showed that chloral hydrate affects centrosome structure, which
results in the inability to reform normal microtubule formations and causes abnormal fertilization
and mitosis of sea urchin embryos. Based on the existing array of data, CH has the potential to
be genotoxic, particularly when aneuploidy is considered in the weight of evidence for genotoxic
potential. Chloral hydrate appears to act through a mechanism of spindle poisoning and resulting
in numerical changes in the chromosomes. These results are consistent with TCE, albeit there
are limited data on TCE for these genotoxic endpoints.
DCVC, and to a lesser degree DCVG, has demonstrated bacterial mutagenicity based on
consistent results in a number of available studies (Dekant et al., 1986c; Vamvakas et al., 1987;
Vamvakas et al., 1988b). DCVC has demonstrated a strong, direct-acting mutagenicity both
with and without the presence of mammalian activation enzymes. It is known that these
metabolites are formed in vivo following TCE exposure, so they have the potential to contribute
to the genotoxicity of TCE. The lack of similar response in bacterial assays with TCE is likely
the result of the small yield (if any) of DCVC under in vitro conditions, since in vivo, DCVC is
likely formed predominantly in situ in the kidney (S9 fractions are typically derived from the
liver). DCVC and DCVG have not been evaluated extensively in other genotoxicity assays, but
the available in vitro and in vivo data are predominantly positive. For instance, several studies
have reported the DCVC can induce primary DNA damage in mammalian cells in vitro and in
vivo (Clay, 2008; Jaffe et al., 1985; Vamvakas et al., 1989). Long-term exposure to DCVC
induced de-differentiation of cells (Vamvakas et al., 1996). It has been shown to induce
expression of the protooncogene c-fos (Vamvakas et al., 1996) and cause cell transformation in
rat kidney cells (Mally et al., 2006). In LLC-PK1 cell clones, DCVC was reported in induce
unscheduled DNA synthesis, but not micronuclei (Vamvakas et al., 1988a). Finally, DCVC
induced transformation in kidney epithelial cells isolated from Eker rats carrying the
heterozygous Tsc-2 mutations (Mally et al., 2006). Moreover, the lack of LOH at the Tsc-2 locus
observed in exposed cells does not constitute negative evidence of DCVC genotoxicity, as none
of renal tumors induced in Eker rats by the genotoxic carcinogen A'-ethyl-A'-nitrosourea showed
LOH (Kubo et al., 1994).
In support of the importance of metabolism, there is some concordance between effects
observed from TCE and those from several metabolites. For instance, both TCE and chloral
This document is a draft for review purposes only and does not constitute Agency policy.
97 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
hydrate have been shown to induce micronucleus in mammalian systems, but chromosome
aberrations have been more consistently observed with chloral hydrate than with TCE. The role
of TCA in TCE genotoxicity is less clear, as there is less concordance between the results from
these two compounds. Finally, several other TCE metabolites show at least some genotoxic
activity, with the strongest data from DCA, DCVG, and DCVC. While quantitatively smaller in
terms of flux as compared to TCA and TCOH (for which there is almost no genotoxicity data),
these metabolites may still be toxicologically important.
Thus, uncertainties with regard to the characterization of TCE genotoxicity remain,
particularly because not all TCE metabolites have been sufficiently tested in the standard
genotoxicity screening battery to derive a comprehensive conclusion. However, the metabolites
that have been tested particularly DCVC have predominantly resulted in positive data although
to a lesser extent in DCVG and NAcDCVC, supporting the conclusion that these compounds are
genotoxic, particularly in the kidney, where in situ metabolism produces and/or bioactivates
these TCE metabolites.
4.3. CENTRAL NERVOUS SYSTEM (CNS) TOXICITY
TCE exposure results in central nervous system (CNS) effects in both humans and
animals that can result from acute, subchronic, or chronic exposure. There are studies indicating
that TCE exposure results in CNS tumors and this discussion can be found in Section 4.9. The
studies discussed in this section focus on the most critical neurological effects that were
extracted from the neurotoxicological literature. Although there are several studies and reports
that have evaluated TCE as an anesthetic, those studies were not included in this section because
of the high exposure levels in comparison to the selected critical neurological effects described
below. The critical neurological effects are nerve conduction changes, sensory effects, cognitive
deficits, changes in psychomotor function, and changes in mood and sleep behaviors. The
selection criteria that were used to determine study importance included study design and
validity, pervasiveness of neurological effect, and for animal studies, the relevance of these
reported outcomes in humans. More detailed information on human and animal neurological
studies with TCE can be found in Appendix D.
This document is a draft for review purposes only and does not constitute Agency policy.
98 DRAFT—DO NOT CITE OR QUOTE
-------
4.3.1. Alterations in Nerve Conduction
4.3.1.1.1. Trigeminal Nerve Function: Human Studies
1 A number of human studies have been conducted that examined the effects of
2 occupational or drinking water exposures to TCE on trigeminal nerve function (see Table 4-20).
3 Many studies reported that humans exposed to TCE present trigeminal nerve function
4 abnormalities as measured by blink reflex and masseter reflex test measurements (Feldman et al.,
5 1988; Feldman et al., 1992; Kilburn, 2002b; Kilburn and Warshaw, 1993)Ruitjen et al., 1991.
6 The blink and masseter reflexes are mediated primarily by the trigeminal nerve and changes in
7 measurement suggest impairment in nerve conduction. Other studies measured the trigeminal
8
This document is a draft for review purposes only and does not constitute Agency policy.
99 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-20. Summary of human trigeminal nerve and nerve conduction
2 velocity studies
3
Reference
Subjects
Exposure
Effect
Barret et al.
(1982)
11 workers with chronic
TCE exposure.
Controls: 20 unexposed
subjects.
Presence of TCE and TCA found
through urinalysis. Atmospheric
TCE concentrations and duration of
exposure not reported in paper.
Following stimulation of the
trigeminal nerve, significantly higher
voltage stimuli was required to obtain
a normal response and there was a
significant increase in latency for
response and decreased response
amplitude.
Barret et al.
(1984)
188 factory workers.
No unexposed controls;
lowest exposure group
used as comparison.
>150 ppm; n = 54 < 150 ppm;
n= 134.
7 h/d for 7 yr.
Trigeminal nerve and optic nerve
impairment, asthenia and dizziness
were significantly increased with
exposure.
Barret et al.
(1987)
104 degreaser machine
operators.
Controls: 52 unexposed
subjects
Mean age 41.6 yr.
Mean duration, 8.2 yr, average daily
exposure 7 h/d.
Average TCOH range = 162-245
mg/g creatinine.
Average TCA range = 93-131 mg/g
creatinine.
Evoked trigeminal responses were
measured following stimulation of the
nerve and revealed increased latency
to respond, amplitude or both and
correlated with length of exposure
(p < 0.01) and with age (p < 0.05), but
not concentration.
El-Ghawabi
et al. (1973)
30 money printing shop
workers.
Controls: 20 nonexposed
males .
10 workers exposed to
inks not containing TCE.
Mean TCE air
concentrations ranged from 41
ppm to 163 ppm. Exposure
durations:
Less than 1 yr: n = 3
1 yr: n = 1
2 yr: n = 2
3 yr: n = 11
4 yr: n = 4
5 yr or greater: n = 9
No effect on trigeminal nerve function
was noted.
Feldman
et al. (1988)
21 Woburn, MA
residents.
27 controls.
TCE maximum
reported concentration in well
water was 267 ppb; other
solvents also present.
Exposure duration
ranged from 1-12 yr.
Measurement of the blink reflex as
mediated by the trigeminal nerve
resulted in significant increases in the
latency of reflex components
(p< 0.001).
This document is a draft for review purposes only and does not constitute Agency policy.
100 DRAFT—DO NOT CITE OR QUOTE
-------
Feldman
etal. (1992)
18 workers.
30 controls.
TCE exposure
categories of "extensive",
"occasional," and "chemical
other than TCE".
"Extensive" =
chronically exposed (>1 yr) to
TCE for 5 d/wk and >50%
workday.
"Occupational" =
chronically exposed to TCE for
1-3 d/wk and >50% workday.
The blink reflex as mediated by the
trigeminal was measured. The
"extensive" group revealed latencies
greater than three SD above the
nonexposed group mean on blink
reflex components.
This document is a draft for review purposes only and does not constitute Agency policy.
101 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-20. Summary of human trigeminal nerve and nerve conduction
velocity studies (continued)
Reference
Subjects
Exposure
Effect
Kilburn and
Warshaw
(1993)
160 residents living in
Southwest Tucson with
TCE, other solvents, and
chromium in
groundwater.
Control: 113 histology
technicians from a
previous study (Kilburn
et al., 1987; Kilburn and
Warshaw, 1992).
>500 ppb of TCE in
well water before 1981 and
25-100 ppb afterwards.
Duration ranged from
1-25 yr.
Significant impairments in sway speed
with eyes open and closed and blink
reflex latency (R-l) which suggests
trigeminal nerve impairment.
Kilburn
(2002a)
236 residents near a
microchip plant in
Phoenix, AZ.
Controls: 161 regional
referents from
Wickenburg, AZ and
67 referents in
northeastern Phoenix.
<0.2-10,000 ppb of
TCE, <0.2-260,000 ppb TCA,
<0.2-6,900 ppb 1,1-DCE,
<0.2-1,600 1,2-DCE,
<0.2-23,000 ppb PCE,
<0.02-330 ppb VC in well
water.
Exposure duration
ranged from 2-37 yr.
Trigeminal nerve impairment as
measured by the blink reflex test; both
right and left blink reflex latencies
(R-l) were prolonged. Exposed group
mean 14.2 + 2.1 ms (right) or
13.9 + 2.1 ms (left) vs. referent group
mean of 13.4 + 2.1 ms (right) or
13.5 + 2.1ms (left),p = 0.0001 (right)
and 0.008 (left).
Mhiri et al.
(2004)
23 phosphate industry
workers.
Controls: 23 unexposed
workers.
Exposure ranged from
50-150 ppm, for 6 h/d for at
least 2 yr.
Mean urinary
trichloroethanol and
trichloroacetic acid levels were
79.3 ± 42 and 32.6 ± 22 mg/g
creatinine.
TSEPs were recorded. Increase in
the TSEP latency was observed in 15
out of 23 (65%) workers.
This document is a draft for review purposes only and does not constitute Agency policy.
102 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-20. Summary of human trigeminal nerve and nerve conduction
velocity studies (continued)
Reference
Subjects
Exposure
Effect
Rasmussen
et al.
(1993d)
96 Danish metal
degreasers.
Age range: 19-68.
No unexposed controls;
low exposure group used
as comparison.
Average exposure
duration: 7.1 yr.); range of
full-time degreasing: 1 mo to
36 yr. Exposure to TCE or to
CFC113 .
1) Low exposure: n =
19, average full-time exposure
0.5 yr.
2) Medium exposure: n
= 36, average full-time
exposure 2.1 yr.
3) High exposure: n =
41, average full-time exposure
11 yr. TCA in high exposure
group = 7.7 mg/L (max = 26.1
mg/L).
No statistically significant trend on
trigeminal nerve function, although
some individuals had abnormal
function.
Ruijten
et al. (1991)
31 male printing
workers. Mean age
44 yr; mean duration
16 yr.
Controls: 28
unexposed; mean age
45 yr.
Mean cumulative
exposure = 704 ppm x yr (SD
583, range: 160-2,150 ppm x
yr.
Mean, 17 ppm at time
of study; historic TCE levels
from 1976-1981, mean of 35
ppm.
Mean duration of 16 yr.
Measurement of trigeminal nerve
function by using the blink reflex
resulted in no abnormal findings.
Increased latency in the masseter
reflex is indicative of trigeminal
nerve impairment.
This document is a draft for review purposes only and does not constitute Agency policy.
103 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Table 4-20. Summary of human trigeminal nerve and nerve conduction
velocity studies (continued)
Reference
Subjects
Exposure
Effect
Triebig et al.
(1982)
24 workers (20 males,
4 females)
occupationally
exposed—ages 17-56.
Controls:
144 individuals to
establish normal nerve
conduction parameters.
Matched group:
24 unexposed workers
(20 males, 4 females).
Exposure duration of 1
mo to 258 mo (mean 83 mo).
Air exposures were between
5-70 ppm.
No statistically significant difference
in nerve conduction velocities between
the exposed and unexposed groups.
Triebig et al.
(1983)
66 workers
occupationally exposed.
Control: 66 workers not
exposed to solvents.
Subjects were exposed
to a mixture of solvents,
including TCE.
Exposure-response relationship
observed between length of solvent
exposure and statistically significant
reduction in mean sensory ulnar nerve
conduction velocities.
Bolded study(ies) carried forward for consideration in dose-response assessment (see Section 5).
DCE = dichloroethylene, PCE = perchloroethylene, SD = standard deviation, VC = vinyl chloride.
somatosensory evoked potential (TSEP) following stimulation of the trigeminal nerve and
reported statistically significantly delayed response on evoked potentials among exposed subjects
compared to nonexposed individuals (Barret et al., 1982; Barret et al., 1984; Barret et al., 1987;
Mhiri et al., 2004). Two studies which also measured trigeminal nerve function did not find any
effect (El Ghawabi et al., 1973; Rasmussen et al., 1993d) but the methods were not provided in
either study (El Ghawabi et al., 1973; Rasmussen et al., 1993d) or an appropriate control group
was not included (Rasmussen et al., 1993d). These studies and results are described below and
summarized in detail in Table 4-20.
Integrity of the trigeminal nerve is commonly measured using blink and masseter
reflexes. Five studies (Barret et al., 1984; Feldman et al., 1988; Feldman et al., 1992; Kilburn,
2002b; Kilburn and Warshaw, 1993) significant increase in the latency to respond to the stimuli
generating the reflex. The latency increases in the blink reflex ranged from 0.4 ms (Kilburn,
2002b) to up to 3.44 ms (Feldman et al., 1988). The population groups in these studies were
exposed by inhalation occupationally (Barret et al., 1984) and through drinking water
environmentally (Feldman et al., 1988; Kilburn, 2002b; Kilburn and Warshaw, 1993).
Feldman et al. (1992) demonstrated persistence in the increased latency of the blink reflex
response. In one subject, exposure to TCE (levels not reported by authors) occurred through a
This document is a draft for review purposes only and does not constitute Agency policy.
104 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
degreasing accident (high and acute exposure), and increased latency response times persisted
20 years after the accident. Another two subjects, evaluated at 9 months and 1 month following
a high occupational exposure (exposure not reported by authors), also had higher blink reflex
latencies with an average increase of 2.8 ms over the average response time in the control group
used in the study. Although one study (Ruijten et al., 1991) did not find these increases in male
printing workers exposed to TCE, this study did find a statistically significant average increase
of 0.32 ms (p < 0.05) in the latency response time in TCE-exposed workers on the masseter
reflex test, another test commonly used to measure the integrity of the trigeminal nerve.
Three studies (Barret et al., 1982; Barret et al., 1987; Mhiri et al., 2004) adopting TSEPs
to measure trigeminal nerve function found significant abnormalities in these evoked potentials.
These studies were conducted on volunteers who were occupationally exposed to TCE through
metal degreasing operations (Barret et al., 1982; Barret et al., 1987) or through cleaning tanks in
the phosphate industry (Mhiri et al., 2004). Barret et al. (1982) reported that in 8 of the 11
workers, an increased voltage ranging from a 25 to a 45 volt increase was needed to generate a
normal TSEP and two of workers had an increased TSEP latency. Three out of 11 workers had
increases in TSEP amplitudes. In a later study, Barret et al. (1987) also reported abnormal
TSEPs (increased latency and/or increased amplitude) in 38% of the degreasers that were
evaluated. The individuals with abnormal TSEPs were significantly older (45 vs. 40.1 years;
p < 0.05) and were exposed to TCE longer (9.9 vs. 5.6 years;p < 0.01). Mhiri et al. (2004) was
the only study to evaluate individual components of the TSEP and noted significant increases in
latencies for all TSEP potentials (Nl, PI, N2, P2, N3; p < 0.01) and significant decreases in
TSEP amplitude (PI, p < 0.02; N2,p < 0.05). A significant positive correlation was
demonstrated between exposure duration and increased TSEP latency (p < 0.02).
Two studies reported no statistically significant effect of TCE exposure on trigeminal
nerve function (El Ghawabi et al., 1973; Rasmussen et al., 1993a). El-Ghawabi et al. (1973)
conducted a study on 30 money printing shop workers occupationally exposed to TCE.
Trigeminal nerve involvement was not detected, but the authors did not include the experimental
methods that were used to measure trigeminal nerve involvement and did not provide any data as
to how this assessment was made. Rasmussen et al. (1993a) conducted an historical cohort study
on 99 metal degreasers, 70 exposed to TCE and 29 to the fluorocarbon, CFC113. It was reported
that 1 out of 21 people (5%) in the low exposure, 2 out of 37 (5%) in the medium exposure and 4
out of 41 (10%) in the high exposure group experienced abnormalities in trigeminal nerve
sensory function, with a linear trend test />value of 0.42. The mean urinary trichloroacetic acid
concentration was reported for the high exposure group only and was 7.7 mg/L (maximum
concentration, 26.1 mg/L). The trigeminal nerve function findings of high exposure group
This document is a draft for review purposes only and does not constitute Agency policy.
105 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
subjects was compared to that of low exposure group since this study did not include an
unexposed or no TCE exposure group, and decreased the sensitivity of the study.
4.3.1.1.2. Nerve Conduction Velocity—Human Studies
Two occupational studies assessed ulnar and median nerve function using tests of
conduction latencies (Triebig et al., 1983; Triebig et al., 1982) (see Table 4-20). The ulnar nerve
and median nerves are major nerves located in the arm and forearm. Triebig (1982) studied 24
healthy workers (20 males, 4 females) exposed to TCE occupationally (5-70 ppm) at three
different plants and did not find statistically significant differences in ulnar or median nerve
conduction velocities between exposed and unexposed subjects. This study has measured
exposure data, but exposures/responses are not reported by dose levels. The Triebig (1983)
study is similar in design to the previous study (Triebig et al., 1982) but of a larger number of
subjects. In this study, a dose-response relationship was observed between lengths of exposure
to mixed solvents that included TCE (at unknown concentration). A statistically significant
reduction in nerve conduction velocities was observed for the medium- and long-term exposure
groups for the sensory ulnar nerve as was a statistically significant reduction in mean nerve
conduction velocity observed between exposed and control subjects.
4.3.1.1.3. Trigeminal Nerve Function: Laboratory Animal Studies
There is little evidence that TCE disrupts trigeminal nerve function in animal studies.
Two studies demonstrated TCE produces morphological changes in the trigeminal nerve at a
dose of 2,500 mg/kg-day for 10 weeks (Barret et al., 1992; Barret et al., 1991). However,
dichloroacetylene, a degradation product formed during the volatilization of TCE was found to
produce more severe morphological changes in the trigeminal nerve and at a lower dose of
17 mg/kg-day (Barret et al., 1992; Barret et al., 1991). Only one study (Albee et al., 2006) has
evaluated the effects of TCE on trigeminal nerve function and a subchronic inhalation exposure
did not result in any significant functional changes. A summary of these studies is provided in
Table 4-21.
Barret et al. (1992; 1991) conducted two studies evaluating the effects of both TCE and
dichloroacetylene on trigeminal nerve fiber diameter and internodal length as well as several
markers for fiber myelination. Female Sprague-Dawley rats (n = 7/group) were dosed with
2,500 mg/kg TCE or 17 mg/kg-day dichloroacetylene by gavage for 5 days/week for lOweeks.
This document is a draft for review purposes only and does not constitute Agency policy.
106 DRAFT—DO NOT CITE OR QUOTE
-------
1 TCE-dosed animals only exhibited changes in the smaller Class A fibers where internode length
2 increased marginally (<2%) and fiber diameter increased by 6%. Conversely,
3
This document is a draft for review purposes only and does not constitute Agency policy.
107 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-21. Summary of animal trigeminal nerve studies
2
Reference
Exposure
route
Species/strain/
sex/number
Dose level/
exposure
duration
NOAEL;
LOAEL3
Effects
Barret
et al.
(1991)
Direct gastric
administration
Rat, Sprague-
Dawley, female,
7/group
0, 2.5 g/kg, acute
administration.
17 mg/kg
dichloroacetylene.
LOAEL:
2.5 g/kg
Morphometric analysis was
used for analyzing the
trigeminal nerve. Increase in
external and internal fiber
diameter as well as myelin
thickness was observed in the
trigeminal nerve after TCE
treatment.
Barret
et al.
(1992)
Direct gastric
administrat-
ion
Rat, Sprague-
Dawley, female,
7/group
0,2.5 g/kg;
one dose/d, 5
d/wk, 10 wk.
17 mg/kg
dichloroacetylene
LOAEL:
2.5 g/kg
Trigeminal nerve analyzed
using morphometric analysis.
Increased internode length
and fiber diameter in class A
fibers of the trigeminal nerve
observed with TCE
treatment. Changes in fatty
acid composition also noted.
Albee et al.
(1997)
Inhalation
Rat, Fischer 344,
male, 6
0 or 300-ppm
dichloro-
acetylene, 2.25 h
LOAEL:
300 ppm
dichloro-
acetylene
Dichloroacetylene (TCE
byproduct) exposure impaired
the TSEP up to 4 d
postexposure.
Albee et al.
(2006)
Inhalation
Rat,
Fischer 344,
male and
female,
10/sex/group
0, 250,
800, or 2,500
ppm
N
OAEL:
2,500
ppm
No effect on TSEPs was noted
at any exposure level.
3
4 a NOAEL = no-observed-adverse-effect level, LOAEL = lowest-observed-adverse-effect-level.
5 Bolded study(ies) carried forward for consideration in dose-response assessment (see Section 5).
6
7
8 dichloroacetylene-treated rats exhibited significant and more robust decreases in internode length
9 and fiber diameter in both fiber classes A (decreased 8%) and B (decreased 4%).
10 Albee et al. (2006) evaluated the effects of a subchronic inhalation TCE exposure in
11 Fischer 344 rats (10/sex/group). Rats were exposed to 0, 250, 800, and 2,500 ppm TCE for
12 6 hours/day, 5 days/week for 13 weeks. TCE exposures were adequate to produce permanent
13 auditory impairment even though TSEPs were unaffected. While TCE appears to be negative in
14 disrupting the trigeminal nerve, the TCE breakdown product, dichloroacetylene, does impair
15 trigeminal nerve function. Albee et al. (1997) showed that a single inhalation exposure of rats to
This document is a draft for review purposes only and does not constitute Agency policy.
108 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
300-ppm dichloroacetylene, for 2.25 hours, disrupted trigeminal nerve evoked potentials for at
least 4 days post exposure.
4.3.1.1.4. Discussion and Conclusions: Trichloroethylene (TCE)-Induced Trigeminal Nerve
Impairment
Epidemiologic studies of exposure to TCE found impairment of trigeminal nerve
function, assessed by the blink reflex test or the TSEP, in humans exposed occupationally by
inhalation or environmentally by ingestion (see Table 4-20). Mean inhalational exposures
inferred from biological monitoring or from a range of atmospheric monitoring in occupational
studies was approximately 50 ppm to <150 ppm TCE exposure. Residence location is the
exposure surrogate in geographical-base studies of contaminated water supplies with several
solvents. Well water contaminant concentrations of TCE ranged from <0.2 ppb to 10,000 ppb
and do not provide an estimate of TCE concentrations in drinking water to studied individuals.
Two occupational studies, each including more than 100 subjects, reported statistically
significant dose-response trends based on ambient TCE concentrations, duration of exposure,
and/or urinary concentrations of the TCE metabolite TCA (Barret et al., 1984; Barret et al.,
1987). Three geographical-based studies of environmental exposures to TCE via contaminated
drinking water are further suggestive of trigeminal nerve function decrements; however, these
studies are more limited than occupational studies due to questions of subject selection. Both
exposed subjects who were litigants and control subjects who may not be representative of
exposed (Kilburn, 2002a, b; Kilburn and Warshaw, 1993); referents in Kilburn and Warshaw
(1993) were histology technicians and subjects in a previous study of formaldehyde and other
solvent exposures and neurobehavioral effects (Kilburn et al., 1987; Kilburn and Warshaw,
1992). Results were mixed in a number of smaller studies. Two of these studies reported
changes in trigeminal nerve response (Barret et al., 1982; Mhiri et al., 2004), including evidence
of a correlation with duration of exposure and increased latency in one study (Mhiri et al., 2004).
Ruijten et al. (1991) reported no significant change in the blink reflex, but did report an increase
in the latency of the masseter reflex, which also may reflect effects on the trigeminal nerve.
Two other studies reported no observed effect on trigeminal nerve impairment, but the authors
failed to provide assessment of trigeminal nerve function (El Ghawabi et al., 1973; Rasmussen et
al., 1993a) or there was not a control (nonexposed) group included in the study (Rasmussen et
al., 1993a). Therefore, because of limitations in statistical power, the possibility of exposure
misclassification, and possible differences in measurement methods, these studies are not judged
to provide substantial evidence against a causal relationship between TCE exposure and
This document is a draft for review purposes only and does not constitute Agency policy.
109 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
trigeminal nerve impairment. Overall, the weight of evidence supports a relationship between
TCE exposure and trigeminal nerve dysfunction in humans.
Impairment of trigeminal nerve function is observed in studies of laboratory animal
studies. Although one subchronic animal study demonstrated no significant impairment of
trigeminal nerve function following TCE exposure up to 2,500 ppm (no observed-adverse-effect
level [NOAEL]; Albee et al., 2006), morphological analysis of the nerve revealed changes in its
structure (Barret et al., 1992; Barret et al., 1991). However, the dose at which an effect was
observed by Barret et al. (1992; 1991) was high (2,500 mg/kg-day—lowest-observed-adverse-
effect level [LOAEL]) compared to any reasonable occupational or environmental setting,
although no lower doses were used. The acute or subchronic duration of these studies, as
compared to the much longer exposure duration in many of the human studies, may also
contribute to the apparent disparity between the epidemiologic and (limited) laboratory animal
data.
The subchronic study of Barret et al. (1992) and the acute exposure study of Albee et al.
(Albee et al., 1997) also demonstrated that dichloroacetylene, a (ex vivo) TCE degradation
product, also induces trigeminal nerve impairment, at much lower doses than TCE. It is possible
that under some conditions, coexposure to dichloroacetylene from TCE degradation may
contribute to the changes observed to be associated with TCE exposure in human studies, and
this issue is discussed further below in Section 4.3.10.
Overall evidence from numerous epidemiologic studies supports a conclusion that TCE
exposure induces trigeminal nerve impairment in humans. Laboratory animal studies provide
limited additional support, and do not provide strong contradictory evidence. Persistence of
these effects after cessation of exposure cannot be determined since exposure was ongoing in the
available human and laboratory animal studies.
4.3.2. Auditory Effects
4.3.2.1.1. Auditory Function: Human Studies
The TCE Subregistry from the National Exposure Registry developed by the ATSDR was
the subject of three studies (ATSDR, 2003b; Burg and Gist, 1999; Burg et al., 1995). A
fourth study (Rasmussen et al., 1993a) of degreasing workers exposed to either TCE or CFC113
also indirectly evaluated auditory function. These studies are discussed below and presented in
detail in Table 4-22.
This document is a draft for review purposes only and does not constitute Agency policy.
110 DRAFT—DO NOT CITE OR QUOTE
-------
1 Burg et al. (1999; 1995) reviewed the effects of TCE on 4,281 individuals (TCE
2 Subregistry) residentially exposed to this solvent for more than 30 consecutive days. Face-to-
3 face interviews were conducted with the TCE subregistry population and self-reported hearing
4 loss was evaluated based on personal assessment through the interview (no clinical evaluation
5 was conducted). TCE registrants that were 9 years old or younger had a statistically significant
6 increase in hearing impairment as reported by the subjects. The relative risk (RR) in this age
7 group for hearing impairments was 2.13 (95% confidence interval [CI]: 1.12-4.06) which
8 decreased to 1.12 (95% CI: 0.52-2.24) for the 10-17 age group and 0.32 (95% CI: 0.10-1.02)
9
This document is a draft for review purposes only and does not constitute Agency policy.
111 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-22. Summary of human auditory function studies
2
Reference
Subjects
Exposure
Effect
ATSDR (2003b)
116 children, under
10 yr of age, residing
near six Superfund sites.
Further study of children
in Burg et al. (1999;
1995).
Control: 182 children.
TCE and other solvents in
ground water supplies.
Exposures were modeled
using tap water TCE
concentrations and GIS for
spatial interpolation, and
LaGrange for temporal
interpolation to estimate
exposures from gestation to
1990 across the area of
subject residences.
Control = 0 ppb; low
exposure group = 0 < 23
ppb-yr; and high exposure
group = >23 ppb-yr.
Auditory screening revealed increased
incidence of abnormal middle ear
function in exposed groups as indicated
from acoustic reflex test. Adjusted odds
ratios for right ear ipsilateral acoustic
reflects control, OR: 1.0, low exposure
group, OR: 5.1,/? < 0.05; high exposure
group, OR: 1.2, p < 0.05. ORs adjusted
for age, sex, medical history and other
chemical contaminants. No significant
decrements reported in the pure tone and
typanometry screening.
Burgetal. (1995)
From an NHIS TCE
subregistry of 4,281
(4,041 living and
240 deceased) residents.
Environmentally exposed
to TCE and other solvents
via well water in Indiana,
Illinois, and Michigan.
Increase in self-reported hearing
impairments for children <9 yr.
Burgetal. (1999)
3,915 white registrants.
Mean age 34 yr (SD =
19.9 yr).
Cumulative TCE exposure
subgroups: <50 ppb,
n = 2,867; 50-500 ppb,
n = 870; 500-5,000 ppb,
n = 190; >5,000 ppb,
n = 35.
Exposure
duration subgroups:
<2 yr, 2-5 yr, 5-10
yr., >10 yr.
A statistically significant
association (adjusted for age and sex)
between duration of exposure and
self-reported hearing impairment
was found.
This document is a draft for review purposes only and does not constitute Agency policy.
112 DRAFT—DO NOT CITE OR QUOTE
-------
Rasmussen et al.
(1993c)
96 Danish metal
degreasers. Age range:
19-68 yr; .
No unexposed controls;
low exposed group is
referent.
Average
exposure duration: 7.1
yr.); range of full-time
degreasing: 1 mo to 36
yr. Exposure to TCE
or and CFC113.
(1) Low
exposure: n = 19,
average full-time
exposure 0.5 yr.
(2) Medium
exposure: n = 36,
average full-time
exposure 2.1 yr.
(3) High
exposure: n = 41,
average full-time
exposure 11 yr. Mean
U-TCA in high
exposure group = 7.7
mg/L (max = 26.1
mg/L).
Auditory impairments noted through
several neurological tests.
Significant relationship of exposure was
found with Acoustic-motor function
(p < 0.001), Paced Auditory Serial
Addition Test (p < 0.001), Rey Auditory
Verbal Learning Test (p< 0.001).
1
2 GIS = geographic information system, NHIS = National Health Interview Survey, SD = standard deviation,
3 U-TCA = urinary trichloroacetic acid.
4
This document is a draft for review purposes only and does not constitute Agency policy.
113 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
for all older age groups. A statistically significant association (when adjusted for age and sex)
was found between duration of exposure, in these studies this was length of residency, and
reported hearing impairment. The odds ratio (OR) was 2.32 (95% CI: 1.18-4.56) for subjects
exposed to TCE >2 years and <5 years, 1.17 (95% CI: 0.55-2.49) for exposure >5 years and
<10 years, 2.46 (95% CI: 1.30-5.02) for exposure durations greater than 10 years.
ATSDR (2003b) conducted a follow-up study to the TCE subregistry findings (Burg and
Gist, 1999; Burg et al., 1995) and focused on the subregistry children located in Elkhart, IN,
Rockford, IL and Battle Creek, MI using clinical tests for oral motor, speech, and hearing
function. Exposures were modeled using tap water TCE concentrations and geographic
information system (GIS) for spatial interpolation, and LaGrange for temporal interpolation to
estimate exposures from gestation to 1990 across the area of subject residences. Modeled data
were used to estimate lifetime exposures (ppb-years) to TCE in residential wells. The median
TCE exposure for the children was estimated from drinking water as 23 ppb/year of exposure
(ranging from 0-702 ppb/year). Approximately 20% (ranged from 17-21%) depending on
ipsilateral or contralateral test reflex) of the children in the TCE subregistry and 5—7% in the
control group exhibited an abnormal acoustic reflex (involuntary muscle contraction that
measures movement of the stapedius muscle in the middle ear following a noise stimulus) which
was statistically significant (p = 0.003). Abnormalities in this reflex could be an early indicator
of more serious hearing impairments. No significant decrements were reported in the pure tone
and typanometry screening.
Rasmussen et al. (1993c) used a psychometric test to measure potential auditory effects
of TCE exposure in an occupational study. Results from 96 workers exposed to TCE and other
solvents were presented in this study. Details of the exposure groups and exposure levels are
provided in Table 4-22. The acoustic motor function test was used for evaluation of auditory
function. Significant decrements (p < 0.05) in acoustic motor function performance scores
(average decrement of 2.5 points on a 10-point scale) was reported for TCE exposure.
4.3.2.1.2. Auditory Function: Laboratory Animal Studies
The ability of TCE to permanently disrupt auditory function and produce abnormalities in
inner ear histopathology has been demonstrated in several studies using a variety of test methods.
Two different laboratories have identified NOAELs following inhalation exposure for auditory
function of 1,600 ppm for 12 hours/day for 13 weeks in Long Evans rats (n = 6-10) (Rebert et
al., 1991) and 1,500 ppm for 18 hours/day, 5 days/week for 3 weeks in Wistar-derived rats
(n = 12) (Jaspers et al., 1993). The LOAELs identified in these and similar studies are
This document is a draft for review purposes only and does not constitute Agency policy.
114 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
2,500-4,000 ppm TCE for periods of exposure ranging from 4 hours/day for 5 days to
12 hours/day for 13 weeks, (e.g., Albee et al., 2006; Boyes et al., 2000; Crofton and Zhao, 1997;
Crofton et al., 1994; Fechter et al., 1998; Muijser et al., 2000; Rebert et al., 1993; Rebert et al.,
1995). Rebert et al. (1993) estimated acute blood TCE levels associated with permanent hearing
impairment at 125 ng/mL by methods that probably underestimated blood TCE values (rats were
anaesthetized using 60% carbon dioxide [CO2]). A summary of these studies is presented in
Table 4-23.
Reflex modification was used in several studies to evaluate the auditory function in
TCE-exposed animals (Boyes et al., 2000; Crofton and Zhao, 1997; Crofton et al., 1994; Crofton
and Zhao, 1993; Fechter et al., 1998; Jaspers et al., 1993; Muijser et al., 2000; Yamamura et al.,
1983). These studies collectively demonstrate significant decreases in auditory function at
midfrequency tones (8-20 kHz tones) for TCE exposures greater than 1,500 ppm after acute,
short-term, and chronic durations. Only one study (Yamamura et al., 1983) did not demonstrate
impairment in auditory function from TCE exposures as high as 17,000 ppm for 4 hours/day over
5 days. This was the only study to evaluate auditory function in guinea pigs, whereas the other
studies used various strains of rats. Despite the negative finding in Yamamura et al. (1983),
auditory testing was not performed in an audiometric sound attenuating chamber and extraneous
noise could have influenced the outcome. It is also important to note that the guinea pig has
been reported to be far less sensitive than the rat to the effects of ototoxic aromatic hydrocarbons
such as toluene.
Crofton and Zhao (1997) also presented a benchmark dose for which the calculated dose
of TCE would yield a 15 dB loss in auditory threshold. This benchmark response was selected
because a 15 dB threshold shift represents a significant loss in threshold sensitivity for humans.
The benchmark concentrations for a 15 dB threshold shift are 5,223 ppm for 1 day, 2,108 ppm
for 5 days, 1,418 ppm for 20 days and 1,707 ppm for 65 days of exposure. While more sensitive
test methods might be used and other definitions of a benchmark effect chosen with a strong
rationale, these data provide useful guidance for exposure concentrations that do yield hearing
loss in rats.
Brainstem auditory-evoked potentials (BAERs) were also measured in several studies
(Albee et al., 2006; Rebert et al., 1993; Rebert et al., 1991; Rebert et al., 1995) following at
exposures ranging from 3-13 weeks. Rebert et al. (1991) measured BAERs in male Long Evans
rats (n= 10) and F344 rats (// = 4-5) following stimulation with 4, 8, and 16 kHz sounds. The
Long-Evans rats were exposed to 0, 1,600, or 3,200 ppm TCE, 12 hours/day for 12 weeks and
the F344 rats were exposed to 0, 2,000, or 3,200 ppm TCE, 12 hours/day for 3 weeks. BAER
This document is a draft for review purposes only and does not constitute Agency policy.
115 DRAFT—DO NOT CITE OR QUOTE
-------
1 amplitudes were significantly decreased at all frequencies for F344 rats exposed to 2,000 and
2 3,000 ppm TCE and for Long Evans rats exposed to 3,200 ppm TCE. These data identify a
3
This document is a draft for review purposes only and does not constitute Agency policy.
116 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-23. Summary of animal auditory function studies
2
Reference
Exposure
route
Species/strain/
sex/number
Dose level/
exposure
duration
NOAEL;
LOAELa
Effects
Rebert et al.
(1991)
Inhalation
Rat, Long Evans,
male, 10/group
Long Evans: 0,
1,600, and 3,200
ppm; 12 h/d,
12 wk
Long
Evans:
NOAEL:
1,600 ppm;
LOAEL:
3,200 ppm
BAERs were measured.
Significant decreases in BAER
amplitude and an increase in
latency of appearance of the
initial peak (PI).
Rat, F344, male,
4-5/group
F344: 0, 2,000,
3,200 ppm;
12 h/d, 3 wk
F344:
LOAEL:
2,000 ppm
Rebert et al.
(1993)
Inhalation
Rat, Long Evans,
male, 9/group
0, 2,500, 3,000,
3,500 ppm;
8 h/d, 5 d
NOAEL:
2,500 ppm;
LOAEL:
3,000 ppm.
BAERs were measured 1-2 wk
postexposure to assess auditory
function. Significant decreases in
BAERs were noted with TCE
exposure.
Rebert et al.
(1995)
Inhalation
Rat, Long Evans,
male, 9/group
0, 2,800 ppm;
8 h/d, 5 d
LOAEL:
2,800 ppm
BAER measured 2-14 ds
postexposure at a 16 kHz tone.
Hearing loss ranged from
55-85 dB.
Crofton et al.
(1994)
Inhalation
Rat, Long Evans,
male, 7-8/group
0, 3,500 ppm
TCE; 8 h/d, 5 d
LOAEL:
3,500 ppm
BAER measured and auditory
thresholds determined 5-8 wk
postexposure. Selective
impairment of auditory function
for mid-frequency tones (8 and
16 kHz).
Crofton and
Zhou
(1997);
Boyes et al.
(2000)
Inhalation
Rat, Long Evans,
male, 9-12/group
0, 4,000, 6,000,
8,000 ppm; 6 h
NOAEL:
6,000 ppm;
LOAEL:
8,000 ppm
Auditory thresholds as
measured by BAERs for the
16 kHz tone increased with
TCE exposure. Measured
3-5 wk post exposure.
Rat, Long Evans,
male, 8-10/group
0, 1,600, 2,400,
and 3,200 ppm;
6 h/d, 5 d
NOAEL:
2,400 ppm;
LOAEL:
3,200 ppm
Rat, Long Evans,
male, 8-10/group
0, 800, 1,600,
2,400, and
3,200 ppm;
6 h/d, 5 d/wk,
4 wk
NOAEL:
2,400 ppm;
LOAEL:
3,200 ppm
Rat, Long Evans,
male, 8-10/group
0, 800,1,600,
2,400, and
3,200 ppm;
6 h/d, 5 d/wk,
13 wk
NOAEL:
1,600 ppm;
LOAEL:
2,400 ppm
3
This document is a draft for review purposes only and does not constitute Agency policy.
117 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-23. Summary of animal auditory function studies (continued)
Reference
Exposure
route
Species/strain/
sex/number
Dose level/
exposure
duration
NOAEL;
LOAEL
Effects
Fechter et al.
(1998)
Inhalation
Rat, Long Evans,
male, 12/group
0, 4,000 ppm;
6 h/d, 5 d
LOAEL:
4,000 ppm
Cochlear function measured
5-7 wk after exposure. Loss of
spiral ganglion cells noted.
Three wk postexposure, auditory
function was significantly
decreased as measured by
compound action potentials and
reflex modification.
Jaspers et al.
(1993)
Inhalation
Rat, Wistar
derived WAG-
Rii/MBL, male,
12/group
0, 1,500, and
3,000 ppm;
18 h/d, 5 d/wk,
3 wk
NOAEL:
1,500 ppm
Auditory function assessed
repeatedly 1-5 wk postexposure
for 5, 20, and 35 kHz tones; no
effect at 5 or 35 kHz; decreased
auditory sensitivity at 20 kHz,
3,000 ppm.
Muijseret al.
(2000)
Inhalation
Rat, Wistar
derived WAG-
Rii/MBL, male, 8
0, 3,000 ppm;
18 h/d, 5 d/wk,
3 wk
LOAEL:
3,000 ppm
Auditory sensitivity decreased
with TCE exposure at 4, 8, 16, and
20 kHz tones. White noise
potentiated the decrease in
auditory sensitivity.
Al
bee et al.
(2006)
In
halation
Rat,
Fischer 344,
male and
female,
10/sex/group
0, 250,
800, 2,500
ppm; 6 h/d, 5
d/wk, 13 wk
N
OAEL:
800 ppm;
LO
AEL:
2,500 ppm
Mild frequency specific hearing
deficits; focal loss of cochlear
hair cells.
Ya
mamura
et al.
(1983)
In
halation
Guinea
Pig, albino
Hartley, male,
7-10/group
0, 6,000, 12,000,
17,000 ppm;
4 h/d, 5 d
N
OAEL:
17,000
ppm
No change in auditory sensitivity
at any exposure level as measured
by cochlear action potentials and
microphonics. Study was
conducted in guinea pig and
species is less sensitive to
auditory toxicity than rats.
Studies were also not conducted
in a sound-isolation chamber and
effects may be impacted by
background noise.
2 Bolded study(ies) carried forward for consideration in dose-response assessment (see Section 5).
3
4 BAER = brainstem auditory-evoked potential.
5
6
This document is a draft for review purposes only and does not constitute Agency policy.
118 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
LOAEL at 2,000 ppm for the F344 rats and a NOAEL at 1,600 ppm for the Long Evans
rats. In subsequent studies Rebert et al. (1993; 1995) again demonstrated TCE significantly
decreases BAER amplitudes and also significantly increases the latency of appearance. Similar
results were obtained by Albee et al. (2006) for male and female F344 rats exposed to TCE for
13 weeks. The NOAEL for this study was 800 ppm based on ototoxicity at 2,500 ppm.
Notable physiological changes were also reported in a few auditory studies. Histological
data from cochleas in Long-Evans rats exposed to 4,000 ppm TCE indicated that there was a loss
in spiral ganglion cells (Fechter et al., 1998). Similarly, there was an observed loss in hair cells
in the upper basal turn of the cochlea in F344 rats exposed to 2,500-ppm TCE (Albee et al.,
2006).
4.3.2.1.3. Summary and Conclusion of Auditory Effects
Human and animal studies indicated that TCE produces decrements in auditory function.
In the human epidemiological studies (ATSDR, 2003b; Burg and Gist, 1999; Burg et al., 1995;
Rasmussen et al., 1993a) it is suggested that auditory impairments result from both an inhalation
and oral TCE exposure. A LOAEL of approximately 23 ppb-years TCE (extrapolated from
<23 ppb-years group in the ATSDR (2003b)) from oral intake is noted for auditory effects in
children. The only occupational study where auditory effects were seen reported mean urinary
trichloroacetic acid concentration, a nonspecific metabolite of TCE, of 7.7 mg/L for the high
cumulative exposure group only (Rasmussen et al., 1993a). A NOAEL or a LOAEL for auditory
changes resulting from inhalational exposure to TCE cannot be interpolated from average urinary
trichloroacetic acid (U-TCA) concentration of subjects in the high exposure group because of a
lack of detailed information on long-term exposure levels and duration (Rasmussen et al.,
1993a). Two studies (Burg and Gist, 1999; Burg et al., 1995) evaluated self-reported hearing
effects in people included in the TCE subregistry comprised of people residing near Superfund
sites in Indiana, Illinois, and Michigan. In Burg et al. (1995), interviews were conducted with
the TCE exposed population and it was found that children aged 9 years or younger had
statistically significant hearing impairments in comparison to nonexposed children. This
significant increase in hearing impairment was not observed in any other age group that was
included in this epidemiological analysis. This lack of effect in other age groups may suggest
association with another exposure other than drinking water ; however, it may also suggest that
children may be more susceptible than adults. In a follow-up analysis, Burg et al. (1999)
adjusted the statistical analysis of the original data (Burg et al., 1995) for age and sex. When
these adjustments were made, a statistically significant association was reported self-reported for
This document is a draft for review purposes only and does not constitute Agency policy.
119 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
auditory impairment and duration of residence. These epidemiological studies provided only
limited information given their use of an indirect exposure metric of residence location, no
auditory testing of this studied population and self-reporting of effects. ATSDR (2003b) further
tested the findings in the Burg studies (Burg and Gist, 1999; Burg et al., 1995) by contacting the
children that were classified as having hearing impairments in the earlier study and conducting
several follow-up auditory tests. Significant abnormalities were reported for the children in the
acoustic reflex test which suggested effects to the lower brainstem auditory pathway with the
large effect measure, the odds ratio, was reported for the high cumulative exposure group.
Strength of analyses was its adjustment for potential confounding effects of age, sex, medical
history and other chemical contaminants in drinking water supplies. The ATSDR findings were
important in that the results supported Burg et al. (1999; 1995). Rasmussen et al. (1993c) also
evaluated auditory function in metal workers with inhalation exposure to either TCE or CFC113.
Results from tasks including an auditory element suggested that these workers may have some
auditory impairment. However, the tasks did not directly measure auditory function.
Animals strongly indicated that TCE produces deficits in hearing and provides biological
context to the epidemiological study observations. Although there is a strong association
between TCE and ototoxicity in the animal studies, most of the effects began to occur at higher
inhalation exposures. NOAELs for ototoxicity ranged from 800-1,600 ppm for exposure
durations of at least 12 weeks (Albee et al., 2006; Boyes et al., 2000; Crofton and Zhao, 1997;
Rebert et al., 1991). Inhalation exposure to TCE was the route of administration in all the animal
studies. These studies either used reflex modification audiometry (Crofton and Zhao, 1997;
Crofton et al., 1994; Jaspers et al., 1993; Muijser et al., 2000) procedures or measured brainstem
auditory evoked potentials (Rebert et al., 1993; Rebert et al., 1991; Rebert et al., 1995) to
evaluate hearing in rats. Collectively, the animal database demonstrates that TCE produces
ototoxicity at midfrequency tones (4-24 kHz) and no observed changes in auditory function
were observed at either the low (<4 kHz) or high (>24 kHz) frequency tones. Additionally,
deficits in auditory effects were found to persist for at least 7 weeks after the cessation of TCE
exposure (Boyes et al., 2000; Crofton and Zhao, 1997; Fechter et al., 1998; Jaspers et al., 1993;
Rebert et al., 1991). Decreased amplitude and latency were noted in the BAERs (Rebert et al.,
1993; Rebert et al., 1991; Rebert et al., 1995) suggesting that TCE exposure affects central
auditory processes. Decrements in auditory function following reflex modification audiometry
(Crofton and Zhao, 1997; Crofton et al., 1994; Jaspers et al., 1993; Muijser et al., 2000)
combined with changes observed in cochlear histopathology (Albee et al., 2006; Fechter et al.,
1998) suggest that ototoxicity is occurring at the level of the cochlea and/or brainstem.
This document is a draft for review purposes only and does not constitute Agency policy.
120 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Changes in auditory function are noteworthy considering that TCE exposure is also
associated with immunotoxicity and inflammatory-based diseases (discussed in Section 4.6).
Autoimmune sensorineural hearing loss is a rare condition, sometimes seen with systemic
autoimmune diseases (Bovo et al., 2006; Ruckenstein, 2004). The potential role of
immunotoxicity in the observed auditory impairment seen with TCE is an area that requires
additional research.
4.3.3. Vestibular Function
4.3.3.1.1. Vestibular Function: Human Studies
The earliest reports of neurological effects resulting from TCE exposures focused on
subjective vestibular system symptoms, such as headaches, dizziness, and nausea. These
symptoms are subjective and self-reported. However, there is little doubt that these effects can
be caused by exposures to TCE, as they have been reported extensively in the literature, resulting
from occupational exposures (Grandjean et al., 1955; Liu et al., 1988; Rasmussen and Sabroe,
1986; Smith, 1970), environmental exposures (Hirsch et al., 1996), and in chamber studies
(Smith, 1970; Stewart et al., 1970).
Kylin et al. (1967) exposed 12 volunteers to 1,000 ppm (5,500 mg/m3) TCE for 2 hours
in a 1.5 x 2 x 2 meters chamber. Volunteers served as their own controls since 7 of the 12 were
pretested prior to exposure and the remaining 5 were post-tested days after exposure. Subjects
were tested for optokinetic nystagmus, which was recorded by electronystogmography, that is,
"the potential difference produced by eye movements between electrodes placed in lateral angles
between the eyes." Venous blood was also taken from the volunteers to measure blood TCE
levels during the vestibular task. The authors concluded that there was an overall reduction in
the limit ("fusion limit") to reach optokinetic nystagmus when individuals were exposed to TCE.
Reduction of the "fusion limit" persisted for up to 2 hours after the TCE exposure was stopped
and the blood TCE concentration was 0.2 mg/100 mL.
4.3.3.1.2. Vestibular Function: Laboratory Animal Data
The effect of TCE on vestibular function was evaluated by either (1) promoting
nystagmus (vestibular system dysfunction) and comparing the level of effort required to achieve
nystagmus in the presence and absence of TCE or (2) using an elevated beam apparatus and
This document is a draft for review purposes only and does not constitute Agency policy.
121 DRAFT—DO NOT CITE OR QUOTE
-------
1 measuring the balance. Overall, it was found that TCE disrupts vestibular function as presented
2 below and summarized in Table 4-24.
3 Niklasson et al. (1993) showed acute impairment of vestibular function in male- and
4 female-pigmented rats during acute inhalation exposure to TCE (2,700-7,200 ppm) and to
5 tricholoroethane (500-2,000 ppm). Both of these agents were able to promote nystagmus during
6 optokinetic stimulation in a dose related manner. While there were no tests performed to assess
7 persistence of these effects, Tham et al. (1984; 1979) did find complete recovery of vestibular
8 function in rabbits (n = 19) and female Sprague-Dawley rats (n= 11) within minutes of
9 terminating a direct arterial infusion with TCE solution.
This document is a draft for review purposes only and does not constitute Agency policy.
122 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-24. Summary of vestibular system studies
2
Reference
Exposure route
Species/strain/
sex/number
Dose level/
exposure duration
NOAEL;
LOAEL
Effects
Vestibular system studies—humans
Kylin et al.
(1967)
Inhalation
Humans, male
and female, 12
1,000 ppm; 2 h
LOAEL:
1,000 ppm
Reduction in potential to
reach nystagmus following
TCE exposure.
Vestibular system studies—animals
Tham et al.
(1979)
Intravenous
Rabbit, strain
unknown, sex
unspecified, 19
1-5 mg/kg/min
Positional nystagmus
developed once blood
levels reached 30 ppm.
Tham et al.
(1984)
Intravenous
Rat, Sprague-
Dawley, female,
11
80 (ig/kg/min
Excitatory effects on the
vestibule-oculomotor
reflex. Threshold effect at
blood (TCE) of 120 ppm or
0.9 mM/L.
Niklasson
et al.
(1993)
Inhalation
Rat, strain
unknown, male
and female, 28
0, 2,700, 4,200,
6,000, 7,200 ppm; 1
h
LOAEL:
2,700 ppm
Increased ability to produce
nystagmus.
Umezu
et al.
(1997)
Intraperitoneal
Mouse, ICR,
male, 116
0, 250, 500, or 1,000
mg/kg, single dose
and evaluated 30 min
postadministration
NOAEL:
250 mg/kg
LOAEL:
500 mg/kg
Decreased equilibrium and
coordination as measured by
the Bridge test (staying time
on an elevated balance
beam).
3
4
5 The finding that trichloroethylene can yield transient abnormalities in vestibular function
6 is not unique. Similar impairments have also been shown for toluene, styrene, along with
7 trichloroethane (Niklasson et al., 1993) and by Tham et al. (1984) for a broad range of aromatic
8 hydrocarbons. The concentration of TCE in blood at which effects were observed for TCE (0.9
9 mM/L) was quite close to that observed for most of these other vestibulo-active solvents.
10
4.3.3.1.3. Summary and Conclusions for the Vestibular Function Studies
11 Studies of TCE exposure in both humans and animals reported abnormalities in vestibular
12 function. Headaches, dizziness, nausea, motor incoordination, among other subjective symptoms
13 are reported in occupational epidemiological studies of TCE exposure (Grandjean et al., 1955;
14 Hirsch et al., 1996; Liu et al., 1988; Rasmussen and Sabroe, 1986; Smith, 1970; Stewart et al.,
15 1970). One human exposure study (Kylin et al., 1967) found that vestibular function was
16 affected following an acute exposure to 1,000-ppm TCE (LOAEL). Individuals had a decreased
This document is a draft for review purposes only and does not constitute Agency policy.
123 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
threshold to reach nystagmus than when exposed to TCE than to air. Animal studies also
evaluated the threshold to reach nystagmus and reported that TCE decreased the threshold to
produce nystagmus in rats (LOAEL: 2,700 ppm; Niklasson et al., 1993; Tham et al., 1984) and
rabbits (Tham et al., 1984).
4.3.4. Visual Effects
4.3.4.1.1. Visual Effects: Human Studies
Visual impairment in humans has been demonstrated following exposures through
groundwater (Kilburn, 2002b; Reif et al., 2003) from occupational exposure through inhalation
(Rasmussen et al., 1993c; Troster and Ruff, 1990) and from a controlled inhalation exposure
study (Vernon and Ferguson, 1969). Visual functions such as color discrimination and
visuospatial learning tasks are impaired in TCE-exposed individuals. Additionally, an acute
exposure can impair visual depth perception. Details of the studies are provided below and
summarized in Table 4-25.
Geographical-based studies utilized color discrimination and contrast sensitivity tests to
determine the effect of TCE exposure on vision. In these studies it was reported that TCE
exposure significantly increased color discrimination errors (Kilburn, 2002b) or decreases in
contrast sensitivity tests approached statistical significance after adjustments for several possible
confounders (p = 0.06 or 0.07; Reif et al., 2003). Exposure in Kilburn (2002b) is poorly
characterized, and for both studies, TCE is one of several contaminants in drinking water
supplies; neither study provides an estimate of an individual's exposure to TCE.
Rasmussen et al. (1993c) evaluated visual function in 96 metal workers, working in
degreasing at various factories and with exposure to TCE or CFC113. Visual function was tested
through the visual gestalts test (visual perception) and a visual recall test. In the visual gestalts
test, the number of total errors significantly increased from the low group (3.4 errors) to the high
exposure group (6.5 errors; p = 0.01). No significant changes were observed in the visual recall
task. Troster and Ruff (1990) presented case studies conducted on two occupationally exposed
workers to TCE. Both patients presented with a visual-spatial task and neither could complete
the task within the number of trials allowed suggesting visual function deficits as a measure of
impaired visuospatial learning.
In a chamber exposure study (Vernon and Ferguson, 1969), eight male volunteers (ages
21-30) were exposed to 0, 100, 300, and 1,000-ppm TCE for 2 hours. Each individual was
exposed to all TCE concentrations and a span of at least 3 days was given between exposures.
This document is a draft for review purposes only and does not constitute Agency policy.
124 DRAFT—DO NOT CITE OR QUOTE
-------
1 When the individuals were exposed to 1,000-ppm TCE (5,500 mg/m ), significant abnormalities
2 were noted in depth perception as measured by the Howard-Dolman test (p < 0.01). There were
3 no effects on the flicker fusion frequency test (threshold frequency at which the individual sees a
This document is a draft for review purposes only and does not constitute Agency policy.
125 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-25. Summary of human visual function studies
2
Reference
Subjects
Exposure
Effect
Kilburn et
al. (2002b)
236 residents near a
microchip plant in
Phoenix, AZ.
Controls: 67 local
referents from Phoenix,
AZ and 161 regional
referents from
Wickenburg, AZ.
TCE, TCA, 1,1-DCE,
1,2-DCE, PCE, and VC
detected in well water up to
260,000 ppm; TCE
concentrations in well water
were 0.2-10,000 ppb.
Exposure duration ranged
from 2-37 yr.
Exposure duration
ranged from 2-37 yr.
Color discrimination errors were
increased among residents
compared to regional referents
(p < 0.01). No adjustment for
possible confounding factors.
Reif et al.
(2003)
143 residents of the
Rocky Mountain Arsenal
community of Denver.
Referent group at lowest
concentration (<5 ppb).
Exposure modeling of TCE
concentrations in groundwater and
in distribution system to estimate
mean TCE concentration by census
block of residence. High exposure
group >15 ppb.
Medium exposure group >5 ppb and
<15 ppb.
Low exposure referent group
<5 ppb.
Contrast sensitivity test
performances (C and D) was
marginally statistically significant
(p = 0.06 and 0.07, respectively).
No significant effects reported for
the Benton visual retention test.
Significant decrements (p = 0.02)
were reported in the Benton visual
retention test when stratified with
alcohol consumption.
This document is a draft for review purposes only and does not constitute Agency policy.
126 DRAFT—DO NOT CITE OR QUOTE
-------
Rasmussen
et al.
(1993c)
96 Danish metal
degreasers. Age range:
19-68; no unexposed
controls; low exposure
group was referent.
Average exposure
duration: 7.1 yr); range of
full-time degreasing: 1 mo to
36 yr. Exposure to TCE or
CFC113.
1) Low exposure:
n= 19, average full-time expo
0.5 yr.
2) Medium exposure:
n = 36, average full-time
exposure 2.1 yr.
3) high exposure:
n = 41, average full-time
exposure 11 yr. TCA in high
exposure group = 7.7 mg/L
(max = 26.1 mg/L).
Statistically significant relationship
of exposure was found with the
Visual Gestalts learning and
retention test (cognitive test)
indicating deficits in visual
performance.
Troster and
Ruff (1990)
Two occupationally
TCE-exposed workers.
Controls: two groups of
n = 30 matched controls;
(all age and education
matched).
Exposure
concentration unknown.
Exposure duration, 3-8 mos.
Both workers experienced impaired
visuospatial learning.
Vernon and
Ferguson
(1969)
8 male volunteers age
range 21-30; self
controls.
0, 100, 300, and 1,000
ppm of TCE for 2 h.
Statistically significant effects on
visual depth perception as measured
by the Howard-Dolman test.
NOAEL: 300 ppm; LOAEL: 1,000
ppm; No significant changes in any
of the other visual test
measurements.
1
2 DCE = dichloroethylene.
3
4 flicker as a single beam of light) or on the form perception illusion test (volunteers presented
5 with an illusion diagram).
6
4.3.4.1.2. Visual Effects: Laboratory Animal Data
7 Changes in visual function have been demonstrated in animal studies during acute
8 (Boyes et al., 2003, 2005) and subchronic exposure (Blain et al., 1994; Rebert et al., 1991). In
This document is a draft for review purposes only and does not constitute Agency policy.
127 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
these studies, the effect of TCE on visual evoked responses to patterns (Boyes et al., 2003, 2005;
Rebert et al., 1991) or a flash stimulus (Blain et al., 1994; Rebert et al., 1991) were evaluated.
Overall, the studies demonstrated that exposure to TCE results in significant changes in the
visual evoked response, which is reversible once TCE exposure is stopped. Details of the studies
are provided below and are summarized in Table 4-26.
Boyes et al. (2003; 2005) exposed adult, male Long-Evans rats were to TCE in a
head-only exposure chamber while pattern onset/offset visual evoked potentials (VEPs) were
recorded. Exposure conditions were designed to provide concentration x time products of
0 ppm/hours (0 ppm for 4 hours) or 4,000 ppm/hours (see Table 4-26 for more details). VEP
amplitudes were depressed by TCE exposure during the course of TCE exposure. The degree of
VEP depression showed a high correlation with the estimated brain TCE concentration for all
levels of atmospheric TCE exposure.
In a subchronic exposure study, Rebert et al. (1991) exposed male Long Evans rats to
1,600- or 3,200-ppm TCE, for 12 weeks, 12 hours/day. No significant changes in flash evoked
potential measurements were reported following this exposure paradigm. Decreases in pattern
reversal visual evoked potentials (N1P1 amplitude) reached statistical significance following 6,
9, and 12 weeks of exposure. The drop in response amplitude ranged from approximately 20%
after 8 weeks to nearly 50% at Week 14 but recovered completely within 1 week postexposure.
This transient effect of TCE on the peripheral visual system has also been reported by
Blain (1994) in which New Zealand albino rabbits were exposed by inhalation to 350- and
700-ppm TCE 4 hours/day, 4 days/week for 12 weeks. Electroretinograms (ERG) and
oscillatory potentials (OPs) were recorded weekly under mesopic conditions. Recordings from
the 350- and 700-ppm exposed groups showed a significant increase in the amplitude of the a -
and b-waves (ERG). The amplitude of the OPs was significantly decreased at 350 ppm {51%)
and increased at 700 ppm (117%). These electroretinal changes returned to pre-exposure
conditions within 6 weeks after the inhalation stopped.
This document is a draft for review purposes only and does not constitute Agency policy.
128 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-26. Summary of animal visual system studies
2
Reference
Exposure route
Species/strain/
sex/number
Dose level/
exposure duration
NOAEL;
LOAEL
Effects
Rebert
et al.
(1991)
Inhalation
Rat, Long Evans,
male, 10/group
0, 1,600, and
3,200 ppm; 12 h/d,
12 wk
NOAEL:
1,600 ppm
Significant amplitude
decreases in pattern reversal
evoked potentials (N1P1
amplitude) at 6, 9, and
12 wk.
Boyes et al.
(2003)
Inhalation
Rat, Long Evans,
male,
9-10/group
0 ppm, 4 h;
1,000 ppm, 4;
2,000 ppm, 2 h;
3,000 ppm, 1.3 h
4,000 ppm, 1 h
LOAEL:
1,000 ppm,
4 h
Visual function
significantly affected as
measured by decreased
amplitude (F2) in Fourier-
transformed visual evoked
potentials. Peak brain TCE
concentration correlated
with dose-response.
Boyes et al.
(2005)
Inhalation
Rat, Long Evans,
male,
8-10/group
0 ppm, 4 h;
500 ppm, 4 h;
1,000 ppm, 4 h;
2,000 ppm, 2 h;
3,000 ppm, 1.3 h
4,000 ppm, 1 h;
5,000 ppm, 0.8 h
LOAEL:
500 ppm,
4 h
Visual function
significantly affected as
measured by decreased
amplitude (F2) in Fourier-
transformed visual evoked
potentials. Peak brain TCE
concentration correlated
with dose-response.
Blain et al.
(1994)
Inhalation
Rabbit, New
Zealand albino,
male, 6-8/group
0,350, 700 ppm;
4 h/d, 4 d/wk, 12 wk
LOAEL:
350 ppm
Significant effects noted in
visual function as
measured by ERG and
OPs immediately after
exposure. No differences
in ERG or OP
measurements were noted
at 6 wk post-TCE
exposure.
3
4 Bolded study(ies) carried forward for consideration in dose-response assessment (see Section 5).
5
6 ERG = electroretinogram, OP = oscillatory potential.
7
8
4.3.4.1.3. Summary and Conclusion of Visual Effects
9 Changes in visual function are reported in human studies. Although central visual function
10 was not evaluated in the human studies (such as electroretinograms, evoked potential
11 measurements), clinical tests indicated deficits in color discrimination (Kilburn, 2002b), visual
12 depth perception (Vernon and Ferguson, 1969) and contrast sensitivity (Reif et al., 2003). These
13 changes in visual function were observed following both an acute exposure (Vernon and Ferguson,
This document is a draft for review purposes only and does not constitute Agency policy.
129 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
1969) and residence in areas with groundwater contamination with TCE and other chemicals
(Kilburn, 2002b; Reif et al., 2003). The exposure assessment approach of Reif et al., who adopted
exposure modeling and information on water distribution patterns, is considered superior to that of
Kilburn (2002b) who used residence location as a surrogate for exposure. In the one acute,
inhalation study (Vernon and Ferguson, 1969), a NOAEL of 300 ppm and a LOAEL of 1,000 ppm
for 2 hours was reported for visual effects. A NOAEL is not available from the drinking water
studies since well water TCE concentration is a poor surrogate for an individual's TCE ingestion
(2002b) and limited statistical analysis comparing high exposure group to low exposure group
(Reifetal., 2003).
Animal studies have also demonstrated changes in visual function. All of the studies
evaluated central visual function by measuring changes in evoked potential response following a
visual stimulus that was presented to the animal. Two acute exposure inhalation studies (Boyes et
al., 2003; Boyes et al., 2005) exposed Long Evans rats to TCE based on a concentration x time
schedule (Haber's law) and reported decreases in visual evoked potential amplitude. All of the
exposures from these two studies resulted in decreased visual function with a LOAEL of 500 ppm
for 4 hours. Another important finding that was noted is the selection of the appropriate
dose-metric for visual function changes following an acute exposure. Boyes et al. (2003; 2005)
found that among other potential dose-metrics, brain TCE concentration was best correlated with
changes in visual function as measured by evoked potentials under acute exposure conditions.
Two subchronic exposure studies (Blain et al., 1994; Rebert et al., 1991) demonstrated visual
function changes as measured by pattern reversal evoked potentials (Rebert et al., 1991) or
electroretinograms/oscillatory potentials (Blain et al., 1994). Unlike the other three visual function
studies conducted with rats, Blain et al., demonstrated these changes in rabbits. Significant
changes in ERGs and oscillatory potentials were noted following a 12-week exposure at 350 ppm
(LOAEL) in rabbits (Blain et al., 1994) and in rats exposed to 3,200-ppm TCE for 12 weeks there
were significant decreases in pattern reversal evoked potentials but no effect was noted in the
1,600-ppm exposure group (Rebert et al., 1991). Both subchronic studies examined visual
function following an exposure-free period of either 2 weeks (Rebert et al., 1991) or 6 weeks
(Blain et al., 1994) and found that visual function returned to pre-exposure levels and the changes
are reversible.
This document is a draft for review purposes only and does not constitute Agency policy.
13 0 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
4.3.5. Cognitive Function
4.3.5.1.1. Cognitive Effects: Human Studies
Effects of TCE on learning and memory have been evaluated in populations
environmentally exposed to TCE through well water, in workers occupationally exposed through
inhalation and under controlled exposure scenarios. Details of the studies are provided in
Table 4-27 and discussed briefly below. In the geographical-based studies (Kilburn, 2002b;
Kilburn and Warshaw, 1993) cognitive function was impaired in both studies and was evaluated
by testing verbal recall and digit span memory among other measures. In Arizona residents
involved in a lawsuit (Kilburn and Warshaw, 1993), significant impairments in all
three cognitive measures were reported; verbal recall (p = 0.001), visual recall (p = 0.03) and
digit span test (p = 0.07), although a question exists whether the referent group was comparable
to exposed subjects and the study's lack of consideration of possible confounding exposures in
statistical analyses. Significant decreases in verbal recall ability was also reported in another
environmental exposure study where 236 residents near a microchip plant with TCE
concentration in well water ranging from 0.2-10,000 ppb (Kilburn, 2002b).
Cognitive impairments are assessed in the occupational exposure and case studies
(Rasmussen et al., 1993c; Rasmussen et al., 1993d; Troster and Ruff, 1990). In metal degreasers
occupationally exposed to TCE and CFC113, significant cognitive performance decreases were
noted in verbal recall testing (p = 0.03) and verbal learning (p = 0.04; Rasmussen et al., 1993a).
No significant effects were found in the visual recall or digit span test for these workers. Troster
and Ruff (1990) reported decrements (no statistical analysis performed) in cognitive performance
as measured in verbal and visual recall tests that were conducted immediately after presentation
(learning phase) and one hour after original presentation (retention/memory phase) for two case
studies.
Several controlled (chamber) exposure studies were conducted to cognitive ability during
TCE exposure and most did not find any significant decrements in the neurobehavioral
measurement. Only Salvini et al. (1971) found significant decrements in cognitive function.
Six males were exposed to 110 ppm (550 mg/m3) TCE for 4-hour intervals, twice per day.
Statistically significant results were observed for perception tests learning (p < 0.001), mental
fatigue (p < 0.01), subjects (p < 0.05); and choice reaction time (CRT) learning (p < 0.01),
mental fatigue (p < 0.01), subjects (p < 0.05). Triebig et al. (1977a, b) exposed seven total
subjects (male and female) to 100 ppm TCE for 6 hours/day, 5 days/week and did not report any
decreases in cognition but details on the experimental procedures were not provided.
Additionally, Gamberale et al. (1976) found that subjects exposed to TCE as high as 194 ppm for
This document is a draft for review purposes only and does not constitute Agency policy.
131 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
70 minutes did not exhibit any impairments on a short term memory test in comparison to an air
exposure.
4.3.5.1.2. Cognitive Effects: Laboratory Animal Studies
Many reports have demonstrated significant differences in performance of learning tasks
such as the speed to complete the task. However, there is little evidence that learning and
memory function are themselves impaired by exposure. There are also limited data that suggest
alterations in the hippocampus of laboratory animals exposed to TCE. Given the important role
that this structure plays in memory formation, such data may be relevant to the question of
whether TCE impairs memory. The studies are briefly discussed below and details are provided
in Table 4-28.
Two studies (Kulig, 1987; Umezu et al., 1997) reported decreased performance in
operant-conditioning cognitive tasks for rodents. Kishi et al. (1993) acutely exposed Wistar rats
to TCE at concentrations of 250, 500, 1,000, 2,000, and 4,000 ppm for 4 hours. Rats exposed to
250 ppm TCE and higher showed a significant decrease both in the total number of lever presses
and in avoidance responses compared with controls. The rats did not recover their pre-exposure
performance until about 2 hours after exposure. Likewise, Umezu et al. (1997) reported a
depressed rate of operant responding in male ICR strain mice (n = 6, exposed to all TCE doses,
see Table 4-28) in a conditioned avoidance task that reached significance with i.p. injections of
1,000 mg/kg. Increased responding during the signaled avoidance period at lower doses (250
and 500 mg/kg) suggests an impairment in ability to inhibit responding or failure to attend to the
signal.
Although cognitive impairments are noted, two additional studies indicate no change in
cognition with continuous TCE exposure or improvements in cognitive tasks. No decrements in
cognitive function as measured by the radial arm maze were observed in Mongolian gerbils
exposed continuously by inhalation to 320 ppm TCE for 9 months (Kjellstrand et al., 1980).
Improved performance was noted in a Morris swim test for weanling rats orally dosed with
5.5 mg/day for 4 weeks followed by 2 weeks of no exposure and an additional 2 weeks of
8.5 mg/day (Isaacson et al., 1990). This improved performance occurred despite a loss in
hippocampal myelination.
This document is a draft for review purposes only and does not constitute Agency policy.
132 DRAFT—DO NOT CITE OR QUOTE
-------
4.3.5.1.3. Summary and Conclusions of Cognitive Function Studies
1 Human environmental and occupational exposure studies suggest impairments in
2 cognitive function. Kilburn and Warshaw (1993) and Kilburn (2002b) reported memory deficits
3 individuals although a question exists whether the referent group was comparable to exposed
4 subjects and these studies lack of consideration of possible confounding exposures in statistical
5 analyses. Significant impairments were found in visual and verbal recall and with the digit span
6 test. Similarly, in occupational exposure studies (Rasmussen et al., 1993c; Rasmussen et al.,
7 1993d; Troster and Ruff, 1990), short term memory tests indicated that immediate memory and
8 learning were impaired in the absence of an effect on digit span performance. In controlled
9 exposure and/or chamber studies, two studies did not report any cognitive impairment
This document is a draft for review purposes only and does not constitute Agency policy.
13 3 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-27. Summary of human cognition effect studies
2
Reference
Subjects
Exposure
Effect
Kilburn and
Warshaw
(1993)
170 residents living in
Southwest Tucson with
TCE, other solvents, and
chromium in
groundwater.
Control: 68 residential
referents matched to
subjects from two
previous studies of waste
oil and oil refinery
exposures.
>500 ppb of TCE
in well water before 1981
and 25-100 ppb
afterwards.
Exposure duration
ranged from 1-25 yr.
Decreased performance in the digit span
memory test and story recall ability.
Kilburn 2002a
236 residents near a
microchip plant.
Controls: 67 local
referents from Phoenix,
AZ and 161 regional
referents from
Wickenburg, AZ.
<0.2-10,000 ppb
of TCE, <0.2-260,000
ppb TCA, <0.2-6,900
ppb 1,1-DCE, <0.2-1,600
1,2-DCE, <0.2-23,000
ppb PCE, <0.02-330 ppb
VC in well water.
Exposure duration
ranged from 2-37 yr.
Exposure duration ranged
from 2-37 yr.
Cognitive effects decreased as measured
by lower scores on Culture Fair 2A,
vocabulary, grooved pegboard
(dominant hand), trail making test, and
verbal recall (i.e., memory).
This document is a draft for review purposes only and does not constitute Agency policy.
134 DRAFT—DO NOT CITE OR QUOTE
-------
Rasmussen
(1993c; 1993d)
96 Danish metal
degreasers. Age range:
19-68; No external
controls.
Average exposure
duration: 7.1 yr.); range
of full-time degreasing: 1
mo to 36 yr.
1) Low exposure:
n = 19, average full-time
expo 0.5 yr.
2) Medium
exposure: n = 36, average
full-time exposure 2.1 yr.
3) High exposure:
n = 41, average full-time
exposure 11 yr. TCA in
high exposure group =
7.7 mg/L (max = 26.1
mg/L).
Cognitive impairment (psycho-organic
syndrome) prevalent in exposed
individuals. The incidence of this
syndrome was 10.5% in the low
exposure, 39.5% for medium exposure,
and 63.4% for high exposure. Age is a
confounder. Dose-response with 9 of
15 tests; Controlling for confounds,
significant relationship of exposure was
found with Acoustic-motor function
(p < 0.001), Paced Auditory Serial
Addition Test (p < 0.001), Rey Auditory
Verbal-Learning Test (p < 0.001),
vocabulary (p < 0.001) and visual
gestalts (p < 0.001); significant age
effects. Age is a confounder.
Troster and
Ruff (1990)
Two occupationally
TCE-exposed workers.
Controls: two groups of
n = 30 matched controls;
(all age and education
matched.
Exposure
concentration unknown;
Exposure duration, 3-8
mo.
Both TCE cases exhibited significant
deficits in verbal recall and visuospatial
learning.
This document is a draft for review purposes only and does not constitute Agency policy.
13 5 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-27. Summary of human cognition effect studies (continued)
Reference
Subjects
Exposure
Effect
Triebig (1976)
Controlled exposure
study four females, three
males.
Controls: four females,
three males.
0, 100 ppm (550
mg/m3), 6 h/d, 5 d.
There was no correlation seen between
exposed and unexposed subjects for any
measured psychological test results. No
methods description was provided.
Triebig (1977b)
Seven men and one
woman occupationally
exposed with an age
range from 23-38 yr.
No control group.
50 ppm (260
mg/m ). Exposure
duration not reported.
The psychological tests showed no
statistically significant difference in the
results before or after the exposure-free
time period. No methods description
was provided.
Triebig (1977a)
Controlled exposure
study on three male and
four female students.
Control: three male and
four female students.
0, 100 ppm (550
mg/m3), 6 h/d, 5 d
No significantly different changes were
obtained. No methods description was
provided.
Salvini et al.
(1971)
Controlled exposure
study six students, male.
Self used as control.
TCE
concentration was
110 ppm for 4-h intervals,
twice per d. 0 ppm
control exposure for all as
self controls
Statistically significant results were
observed for perception tests learning
(p < 0.001) and CRT learning
(p<0.01).
Gamberale et al.
(1976)
15 healthy men aged
20-31 yr old.
Controls: Within
Subjects (15
self-controls).
0 mg/m3, 540
mg/m3 (97 ppm), 1,080
mg/m3 (194 ppm), 70 min
Repetition of the testing led to
a pronounced improvement in
performance as a result of the
training effect; No interaction effects
between exposure to TCE and training.
Stewart et al.
(1970)
130 (108 males, 22
females).
Controls: 63 unexposed
men.
TCA metabolite
levels in urine were
measured: 60.8% had
levels up to 20 mg/L, and
82.1% had levels up to
60 mg/L
No significant effect on cognitive tests
noted, but more effort required to
perform the test in exposed group.
Chalupa (1960)
Case study—6 subjects.
Average age 38.
No exposure data were
reported
80% of those with pathological EEG
displayed memory loss; 30% of those
with normal EEGs displayed memory
loss.
1
2 DCE = dichloroethylene, EEG = electroencephalogram.
3
This document is a draft for review purposes only and does not constitute Agency policy.
13 6 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7 Table 4-28. Summary of animal cognition effect studies
8
Reference
Exposure
route
Species/strain/
sex/number
Dose level/
exposure duration
NOAEL;
LOAEL
Effects
Kjellstrand
et al.
(1980)
Inhalation
Gerbil,
Mongolian,
males and
females,
12/sex/dose
0, 320 ppm; 9 mo,
continuous (24 h/d) except
1-2 h/wk for cage cleaning
NOAEL: 320
ppm
No significant effect
on spatial memory
(radial arm maze).
Isaacson
et al.
(1990)
Oral,
drinking
water
Rat, Sprague-
Dawley, male
weanlings,
12/dose
(1) 0 mg/kg-day, 8 wk.
(2) 5.5 mg-day
(47 mg/kg-daya), 4 wk + 0
mg/kg-day, 4 wk.
(3) 5.5 mg/dd, 4 wk
(47 mg/kg-dayb) + 0
mg/kg-day, 2 wk + 8.5
mg/dd (24 mg/kg-dayb),
2 wk
NOAEL:
5.5 mg/d,
4 wk—spatial
learning
LOAEL:
5.5 mg/d—
hippocampal
demyelination
Decreased latency
to find platform in
the Morris water
maze (Group #3);
Hippocampal
demyelination
observed in all
TCE-treated
groups.
Kishi et al.
(1993)
Inhalation
Rats,
Wistar, male,
number not
specified
0, 250,500,
1,000, 2,000, and 4,000
ppm, 4 h
LOAE
L: 250 ppm
Decreased lever
presses and
avoidance responses
in a shock avoidance
task.
Umezu
et al.
(1997)
Intra-
peritoneal
Mouse, ICR,
male, six
exposed to all
treatments
(repeated
exposure)
0, 125, 250, 500, and 1,000
mg/kg, single dose and
evaluated 30 min
postadministration
NOAEL: 500
mg/kg
LOAEL: 1,000
mg/kg
Decreased response
rate in an operant
response—condition
avoidance task.
Oshiro
et al.
(2004)
Inhalation
Rat,
Long Evans,
male, 24
0, 1,600, and
2,400 ppm; 6 h/d, 5
d/wk, 4 wk
NOA
EL: 2,400
ppm
No change in
reaction time in
signal detection task
and when challenged
with amphetamine,
no change in
response from
control.
9
10 amg/kg-day conversion estimated from average male Sprague-Dawley rat body weight from ages 21-49 d (118 g)
11 for the 5.5 mg dosing period and ages 63-78 d (354 g) for the 8.5 mg dosing period.
12 Bolded study(ies) carried forward for consideration in dose-response assessment (see Section 5).
This document is a draft for review purposes only and does not constitute Agency policy.
13 7 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
(Gamberale et al., 1976; Stewart et al., 1970) and one study (Salvini et al., 1971) reported
significant impairments in learning memory and complex choice reaction tasks. All of the
controlled exposure studies were acute and/or short-term exposure studies and the sensitivity of
test procedures is unknown due to the lack of methodologic information provided in the reports.
Despite identified study deficiencies, these studies collectively suggest cognitive function
impairment.
The animal studies measured cognitive function through spatial memory and operant
responding tasks. In the two studies where spatial memory was evaluated, there was either no
effect at 320 ppm TCE (Kjellstrand et al., 1980) or improved cognitive performance in weanling
rats at a dose of 5.5 mg/day for four weeks (Isaacson et al., 1990). Improved cognitive
performance was observed in weanling rats (Isaacson et al., 1990) and could be due to
continuing neurodevelopment as well as compensation from other possible areas in the brain
since there was a significant loss in hippocampal myelination. Significant decreases in operant
responding (avoidance/punished responding) during TCE exposure were reported in two studies
(Kishi et al., 1993; Umezu et al., 1997). When TCE exposure was discontinued operant
responding return to control levels and it is unclear if the significant effects are due to decreased
motor function or decreased cognitive ability.
4.3.6. Psychomotor Effects
There is considerable evidence in the literature for both animals and humans on
psychomotor testing although human and laboratory animal studies utilize very different
measures of motor behavior. Generally, the human literature employs a wide variety of
psychomotor tasks and assesses error rates and reaction time in the performance of the task. The
laboratory animal data, by contrast, tend to include unlearned naturalistic behaviors such as
locomotor activity, gait changes, and foot splay to assess neuromuscular ability.
This document is a draft for review purposes only and does not constitute Agency policy.
13 8 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
4.3.6.1.1. Psychomotor Effects: Human Studies
The effects of TCE exposure on psychomotor response have been studied primarily as a
change in reaction time (RT) with studies on motor dyscoordination resulting from TCE
exposure providing subjective reporting.
4.3.6.1.2. Reaction time
Several studies have evaluated the effects of TCE on reaction time using simple and
choice reaction time tasks (simple reaction time [SRT] and CRT tasks). The studies are
presented below and summarized in more detail in Table 4-29.
Increases in reaction time were observed in environmental exposure studies by Kilburn
(2002b), Kilburn and Warshaw (1993), and Kilburn and Thornton (1996) as well as in an
occupational exposure study by Gun et al. (1978). All populations except that of Gun et al.
(1978) were exposed through groundwater contaminated as the result of environmental spills and
the exposure duration was for at least 1 year and exposure levels ranged from 0.2-10,000 ppb for
the three studies. Kilburn and Warshaw (1993) reported that SRT significantly increased from
281 ± 55 msec to 348 ± 96 msec in individuals (p < 0.0001). CRT of the exposed subjects was
93 msec longer (p < 0.0001) than referents. Kilburn and Thornton (1996) evaluated SRT and
CRT function and also found similar increases in reaction time. The average SRT and CRT for
the combined control groups were 276 msec and 532 msec, respectively. These reaction times
increased in the TCE exposure group where the average SRT was 334 msec and CRT was
619 msec. Similarly, Kilburn (2002b) compared reaction times between 236 TCE-exposed
persons and the 161 unexposed regional controls. SRTs significantly increased from
283 ± 63 msec in controls to 334 ±118 msec in TCE exposed individuals (p < 0.0001).
Similarly, CRTs also increased from 510 ± 87 msec to 619 ± 153 msec with exposure to TCE
(p< 0.0001).
No effect on SRT was reported in a geographical-based study by Reif et al. (2003). SRTs
were 301msec for the lowest exposure group and 316 msec for the highest exposure group
(p = 0.42). When the SRT data were analyzed individuals that consumed at least on alcoholic
drink per month (n = 80), a significant increase (18% ,p< 0 .04) in SRT times were observed
between the lowest exposure and the highest exposure groups. In TCE exposed individuals who
did not consume alcohol (n = 55), SRTs decreased from 321 msec in the lowest exposed group to
296 msec in the highest exposed group, but this effect was not statistically significantly different.
A controlled exposure (chamber study) of 15 healthy men aged 20-31 years old, were exposed to
-3
0, 540, and 1,080 mg/m TCE for 70 minutes or served as his own control, reported no
statistically significant differences with the SRT or CRT tasks. However, in the RT-Addition
This document is a draft for review purposes only and does not constitute Agency policy.
13 9 DRAFT—DO NOT CITE OR QUOTE
-------
1 test the level of performance varied between the different exposure conditions (F(2.24) = 4.35;
2 p< 0.05) and between successive measurement occasions (F(2.24) = 19.25; p < 0.001).
3
This document is a draft for review purposes only and does not constitute Agency policy.
140 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-29. Summary of human choice reaction time studies
2
Reference
Subjects
Exposure
Effect
Kilburn
(2002b)
236 residents near a microchip
plant in Phoenix, AZ.
Controls:
161 regional referents from
Wickenburg, AZ.
67 referents from Phoenix, AZ
not residing near a plant.
0.2-10,000
ppb of TCE,
chronic exposure.
Simple and choice reaction times were
increased in the exposed group
(p < 0.05).
Kilburn and
Warshaw
(1993)
160 residents living in Southwest
Tucson with TCE and other
solvents in groundwater.
Control: 68 residential referents
matched to subjects from
two previous studies of waste oil
and oil refinery exposures.
>500 ppb of
TCE in well-water
before 1981 and
25-100 ppb
afterwards.
Exposure
duration ranged
from 1-25 yr.
Mean simple reaction time was
67 milliseconds (msec) longer than the
referent group p < 0.0001).
CRT of the exposed subjects was
between 93-100 msec longer in three
different trials (p < 0.0001) compared to
referents.
Reif et al.
(2003)
143 residents of the Rocky
Mountain Arsenal community of
Denver.
Referent group at lowest
concentration (<5 ppb).
High exposure group
>15 ppb.
Medium exposure
group >5 ppb and <15
ppb.
Low exposure referent
group <5 ppb.
Significant increase in reaction time as
measured by the simple reaction time
test (p < 0.04) in only among subjects
who reported alcohol use (defined as
having at least one drink per mo).
Kilburn and
Thornton
(1996)
Group A: Registered voters from
Arizona and Louisiana with no
exposure to TCE: n = 264, aged
18-83. Group B volunteers
from California n = 29 (17 males
and 12 females). Group C:
exposed to TCE and other
chemicals for 5 yr or more
n = 217.
No exposure
or groundwater
analyses reported.
Significant increase in simple and
choice reaction time in exposed group
compared to the unexposed populations.
Gamberale
et al. (1976)
15 healthy men aged 20-31 yr
old.
Controls: Within subjects
(15 self-controls).
0 mg/m3,
540 mg/m3 (97
ppm), 1,080 mg/m3
(194 ppm), 70 min.
No change in CRT or SRT. Increase in
time required to perform the RT-
Addition Test (task for adding numbers)
(p < 0.05).
3
This document is a draft for review purposes only and does not constitute Agency policy.
141 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Table 4-29. Summary of human choice reaction time studies (continued)
Reference
Subjects
Exposure
Effect
Gun et al.
(1978)
Four female workers from one
plant exposed to TCE and four
female workers from another
plant exposed to TCE +
nonhalogenated hydrocarbon
solvent.
Control: (n = 8) four unexposed
female workers from each plant.
3-419 ppm,
duration not
specified.
TCE-only exposure increased reaction
time in comparison to controls. In
TCE + solvent group, ambient TCE was
lower and mean reaction time shortened
in Session 2, then rose subsequently to
be greater than at the start.
4.3.6.1.3. Muscular dyscoordination
Three studies examined motor dyscoordination effects from TCE exposure using
subjective and self-reported individual assessment. Rasmussen et al. (1993a) presented findings
on muscular dyscoordination for 96 metal degreasers exposed to either TCE or CFC113. A
statistically significant increasing trend of dyscoordination with TCE exposure was observed
(p = 0.01) in multivariate regression analyses which adjusted for the effects of age, neurological
disease, arteriosclerotic disease, and alcohol abuse. Furthermore, a greater number of abnormal
coordination tests were observed in the higher exposure group compared to the low exposure
group (p = 0.003).
Gash et al. (2008) reported fine motor hand movement times in subjects who had filed
workman compensation claims were significantly slower (p < 0.0001) than age-matched
nonexposed controls. Exposures were based on self-reported information, and no information on
the control group is presented. Troster and Ruff (1990) reported a case study conducted on
two occupationally exposed workers to TCE. Mild deficits in motor speed were reported for
both cases. In the first case, manual dexterity was impaired in a male exposed to TCE (unknown
concentration) for eight months. In the second case study where a female was exposed to TCE
(low concentration; exact level not specified) for 3 months, there was weakness in the quadriceps
muscle as evaluated in a neurological exam and a decreased sensation to touch on one hand.
Both Gash et al. (2008) and Troster and Ruff (1990) provide very limited information given their
deficiencies related to lack of exposure data, self-reported information, and limited reporting of
referents and statistical analysis.
This document is a draft for review purposes only and does not constitute Agency policy.
142 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
4.3.6.1.4. Psychomotor Effects: Laboratory Animal Data
Several animal studies have demonstrated that TCE exposure produces changes in
psychomotor function. At high doses (>2,000 mg/kg) TCE causes mice to lose their righting
reflex when the compound is injected intraperitoneally (Shih et al., 2001; Umezu et al., 1997).
At lower exposures (inhalation and oral), TCE produces alterations in neurobehavioral measures
including locomotor activity, gait, operant responding, and reactivity. The studies are described
in Sections 4.3.6.2.1-4.3.6.2.3 and summarized in Tables 4-30 and 4-31.
4.3.6.1.5. Loss of righting reflex
Umezu et al. (1997) studied disruption of the righting reflex following acute injection
(i.p.) of 2,000, 4,000, and 5,000 mg/kg TCE in male ICR mice. TCE disrupted the righting
reflex at doses of 2,000 mg/kg and higher. At 2,000 mg/kg, loss of righting reflex (LORR) was
observed in only 2/10 animals injected. At 4,000 mg/kg, 9/10 animals experienced LORR and
100% of the animals experienced LORR at 5,000 mg/kg.
Shih et al. (2001) reported impaired righting reflexes at exposure doses of 5,000 mg/kg
(i.p.) in male MF1 mice. Mice pretreated with dimethyl sulfoxide or disulfuram (CYP2E1
inhibitor) delayed LORR in a dose related manner. By contrast, the alcohol dehydrogenase
inhibitor, 4-metylpyradine did not delay LORR that resulted from 5,000 mg/kg TCE. These data
suggest that the anesthetic properties of TCE involve its oxidation via CYP2E1 to an active
metabolite.
4.3.6.1.6. Activity, sensory-motor and neuromuscular function
Changes in sensory-motor and neuromuscular activity was reported in three studies
(Kishi et al., 1993; Moser et al., 1995; Moser et al., 2003). Kishi et al. (1993) exposed male
Wistar rats to 250, 500, 1,000, 2,000, and 4,000 ppm TCE for 4 hours. Rats exposed to 250-ppm
TCE showed a significant decrease both in the total number of lever presses and in avoidance
responses at 140 minutes of exposure compared with controls. Moser et al. (1995) evaluated the
effects of acute and short-term (14 day) administration of TCE in adult female Fischer 344 rats
(n = 8-10/dose) on activity level, neuromuscular function and sensorimotor function as part of a
larger functional observational battery (FOB) testing. The NOAEL levels identified by the
authors are 500 mg/kg (10% of the limit dose) for the acute treatment and 150 mg/kg (3% of the
limit dose) for the 14-day study. In the acute study, TCE produced the most significant effects in
This document is a draft for review purposes only and does not constitute Agency policy.
143 DRAFT—DO NOT CITE OR QUOTE
-------
1 motor activity (activity domain), gait (neuromuscular domain), and click response (sensorimotor
2 domain). In the 14-day study, only the activity domain (rearing) and neuromuscular domain
3 (forelimb grip strength) were significantly different (p < 0.05) from control animals. In a
4
This document is a draft for review purposes only and does not constitute Agency policy.
144 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-30. Summary of animal psychomotor function and reaction time
2 studies
3
Reference
Exposure
route
Species/strain/
sex/number
Dose level/
exposure duration
NOAEL; LOAEL
Effects
Savolainen
etal. (1977)
Inhalation
Rat, Sprague-
Dawley, male,
10
0, 200 ppm; 6 h/d,
4 d
LOAEL: 200 ppm
Increased frequency of
preening, rearing, and
ambulation. Increased
preening time.
Kishi et al.
(1993)
Inhalation
Rats,
Wistar, male,
number not
specified
o,
250,500, 1,000,
2,000, and 4,000
ppm, 4 hours
LOAEL:
250 ppm
Decreased lever presses
and increased responding
when lever press coupled
with a 10-s electric shock
(decreased avoidance
response).
Kulig et al.
(1987)
Inhalation
Rat, Wistar,
male, 8/dose
0, 500, 1,000, and
1,500 ppm; 16 h/d,
5 d/wk, 18 wk
NOAEL: 1,500 ppm
No change in spontaneous
activity, grip strength, or
hindlimb movement.
Moser et al.
(1995)
Oral
Rat, Fischer
344, female,
8/dose
0,150, 500,1,500,
and 5,000 mg/kg,
one dose
NOAEL: 500 mg/kg
LOAEL: 1,500
mg/kg
Decreased motor
activity; Neuro-
muscular and
sensorimotor
impairment.
0,50,150,500,
and 1,500
mg/kg-day, 14 d
NOAEL:
150 mg/kg-day
LOAEL: 500
mg/kg-day
Increased rearing
activity and decreased
forelimb grip strength.
Bushnell
(1997)
Inhalation
Rat, Long
Evans, male, 12
0, 400, 800, 1,200,
1,600, 2,000, or
2,400 ppm, 1 h/test
d, 4 consecutive
test days, 2 wk
NOAEL: 800 ppm
LOAEL: 1,200 ppm
Decreased sensitivity and
increased response time in
the signal detection task.
Shih et al.
(2001)
Intra-
peritoneal
Mouse, MF1,
male, 6
0, 5,000 mg/kg,
acute
LOAEL: 5,000
mg/kg
Impairment of righting
reflex.
Umezu et al.
(1997)
Intra-
peritoneal
Mouse, ICR,
male, 10/group
0, 2,000, 4,000,
5,000 mg/kg—loss
of righting reflex
measure
LOAEL: 2,000
mg/kg— loss of
righting reflex
Loss of righting reflex.
Mouse, ICR,
male,
6-10/group
0, 62.5, 125, 250,
500, and 1,000
mg/kg, single dose
and evaluated 30
min
postadministration
NOAEL: 500 mg/kg
LOAEL: 1,000
mg/kg—operant
behavior
NOAEL: 125 mg/kg
LOAEL: 250
mg/kg—punished
responding
Decreased responses
(lever presses) in an
operant response task for
food reward.
Increased responding
when lever press coupled
with a 20-V electric shock
(punished responding).
4
This document is a draft for review purposes only and does not constitute Agency policy.
145 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-30. Summary of animal psychomotor function and reaction time
studies (continued)
Reference
Exposure
route
Species/strain/
sex/number
Dose level/
exposure duration
NOAEL; LOAEL
Effects
Bushnell and
Oshiro
(2000)
Inhalation
Rat, Long
Evans, male, 32
0, 2,000, 2,400
ppm; 70 min/d, 9 d
LOAEL: 2,000 ppm
Decreased performance
on the signal detection
task. Increased response
time and decreased
response rate.
Nunes et al.
(2001)
Oral
Rat, Sprague-
Dawley, male,
10/group
0, 2,000
mg/kg-day, 7 d
LOAEL: 2,000
mg/kg-day
Increased foot splay. No
change in any other FOB
parameter (e.g.,
piloerection, activity,
reactivity to handling).
Moser et al.
(2003)
Oral
Rat, Fischer
344, female,
10/group
0, 40, 200, 800,
and 1,200
mg/kg-day, 10 d
Decreased motor activity;
Decreased sensitivity to
tail pinch; Increased
abnormality in gait;
Decreased grip strength;
Adverse changes in
several FOB parameters.
Albee et al.
(2006)
Inhalation
Rat,
Fischer 344,
male and
female,
10/sex/group
0, 250,
800, 2,500 ppm;
6 h/d, 5 d/wk,
13 wk
NOAEL:
2,500 ppm
No change in any FOB
measured parameter.
2 Bolded study(ies) carried forward for consideration in dose-response assessment (see Section 5).
3
4 FOB = functional observational battery.
5
This document is a draft for review purposes only and does not constitute Agency policy.
146 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-31. Summary of animal locomotor activity studies
2
Reference
Exposure
route
Species/strain/
sex/number
Dose level/
exposure duration
NOAEL;
LOAEL
Effects
Wolff and
Siegmund
(1978)
Intra-
peritoneal
Mouse, AB,
male, 18
0, 182 mg/kg, tested
30 min after
injection
LOAEL:
182 mg/kg
Decreased spontaneous motor
activity.
Kulig et al.
(1987)
Inhalation
Rat, Wistar,
male, 8/dose
0,500,1,000, and
1,500 ppm; 16 h/d,
5 d/wk, 18 wk
NOAEL:
500 ppm
LOAEL:
1,000 ppm
No change in spontaneous
activity, grip strength or
hindlimb movement.
Increased latency time in the
2-choice visual discrimination
task (cognitive disruption
and/or motor activity related
effect).
Moser et al.
(1995)
Oral
Rat, Fischer
344, female,
8/dose
0, 150, 500, 1,500,
and 5,000 mg/kg,
one dose
NOAEL:
500 mg/kg
LOAEL:
1,500 mg/kg
Decreased motor activity;
Neuro-muscular and
sensorimotor impairment.
0, 50, 150, 500, and
1,500 mg/kg-day,
14 d
NOAEL:
150
mg/kg-day
LOAEL: 500
mg/kg-day
Increased rearing activity.
Waseem et al.
(2001)
Oral
Rat, Wistar,
male, 8/group
0, 350, 700, and
1,400 ppm in
drinking water for
90 d
NOAEL:
1,400 ppm
No significant effect on
spontaneous locomotor activity.
Inhalation
Rat, Wistar,
male, 8/group
0,376 ppm for up
to 180 d; 4 h/d,
5 d/wk
LOAEL:
376 ppm
Changes in locomotor activity
and vary by timepoint when
measured over the 180-d
period.
Moser et al.
(2003)
Oral
Rat, Fischer
344, female,
10/group
0, 40, 200, 800, and
1,200 mg/kg-day,
10 d
Decreased motor activity;
Decreased sensitivity; Increased
abnormality in gait; Adverse
changes in several FOB
parameters.
3
4 Bolded study(ies) carried forward for consideration in dose-response assessment (see Section 5).
5
6 FOB = functional observational battery.
7
8
9 separate 10-day study (Moser et al., 2003), TCE administration significantly (p < 0.05) reduced
10 motor activity, tail pinch responsiveness, reactivity to handling, hind limb grip strength and body
11 weight. Significant increases (p < 0.05) in piloerection, gait scores, lethality, body weight loss,
12 and lacrimation was also reported in comparison to controls.
This document is a draft for review purposes only and does not constitute Agency policy.
147 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
There are also two negative studies which used adequate numbers of subjects in their
experimental design but used lower doses than did Moser et al. (2003). Albee et al. (2006)
exposed male and female Fischer 344 rats (n = 10/sex) to TCE by inhalation at exposure doses of
250, 800, and 2,500 ppm, for 6 hours/day, 5 days/week, for 13 weeks. The FOB was performed
monthly although it is not certain how much time elapsed from the end of exposure until the
FOB test was conducted. No treatment related differences in grip strength or landing foot splay
were demonstrated in this study. Kulig et al. (1987) also failed to show significant effects of
TCE inhalation exposure on markers of motor behavior. Wistar rats (n = 8) exposed to 500,
1,000, and 1,500 ppm, for 16 hours/day, 5 days/week, for 18 weeks failed to show changes in
spontaneous activity, grip strength, or coordinated hind limb movement. Measurements were
made every three weeks during the exposure period and occurred between 45 and 180 minutes
following the previous TCE inhalation exposure.
4.3.6.1.7. Locomotor activity
The data, with regard to locomotor activity, are inconsistent. Several studies showed that
TCE exposure can decrease locomotor activity including Wolff and Siegmund (1978) where AB
mice (n = 18) were treated acutely with a dose of 182 mg/kg, i.p. at one of four time points
during a 24-hour day. Moser et al. (1995; 2003) reported reduced locomotor activity in female
Fischer 344 rats (n = 8-10) gavaged with TCE over an acute (LOAEL = 5,000 mg/kg TCE) or
subacute period (LOAEL = 500 but no effect at 5,000 mg/kg). In the Moser et al. (2003), it
appears that 200-mg/kg TCE yielded a significant reduction in locomotor activity and that the
degree of impairment at this dose represented a maximal effect on this measure. That is, higher
doses of TCE appear to have produced equivalent or slightly less of an effect on this behavior.
While this study identifies a LOAEL of 200-mg/kg TCE by gavage over a 10-day period, this is
a much more lower dose effect than that reported in Moser et al. (1995). Both studies (Moser et
al., 1995; Moser et al., 2003) demonstrate a depression in motor activity that occurs acutely
following TCE administration. Kulig et al. (1987) demonstrated that rats had increased response
latency to a two choice visual discrimination following 1,000- and 1,500-ppm TCE exposures for
18 weeks. However, no significant changes in grip strength, hindlimb movement, or any other
motor activity measurements were noted.
There are also a few studies (Fredriksson et al., 1993; Waseem et al., 2001) generally
conducted using lower exposure doses that failed to demonstrate impairment of motor activity or
ability following TCE exposure. Waseem et al. (2001) failed to demonstrate changes in
locomotor activity in male Wistar rats (n = 8) dosed with TCE (350, 700, and 1,400 ppm) in
This document is a draft for review purposes only and does not constitute Agency policy.
148 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
drinking water for 90 days. Wistar rats (n = 8) exposed to 500, 1,000, and 1,500 ppm for
16 hours/day, 5 days/week, for 18 weeks failed to show changes in spontaneous activity. No
changes in locomotor activity were observed for 17-day-old male NMRI mice that were dosed
postnatally with 50 or 290 mg/kg-day from Day 10-16 (Fredriksson et al., 1993). However,
rearing activity was significantly decreased in the NMRI mice at Day 60.
4.3.6.1.8. Summary and Conclusions for Psychomotor Effects
In human studies, psychomotor effects such as reaction time and muscular
dyscoordination have been examined following TCE exposure. In the reaction time studies,
statistically significant increases in CRT and SRT were reported in the Kilburn studies (Kilburn,
2002b; Kilburn and Thornton, 1996; Kilburn and Warshaw, 1993). All of these studies were
geographically based and it was suggested that the results were used for litigation and the
differences between exposed and referent groups on other factors influencing reaction speed time
may introduce a bias to the findings. Additionally, in these studies exposure to TCE and other
chemicals occurred through drinking water for at least 1 year and TCE concentrations in well
water ranged from 0.2 ppb to 10,000 ppb. Reif et al. (2003) whose exposure assessment
approach included exposure modeling of water distribution system to estimate TCE
concentrations in tap water at census track of residence found that residents with drinking water
containing TCE (up to >15 ppb—the highest level not specified) and other chemicals did not
significantly increase CRTs or SRTs. Inhalation studies also demonstrated increased reaction
times. An acute exposure chamber study (Gamberale et al., 1976) tested for CRT, SRT, and
RT-addition following a 70-minute exposure to TCE. A concentration-dependent significant
decrease in performance was observed with the RT-addition test and not for CRT or SRT tasks.
An occupational exposure study on eight female workers exposed to TCE (Gun et al., 1978) also
reported increased reaction time in the females exposed to TCE-only. Muscular dyscoordination
for humans following TCE exposure has been reported in a few studies as a subjective
observation. The studies indicated that exposure resulted in decreased motor speed and dexterity
(Rasmussen et al., 1993a; Troster and Ruff, 1990) and self-reported faster asymptomatic fine
motor hand movements (Gash et al., 2008).
Animal studies evaluated psychomotor function by examining locomotor activity, operant
responding, changes in gait, loss of righting reflex, and general motor behavior (see Tables 4-30
and 4-31 for references). Overall, the studies demonstrated that TCE causes loss of righting
reflex at injection doses of 2,000 mg/kg or higher (Shih et al., 2001; Umezu et al., 1997).
Regarding general psychomotor testing, significant decreases in lever presses and avoidance
This document is a draft for review purposes only and does not constitute Agency policy.
149 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
were observed at inhalation exposures as low as 250 ppm for 4 hours (LOAEL; Kishi et al.,
1993). Following subchronic inhalation exposures, no significant changes in psychomotor
activity were noted at up to 2,500 ppm for 13 weeks (Albee et al., 2006) or at 1,500 ppm for
18 weeks (Kulig, 1987). In the oral administration studies (Moser et al., 1995; Moser et al.,
2003), psychomotor effects were evaluated using an FOB. More psychomotor domains were
significantly affected by TCE treatment in the acute study in comparison to the 14-day study, but
a lower NOAEL (150 mg/kg-day) was reported for the 14-day study in comparison to the acute
study (500 mg/kg; Moser et al., 1995). Upon closer examination of the data, a biphasic effect in
one measure of the FOB (rearing) was resulting in the lower NOAEL for the 14-day study and
doses that were higher and lower than the NOAEL did not produce a statistically significant
increase in the number of rears. Therefore, it can be surmised that acute exposure to TCE results
in significant changes in psychomotor function. However, there may be some tolerance to these
psychomotor changes in increased exposure duration to TCE as evidenced by the results noted in
the short-term and subchronic exposure studies.
4.3.7. Mood Effects and Sleep Disorders
4.3.7.1.1. Effects on Mood: Human Studies
Reports of mood disturbance (depression, anxiety) resulting from TCE exposure are
numerous in the human literature. These symptoms are subjective and difficult to quantify.
Studies by Gash et al. (2008), Kilburn and Warshaw (1993), Kilburn (2002a, b), McCunney et al.
(1988), Mitchell et al. (1969), Rasmussen and Sabroe (1986), and Troster and Ruff (1990)
reported mood disturbances in humans. Reif et al. (2003) and Triebig (1976) reported no effect
on mood following TCE exposures.
4.3.7.1.2. Effects on Mood: Laboratory Animal Findings
It is difficult to obtain comparable data of emotionality in laboratory studies. However,
Moser et al. (2003) and Albee et al. (2006) both report increases in handling reactivity among
rats exposed to TCE. In the Moser study, female Fischer 344 rats received TCE by oral gavage
for periods of 10 days at doses of 0, 40, 200, 800, and 1,200 while Albee et al. (2006) exposed
Fischer 344 rats to TCE by inhalation at exposure doses of 250, 800, and 2,500 ppm for
This document is a draft for review purposes only and does not constitute Agency policy.
150 DRAFT—DO NOT CITE OR QUOTE
-------
1 6 hours/day, 5 days/week, for 13 weeks. These studies are summarized and described in
2 Table 4-32.
3
4.3.7.1.3. Sleep Disturbances
4 Arito et al. (1994) exposed male Wistar rats to 50-, 100-, and 300-ppm TCE for
5 8 hour/day, 5 days/week, for 6 weeks and measured electroencephalographic (EEG) responses
6 (see Table 4-32). EEG responses were used as a measure to determine the number of awake
7 Table 4-32. Summary of animal mood effect and sleep disorder studies
Reference
Exposure
route
Species/strain/
sex/number
Dose level/
exposure duration
NOAEL; LOAEL
Effects
Mood Effects
Albee et al.
(2006)
Inhalation
Rat,
Fischer 344,
male and
female,
10/sex/group
0, 250,
800, 2,500 ppm;
6 h/d, 5 d/wk,
13 wk
NOAEL:
800 ppm
Increased handling
reactivity.
Moser et al.
(2003)
Oral
Rat, Fischer
344, female,
10/group
0, 40, 200, 800, and
1,200 mg/kg-day,
10 d
Decreased handling
reactivity score.
Sleep Disorder
Arito et al.
(1994)
Inhalation
Rat,
Wistar, male,
5/group
0,50,
100, and
300 ppm; 8 h/d,
5 d/wk, for 6 wk
LOAEL:
50 ppm
Significant changes in
sleep cycle as measured
through EEG changes;
Significant decreases in
wakefulness.
9
10 Bolded study(ies) carried forward for consideration in dose-response assessment (see Section 5).
11
12
13 (wakefulness hours) and sleep hours. Exposure to all the TCE levels significantly decreased
14 amount of time spent in wakefulness (W) during the exposure period. Some carry over was
15 observed in the 22 hours post exposure period with significant decreases in wakefulness seen at
16 100-ppm TCE. Significant changes in W-sleep elicited by the long-term exposure appeared at
17 lower exposure levels. These data seem to identify a low dose effect of TCE and established a
18 LOAEL of 50 ppm for sleep changes.
19
This document is a draft for review purposes only and does not constitute Agency policy.
151 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
4.3.8. Developmental Neurotoxicity
4.3.8.1.1. Human Studies
In humans, CNS birth defects were observed in a few studies (ATSDR, 2001; Bove,
1996; Bove et al., 1995; Lagakos et al., 1986). Postnatally, observed adverse effects in humans
include delayed newborn reflexes following exposure to TCE during childbirth (Beppu, 1968),
impaired learning or memory (Bernad et al., 1987, abstract; White et al., 1997); aggressive
behavior (Bernad et al., 1987, abstract); hearing impairment (Burg and Gist, 1999); speech
impairment (Burg and Gist, 1999; White et al., 1997); encephalopathy (White et al., 1997);
impaired executive and motor function (White et al., 1997); attention deficit (Bernad et al., 1987,
abstract; White et al., 1997); and autism spectrum disorder (Windham et al., 2006). The human
developmental neurotoxicity studies are discussed in more detail in Section 4.8.2.1.2, and
summarized in Table 4-33.
Table 4-33. Summary of human developmental neurotoxicity associated with
TCE exposures
Finding
Species
Citations
CNS defects, neural tube defects
Human
ATSDR (2001)
Bove (1996); Bove et al. (1995)
Lagakos et al. (1986)
Delayed newborn reflexes
Human
Beppu (1968)
Impaired learning or memory
Human
Bernad et al. (1987, abstract)
White etal. (1997)
Aggressive behavior
Human
Bernad et al., (1987, abstract)
Hearing impairment
Human
Burg and Gist (1999)
Speech impairment
Human
Burg and Gist (1999)
White etal. (1997)
Encephalopathy
Human
White etal. (1997)
Impaired executive function
Human
White etal. (1997)
Impaired motor function
Human
White etal. (1997)
Attention deficit
Human
White etal. (1997)
Human
Bernad et al. (1987, abstract)
Autism spectrum disorder (ASD)
Human
Windham et al. (2006)
This document is a draft for review purposes only and does not constitute Agency policy.
152 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
4.3.8.1.2. Animal Studies
There are a few studies demonstrating developmental neurotoxicity following
trichloroethylene exposure (range of exposures) to experimental animals. These studies
collectively suggest that developmental neurotoxicity result from TCE exposure, however, some
types of effects such as learning and memory measures have not been evaluated. Most of the
studies demonstrate either spontaneous motor activity changes (Taylor et al., 1985) or
neurochemical changes such as decreased glucose uptake and changes in the specific gravity of
the cortex and cerebellum (Isaacson and Taylor, 1989; Noland-Gerbec et al., 1986; Westergren et
al., 1984). In addition, in most of these studies there is no assessment of the exposure to TCE or
metabolites in the pups/offspring. Details of the studies are presented below and summarized in
Table 4-34.
Taylor et al. (1985) administered TCE to female Sprague-Dawley rats in their drinking
water from 14 days before breeding throughout gestation and until pups were weaned at 21 days.
Measured TCE concentrations in the dams ranged from 312-646 mg/L, 625-1,102 mg/L, and
This document is a draft for review purposes only and does not constitute Agency policy.
153 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-34. Summary of mammalian in vivo developmental neurotoxicity
2 studies—oral exposures
3
Reference
Species/strain/
sex/number
Dose level/
exposure duration
NOAEL;
LOAEL a
Effects
Fredriksson
et al. (1993)
Mouse, NMRI, male
pups, 12 pups from 3—4
different litters/group
0, 50, or 290 mg/kg-day
PND 10-16
LOAEL:
50 mg/kg-day
Rearing activity sig. X at both
dose levels on PND 60.
George
et al. (1986)
Rat, F334, male and
female,
20 pairs/treatment
group,
40 controls/sex
0, 0.15, 0.30, or 0.60%
microencapsulated TCE in
diet.
Breeders exposed 1 wk
premating, then for 13 wk;
pregnant females
throughout pregnancy (i.e.,
18 wk total).
LOAEL:
0.15%
Open field testing in pups: a
sig. dose-related trend toward
T time required for male and
female pups to cross the first
grid in the test device.
Isaacson
and Taylor
(1989)
Rat, Sprague-Dawley,
females, six
dams/group
0, 312, or 625 mg/L
(0, 4.0, or 8.1 mg/d)b.
Dams (and pups) exposed
from 14 d prior to mating
until end of lactation.
LOAEL:
312 mg/L
Sig. X myelinated
fibers in the stratum
lacunosum-moleculare of
pups. Reduction in
myelin in the CA1 region
of the hippocampus.
Noland-
Gerbec et al.
(1986)
Rat, Sprague-Dawley,
females, 9-11
dams/group
0, 312 mg/L.
(Avg. total intake of dams:
825 mg TCE over 61 d.).
Dams (and pups) exposed
from 14 d prior to mating
until end of lactation.
LOAEL:
312 mg/L
Sig. i uptake of 3H-2-DG in
whole brains and cerebella (no
effect in hippocampus) of
exposed pups at 7, 11, and 16 d,
but returned to control levels by
21 d.
Ta
ylor et al.
(1985)
Rat,
Sprague-Dawley,
females, no.
dams/group not
reported
0, 312, 625,
and 1,250 mg/L in
drinking water.
Dams (and pups) exposed
from 14 d prior to mating
until end of lactation.
LO
AEL:
312 mg/L
Exploratory behavior sig. f in
60- and 90-d old male rats at
all treatment levels.
Locomotor activity (measured
through the wheel-running tasks)
was higher in rats from dams
exposed to 1,250 mg/L TCE.
Blossom
et al. (2008)
Mouse, MRL +/+, dams
and both sexes offspring,
eight litters/group;
three-eight pups/group
Drinking water, from GD 0
to PND 42; 0 or (XI mg/mL;
maternal dose = 25.7
mg/kg-day; offspring PND
24-42 dose =
31.0 mg/kg-day.
LOAEL:
31 mg/kg-day
for offspring
Righting reflex, bar holding, and
negative geotaxis were not
impaired. Significant association
between impaired nest quality
and TCE exposure.
Lower GSH levels and
GSH:GSSG ratios with TCE
exposure.
4
5 Bolded study(ies) carried forward for consideration in dose-response assessment (see Section 5).
6
7 " LOEL (lowest-observed-effect level) are based upon reported study findings.
8 b Dose conversions provided by study author(s).
This document is a draft for review purposes only and does not constitute Agency policy.
154 DRAFT—DO NOT CITE OR QUOTE
-------
1
2 GD = gestation day, GSSG = oxidized GSH, PND = postnatal day.
3
This document is a draft for review purposes only and does not constitute Agency policy.
15 5 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
1,250-1,991 mg/L in the low, mid, and high-dose groups as measured from the drinking water.
Pups were evaluated for exploratory activity at 28, 60, or 90 days. No significant differences
were noted between control and treated pups at 28 days. At 60 days, all TCE-treated animals
had significantly increased exploratory activity in comparison to age-matched controls, but only
the high group had increased activity at 90 days. A significant increase in spontaneous motor
activity (as measured by a wheel-running task) was noted in only the high dose TCE
(1,250-1,991 mg/L) group during the onset of the darkness period. This study demonstrated that
both spontaneous and open field activities are significantly affected by developmental TCE
exposure.
Spontaneous behavioral changes were also investigated in another study by Fredriksson
et al. (1993). Male and female NMRI pups (mice) were orally administered 50 or
290 mg/kg-day for 7 days starting at PND 10. Spontaneous motor activity was investigated in
male mice at ages 17 and 60 days. TCE-treated animals tested at Day 17 did not demonstrate
changes in any spontaneous activity measurements in comparison to control animals. Both doses
of TCE (50 and 290 mg/kg-day) significantly decreased rearing in 60 day-old male mice.
Westergren et al. (1984) examined the brain specific gravity of litters from mice exposed
to TCE. NMRI mice (male and female) were exposed to 150-ppm TCE (806.1 mg/m3) for
30 days prior to mating. Exposure in males continued until the end of mating and females were
exposed until the litters were born. Brains were removed from the offspring at either PNDs 1,
10, 20-22, or 29-31. At PNDs 1 and 10, significant decreases were noted in the specific gravity
of the cortex. Significant decreases in the specific gravity of the cerebellum were observed at
PND 10 (decrease from 1.0429 ± 0.00046-1.0405 ± 0.00030) and 20-22 (decrease from 1.0496
± 0.00014-1.0487 ± 0.00060). Cerebellum measurements were not reported for PND 29-31
animals. Neurobehavioral assessments were not conducted in this study. Additionally,
decreased brain specific gravity is suggestive of either decreased brain weight or increased brain
volume (probably from edema) or a combination of the two factors and is highly suggestive of an
adverse neurological effect. The effects of TCE on the cortical specific gravity were not
persistent since cortices from PND 29-31 animals did not exhibit any significant changes. It is
unclear if the effects on the cerebellum were persistent since results were not reported for the
PND 29-31 animals. However, the magnitude of the change in the specific gravity of the
cerebellum is decreased from PND 10 to PND 20-22 suggesting that the effect may be reversible
given a longer recovery period from TCE.
The effect of TCE on glucose uptake in the brain was evaluated in rat pups exposed to
TCE during gestation and through weaning. The primary source of energy utilized in the CNS is
glucose. Changes in glucose uptake in the brain are a good indicator for neuronal activity
This document is a draft for review purposes only and does not constitute Agency policy.
156 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
modification. Noland-Grebec et al. (1986) administered 312 mg/L TCE through drinking water
to female Sprague-Dawley rats from 2 weeks before breeding and up until pups reached 21 days
of age. To measure glucose uptake, 2-deoxyglucose was administered intraperitoneally to male
pups at either postnatal day (PND) 7, 11, 16, or 21. Significant decreases in glucose uptake were
noted in whole brain and cerebellum at all postnatal days tested. Significant decreases in glucose
uptake were also observed in the hippocampus except for animals tested at PND 21. The
observed decrease in glucose uptake suggests decreased neuronal activity.
Female Sprague-Dawley rats (70 days old) were administered TCE in drinking water at a
level of either 4.0 or 8.1 mg/day for 14 days prior to mating and continuing up through lactation
(Isaacson and Taylor, 1989). Only the male pups were evaluated in the studies. At PND 21,
brains were removed from the pups, sectioned, and stained to evaluate the changes in myelin.
There was a significant decrease (40% decrease) in myelinated fibers in the CA1 region of the
hippocampus of the male pups. This effect appeared to be limited to the CA1 region of the
hippocampus since other areas such as the optic tract, fornix, and cerebral peduncles did not have
decreases in myelinated fibers.
Neurological changes were found in pups exposed to TCE in a study conducted by the
National Toxicology Program (NTP) in Fischer 344 rats (George et al., 1986). TCE was
administered to rats at dietary levels of 0, 0.15, 0.30, or 0.60%. No intake calculations were
presented for the rat study and therefore, a dose rate is unavailable for this study. Open field
testing revealed a significant (p < 0.05) dose-related trend toward an increase in the time required
for male and female F1 weanling pups (PND 21) to cross the first grid in the testing device,
suggesting an effect on the ability to react to a novel environment.
Blossom et al. (2008) treated male and female MRL +/+ mice with 0 or 0.1 mg/mL TCE
in the drinking water. Treatment was initiated at the time of mating, and continued in the
females (8/group) throughout gestation and lactation. Behavioral testing consisted of righting
reflex on PNDs 6, 8, and 10; bar-holding ability on PNDs 15 and 17; and negative geotaxis on
PNDs 15 and 17. Nest building was assessed and scored on PND 35, the ability of the mice to
detect and distinguish social odors was examined with an olfactory habituation/dishabituation
method at PND 29, and a resident intruder test was performed at PND 40 to evaluate social
behaviors. Righting reflex, bar holding, and negative geotaxis were not impaired by treatment.
There was a significant association between impaired nest quality and TCE exposure in tests of
nest-building behavior; however, TCE exposure did not have an effect on the ability of the mice
to detect social and nonsocial odors using habituation and dishabituation methods. Resident
intruder testing identified significantly more aggressive activities (i.e., wrestling and biting) in
TCE-exposed juvenile male mice as compared to controls, and the cerebellar tissue from the
This document is a draft for review purposes only and does not constitute Agency policy.
157 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
male TCE-treated mice had significantly lower GSH levels and GSH:oxidized GSH
(GSH:GSSG) ratios, indicating increased oxidative stress and impaired thiol status, which have
been previously reported to be associated with aggressive behaviors (Franco et al., 2006).
Histopathological examination of the brain did not identify alterations indicative of neuronal
damage or inflammation.
4.3.8.1.3. Summary and Conclusions for the Developmental Neurotoxicity Studies
Gestational exposure to TCE in humans has resulted in several developmental
abnormalities. These changes include neuroanatomical changes such as neural tube defects
(ATSDR, 2001; Bove et al., 1995, 1996; Lagakos et al., 1986) and encephalopathy (White et al.,
1997). Clinical neurological changes such as impaired cognition (Bernad et al., 1987; White et
al., 1997), aggressive behavior (Bernad et al., 1987), and speech and hearing impairment (Burg
and Gist, 1999; White et al., 1997) are also observed when TCE exposure occurs in utero.
In animal studies, anatomical and clinical developmental neurotoxicity is also observed.
Following inhalation exposures of 150 ppm to mice during mating and gestation, the specific
gravity of offspring brains was significantly decreased at postnatal time points through the age of
weaning; this effect did not persist to 1 month of age (Westergren et al., 1984). In studies
reported by Taylor et al. (1985), Isaacson and Taylor (1989), and Noland-Gerbec et al. (1986),
312 mg/L exposures in drinking water that were initiated 2 weeks prior to mating and continued
to the end of lactation resulted in (a) significant increase in exploratory behavior at PNDs 60 and
90, (b) reductions in myelination in the CA1 hippocampal region of offspring at weaning, and
(c) significantly decreased uptake of 2-deoxyglucose in the rat brain at PND 21. Gestational
exposures to mice (Fredriksson et al., 1993) resulted in significantly decreased rearing activity
on PND 60, and dietary exposures during the course of a continuous breeding study in rats
(George et al., 1986) found a significant trend toward increased time to cross the first grid in
open field testing. In a study by Blossom et al. (2008), male mice exposed gestationally to TCE
exhibited lower GSH levels and lower GSH:GSSG ratios which is also observed in mice that
have more aggressive behaviors (Franco et al., 2006).
This document is a draft for review purposes only and does not constitute Agency policy.
15 8 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
4.3.9. Mechanistic Studies of Trichloroethylene (TCE) Neurotoxicity
4.3.9.1.1. Dopamine Neuron Disruption
There are very recent laboratory animal findings resulting from short-term TCE
exposures that demonstrate vulnerability of dopamine neurons in the brain to this chlorinated
hydrocarbon. The key limitation of these laboratory animal studies is that only one dosing
regimen was included in each study. Moreover, there has been no systematic body of data to
show that other chlorinated hydrocarbons such as tetrachloroethylene or aromatic solvents
similarly target this cell type. Confidence in the limited data regarding dopamine neuron death
and in vivo TCE exposure would be greatly enhanced by identifying a dose-response
relationship. If indeed TCE can target dopamine neurons it would be anticipated that human
exposure to this agent would result in elevated rates of parkinsonism. There are no systematic
studies of this potential relationship in humans although one limited report attempted to address
this possibility. Difficulties in subject recruitment into that study limit the weight that can be
given to the results.
Endogenously formed chlorinated tetrahydro-beta-carbolines (TaClo) have been
suggested to contribute to the development of Parkinson-like symptoms Bringmann et al. 1992
(Bringmann et al., 1995; Kochen et al., 2003; Riederer et al., 2002). TaClo can be formed
endogenously from metabolites of TCE such as trichloroacetaldehyde. TaClo has been
characterized as a potent neurotoxicant to the dopaminergic system. Some research groups have
hypothesized that Parkinson-like symptoms resulting from TCE exposure may occur through the
formation of TaClo, but not enough evidence is available to determine if this mechanism occurs.
4.3.9.1.2. Dopamine neuron disruption: human studies
There are no human studies that present evidence of this effect. Nagaya et al. (1990)
examined serum dopamine P-hydroxylase activity without differences observed in mean
activities between control and exposed subjects. In the study, 84 male workers exposed to TCE
were compared to 83 male age-matched controls. The workers had constantly used TCE in their
jobs and their length of employment ranged from 0.1-34 years.
4.3.9.1.3. Dopamine neuron disruption: animal studies
There are limited data from mice and rats that suggest the potential for TCE to disrupt
dopamine neurons in the basal ganglia (see Table 4-35). Gash et al. (2008) showed that TCE
This document is a draft for review purposes only and does not constitute Agency policy.
159 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
gavage in Fischer 344 rats (n = 9) at an exposure level of 1,000 mg/kg-day, 5 days/week, for
6 weeks yielded degeneration of dopamine neurons in the substantia nigra and alterations in
dopamine turnover as reflected in a shift in dopamine metabolite to parent compound ratios.
Guehl et al. (1999) reported similar findings in OF1 mice (n = 10) that were injected i.p. with
400 mg/kg-day TCE 5 days/week for 4 weeks. Each of these studies evaluated only a single dose
level of TCE so that establishing a dose-response relationship is not possible. Consequently,
these data are of limited utility in risk assessment because they do not establish the potency of
TCE to damage dopamine neurons. They are important, however, in identifying a potential
permanent impairment that might occur following TCE exposure at relatively high exposure
doses. They also identify a potential mechanism by which TCE could produce CNS injury.
Table 4-35. Summary of animal dopamine neuronal studies
Reference
Exposure route
Species/strain/
sex/number
Dose level/
exposure
duration
NOAEL;
LOAEL
Effects
Guehl et al.
(1999)
Intraperitoneal
Administration
Mouse, OF1, male,
10
0 and 400
mg/kg; 5 d/wk,
4 wk
LOAEL:
400 mg/kg
Significant dopaminergic
neuronal death in substantia
nigra.
Gash et al.
(2008)
Oral gavage
Rat, Fischer 344,
male, 9/group
0 and 1,000
mg/kg; 5 d/wk,
6 wk
LOAEL:
1,000
mg/kg
Degeneration of dopamine-
containing neurons in
substantia nigra.
Change in dopamine
metabolism.
Bolded study(ies) carried forward for consideration in dose-response assessment (see Section 5).
4.3.9.1.4. Summary and conclusions of dopamine neuron studies
Only two animal studies have reported changes in dopamine neuron effects from TCE
exposure (Gash et al., 2008; Guehl et al., 1999). Both studies demonstrated toxicity to
dopaminergic neurons in the substantia nigra in rats or mice. LOAELs of 400 mg/kg (mice;
Guehl et al., 1999) and 1,000 mg/kg (rats; Gash et al., 2008) were reported for this effect.
Dopaminergic neuronal degeneration following TCE exposure has not been studied in humans.
However, there were no changes in serum dopamine P-hydroxylase activity in TCE-exposed and
control individuals (Nagaya et al., 1990). Loss of dopaminergic neurons in the substantia nigra
also occurs in patients with Parkinson's disease and the substantia nigra is an important region in
helping to control movements. As a result, loss of dopaminergic neurons in the substantia nigra
This document is a draft for review purposes only and does not constitute Agency policy.
160 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
may be one of the potential mechanisms involved in the clinical psychomotor effects that are
observed following TCE exposure.
4.3.9.1.5. Neurochemical and Molecular Changes
There are limited data obtained only from laboratory animals that TCE exposure may
have consequences on GABAergic (gamma-amino butyric acid [GAB A]) and glutamatergic
neurons (Briving et al., 1986; Shih et al., 2001; see Table 4-36). However, the data obtained are
limited with respect to brain region examined, persistence of effect, and whether there might be
functional consequences to these changes. The data of Briving et al. (1986) demonstrating
changes in cerebellar high affinity uptake for GABA and glutamate following chronic low level
(50 and 150 ppm) TCE exposure do not appear to be reflected in the only other brain region
evaluated (hippocampus). However, glutamate levels were increased in the hippocampus. The
data of Shih et al. (2001) are indirect in that it shows an altered response to GABAergic
Table 4-36. Summary of neurophysiological, neurochemical, and
neuropathological effects with TCE exposure
Reference
Exposure
route
Species/strain/
sex/number
Dose level/
exposure duration
NOAEL; LOAEL
Effects
Neurophysiological studies
Shih et al.
(2001)
Intra-
peritoneal
Mouse, MF1,
male, 6/group
0, 250 500, 1,000, or
2,000 mg/kg, 15 min;
followed by tail
infusion of PTZ
(5 mg/mL), picrotoxin
(0.8 mg/mL),
bicuculline
(0.06 mg/mL),
strychnine
(0.05 mg/mL), 4-AP
(2 mg/mL), or
NMDA (8 mg/mL).
Increased threshold for
seizure appearance
with TCE pretreatment
for all convulsants.
Effects strongest on
the GABAa
antagonists, PTZ,
picrotoxin, and
bicuculline suggesting
GABAa receptor
involvement. NMDA
and glycine Rc
involvement also
suggested.
Ohta et al.
(2001)
Intra-
peritoneal
Mouse, ddY,
male, 5/group
0, 300, or
1,000 mg/kg,
sacrificed 24 h after
injection.
LOAEL: 300 mg/kg
Decreased response
(LTP response) to
tetanic stimulation in
the hippocampus.
Neurochemical studies
This document is a draft for review purposes only and does not constitute Agency policy.
161 DRAFT—DO NOT CITE OR QUOTE
-------
Briving et al.
(1986)
Inhalation
Gerbils,
Mongolian,
male and
female, 6/group
0, 50, or 150 ppm,
continuous, 24 h/d,
12 mos.
NOAEL: 50 ppm;
LOAEL: 150 ppm
for glutamate levels
in hippocampus.
NOAEL: 150 ppm
for glutamate and
GABA uptake in
hippocampus.
LOAEL: 50 ppm
for glutamate and
GABA uptake in
cerebellar vermis.
Increased glutamate
levels in the
hippocampus.
Increased glutamate
and GABA uptake in
the cerebellar vermis.
Subramoniam
et al. (1989)
Oral
Rat, Wistar,
female
0 or 1,000 mg/kg, 2 or
20 h.
0 or 1,000 mg/kg-day,
5 d/wk, 1 yr.
PI and PIP2 decreased
by 24 and 17% at 2 h.
PI and PIP2 increased
by 22 and 38% at 20 h.
PI, PIP, and PIP2
reduced by 52, 23, and
45% in 1 yr study.
This document is a draft for review purposes only and does not constitute Agency policy.
162 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-36. Summary of neurophysiological, neurochemical, and
neuropathological effects with TCE exposure (continued)
Reference
Exposure
route
Species/strain/
sex/number
Dose level/
exposure duration
NOAEL; LOAEL
Effects
Haglid et al.
(1981)
Inhalation
Gerbil,
Mongolian,
male and
female,
6-7/group
0, 60, or 320 ppm,
24 h/d, 7 d/wk, 3 mo.
LOAEL: 60 ppm,
brain protein
changes.
NOAEL: 60 ppm;
LOAEL: 320 ppm,
brain DNA changes.
(1) Decreases in total
brain soluble protein
whereas increase in
S100 protein.
(2) Elevated DNA in
cerebellar vermis and
sensory motor cortex.
Neuropathological studies
Kjellstrand
et al. (1987)
Inhalation
Mouse,
NMRI, male
0, 150, or 300 ppm,
24 h/d, 4 or 24 d.
LOAEL: 150 ppm,
4 and 24 d
Sciatic nerve
regeneration was
inhibited in both
mice and rats.
Rat,
Sprague-
Dawley,
female
0,300 ppm, 24 h/d,
4 or 24 d.
NOAEL: 300 ppm,
4 d.
LOAEL: 300 ppm,
24 d.
Isaacson and
Taylor (1989)
Oral
Rat, Sprague-
Dawley,
females,
six dams/grou
P
0, 312, or 625 mg/L.
(0, 4.0, or 8.1 mg/d).
Dams (and pups)
exposed from 14 d
prior to mating until
end of lactation.
LOAEL:
312 mg/L
Significant 1
myelinated fibers in
the stratum
lacuno sum-moleculare
of pups. Reduction in
myelin in the
hippocampus.
1
2 Bolded study(ies) carried forward for consideration in dose-response assessment (see Section 5).
3
4 NMDA = N-nitrosodimethylamine, PTZ = pentylenetetrazole.
5
6
7 antagonist drugs in mice treated by acute injection with 250, 500, 1,000, and 2,000 mg/kg TCE.
8 However, these data do show some dose dependency with significant findings observed with
9 TCE exposure as low as 250 mg/kg.
10 The development and physiology of the hippocampus has also been evaluated in
11 two different studies (Isaacson and Taylor, 1989; Ohta et al., 2001). Isaacson and Taylor (1989)
12 found a 40% decrease in myelinated fibers from hippocampi dissected from neonatal
13 Sprague-Dawley rats (// = 2-3) that were exposed to TCE (4 and 8.1 mg/day) in utero and during
14 the preweaning period. Ohta et al. (2001) injected male ddY mice with 300 mg/kg TCE and
15 found a significant reduction in response to titanic stimuli in excised hippocampal slices. Both
16 of these studies demonstrated that there is some interaction with TCE and the hippocampal area
17 in the brain.
This document is a draft for review purposes only and does not constitute Agency policy.
163 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Impairment of sciatic nerve regeneration was demonstrated in mice and rats exposed to
TCE (Kjellstrand et al., 1987). Under heavy anesthesia, the sciatic nerve of the animals was
artificially crushed to create a lesion. Prior to the lesion, some animals were pre-exposed to TCE
for 20 days and then for an additional 4 days after the lesion. Another set of animals were only
exposed to TCE for 4 days following the sciatic nerve lesion. For mice, regeneration of the
sciatic nerve in comparison to air-exposed animals was 20 and 33% shorter in groups exposed to
150- and 300-ppm TCE for 4 days, respectively. This effect did not significantly increase in
mice pre-exposed to TCE for 20 days, and the regeneration was 30% shorter in the 150-ppm
group and 22% shorter in the 300-ppm group. Comparatively, a 10% reduction in sciatic nerve
regeneration length was observed in rats exposed to TCE for 20 days prior to the lesion plus the
4 days after the sciatic nerve lesion.
There are also a few in vitro studies (summarized in Table 4-37) that have demonstrated
that TCE exposure alters the function of inhibitory ion channels such as GABAa and glycine
receptors (Beckstead et al., 2000; Krasowski and Harrison, 2000), and serotonin receptors
(Lopreato et al., 2003). Krasowski and Harrison (2000) and Beckstead et al. (2000) were able to
demonstrate that human GABAa and glycine receptors could be potentiated by TCE when a
receptor agonist was coapplied. Krasowski and Harrison (2000) conducted an additional
experiment in order to determine if TCE was interacting with the receptor or perturbating the
cellular membrane (bilipid layer). Specific amino acids on the GABAa and glycine receptors
were mutated and in the presence of a receptor agonist (GABA for GABAa and glycine for
glycine receptors) and in these mutated receptors TCE-mediated potentiation was significantly
decreased or abolished suggesting that there was an interaction between TCE and these
receptors. Lopreato et al. (2003) conducted a similar study with the 5HT3A serotonin receptor
and found that when TCE was coapplied with serotonin, there was a potentiation in receptor
response. Additionally, TCE has been demonstrated to alter the function of voltage sensitive
calcium channels (VSCCs) by inhibiting the calcium mediated-current at a holding potential of
-70 mV and shifting the activation of the channels to a more hyperpolarizing potential (Shafer et
al., 2005).
4.3.10. Potential Mechanisms for Trichloroethylene (TCE)-Mediated Neurotoxicity
The mechanisms of TCE neurotoxicity have not been established despite a significant
level of research on the outcomes of TCE exposure. Results from several mechanistic studies
can be used to help elucidate the mechanism(s) involved in TCE-mediated neurological effects.
This document is a draft for review purposes only and does not constitute Agency policy.
164 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-37. Summary of in vitro ion channel effects with TCE exposure
2
Reference
Cellular
system
Neuronal channel/
receptor
Concentrations
Effects
In vitro studies
Shafer et al.
(2005)
PC12 cells
vscc
0, 500, 1,000,
1,500, or
2,000 (iM
Shift of VSCC activation to a more
hyperpolarizing potential.
Inhibition of VSCCs at a holding
potential of -70 mV.
Beckstead
et al. (2000)
Xenopus
oocytes
Human
recombinant:
glycine receptor al,
GAB Aa receptors,
aipi, aip2y2L
0 or 390 |iM
50% potentiation of the GABAa
receptors; 100% potentiation of the
glycine receptor.
Lopreato et al.
(2003)
Xenopus
oocytes
Human recombinant
serotonin 3 A
receptor
0 or 390 |iM
Potentiation of serotonin receptor
function.
Krasowski and
Harrison
(2000)
Human
embryonic
kidney
293 cells
Human recombinant
Glycine receptor al,
GAB Aa receptors
a2pi
Not provided
Potentiation of glycine receptor function
with an EC50 of 0.65 ± 0.05 mM.
Potentiation of GAB Aa receptor function
with an EC50 of 0.85 ± 0.2 mM.
3
4 EC50 = concentration of the chemical at which 50% of the maximal effect is produced.
5
6
7 The disruption of the trigeminal nerve appears to be a highly idiosyncratic outcome of
8 TCE exposure. There are limited data to suggest that it might entail a demyelination
9 phenomenon, but similar demyelination does not appear to occur in other nerve tracts. In this
10 regard, then, TCE is unlike a variety of hydrocarbons that have more global demyelinating
11 action. There are some data from central nervous system that focus on shifts in lipid profiles as
12 well as data showing loss of myelinated fibers in the hippocampus. However, the changes in
13 lipid profiles are both quite small and, also, inconsistent. And the limited data from
14 hippocampus are not sufficient to conclude that TCE has significant demyelinating effects in this
15 key brain region. Indeed, the bulk of the evidence from studies of learning and memory function
16 (which would be tied to hippocampal function) suggests no clear impairments due to TCE.
17 Some researchers (Albee et al., 1997; Albee et al., 2006; Barret et al., 1992; Barret et al.,
18 1991; Laureno, 1993) Laureno, 1988 have indicated that changes in trigeminal nerve function
19 may be due to dichloroacetylene which is formed under nonbiological conditions of high
20 alkalinity or temperature during volatilization of TCE. In experimental settings, trigeminal nerve
21 function (Albee et al., 1997) and trigeminal nerve morphology (Barret et al., 1992; Barret et al.,
22 1991) was found to be more altered following a low exposure to dichloroacetylene in comparison
This document is a draft for review purposes only and does not constitute Agency policy.
165 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
to the higher TCE exposure. Barret et al. (1992; 1991) also demonstrated that TCE
administration results in morphological changes in the trigeminal nerve. Thus, dichloroacetylene
may contribute to trigeminal nerve impairment may be plausible following an inhalation
exposure under conditions favoring its formation. Examples of such conditions include passing
through a carbon dioxide scrubber containing alkaline materials, application to remove a wax
coating from a concrete-lined stone floor, or mixture with alkaline solutions or caustic (Saunders,
1967) Greim et al., 1984 Bingham et al., 2001. However, dichloroacetylene exposures have not
been identified or measured in human epidemiologic studies with TCE exposure, and thus, do
not appear to be common to occupational or residential settings (Lash and Green, 1993).
Moreover, changes in trigeminal nerve function have also been consistently reported in humans
exposed to TCE following an oral exposure (Kilburn, 2002a); across many human studies of
occupational and drinking water exposures under conditions with highly varying potentials for
dichloroacetylene formation individuals (Barret et al., 1982; Barret et al., 1984; Barret et al.,
1987; Feldman et al., 1988). As a result, the mechanism(s) for trigeminal nerve function
impairment following TCE exposure is unknown(Kilburn, 2002b; Kilburn and Warshaw, 1993;
Mhiri et al., 2004; Ruijten et al., 1991). The varying dichloroacetylene exposure potential across
these studies suggests TCE exposure, which is common to all of them, as the most likely
etiologic agent for the observed effects.
The clearest consequences of TCE are permanent impairment of hearing in animal
models and disruption of trigeminal nerve function in humans with animal models showing
comparable changes following administration of a TCE metabolite. With regard to hearing loss,
the effect of TCE has much in common with the effects of several aromatic hydrocarbons
including ethylbenzene, toluene, and ^-xylene. Many studies have attempted to determine how
these solvents damage the cochlea. Of the hypotheses that have been advanced, there is little
evidence to suggest oxidative stress, changes in membrane fluidity, or impairment of central
efferent nerves whose endings innervate receptor cells in the cochlea. Rather, for reasons that
are still uncertain these solvents seem to preferentially target supporting cells in the cochlea
whose death then alters key structural elements of the cochlea resulting ultimately in hair cell
displacement and death. Recently, potential modes of action resulting in ototoxicity have been
speculated to be due to blockade of neuronal nicotinic receptors present on the auditory cells
(Campo et al., 2007) and potentially changes in calcium transmission (Campo et al., 2008) from
toluene exposure. Although these findings were reported following an acute toluene exposure, it
is speculated that this mechanism may be a viable mechanism for TCE-mediated ototoxicity.
A few studies have tried to relate TCE exposure with selective impairments of dopamine
neurons. Two studies (Gash et al., 2008; Guehl et al., 1999) demonstrated dopaminergic
This document is a draft for review purposes only and does not constitute Agency policy.
166 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
neuronal death and/or degeneration following an acute TCE administration. However, the only
human TCE exposure study examining dopamine neuronal activity found no changes in serum
dopamine P-hydroxylase activity in comparison to nonexposed individuals (Nagaya et al., 1990).
It is thought that TaClo, which can be formed from TCE metabolites such as
trichloroacetaldehyde, may be the potent neurotoxicant that selectively targets the dopaminergic
system. More studies are needed to confirm the dopamine neuronal function disruption and if
this disruption is mediated through TaClo.
There is good evidence that TCE and certain metabolites such as choral hydrate have
CNS depressant properties and may account for some of the behavioral effects (such as
vestibular effects, psychomotor activity changes, central visual changes, sleep and mood
changes) that have been observed with TCE. Specifically, in vitro studies have demonstrated
that TCE exposure results in changes in neuronal receptor function for the GABAa, glycine, and
serotonin receptors (Beckstead et al., 2000; Krasowski and Harrison, 2000; Lopreato et al.,
2003). All of these inhibitory receptors that are present in the CNS are potentiated when
receptor-specific agonist and TCE are applied. These results are similar to other anesthetics and
suggest that some of the behavioral functions are mediated by modifications in ion channel
function. However, it is quite uncertain whether there are persistent consequences to such high
dose TCE exposure. Additionally, with respect to the GABAergic system, acute administration
of TCE increased the seizure threshold appearance and this effect was the strongest with
convulsants that were GABA receptor antagonists (Shih et al., 2001). Therefore, this result
suggests that TCE interacts with the GABA receptor and that was also verified in vitro
(Beckstead et al., 2000; Krasowski and Harrison, 2000).
Also, TCE exposure has been linked to decreased sensitivity to titanic stimulation in the
hippocampus (Ohta et al., 2001) as well as significant reduction in myelin in the hippocampus in
a developmental exposure (Isaacson et al., 1990). These effects are notable since the
hippocampus is highly involved in memory and learning functions. Changes in the hippocampal
physiology may correlate with the cognitive changes that were reported following TCE
exposure.
4.3.11. Overall Summary and Conclusions—Weight of Evidence
Both human and animal studies have associated TCE exposure with effects on several
neurological domains. The strongest neurological evidence of hazard in humans is for changes
in trigeminal nerve function or morphology and impairment of vestibular function. Fewer and
more limited evidence exists in humans on delayed motor function, and changes in auditory,
This document is a draft for review purposes only and does not constitute Agency policy.
167 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
visual, and cognitive function or performance. Acute and subchronic animal studies show
morphological changes in the trigeminal nerve, disruption of the peripheral auditory system
leading to permanent function impairments and histopathology, changes in visual evoked
responses to patterns or flash stimulus, and neurochemical and molecular changes. Additional
acute studies reported structural or functional changes in hippocampus, such as decreased
myelination or decreased excitability of hippocampal CA1 neurons, although the relationship of
these effects to overall cognitive function is not established. Some evidence exists for
motor-related changes in rats/mice exposed acutely/subchronically to TCE, but these effects have
not been reported consistently across all studies.
Epidemiologic evidence supports a relationship between TCE exposure and trigeminal
nerve function changes, with multiple studies in different populations reporting abnormalities in
trigeminal nerve function in association with TCE exposure (Barret et al., 1982; Barret et al.,
1984; Barret et al., 1987; Feldman et al., 1988; Feldman et al., 1992; Kilburn, 2002b; Kilburn
and Warshaw, 1993; Mhiri et al., 2004) Ruitjen et al., 1991. Of these, two well conducted
occupational cohort studies, each including more than 100 TCE-exposed workers without
apparent confounding from multiple solvent exposures, additionally reported statistically
significant dose-response trends based on ambient TCE concentrations, duration of exposure,
and/or urinary concentrations of the TCE metabolite TCA (Barret et al., 1984; Barret et al.,
1987). Limited additional support is provided by a positive relationship between prevalence of
abnormal trigeminal nerve or sensory function and cumulative exposure to TCE (most subjects)
or CFC-113 (<25% of subjects) (Rasmussen et al., 1993a). Test for linear trend in this study was
not statistically significant and may reflect exposure misclassification since some subjects
included in this study did not have TCE exposure. The lack of association between TCE
exposure and overall nerve function in three small studies (trigeminal: El Ghawabi et al., 1973;
ulnar and medial: Triebig et al., 1983; Triebig et al., 1982) does not provide substantial evidence
against a causal relationship between TCE exposure and trigeminal nerve impairment because of
limitations in statistical power, the possibility of exposure misclassification, and differences in
measurement methods. Laboratory animal studies have also shown TCE-induced changes in the
trigeminal nerve. Although one study reported no significant changes in trigeminal
somatosensory evoked potential in rats exposed to TCE for 13 weeks (Albee et al., 2006), there
is evidence of morphological changes in the trigeminal nerve following short-term exposures in
rats (Barret et al., 1992; Barret et al., 1991).
Human chamber, occupational, geographic based/drinking water, and laboratory animal
studies clearly established TCE exposure causes transient impairment of vestibular function.
Subjective symptoms such as headaches, dizziness, and nausea resulting from occupational
This document is a draft for review purposes only and does not constitute Agency policy.
168 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
(Grandjean et al., 1955; Liu et al., 1988; Rasmussen and Sabroe, 1986; Smith, 1970)
environmental (Hirsch et al., 1996), or chamber exposures (Smith, 1970; Stewart et al., 1970)
have been reported extensively. A few laboratory animal studies have investigated vestibular
function, either by promoting nystagmus or by evaluating balance (Niklasson et al., 1993; Tham
et al., 1984; Tham et al., 1979; Umezu et al., 1997).
In addition, mood disturbances have been reported in a number of studies, although these
effects also tend to be subjective and difficult to quantify (Gash et al., 2008; Kilburn, 2002a, b;
Kilburn and Warshaw, 1993; McCunney, 1988; Mitchell and Parsons-Smith, 1969; Rasmussen
and Sabroe, 1986; Troster and Ruff, 1990), and a few studies have reported no effects from TCE
on mood (Reif et al., 2003; Triebig et al., 1976; Triebig et al., 1977a). Few comparable mood
studies are available in laboratory animals, although both Moser et al. (2003) and Albee et al.
(2006) report increases in handling reactivity among rats exposed to TCE. Finally, significantly
increased number of sleep hours was reported by Arito et al. (1994) in rats exposed via
inhalation to 50-300-ppm TCE for 8 hours/day for 6 weeks.
Four epidemiologic studies of chronic exposure to TCE observed disruption of auditory
function. One large occupational cohort study showed a statistically significant difference in
auditory function with cumulative exposure to TCE or CFC-113 as compared to control groups
after adjustment for possible confounders, as well as a positive relationship between auditory
function and increasing cumulative exposure (Rasmussen et al., 1993c). Of the three studies
based on populations from ATSDR's TCE Subregistry from the National Exposure Registry,
more limited than Rasmussen et al. (1993c) due to inferior exposure assessment, Burg et al.
(1995) and Burg and Gist (1999) reported a higher prevalence of self-reported hearing
impairments. The third study reported that auditory screening revealed abnormal middle ear
function in children less than 10-years-of-age, although a dose-response relationship could not be
established and other tests did not reveal differences in auditory function (ATSDR, 2003b).
Further evidence for these effects is provided by numerous laboratory animal studies
demonstrating that high dose subacute and subchronic TCE exposure in rats disrupts the auditory
system leading to permanent functional impairments and histopathology.
Studies in humans exposed under a variety of conditions, both acutely and chronically,
report impaired visual functions such as color discrimination, visuospatial learning tasks, and
visual depth perception in subjects with TCE exposure. Abnormalities in visual depth perception
were observed with a high acute exposure to TCE under controlled conditions (Vernon and
Ferguson, 1969). Studies of lower TCE exposure concentrations also observed visuofunction
effects. One occupational study (Rasmussen et al., 1993c) reported a statistically significant
positive relationship between cumulative exposure to TCE or CFC-113 and visual gestalts
This document is a draft for review purposes only and does not constitute Agency policy.
169 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
learning and retention among Danish degreasers. Two studies of populations living in a
community with drinking water containing TCE and other solvents furthermore suggested
changes in visual function (Kilburn, 2002b; Reif et al., 2003). These studies used more direct
measures of visual function as compared to Rasmussen et al. (1993c), but their exposure
assessment is more limited because TCE exposure is not assigned to individual subjects
(Kilburn, 2002b), or because there are questions regarding control selection (Kilburn, 2002b) and
exposure to several solvents (Kilburn, 2002b; Reif et al., 2003).
Additional evidence of effects of TCE exposure on visual function is provided by a
number of laboratory animal studies demonstrating that acute or subchronic TCE exposure
causes changes in visual evoked responses to patterns or flash stimulus (Blain et al., 1994; Boyes
et al., 2003; Boyes et al., 2005). Animal studies have also reported that the degree of some
effects is correlated with simultaneous brain TCE concentrations (Boyes et al., 2003; Boyes et
al., 2005) and that, after a recovery period, visual effects return to control levels (Blain et al.,
1994; Rebert et al., 1991). Overall, the human and laboratory animal data together suggest that
TCE exposure can cause impairment of visual function, and some animal studies suggest that
some of these effects may be reversible with termination of exposure.
Studies of human subjects exposed to TCE either acutely in chamber studies or
chronically in occupational settings have observed deficits in cognition. Five chamber studies
reported statistically significant deficits in cognitive performance measures or outcome measures
suggestive of cognitive effects (Gamberale et al., 1976; Stewart et al., 1970; Triebig et al., 1976;
Triebig et al., 1977b). Danish degreasers with high cumulative exposure to TCE or CFC-113
had a high risk (OR: 13.7, 95% CI: 2.0-92.0) for psychoorganic syndrome characterized by
cognitive impairment, personality changes, and reduced motivation, vigilance, and initiative
compared to workers with low cumulative exposure. Studies of populations living in a
community with contaminated groundwater also reported cognitive impairments (Kilburn,
2002b; Kilburn and Warshaw, 1993), although these studies carry less weight in the analysis
because TCE exposure is not assigned to individual subjects and their methodological design is
weaker.
Laboratory studies provide some additional evidence for the potential for TCE to affect
cognition, though the predominant effect reported has been changes in the time needed to
complete a task, rather than impairment of actual learning and memory function (Kishi et al.,
1993; Kulig, 1987; Umezu et al., 1997). In addition, in laboratory animals, it can be difficult to
distinguish cognitive changes from motor-related changes. However, several studies have
reported structural or functional changes in the hippocampus, such as decreased myelination
(Isaacson and Taylor, 1989; Isaacson et al., 1990) or decreased excitability of hippocampal CA1
This document is a draft for review purposes only and does not constitute Agency policy.
170 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
neurons (Ohta et al., 2001), although the relationship of these effects to overall cognitive
function is not established.
Two studies of TCE exposure, one chamber study of acute exposure duration and
one occupational study of chronic duration, reported changes in psychomotor responses. The
chamber study of Gamberale et al. (1976) reported a dose-related decrease in performance in a
choice reaction time test in healthy volunteers exposed to 100- and 200-ppm TCE for 70 minutes
as compared to the same subjects without exposure. Rasmussen et al. (1993a) reported a
statistically significant association with cumulative exposure to TCE or CFC-113 and
dyscoordination trend among Danish degreasers. Observations in a third study (Gun et al., 1978)
are difficult to judge given the author's lack of statistical treatment of data. In addition, Gash
et al. (2008) reported that 14 out of 30 TCE-exposed workers exhibited significantly slower fine
motor hand movements as measured through a movement analysis panel test. Studies of
population living in communities with TCE and other solvents detected in groundwater supplies
reported significant delays in simple and choice reaction times in individuals exposed to TCE in
contaminated groundwater as compared to referent groups (Kilburn, 2002b; Kilburn and
Thornton, 1996; Kilburn and Warshaw, 1993). Observations in these studies are more uncertain
given questions of the representativeness of the referent population, lack of exposure assessment
to individual study subjects, and inability to control for possible confounders including alcohol
consumption and motivation. Finally, in a presentation of two case reports, decrements in motor
skills as measured by the grooved pegboard and finger tapping tests were observed (Troster and
Ruff, 1990).
Laboratory animal studies of acute or subchronic exposure to TCE observed psychomotor
effects, such as loss of righting reflex (Shih et al., 2001; Umezu et al., 1997) and decrements in
activity, sensory-motor function, and neuromuscular function (Kishi et al., 1993; Moser et al.,
1995; Moser et al., 2003). However, two studies also noted an absence of significant changes in
some measures of psychomotor function (Albee et al., 2006; Kulig, 1987). In addition, less
consistent results have been reported with respect to locomotor activity in rodents. Some studies
have reported increased locomotor activity after an acute i.p. dosage (Wolff and Siegmund,
1978) or decreased activity after acute or short term oral gavage dosing (Moser et al., 1995,
2003). No change in activity was observed following exposure through drinking water (Waseem
et al., 2001), inhalation (Kulig, 1987) or orally during the neurodevelopment period (Fredriksson
et al., 1993).
Several neurochemical and molecular changes have been reported in laboratory
investigations of TCE toxicity. Kjellstrand et al. (1987) reported inhibition of sciatic nerve
regeneration in mice and rats exposed continuously to 150-ppm TCE via inhalation for 24 days.
This document is a draft for review purposes only and does not constitute Agency policy.
171 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Two studies have reported changes in GABAergic and glutamatergic neurons in terms of GABA
or glutamate uptake (Briving et al., 1986) or response to GABAergic antagonistic drugs (Shih et
al., 2001) as a result of TCE exposure, with the Briving et al. (1986) conducted at 50 ppm for
12 months. Although the functional consequences of these changes is unclear, Tham et al.
(1984; 1979) described central vestibular system impairments as a result of TCE exposure that
may be related to altered GABAergic function. In addition, several in vitro studies have
demonstrated that TCE exposure alters the function of inhibitory ion channels such as receptors
for GABAa glycine, and serotonin (Beckstead et al., 2000; Krasowski and Harrison, 2000;
Lopreato et al., 2003) or of voltage-sensitive calcium channels (Shafer et al., 2005).
4.4. KIDNEY TOXICITY AND CANCER
4.4.1. Human Studies of Kidney
4.4.1.1.1. Nonspecific Markers of Nephrotoxicity
Investigations of nephrotoxicity in human populations show that workers highly exposed
to TCE exhibit evidence of damage to the proximal tubule (NRC, 2006). The magnitude of
exposure needed to produce kidney damage is not clear. Several kidney early biological effect
markers, or biomarkers, are examined in these studies as are less sensitive clinical kidney
outcomes such as glomerular filtration rate and end stage disease. Observation of elevated
excretion of urinary proteins in the four studies of TCE exposure (Bolt et al., 2004; Briining et
al., 1999a; Briining et al., 1999b; Green et al., 2004) indicates the occurrence of a toxic insult
among TCE-exposed subjects compared to unexposed controls. Two studies are of subjects with
previously diagnosed kidney cancer (Bolt et al., 2004; Briining et al., 1999a), with limited
interpretation if effects are associated with exposure or to the disease process. Subjects in
Briining et al. (1999b) and Green et al. (2004) are disease free. Urinary proteins are considered
nonspecific markers of nephrotoxicity and include al-Microglobulin, albumin, and iV-acetyl-P-
D-glucosaminidase (NAG; Lybarger et al., 1999; Price et al., 1996; Price et al., 1999). Four
studies measure al-microglobulin with elevated excretion observed in the German studies (Bolt
et al., 2004; Briining et al., 1999a; Briining et al., 1999b) but not Green et al. (2004). However,
Green et al. (2004) found statistically significant group mean differences in NAG, another
nonspecific marker of tubular toxicity, in disease free subjects. Observations in Green et al.
(2004) provide evidence of tubular damage among workers exposed to trichloroethylene at
32 ppm (mean) (range, 0.5-252 ppm). Elevated excretion of NAG as a nonspecific marker of
This document is a draft for review purposes only and does not constitute Agency policy.
172 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
tubular damage has also been observed with acute TCE poisoning (Carried et al., 2007). These
and other studies relevant to evaluating TCE nephrotoxicity are discussed in more detail below.
Biological monitoring of persons who previously experienced "high" exposures to
trichloroethylene (100-500 ppm) in the workplace show altered kidney function evidenced by
urinary excretion of proteins suggestive of renal tubule damage. Similar results were observed in
the only study available of subjects with TCE exposure at current occupational limits (NRC,
2006). Table 4-38 provides details and results from these studies. Briining et al. (1999a) report a
statistically significantly higher prevalence of elevated proteinuria suggestive of severe tubular
damage (n = 24, 58. 5%,p<0 .01) and an elevated excretion of al-microglobin, another urinary
biomarker of renal tubular function, was observed in 41 renal cell carcinoma cases with prior
trichloroethylene exposure and with pending workman's compensation claims compared with the
nonexposed renal cell cancer patients (n = 14, 28%) and to hospitalized surgical patients n = 2,
2%). Statistical analyses did not adjust for differences in median systolic and diastolic blood
pressure that appeared higher in exposed renal cell carcinoma cases compared to nonexposed
controls. Similarly, severe tubular proteinuria is seen in 14 of 39 workers (35%) exposed to
trichloroethylene in the electrical department, fitters shop and through general degreasing
operations of felts and sieves in a cardboard manufacturing factory compared to no subjects of
46 nonexposed males office and administrative workers from the same factory (p < 0.01)
(Briining et al., 1999b). Furthermore, slight tubular proteinuria is seen in 20% of exposed
workers and in 2% of nonexposed workers (Briining et al., 1999b). Exposed subjects also had
statistically significantly elevated levels of al-microglobulin compared to unexposed controls.
Furthermore, subjects with tubular damage as indicated by urinary protein patterns had higher
GST-alpha concentrations than nonexposed subjects (p < 0.001). Both sex and use of spot or
24-hour urine samples are shown to influence al-microglobulin (Andersson et al., 2008);
however, these factors are not considered to greatly influence observations given only males
were subjects and al-microglobulin levels in spot urine sample are adjusted for creatinine
concentration.
Bolt et al. (2004) measured al-microglobulin excretion in living subjects from the renal
cell carcinoma case-control study by Briining et al. (2003). Some subjects in this study were
highly exposed. Of the 134 with renal cell cancer, 19 reported past exposures that led to narcotic
effects and 18 of the 401 controls, experienced similar effects (OR: 3.71, 95% CI: 1.80-7.54)
(Briining et al., 2003). Bolt et al. (2004) found that al-microglobulin excretion increased in
exposed renal cancer patients compared with nonexposed patients controls. A lower proportion
of exposed cancer patients had normal al-microglobulin excretion, less than 5 mg/L, the
detection level for the assay and the level considered by these investigators as associated with no
This document is a draft for review purposes only and does not constitute Agency policy.
173 DRAFT—DO NOT CITE OR QUOTE
-------
1 clinical or subclinical tubule damage, and a higher proportion of high values, defined as
2 >45 mg/L, compared to cases who did not report TCE occupational exposure and to nonexposed
3 controls (p < 0.05). Exposed cases, additionally, had statistically significantly higher median
4 concentration of al-microglobulin compared to unexposed cases in creatinine-unadjusted spot
5 urine specimens (p < 0.05). Reduced clearance of creatinine attributable to renal cancer does not
6 Table 4-38. Summary of human kidney toxicity studies
7
Subjects
Effect
Exposure
Reference
206 subjects—
104 male workers exposed
to TCE; 102 male controls
(source not identified)
Increased (^-microglobulin and total
protein in spot urine specimen,
(^-microglobulin:
Exposed, 129.0 + 113.3 mg/g
creatinine (Cr)
Controls, 113.6 + 110.6 mg/g Cr.
Total protein:
Exposed, 83.4 + 113.2 mg/g creatinine
(Cr)
Controls, 54.0 + 18.6 mg/g Cr.
TCE exposure was through
degreasing activities in metal
parts factory or semiconductor
industry.
U-total trichloro compounds:
Exposed, 83.4 mg/g Cr (range,
2-66.2 mg/g Cr.
Controls, N.D.
8.4 + 7.9 yr mean employment
duration.
Nagaya et al.
(1989b)
29 metal workers
NAG in morning urine specimen,
0.17 +.0.11 U/mmol Cr.
Breathing zone monitoring,
3 ppm (median) and 5 ppm
(mean).
Selden et al.
(1993)
191 subjects—
41 renal cell carcinoma
cases pending cases
involving compensation
with TCE exposure;
50 unexposed renal cell
carcinoma cases from same
area as TCE-exposed cases;
100 nondiseased control and
hospitalized surgical
patients
Increased urinary proteins patterns,
al-microglobulin, and total protein in
spot urine specimen.
Slight/severe tubular damage:
TCE RCC cases, 93%.
Nonexposed RCC cases, 46%
Surgical controls, 11%.
p < 0.01.
al-microglobulin (mg/g creatinine):
Exposed RCC cases, 24.6 ;t[SD] 13.9
Unexposed RCC cases, 11.3 ;t[SD]
9.8.
Surgical controls, 5.5 ;t[SD] 6.8.
All exposed RCC cases
exposed to 'high" and "very
high" TCE intensity.
18 yr mean exposure duration.
Briining et al.
(1999a)
This document is a draft for review purposes only and does not constitute Agency policy.
174 DRAFT—DO NOT CITE OR QUOTE
-------
85 male workers employed
Increased urinary protein patterns and
"High" TCE exposure to
Briining et al.
in cardboard manufacturing
excretion of proteins in spot urine
workers in the fitters shop and
(1999b)
factory (39 TCE exposed,
specimen.
electrical department.
46) nonexposed office and
Slight/severe tubular damage:
"Very high" TCE exposure to
administrative controls)
TCE exposed, 67%
workers through general
Nonexposed, RCC cases, 9%
degreasing operations in carton
p < 0.001.
machinery section.
al-microglobulin (mg/g creatinine):
Exposed, 16.2 + [SD] 10.3
Unexposed, 7.8 + [SD] 6.9
p < 0.001.
GST-alpha (|ig/g creatinine):
Exposed 6.0 + [SD] 3.3
Unexposed, 2.0 + [SD] 0.57
p < 0.001.
No group differences in total protein
or GST-pi.
1
This document is a draft for review purposes only and does not constitute Agency policy.
175 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-38. Summary of human kidney toxicity studies (continued)
Subjects
Effect
Exposure
Reference
99 renal cell carcinoma
cases and 298 hospital
controls (from Briining
et al. (2003) and alive at the
time of interview)
Increased excretion of
al-microglobulin in spot urine
specimen.
Proportion of subjects with
al-microglobulin <5.0
mg/L:
Exposed cases, 15%
Unexposed cases, 51%
Exposed controls, 55%
Unexposed controls, 55%
p < 0.05, prevalence of exposed cases
compared to prevalences of either
exposed controls or unexposed
controls.
Mean al-microglobulin:
Exposed cases, 18.1 mg/L
Unexposed cases, <5.0 mg/L
p < 0.05.
All exposed RCC cases
exposed to 'high" and "very
high" TCE intensity .
Bolt et al.
(2004)
124 subjects (70 workers
currently exposed to TCE
and 54 hospital and
administrative staff
controls)
Analysis of urinary proteins in spot
urine sample obtained 4 d after
exposure.
Increased excretion of albumin, NAG,
and formate in spot urine specimen.
Albumin (mg/g creatinine):3
Exposed, 9.71 + [SD] 11.6
Unexposed, 5.50 + [SD] 4.27
p < 0.05.
Total NAG (U/g creatinine):
Exposed, 5.27 + [SD] 3.78
Unexposed, 2.41 + [SD] 1.91
p < 0.01.
Format (mg/g creatinine):
Exposed, 9.45 + [SD] 4.78
Unexposed, 5.55 + [SD] 3.00
p < 0.01.
No group mean differences in
GST-alpha, retinol binding protein,
al-microglobulin, (^-microglobulin,
total protein, and methylmalonic acid.
Mean U-TCA of exposed
workers was 64 + [SD] 102
(Range, 1-505).
Mean U-TCOH of exposed
workers was 122 + [SD] 119
(Range, 1-639).
Mean TCE concentration to
exposed subjects was
estimated as 32 ppm (range,
0.5-252 ppm) and was
estimated by applying the
German occupational
exposure limit (maximale
arbeitsplatz konzentration,
MAK) standard to U-TCA and
assuming that the linear
relationship holds for
exposures above 100 ppm.
86% of subjects with exposure
to <50 ppm TCE.
Green et al.
(2004)
This document is a draft for review purposes only and does not constitute Agency policy.
176 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-38. Summary of human kidney toxicity studies (continued)
Subjects
Effect
Exposure
Reference
101 cases or deaths from
ESDR among male and
female subjects in Hill Air
Force Base aircraft
maintenance worker cohort
of Blair et al. (1998)
TCE exposure:
Cox Proportional Hazard Analysis:
Ever exposed to TCE,b
1.86 (1.02,3.39).
Logistic regression:15
No chemical exposure (referent
group): 1.0
<5 unit-yr, 1.73 (0.86, 3.48)
5-25 unit-yr, 1.65 (0.82, 3.35)
>25 unit-yr, 1.65 (0.82,3.35)
Monotonic trend test, p > 0.05.
Indirect low-intermittent TCE
exposure, 2.47 (1.17, 5.19)
Indirect peak/infrequent TCE
exposure 3.55 (1.25, 10.74)
Direct TCE exposure, "not
statistically significant" but hazard
ratio and confidence intervals were
not presented in paper.
Cumulative TCE exposure
(intensity x duration)
identified using three
categories, <5 unit-yr, 5-25
unit-yr, >25 unit-yr per job
exposure matrix of Stewart
et al. (1991).
Radican et al.
(2006)
269 cases of IgA
nephropathy or
membranous nephropathy
glomerulonephritis
followed 5 yr (mean) for
progression to ESRD
TCE exposure:
Cox Proportional Hazard Analysis:
Ever exposed to TCE,b
2.5 (0.9, 6.5)
High exposure level to TCE,b
2.7 (0.7, 10.1).
Exposure to TCE assigned
usingjob title and job-
exposure matrix; two dose
surrogates, ever exposed and
high exposure level.
Jacob et al.
(2007)
1
2 aFor a urine sample, 10-17 mg of albumin per g of creatinine is considered to be suspected albuminuria in males
3 (15-25 in females) (de Jong and Brenner, 2004).
4 b Hazard ratio and 95% CI.
5
6 ESDR = end-stage renal disease, N.D. = not detectable, SD = standard deviation.
7
8
9 explain the lower percentage of normal values among exposed cases given findings of similar
10 prevalence of normal excretion among unexposed renal cell cases and controls.
11 In their study of 70 current employees (58 males, 12 females) of an electronic factory
12 with trichloroethylene exposure and 54 (50 males, 4 females) age-matched subjects drawn from
13 hospital or administrative staff, Green et al. (2004) found that urinary excretion of albumin, total
14 NAG and formate were increased in the exposed group compared with the unexposed group.4
15 No differences between exposed and unexposed subjects were observed in other urinary proteins,
4 Elevation of NAG in urine is a sign of proteinuria, and proteinuria is both a sign and a cause of kidney malfunction
(Zandi-Neiad et al.. 2004). For a urine sample, 10-17 mg of albumin per g of creatinine is considered to be
suspected albuminuria in males (15-25 in females) (de Jong and Brenner. 2004).
This document is a draft for review purposes only and does not constitute Agency policy.
177 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
including al-microglobulin, p2-microglobulin, and GST-alpha. Green et al. (2004) stated that
NAG is not an indicator of nephropathy, or damage, but rather is an indicator of functional
change in the kidney. Green et al. (2004) further concluded that increased urinary albumin or
NAG was not related to trichloroethylene exposure; analyses to examine the exposure-response
relationship found neither NAG or albumin concentration correlated to U-TCA or employment
duration (years). The National Research Council (NRC, 2006) did not consider U-TCA as
sufficiently reliable to use as a quantitative measure of TCE exposure, concluding that the data
reported by Green et al. (2004) were inadequate to establish exposure-response information
because the relationship between U-TCA and ambient TCE intensity is highly variable and
nonlinear, and conclusions about the absence of association between TCE and nephrotoxicity can
not be made based on U-TCA. Moreover, use of employment duration does not consider
exposure intensity differences between subjects with the same employment duration, and bias
introduced through misclassification of exposure may explain the Green et al. (2004) findings.
Selden et al. (1993) in their study of 29 metal workers (no controls) reported a correlation
between NAG and U-TCA (r = 0,48, p < 0.01) but not with other exposure metrics of recent or
long-term exposure. Personal monitoring of worker breath indicated median and mean
time-weighted-average TCE exposures of 3 and 5 ppm, respectively. Individual NAG
concentrations were within normal reference values. Rasmussen et al. (1993b), also, reported a
positive relationship (p = 0.05) between increasing urinary NAG concentration (adjusted for
creatinine clearance) and increasing duration in their study of 95 metal degreasers (no controls)
exposed to either TCE (70 subjects) or CFC113(25 subjects). Multivariate regression analyses
which adjusted for age were suggestive of an association between NAG and exposure duration
(p = 0.011). Mean urinary NAG concentration was higher among subjects with annual exposure
of >30 hours/week, defined as peak exposure, compared to subjects with annual exposure of less
than <30 hours/week (72.4 +_44.1 (_ig/g creatinine compared to 45.9 j^30.0 j_ig/g creatinine,
^<0.01).
Nagaya et al. (1989b) did not observe statistically significant group differences in urinary
P2-microglobulin and total protein in spot urine specimens of male degreasers and their controls,
nor were these proteins correlated with urinary total trichloro-compounds (U-TTC). The paper
lacks details on subject selection, whether urine collection was at start of work week or after
sufficient exposure, and presentation of ^-values and correlation coefficients. The presentation
of urinary protein concentrations stratified by broad age groups is less statistically powerful than
examination of this confounder using logistic regression. Furthermore, although valid for
pharmacokinetic studies, examination of renal function using U-TTC as a surrogate for TCE
exposure is uncertain, as discussed above for Green et al. (2004).
This document is a draft for review purposes only and does not constitute Agency policy.
178 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
4.4.1.1.2. End-Stage Renal Disease
End-stage renal disease is associated with hydrocarbon or organic solvent exposures in
two studies examining this endpoint (Jacob et al., 2007; Radican et al., 2006). Table 4-38
provides details and results from Radican et al. (2006) and Jacob et al. (2007). Radican et al.
(2006) assessed end-stage renal disease in a cohort of aircraft maintenance workers at Hill Air
Force Base (Blair et al., 1998) with strong exposure assessment to trichloroethylene (NRC, 2006)
and reported a twofold risk with overall TCE exposure and end stage renal disease (1.86, 95%
CI: 1.02, 3.39). A second study, the GN-PROGRESS retrospective cohort study, observed a
twofold elevated risk for progression of glomerulonephritis to ESRD from TCE (overall
exposure: 2.5, 95% CI: 0.9-6.5; high level TCE exposure: 2.7, 95% CI: 0.7, 10.1) (Jacob et al.,
2007). Statistical power was more limited in Jacob et al. (2007) because of its smaller number of
exposed cases, 21 with overall exposure, compared to 56 exposed cases in Radican et al. (2006).
Other occupational studies do not examine end-stage renal disease specifically, instead reporting
relative risks associated with deaths due to nephritis and nephrosis (ATSDR, 2004; Boice et al.,
1999; Boice et al., 2006b), all genitourinary system deaths (Costa et al., 1989; Garabrant et al.,
1988; Ritz, 1999a), or providing no information on renal disease mortality in the published paper
(Blair et al., 1998; Chang et al., 2003; Morgan et al., 1998).
4.4.2. Human Studies of Kidney Cancer
Cancer of the kidney and renal pelvis is the 6th leading cause of cancer in the
United States with an estimated 54,390 (33,130 men and 21,260 women) newly diagnosed cases
and 13,010 deaths (Jemal et al., 2008; NCI, 2008). Age-adjusted incidence rates based on cases
diagnosed in 2001-2005 from 17 Surveillance, Epidemiology, and End Results (SEER)
geographic areas are 18.3 per 100,000 for men and 9.2 per 100,000 for women. Age-adjusted
mortality rates are much lower; 6.0 per 100,000 for men and 2.7 for women.
Cohort, case-control, and geographical studies have examined trichloroethylene and
kidney cancer, defined either as cancer of kidney and renal pelvis in cohort and geographic based
studies or as renal cell carcinoma, the most common type of kidney cancer, in case-control
studies. Appendix C identifies these studies' design and exposure assessment characteristics.
Observations in these studies are presented below in Table 4-39. Rate ratios for incidence
studies in Table 4-39 are, generally, larger than for mortality studies.
Additionally, a large body of evidence exists on kidney cancer risk and either job or
industry titles where trichloroethylene usage has been documented. TCE has been used as a
This document is a draft for review purposes only and does not constitute Agency policy.
179 DRAFT—DO NOT CITE OR QUOTE
-------
1 degreasing solvent in a number of jobs, task, and industries, some of which include metal,
2 electronic, paper and printing, leather manufacturing and aerospace/aircraft manufacturing or
3
This document is a draft for review purposes only and does not constitute Agency policy.
180 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-39. Summary of human studies on TCE exposure and kidney cancer
2
Exposure group
Relative risk (95% CI)
No. obs.
events
Reference
Cohort and PMR studies—incidence
Aerospace workers (Rocketdyne)
Zhao et al. (2005)
Any exposure to TCE
Not reported
Low cum TCE score
1.00a
6
Med cum TCE score
1.87 (0.56, 6.20)
6
High TCE score
4.90 (1.23, 19.6)
4
p for trend
p = 0.023
TCE, 20 yr exposure lagb
Low cum TCE score
1.00a
6
Med cum TCE score
1.19(0.22, 6.40)
7
High TCE score
7.40 (0.47, 116)
3
p for trend
p = 0.120
All employees at electronics factory (Taiwan)
Chang et al. (2005)
Males
1.06 (0.45, 2.08)c
8
Females
1.09 (0.56, 1.91)0
12
Females
1.10(0.62, 1.82)c
15
Sung et al. (2008)
Danish blue-collar worker with TCE exposure
Raaschou-Nielsen
Any exposure, all subjects
1.2 (0.98, 1.46)
103
et al. (2003)
Any exposure, males
1.2 (0.97, 1.48)
93
Any exposure, females
1.2(0.55,2.11)
10
Exposure lag time
20 yr
1.3 (0.86, 1.88)
28
Employment duration
<1 yr
0.8 (0.5, 1.4)
16
1-4.9 yr
1.2 (0.8, 1.7)
28
>5 yr
1.6 (1.1,2.3)
32
Subcohort w/higher exposure
Any TCE exposure
1.4 (1.0, 1.8)
53
Employment duration
1-4.9 yr
1.1 (0.7, 1.7)d
23
>5 yr
1.7 (1.1, 2.4)d
30
This document is a draft for review purposes only and does not constitute Agency policy.
181 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-39. Summary of human studies on TCE exposure and kidney cancer
(continued)
Exposure group
Relative risk (95% CI)
No. obs.
events
Reference
Biologically monitored Danish workers
1.1 (0.3,2.8)
4
Hansen etal. (2001)
Any TCE exposure, males
0.9 (0.2, 2.6)
3
Any TCE exposure, females
2.4 (0.03, 14)
1
Cumulative exp (Ikeda)
Not reported
<17 ppm-yr
>17 ppm-yr
Mean concentration (Ikeda)
Not reported
<4 ppm
4+ ppm
Employment duration
Not reported
<6.25 yr
>6.25
Aircraft maintenance workers from Hill Air Force Base
Blair et al. (1998)
TCE subcohort
Not reported
Males, cumulative exp
0
1.0a
<5 ppm-yr
1.4 (0.4, 4.7)
9
5-25 ppm-yr
1.3 (0.3, 4.7)
5
>25 ppm-yr
0.4(0.1,2.3
2
Females, cumulative exp
0
1.0a
<5 ppm-yr
0
5-25 ppm-yr
0
>25 ppm-yr
3.6 (0.5, 25.6)
2
Biologically-monitored Finnish workers
Anttila et al. (1995)
All subjects
0.87 (0.32, 1.89)
6
Mean air-TCE (Ikeda extrapolation)
<6 ppm
Not reported
6+ ppm
Not reported
Cardboard manufacturing workers in Arnsberg, Germany
Henschler et al.
Exposed workers
7.97 (2.59, 8.59)e
5
(1995)
Biologically-monitored Swedish workers
Axelsonetal. (1994)
Any TCE exposure, males
1.16(0.42, 2.52)
6
Any TCE exposure, females
Not reported
This document is a draft for review purposes only and does not constitute Agency policy.
182 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-39. Summary of human studies on TCE exposure and kidney cancer
(continued)
Exposure group
Relative risk (95% CI)
No. obs.
events
Reference
Cardboard manufacturing workers, Atlanta area, GA
Sinks et al. (1992)
All subjects
3.7 (1.4, 8.1)
6
All departments
oo (3.0, co)f
5
Finishing department
16.6 (1.7, 453.l)f
3
Cohort and PMR studies—mortality
Computer manufacturing workers (IBM), NY
Males
1.64 (0.45, 4.21)g
4
Clapp and Hoffman
(2008)
Females
0
Aerospace workers (Rocketdyne)
Any TCE (utility/eng flush)
2.22 (0.89, 4.57)
7
Boice et al. (2006b)
Any exposure to TCE
Not reported
Zhao et al. (2005)
Low cum TCE score
1.00a
7
Med cum TCE score
1.43 (0.49, 4.16)
7
High TCE score
2.13 (0.50, 8.32)
3
p for trend
p = 0.31
TCE, 20 yr exposure lagb
Low cum TCE score
1.00a
10
Med cum TCE score
1.69 (0.29, 9.70)
6
High TCE score
1.82 (0.09, 38.6)
1
p for trend
p = 0.635
View-Master employees
ATSDR (2004)
Males
2.76 (0.34, 9.96)g
2
Females
6.21 (2.68, 12.23)8
8
United States Uranium-processing workers (Fernald)
Ritz (1999a) (as
reported in NRC,
2006)
Any TCE exposure
Not reported
Light TCE exposure, 2-10 yr duration"1
1.94 (0.59, 6.44)
5
Light TCE exposure, >10 yr duration"1
0.76(0.14, 400.0)
2
Mod TCE exposure, >2 yr duration"1
0
This document is a draft for review purposes only and does not constitute Agency policy.
183 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-39. Summary of human studies on TCE exposure and kidney cancer
(continued)
Exposure group
Relative risk (95% CI)
No. obs.
events
Reference
Aerospace workers (Lockheed)
Boice et al. (1999)
Routine Exp
0.99 (040, 2.04)
7
Routine-Intermittent3
Not presented
11
Duration of exposure
0 yr
1.0
22
<1 yr
0.97 (0.37, 2.50)
6
1-4 yr
0.19(0.02, 1.42)
1
>5 yr
0.69 (0.22, 2.12)
4
p for trend
Aerospace workers (Hughes)
Morgan et al. (1998)
TCE subcohort
1.32 (0.57, 2.60)
8
Low intensity (<50 ppm)e
0.47 (0.01, 2.62)
1
High intensity (>50 ppm)e
1.78 (0.72, 3.66)
7
TCE subcohort (Cox analysis)
Never exposed
1.00a
24
Ever exposed
1.14 (0.51, 2.58)h
8
Peak
No/Low
1.00a
24
Med/Hi
1.89 (0.85, 4.23)h
8
Cumulative
Referent
1.00a
24
Low
0.31 (0.04, 2.36)h
1
High
1.59 (0.68, 3.71)h
7
Aircraft maintenance workers (Hill AFB, Utah)
Blair et al. (1998)
TCE subcohort
1.6 (0.5, 5.1)a
15
Males, cumulative exp
0
1.0a
<5 ppm-yr
2.0 (0.5, 7.6)
8
5-25 ppm-yr
0.4(0.1,4.0)
1
>25 ppm-yr
1.2 (0.3, 4.8)
4
Females, cumulative exp
0
1.0a
<5 ppm-yr
0
5-25 ppm-yr
9.8 (0.6, 157)
1
>25 ppm-yr
3.5 (0.2, 56.4)
1
TCE subcohort
1.18(0.47, 2.94)1
18
Radican et al. (2008)
Males, cumulative exp
1.24 (0.41, 3.71)1
16
0
1.01
This document is a draft for review purposes only and does not constitute Agency policy.
184 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-39. Summary of human studies on TCE exposure and kidney cancer
(continued)
Exposure group
Relative risk (95% CI)
No. obs.
events
Reference
Aircraft maintenance workers (Hill AFB, Utah) (continued)
Blair et al. (1998)
<5 ppm-yr
1.87 (0.59, 5.97)1
10
5-25 ppm-yr
0.31 (0.03,2.75)'
1
>25 ppm-yr
1.16 (0.31,4.32)'
5
Females, cumulative exp
0.93 (0.15, 5.76)'
2
0
1.0a
<5 ppm-yr
0
5-25 ppm-yr
2.86 (0.27, 29.85)'
1
>25 ppm-yr
0.97 (0.10, 9.50)'
1
Cardboard manufacturing workers in Arnsberg, Germany
Henschler et al.
TCE exposed workers
3.28 (0.40, 11.84)
2
(1995)
Unexposed workers
(0.00, 5.00)
0
Deaths reported to among GE pension fund (Pittsfield, MA)
0.99 (0.30, 3.32)f
12
Greenland et al.
(1994)
Cardboard manufacturing workers, Atlanta area, GA
Sinks et al. (1992)
1.4 (0.0, 7.7)
1
U.S. Coast Guard employees
Blair et al. (1989)
Marine inspectors
1.06 (0.22,3.10)
3
Noninspectors
1.03 (0.21, 3.01)
3
Aircraft manufacturing plant employees (Italy)
Costa et al. (1989)
All subjects
Not reported
Aircraft manufacturing plant employees (San Diego, CA)
Garabrant et al.
All subjects
0.93 (0.48, 1.64)
12
(1988)
Case-control studies
Population of four countries in central and eastern Europe
Moore et al. (2010)
Any TCE exposure
1.63 (1.04, 2.54)
48
Any TCE exposure (High confidence exposure)
2.05 (1.13, 3.73)
29
Cumulative TCE exposure
<1.58 ppm-yr
1.19(0.61,2.35)
17
>1.58 ppm-vr
2.02 (1.14, 3.59)J
31
p for trend
p = 0.02
Average intensity
<0.076 ppm
1.38 (0.81,2.35)
31
>0.076 ppm
2.34 (1.05, 5.21)
17
p for trend
p = 0.02
This document is a draft for review purposes only and does not constitute Agency policy.
185 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-39. Summary of human studies on TCE exposure and kidney cancer
(continued)
No. obs.
Exposure group
Relative risk (95% CI)
events
Reference
Population of Arve Valley, France
Charbotel et al.
Any TCE exposure
1.64 (0.95, 2.84)
37
(2006; 2009; 2007)
Any TCE exposure (High confidence exposure)
1.88 (0.89, 3.98)
16
Cumulative TCE exposure
Referent/nonexposed
1.00a
49
Low, 62.4 ppm-yr1
1.62 (0.75, 3.47)
12
Medium, 253.2 ppm-yr k
1.15 (0.47, 2.77)
9
High, 925 ppm-yr15
2.16(1.02, 4.60)1
16
Test for trend
p = 0.04
Cumulative TCE exposure + peak
Referent/nonexposed
1.00a
49
Low/medium, no peaks
1.35 (0.69, 2.63)
18
Low/medium + peaks
1.61 (0.36. 7.30)
3
High, no peaks
1.76 (0.65, 4.73)
8
High + peaks
2.73 (1.06, 7.07)1
8
Cumulative TCE exposure, 10-yr lag
Referent/nonexposed
1.00a
49
Low/medium, no peaks
1.44 (0.69, 2.80)
19
Low/medium + peaks
1.38 (0.32, 6.02)
3
High, no peaks
1.50 (0.53,4.21)
7
High + peaks
3.15 (1.19, 8.38)
8
Time-weighted-average TCE exposure™
Referent/nonexposed
1.00a
46
Any TCE without cutting fluid
1.62 (0.76, 3.44)
15
Any cutting fluid without TCE
2.39 (0.52, 11.03)
3
<50 ppm TCE + cutting fluid
1.14(0.49, 2,66)
12
50 + ppm TCE + cutting fluid
2.70(1.02, 7.17)
10
Population of Arnsberg Region, Germany
Briining et al. (2003)
Longestjob held—TCE/PERC
(CAREX)
1.80 (1.01,3.20)
117
Self-assessed exposure to TCE
2.47 (1.36, 4.49)
25
Duration of self-assessed TCE exposure
0
1.00a
109
<10 yr
3.78 (1.54, 9.28)
11
10-20 yr
1.80 (0.67, 4.79)
7
>20 yr
2.69 (0.84, 8.66)
8
This document is a draft for review purposes only and does not constitute Agency policy.
186 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-39. Summary of human studies on TCE exposure and kidney cancer
(continued)
Exposure group
Relative risk (95% CI)
No. obs.
events
Reference
Population in five German Regions
Pesch et al., (2000a)
Any TCE Exposure
Not reported
Males
Not reported
Females
Not reported
TCE exposure (Job Task Exposure Matrix)
Males
Medium
1.3 (1.0, 1.8)
68
High
1.1 (0.8, 1.5)
59
Substantial
1.3 (0.8,2.1)
22
Females
Medium
1.3 (0.7,2.6)
11
High
0.8 (0.4, 1.9)
7
Substantial
1.8 (0.6, 5.0)
5
Population of Minnesota
Dosemeci et al.
Ever exposed to TCE, NCI JEM
(1999)
Males
1.04 (0.6, 1.7)
33
Females
1.96(1.0,4.0)
22
Males + Females
1.30(0.9, 1.9)
55
Population of Arnsberg Region, Germany
Vamvakas et al.
Self-assessed exposure to TCE
10.80 (3.36, 34.75)
19
(1998)
Population of Montreal, Canada
Siemiatycki et al.
Any TCE exposure
0.8 (0.4, 2.0)n
4
(1991)
Substantial TCE exposure
0.8 (0.2, 2.6)n
2
Geographic based studies
Residents in two study areas in Endicott, NY
1.90 (1.06,3.13)
15
ATSDR (2006a)
(2008)
Residents of 13 census tracts in Redlands, CA
0.80 (0.54, 1.12)°
54
Morgan and Cassady
(2002)
Finnish residents
Vartiainen et al.
Residents of Hausjarvi
Not reported
(1993)
Residents of Huttula
Not reported
1
2 a Internal referents, workers not exposed to TCE.
3 b Relative risks for TCE exposure after adjustment for 1st employment, socioeconomic status, age at event, and all
4 other carcinogens, including hydrazine.
5 0 Chang et al. (2005)—urinary organs combined.
6 d SIR for renal cell carcinoma.
7 e Henschler et al. (1995) Expected number of incident cases calculated using incidence rates from the Danish Cancer
8 Registry.
9 f Odds ratio from nested case-control analysis.
10 8 Proportional mortality ratio.
This document is a draft for review purposes only and does not constitute Agency policy.
187 DRAFT—DO NOT CITE OR QUOTE
-------
1
2 Table 4-39. Summary of human studies on TCE exposure and kidney cancer
3 (continued)
4
5 hRisk ratio from Cox Proportional Hazard Analysis, stratified by age, sex and decade (EHS, 1997).
6 1 In Radican et al. (2008), kidney cancer defined as renal cell carcinoma (ICDA 8 code 189.0) and estimated relative
7 risks from Cox proportional hazard models were adjusted for age and sex.
8 J The odds ratio, adjusted for age, sex and center, for subjects with high-confidence exposure assessment with
9 cumulative exposure, >1.58 ppm-yr, was 2.23 (95% CI: 1.07, 4.64) and p-valuc for trend = 0.02.
10 k Mean cumulative exposure score in Charbotel et al. (2006) (personal communication from Barbara Charbotel,
11 University of Lyon, to Cheryl Scott, U.S. EPA, 11 April 2008).
12 1 In Charbotel et al. (2006) analyses adjusted for age, sex, smoking and body mass index. The odds ratio, adjusted
13 for age, sex, smoking, body mass index and exposure to cutting fluids and other petroleum oils, for high
14 cumulative TCE exposure was 1.96 (95% CI: 0.71, 5.37) and for high cumulative + peak TCE exposure was 2.63
15 (95% CI: 0.79, 8.83). The odds ratio for, considering only job periods with high confidence TCE exposure
16 assessment, adjusted for age, sex, smoking and body mass index, for high cumulative dose plus peaks was 3.80
17 (95% CI: 1.27. 11.40).
18 m The exposure surrogate is calculated for one occupational period only and is not the average exposure
19 concentration over the entire employment period.
20 n 90% CI.
21 ° 99% CI.
22
23 GE = General Electric, IBM = International Business Machines Corporation, JEM = job-exposure matrix, NCI =
24 National Cancer Institute, PERC = perchloroethylene, PMR = proportionate mortality ratio.
25
26
27 maintenance industries and job title of degreaser, metal workers, electrical worker, and machinist
28 (IARC, 1995b; Purdue et al., 2011). NRC (2006) identifies characteristics for kidney cancer
29 case-control studies that assess job title or occupation in their Table 3-8. Relative risks and
30 95% CIs reported in these studies are found in Table 4-40 below.
31
4.4.2.1.1. Studies of Job Titles and Occupations with Historical Trichloroethylene (TCE)
Usage
32 Elevated risks are observed in many of the cohort or case-control studies between kidney
33 cancer and industries or job titles with historical use of trichloroethylene (Briining et al., 2003;
34 Charbotel et al., 2006; Mandel et al., 1995; Mattioli et al., 2002; McCredie and Stewart, 1993;
35 Parent et al., 2000a; Partanen et al., 1991; Pesch et al., 2000a; Schlehofer et al., 1995; Wilson et
36 al., 2008; Zhang et al., 2004). Overall, these studies, although indicating association with metal
37 work exposures and kidney cancer, are insensitive for identifying a TCE hazard. The use of job
38 title or industry as a surrogate for exposure to a chemical is subject to substantial
39 misclassification that will attenuate rate ratios due to exposure variation and differences among
40 individuals with the same job title. Several small case-control studies (Auperin et al., 1994;
41 Harrington et al., 1989; Jensen et al., 1988; Parent et al., 2000a; Sharpe et al., 1989; Vamvakas et
This document is a draft for review purposes only and does not constitute Agency policy.
188 DRAFT—DO NOT CITE OR QUOTE
-------
1 al., 1998) have insufficient statistical power to detect modest associations due to their small size
2 and potential exposure misclassification (NRC, 2006). For these reasons, statistical
3 Table 4-40. Summary of case-control studies on kidney cancer and
4 occupation or job title
5
Case ascertainment area/exposure group
Relative risk
(95% CI)
No. exposed
cases
Reference
Swedish Cancer Registry Cases
Wilson etal. (2008)
Machine/electronics industry
1.30 (1.08, 1.55)a [M]
120
1.75 (1.04, 2.76 a[F]
18
Shop and construction metal work
1.19(1.00, 1.40)a [M]
143
Machine assembly
1.62 (0.94, 2.59)a [M]
Metal plating work
2.70 (0.73, 6.92)a [M]
4
Shop and construction metal work
1.66 (0.71, 3.26)a [F]
8
Arve Valley, France
Charbotel et al.
Metal industry
1.02 (0.59, 1.76)
28
(2006)
Metal workers, job title
1.00 (0.56, 1.77)
25
Metal industry, screw-cutting workshops
1.39 (0.75, 2.58)
22
Machinery, electrical and transportation
equipment manufacture
1.19(0.61,2.33)
15
Iowa Cancer Registry Cases
Zhang et al. (2004)
Assemblers
2.5 (0.8, 7.6)
5
>10 yr employment
4.2 (1.2, 15.3)
4
Arnsberg Region, Germany
Briining et al. (2003)
Iron/steel
1.15 (0.29, 4.54)
3
Occupations with contact to metals
1.53 (0.97, 2.43)
46
Longest job held
1.14(0.66, 1.96)
24
Metal greasing/degreasing
5.57 (2.33, 13.32)
15
Degreasing agents
Low exposure
2.11 (0.86, 5.18)
9
High exposure
1.01 (0.40, 2.54)
7
Bologna, Italy
Mattioli et al. (2002)
Metal workers
2.21 (0.99, 5.37)
37
Printers
1.55 (0.17, 13.46)
7
Solvents
0.79 (0.31, 1.98) [M]
17
1.47 (0.12, 17.46) [F]
3
6
7
This document is a draft for review purposes only and does not constitute Agency policy.
189 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-40. Summary of case-control studies on kidney cancer and
occupation or job title (continued)
Case ascertainment area/exposure group
Relative risk
(95% CI)
No. exposed
cases
Reference
Montreal, Canada
Parent et al. (2000a)
Metal fabricating and machining industry
1.0 (0.6, 1.8)
14
Metal processors
1.2 (0.4, 3.4)
4
Printing and publishing industry
1.1 (0.4, 3.0)
4
Printers
3.0 (1.2,7.5)
6
Aircraft mechanics
2.8 (1.0, 8.4)
4
5 Regions in Germany
Pesch et al. (2000a)
Electrical and electronic equipment
3.2 (1.0, 10.3) [M]
5
assembler
2.7 (1.3, 5.8) [F]
11
Printers
3.5 (1.1, 11.2)[M]
5
2.1 (0.4, 11.7) [F]
2
Metal cleaning/degreasing, job task
1.3 (0.7, 2.3) [M]
15
1.5 (0.3, 7.7) [F]
2
New Zealand Cancer Registry
Delahunt et al.
Toolmakers and blacksmiths
1.48 (0.72, 3.03)
No info
(1995)
Printers
0.67 (0.25, 1.83)
Minnesota Cancer Surveillance System
Mandel et al. (1995)
Iron or steel
1.6 (1.2,2.2)
8
Rhein-Neckar-Odenwald Area, Germany
Schlehofer et al.
Metal
(1995)
Industry
1.63 (1.07, 2.48)
71
Occupation
1.38 (0.89, 2.12)
Electronic
Industry
0.51 (0.26, 1.01)
14
Occupation
0.57 (0.25, 1.33)
9
Chlorinated solvents
2.52 (1.23,5.16)
27
Metal and metal compounds
1.47 (0.94, 2.30)
62
Danish Cancer Registry
Mellemgaard et al.
Iron and steel
1.4 (0.8, 2.4) [M]
31
(1994)
1.0 (0.1, 3.2) [F]
1
Solvents
1.5 (0.9, 2.4) [M]
50
6.4 (1.8, 23) [F]
16
This document is a draft for review purposes only and does not constitute Agency policy.
190 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-40. Summary of case-control studies on kidney cancer and
occupation or job title (continued)
Case ascertainment area/exposure group
Relative risk
(95% CI)
No. exposed
cases
Reference
France
Auperin et al.
Machine fitters, assemblers, and precision
instrument makers
0.7 (0.3, 1.9)
16
(1994)
New South Wales, Australia
McCredie and
Iron and steel
1.18 (0.75, 1.85)b
52
Stewart (1993)
2.39 (1.26, 4.52)°
19
Printing or graphics
1.18(0.87, 2.08)b
29
0.82 (0.32, 2.1 l)d
6
Machinist or tool maker
1.15 (0.72, 1.86)b
48
1.83 (0.92, 3.61)°
16
Solvents
1.54 (1.11, 2.14)b
109
1.40 (0.82, 2.40)°
24
Finnish Cancer Registry
Partanen et al.
Iron and metalware work
1.87 (0.94, 3.76)
22
(1991)
Machinists
2.33 (0.83,6.51)
10
Paper and pulp; printing/publishing
2.20 (1.02, 4.72) [M]
18
5.95 (1.21, 29.2) [F]
7
Nonchlorinated solvents
3.46 (0.91, 13.2) [M]
9
West Midlands UK Cancer Registry
Harrington et al.
Organic solvents
(1989)
Ever exposed
1.30 (0.31,8.50)
3
Intermediate exposure
1.54 (0.69, 4.10)
3
Montreal, Canada
Sharpe et al. (1989)
Organic solvents
1.68 (0.83, 2.22)
33
Degreasing solvents
3.42 (0.92, 12.66)
10
Oklahoma
Asal et al. (1988a;
Metal degreasing
1.7 (0.7, 3.8) [M]
19
1988b)
Machining
1.7 (0.7, 4.3) [M]
13
Painter, paint manufacture
1.3 (0.7, 2.6) [M]
22
Missouri Cancer Registry
Brownson (1988)
Machinists
2.2 (0.5, 10.3)
3
Danish Cancer Registry
Jensen etal. (1988)
This document is a draft for review purposes only and does not constitute Agency policy.
191 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Table 4-40. Summary of case-control studies on kidney cancer and
occupation or job title (continued)
Case ascertainment area/exposure group
Relative risk
(95% CI)
No. exposed
cases
Reference
Iron and metal, blacksmith
1.4(0.7, 2.9)d
17
Painter, paint manufacture
1.8 (0.7, 4.6)
10
"Renal pelvis, Wilson et al. (2008).
b Renal cell carcinoma, McCredie and Stewart (1993).
0 Renal pelvis, McCredie and Stewart (1993).
dRenal pelvis and ureter, Jensen et al. (1988).
UK = United Kingdom.
variation in the risk estimate is large and observation of statistically significantly elevated risks
associated with metal work in many of these studies is noteworthy. Some studies also examined
broad chemical grouping such as degreasing solvents or chlorinated solvents. Observations in
studies that assessed degreasing agents or chlorinated solvents reported statistically significant
elevated kidney cancer risk (Asal et al., 1988a; Asal et al., 1988b; Briining et al., 2003;
Harrington et al., 1989; McCredie and Stewart, 1993; Mellemgaard et al., 1994; Pesch et al.,
2000a; Schlehofer et al., 1995). Observations of association with degreasing agents together
with job title or occupations where TCE has been used historically provide a signal and suggest
an etiologic agent common to degreasing activities.
4.4.2.1.2. Cohort and Case-Controls Studies of Trichloroethylene (TCE) Exposure
Cohort and case-controls studies that include job-exposure matrices for assigning TCE
exposure potential to individual study subjects show associations with kidney cancer, specifically
renal cell carcinoma, and trichloroethylene exposure. Support for this conclusion derives from
findings of increased risks in cohort studies (Henschler et al., 1995; Raaschou-Nielsen et al.,
2003; Zhao et al., 2005) and in case-control studies from the Arnsberg region of Germany
(Briining et al., 2003; Pesch et al., 2000a; Vamvakas et al., 1998), the Arve Valley region in
France (Charbotel et al., 2006; Charbotel et al., 2009), the United States (Dosemeci et al., 1999;
Sinks et al., 1992) and the four central and eastern Europe countries of Czech Republic, Poland,
Romania, and Russia (Moore et al., 2010).
A consideration of a study's statistical power and exposure assessment approach is
necessary to interpret observations in Table 4-39. Most cohort studies are underpowered to
This document is a draft for review purposes only and does not constitute Agency policy.
192 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
detect a doubling of kidney cancer risks including the essentially null studies by Greenland et al.
(1994), Axelson et al. (1994 [incidence]), Anttila et al. (1995 [incidence]), Blair et al. (1998
[incidence and mortality]), Morgan et al. (1998), Boice et al. (1999) and Hansen et al. (2001).
Only the exposure duration-response analysis of Raaschou-Nielsen et al. (2003) had over
80% statistical power to detect a doubling of kidney cancer risk (NRC, 2006), and they observed
a statistically significant association between kidney cancer and >5-year employment duration.
Rate ratios estimated in the mortality cohort studies of kidney cancer (e.g., (Axelson et al., 1994;
Blair et al., 1998; Boice et al., 2006b; Garabrant et al., 1988; Greenland et al., 1994; Morgan et
al., 1998; Ritz, 1999a, b; Sinks et al., 1992) are likely underestimated to some extent because
their reliance on death certificates and increased potential of nondifferential misclassification of
outcome in these studies, although the magnitude is difficult to predict (NRC, 2006). Cohort or
proportionate mortality ratio (PMR) studies with more uncertain exposure assessment
approaches, e.g., studies of all subjects working at a factory (ATSDR, 2004; Chang et al., 2003;
Chang et al., 2005; Clapp and Hoffman, 2008; Costa et al., 1989; Garabrant et al., 1988; Sung et
al., 2007), do not show association but are quite limited given their lack of attribution of higher
or lower exposure potentials; risks are likely diluted due to their inclusion of no or low exposed
subjects.
Two studies were carried out in geographic areas with a high frequency and a high degree
of TCE exposure and were designed with a priori hypotheses to test for the effects of TCE
exposure on renal cell cancer risk (Briining et al., 2003; Charbotel et al., 2006; Charbotel et al.,
2009) and a third study carried out in four central and eastern European countries with high renal
cell carcinoma rates unexplained by established risk factors Ferlay et al., 2008 (Moore et al.,
2010). For these reasons, their observations have important bearing to the epidemiologic
evidence evaluation. These studies found a twofold elevated risk with any TCE exposure after
adjustment for several possible confounding factors including smoking (2.47, 95% CI: 1.36,
4.49) for self-assessed exposure to TCE (Briining et al., 2003); any confidence job with high
cumulative TCE exposure, 925 ppm-years (2.16, 95% CI: 1.02, 4.60) with a positive and
statistically significant trend test,/? = 0.04, high confidence jobs with high cumulative TCE
exposure (3.34, 95% CI: 1.27, 8.74) (Charbotel et al., 2006); high confidence assessment of high
TCE cumulative exposure >1.58 ppm-years (2.23, 95% CI: 1.07, 4.64) with a positive and
statistically significant trend test, p = 0.02 (Moore et al., 2010). Furthermore, renal cell
carcinoma risk in Charbotel et al. (2005) increased to over threefold (95% CI: 1.19, 8.38) in
statistical analyses which considered a 10-year exposure lag period. An exposure lag period is
often adopted in analysis of cancer epidemiology to reduce exposure measurement biases
(Salvan et al., 1995). Most exposed cases in this study were exposed to TCE below any current
This document is a draft for review purposes only and does not constitute Agency policy.
193 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
occupational standard (26 of 37 cases [70%]) had held a job with a highest time-weighted
average (TWA [<50 ppm]) (Charbotel et al., 2009). A subsequent analysis of Charbotel et al.
(2009) using an exposure surrogate defined as the highest TWA for any job held, an inferior
surrogate given TCE exposures in other jobs were not considered, reported an almost threefold
elevated risk (2.80, 95% CI: 1.12, 7.03) adjusted for age, sex, body mass index (BMI), and
smoking with exposure to TCE in any job to >50-ppm TWA (Charbotel et al., 2009).
Considering all jobs, Moore et al. (2010) reported a risk of 2.34 (95% CI: 1.05, 5.21) for average
TCE intensity (>0.76 ppm), an exposure metric similar to a TWA exposure category. Zhao et al.
(2005) compared 2,689 TCE-exposed workers at a California aerospace company to nonexposed
workers from the same company as the internal referent population, and found a monotonic
increase in incidence of kidney cancer by increasing cumulative TCE exposure. In addition, a
fivefold increased incidence was associated with high cumulative TCE exposure. This
relationship for high cumulative TCE exposure, lagged 20 years, was accentuated with
adjustment for other occupational exposures (RR = 7.40, 95% CI: 0.47, 116), although the
confidence intervals were increased. An increased confidence interval with adjustments is not
unusual in occupational studies, as exposure is usually highly correlated with them, so that
adjustments often inflate standard error without removing any bias (NRC, 2006). Observed risks
were lower for kidney cancer mortality and because of reliance on cause of death on death
certificates are likely underestimated because of nondifferential misclassification of outcome
(Percy et al., 1981). Boice et al. (2006b), another study of 1,111 workers with potential TCE
exposure at this company and which overlaps with Zhao et al. (2005), found a twofold increase
in kidney cancer mortality (standardized mortality ratio [SMR] = 2.22, 95% CI: 0.89, 4.57). This
study examined mortality in a cohort whose definition date differs slightly from Zhao et al.
(2005), working between 1948-1999 with vital status as of 1999 (Boice et al., 2006b) compared
to working between 1950-1993 with follow-up for mortality as of 2001 (Zhao et al., 2005), and
used a qualitative approach for TCE exposure assessment. Boice et al. (2006b) is a study of
fewer subjects identified with potential TCE exposure, of fewer kidney cancer deaths [7 deaths;
10 incident cases, 10 deaths in Zhao et al. (2005)], of subjects with more recent exposures, and
with a inferior exposure assessment approach compared to Zhao et al. (2005); a finding of a
twofold mortality increase (95% CI: 0.89, 4.57) is noteworthy given the insensitivities.
Zhao et al. (2005), Charbotel et al. (2006) and Moore et al. (2010), furthermore, are three
of the few studies to conduct a detailed assessment of exposure that allowed for the development
of a job-exposure matrix that provided rank-ordered levels of exposure to TCE and other
chemicals. NRC (2006) discussed the inclusion of rank-ordered exposure levels is a strength
increasing precision and accuracy of exposure information compared to more inferior exposure
This document is a draft for review purposes only and does not constitute Agency policy.
194 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
assessment approaches in some other studies such as duration of exposure or a grouping of all
exposed subjects.
The finding in Raaschou-Nielsen et al. (2003) of an elevated renal cell carcinoma risk
with longer employment duration is noteworthy given this study's use of a relatively insensitive
exposure assessment approach. One strength of this study is the presentation of incidence ratios
for a subcohort of higher exposed subjects, those with at least 1-year duration of employment
and first employment before 1980, as a sensitivity analysis for assessing the effect of possible
exposure misclassification bias. Renal cell carcinoma risk was higher in this subcohort
compared to the larger cohort and indicated some potential for misclassification bias in the
grouped analysis. For both the cohort and subcohort analyses, risk appeared to increase with
increasing employment duration, although formal statistical tests for trend are not presented in
the published paper.
4.4.2.1.3. Discussion of controversies on studies in the Arnsberg region of Germany
Two previous studies of workers in this region, a case-control study of Vamvakas et al.
(1998) and Henschler et al. (1995), a study prompted by a kidney cancer case cluster, observed
strong associations between kidney cancer and TCE exposure. A fuller discussion of the studies
from the Arnsberg region and their contribution to the overall weight of evidence on cancer
hazard is warranted in this evaluation given the considerable controversy (Bloemen and
Tomenson, 1995; Cherrie et al., 2001; Green and Lash, 1999; Mandel, 2001; McLaughlin and
Blot, 1997; Swaen, 1995) surrounding Henschler et al. (1995) and Vamvakas et al. (1998).
Criticisms of Henschler et al. (1995) and Vamvakas et al. (1998) relate, in part, to
possible selection biases that would lead to inflating observed associations and limited inferences
of risk to the target population. Specifically, these include (1) the inclusion of kidney cancer
cases first identified from a cluster and the omission of subjects lost to follow-up from Henschler
et al. (1995); (2) use of a Danish population as referent, which may introduce bias due to
differences in coding cause of death and background cancer rate differences (Henschler et al.,
1995); (3) follow-up of some subjects outside the stated follow-up period (Henschler et al.,
1995); (4) differences between hospitals in the identification of cases and controls in Vamvakas
et al. (1998); (5) lack of temporality between case and control interviews (Vamvakas et al.,
1998); (6) lack of blinded interviews (Vamvakas et al., 1998); (7) age differences in Vamvakas
et al. (1998) cases and controls that may lead to a different TCE exposure potential; (8) inherent
deficiencies in Vamvakas et al. (1998) as reflected by its inability to identify other known kidney
cancer risk factors; and, (9) exposure uncertainty, particularly unclear intensity of TCE exposure.
This document is a draft for review purposes only and does not constitute Agency policy.
195 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Overall, NRC (2006) noted that some of the points above may have contributed to an
underestimation of the true exposure distribution of the target population (points 5, 6, and 7),
other points would underestimate risk (points 3), and that these effects could not have explained
the entire excess risk observed in these studies (points 1, 2, and 4). The NRC (2006) furthermore
disagreed with the exposure uncertainty criticism (point 9), and concluded TCE exposures,
although of unknown intensity, were substantial and, clearly showed graded differences on
several scales in Vamvakas et al. (1998) consistent with this study's semiquantitative exposure
assessment.
Briining et al. (2003) was carried out in a broader region in southern Germany, which
included the Arnsberg region and a different set of cases and control identified from a later time
period than Vamvakas et al. (1998). The TCE exposure range in this study was similar to that in
Vamvakas et al. (1998), although at a lower exposure prevalence because of the larger and more
heterogeneous ascertainment area for cases and controls. For "ever exposed" to TCE,
Briining et al. (2003) observed a risk ratio of 2.47 (95% CI: 1.36, 4.49) and a fourfold increase in
risk (95% CI: 1.80, 7.54) among subjects with any occurrence of narcotic symptom and a sixfold
increase in risk (95% CI: 1.46, 23.99) for subjects who had daily occurrences of narcotic
symptoms; risks which are lower than observed in Vamvakas et al. (1998). The lower rate ratio
in Briining et al. (2003) might indicate bias in the Vamvakas et al., study or statistical variation
between studies related to the broader base population included in Briining et al. (2003).
Observational studies such as epidemiologic studies are subject to biases and
confounding which can be minimized but never completely eliminated through a study's design
and statistical analysis methods. While Briining et al. (2003) overcomes many of the
deficiencies of Henschler et al. (1995) and Vamvakas et al. (1998), nonetheless, possible biases
and measurement errors could be introduced through their use of prevalent cases and residual
noncases, use of controls from surgical and geriatric clinics, nonblinding of interviewers, a
2-year difference between cases and controls in median age, use or proxy or next-of-kin
interviews, and self-reported occupational history.
The impact of any one of the above points could either inflate or depress observed
associations. Biases related to a longer period for case compared to control ascertainment could
go in either direction. Next-of-kin interviewers for deceased cases, all controls being alive at the
time of interview, would be expected to underestimate risk if exposures were not fully reported
and thus, misclassified. On the other hand, the control subjects who were enrolled when the
interviews were conducted might not represent the true exposure distribution of the target
population through time and would lead to overestimate of risk. Selection of controls from
clinics is not expected to greatly influence observed associations since these clinics specialized
This document is a draft for review purposes only and does not constitute Agency policy.
196 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
in the type of care they provided (NRC, 2006). Briining et al. (2003) is not the only kidney
case-control study where interviewers were not blinded; in fact, only the study of Charbotel et al.
(2006) included blinding of interviewers. Blinding of interviewers is preferred to reduce
possible bias. Briining et al.'s use of frequency matching using 5-year age groupings is common
in epidemiologic studies and any biases introduced by age difference between cases and controls
is expected to be minimal because the median age difference was 3 years.
Despite these issues, the three studies of the Arnsberg region, with very high apparent
exposure and different base populations showed a significant elevation of risk and all have
bearing on kidney cancer hazard evaluations. The emphasis provided by each study for
identifying a kidney cancer hazard depends on its strengths and weaknesses. Briining et al.
(2003) overcomes many of the deficiencies in Henschler et al. (1995) and Vamvakas et al.
(1998). The finding of a statistically significantly approximately threefold elevated odds ratio
with occupational TCE exposure in Briining et al. (2003) strengthens the signal previously
reported by Henschler et al. (1995) and Vamvakas et al. (1998). A previous study of cardboard
workers in the United States (Sinks et al., 1992), a study like Henschler et al. (1995) which was
prompted by a reported cancer cluster, had observed association with kidney cancer incidence,
particularly with work in the finishing department where TCE use was documented. Henschler
et al. (1995), Vamvakas et al. (1998) and Sinks et al. (1992) are less likely to provide a precise
estimate of the magnitude of the association given greater uncertainty in these studies compared
to Briining et al. (2003). For this reason, Briining et al. (2003) is preferred for meta-analysis
treatment since it is considered to better reflect risk in the target population than the two other
studies. Another study (Charbotel et al., 2006) of similar exposure conditions of a different base
population and of different case and control ascertainment methods as the Arnsberg region
studies has become available since the Arnsberg studies. This study shows a statistically
significant elevation of risk and high cumulative TCE exposure in addition to a positive trend
with rank-order exposure levels. Charbotel et al. (2006) adds evidence to observations from
earlier studies on high TCE exposures in Southern Germany and suggests that peak exposure
may add to risk associated with cumulative TCE exposure.
4.4.2.1.4. Examination of Possible Confounding Factors
Examination of potential confounding factors is an important consideration in the
evaluation of observations in the epidemiologic studies on TCE and kidney cancer. A known
risk factor for kidney cancer is cigarette smoking. Obesity, diabetes, hypertension and
antihypertensive medications, and analgesics are linked to kidney cancer, but causality has not
This document is a draft for review purposes only and does not constitute Agency policy.
197 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
been established (McLaughlin et al., 2006; Moore et al., 2005). On the other hand, fruit and
vegetable consumption is considered protective of kidney cancer risk (McLaughlin et al., 2006).
Studies by Asal et al. (1988a; 1988b), Partanen et al. (1991), McCredie and Stewart (1993),
Auperin et al. (1994), Chow et al. (1994), Mellemgaard et al. (1994), Mandel et al. (1995),
Vamvakas et al. (1998), Dosemeci et al. (1999), Pesch et al. (2000a), Briining et al. (2003), and
Charbotel et al. (2006) controlled for smoking and all studies except Pesch et al. (2000a)
controlled for BMI. Moore et al. (2010) examined but did not find smoking or BMI as potential
counfounders because statistical examination of cigarette smoking and BMI altered risk
estimates for the association between TCE exposure and kidney cancer by less than 10%.
Vamvakas et al. (1998) and Dosemeci et al. (1999) controlled for hypertension and or diuretic
intake in the statistical analysis. Because it is unlikely that exposure to trichloroethylene is
associated with smoking, body mass index, hypertension, or diuretic intake, these possible
confounders do not significantly affect the estimates of risk (NRC, 2006).
Direct examination of possible confounders is less common in cohort studies than in
case-control studies where information is obtained from study subjects or their proxies. Use of
internal controls, such as for Zhao et al. (2005), in general minimizes effects of potential
confounding due to smoking or socioeconomic status since exposed and referent subjects are
drawn from the same target population. Information on possible confounding due to BMI
(obesity) and to diabetes is lacking in cohort studies; however, any uncertainties are likely small
given the generally healthy nature of an employed population and its favorable access to medical
care.
The effect of smoking as a possible confounder may be assessed indirectly through
(1) examination of risk ratios for other smoking-related sites, (2) examination of the expected
contribution by smoking to cancer risks and (3) examination of lung cancer in nine TCE cohort
studies in which there is a high likelihood of TCE exposure in individual study subjects (and
which met, to a sufficient degree, the standards of epidemiologic design and analysis in a
systematic review using meta-analysis methods. Some information on smoking-related lung and
kidney cancer risks may be obtained from IARC (2004b) for indirectly evaluating the expected
magnitude by smoking on kidney cancer risks in TCE cohort studies. Five cohort studies of
cigarette smoking reported risk estimates for both lung and kidney cancers with an observed ratio
of lung:kidney cancer risks of 3.5-10.6 for active smokers, who will have higher
smoking-related risks than former smokers (see Table 4-41). The nNine -cohort studies (Anttila
et al., 1995; Axelson et al., 1994; Boice et al., 1999; Greenland et al., 1994; Hansen et al., 2001;
Morgan et al., 1998; Raaschou-Nielsen et al., 2003; Radican et al., 2008; Zhao et al., 2005)
present lung cancer risks and reported risks for overall TCE exposure range from 0.69 (95% CI:
This document is a draft for review purposes only and does not constitute Agency policy.
198 DRAFT—DO NOT CITE OR QUOTE
-------
1 0.31, 1.30) by Axelson et al. (1994) to 1.4 (95% CI: 1.32, 1.52) by Raaschou-Nielsen et al.
2 (2003) (see Table 4-70). Smoking was more prevalent in the Raaschou-Nielsen et al. (2003)
3 cohort than the background population as suggested by the elevated risks for lung and other
4 smoking-related sites.
This document is a draft for review purposes only and does not constitute Agency policy.
199 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-41. Summary of lung and kidney cancer risks in active smokers
2 (from IARC, 2004b)
3
Cohort
Relative risk
Ratio
lung; kidney
Reference
Lung
Kidney
MRFIT (USA)
1975-1985, men
6.7
1.9
3.5
Kuller et al. (1991)
British Doctor's Study (UK)
1957-1991, men
14.9a
1.4a
10.6
Doll et al. (1994)
U.S. Veterans Study (USA)
1954-1980, men
11.6
1.5
7.7
McLaughlin et al. (1995)
Swedish Census Study (Sweden)
1963-1989, women
4.7
1.1
4.3
Nordlund et al. (1997, 1999)
Cancer Prevention Study II (USA)
1982-1986, women
12.4
1.4
9.1
Garfinkel and Stellman (1988);
Heath etal. (1997)
4
5 a Relative mortality rate compared to nonsmokers.
6
7 UK = United Kingdom.
8
9
10 If smoking fully contributes to the observed 40 % excess lung cancer risk in this study
11 and based on observations in the five smoking cohorts, the expected contribution by smoking to
12 renal cell carcinoma risk is estimated as 1-6 % and far smaller than the 20 and 40% excess in
13 renal cell carcinoma risk in the cohort and subcohort. The use of internal referents who are
14 unexposed subjects drawn from the occupational settings as TCE exposed subjects in
15 three studies will reduce any confounding related to smoking as referents (Morgan et al., 1998;
16 Radican et al., 2008; Zhao et al., 2005). In the other cohort studies lacking direct adjustment for
17 smoking and internal referents, difference in cigarette smoking between exposed and referent
18 subjects is not sufficient to fully explain observed excess kidney cancer risks associated with
19 TCE, particularly, high TCE exposure. Lung cancer risk estimates are lower than or equal to
20 kidney cancer risk estimates and inconsistent with observations in the five smoking cohorts
21 (Axelson et al., 1994; Boice et al., 1999; Hansen et al., 2001).
22 Meta-analysis methods were adopted, additionally, as a tool for examining risk estimates
23 from the nine cohort studies in which there is a high likelihood of TCE exposure in individual
24 study subjects (e.g., based on job-exposure matrices or biomarker monitoring) and which met, to
25 a sufficient degree, the standards of epidemiologic design and analysis in a systematic review
26 reporting lung cancer to assess the presence of potential systematic error related to confounding
27 from smoking. Significant heterogeneity was observed across the nine studies of overall
This document is a draft for review purposes only and does not constitute Agency policy.
200 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
2 2
exposure (/ = 90%) and for six of the nine studies with highest exposure groups (/ = 80%).
Although the appropriateness of conducting any meta-analysis without attempting to explain the
heterogeneity is arguable, the summary estimate from the primary random effects meta-analysis
of the nine studies was 0.96 (95% CI: 0.76, 1.21) for overall TCE exposure, and 0.96 (95% CI:
0.72, 1.27) for the highest group exposure reported by six studies. These observations suggest
potential confounding by smoking of kidney cancer summary risk estimates can be reasonably
excluded in cohort studies of TCE exposure.
Mineral oils such as cutting fluids or hydrazine common to some job titles with potential
TCE exposures (such as machinists, metal workers, and test stand mechanics) were included as
covariates in statistical analyses of Zhao et al. (2005), Boice et al. (2006b) and Charbotel et al.
(2006; 2009) or evaluated as a single exposure for cases and controls in Moore et al. (Karami et
al., 2011; 2010). A TCE effect on kidney cancer incidence was still evident although effect
estimates were often imprecise due to lowered statistical power (Charbotel et al., 2006;
Charbotel et al., 2009; Zhao et al., 2005). Observed associations were similar in analyses
including chemical coexposures in both Zhao et al. (2005) and Charbotel et al. (2006; 2009)
compared to chemical coexposure unadjusted risks. The association or OR between high TCE
score and kidney cancer incidence in Zhao et al. (2005) was 7.71 (95% CI: 0.65, 91.4) after
adjustment for other carcinogens including hydrazine and cutting oils, compared to analyses
unadjusted for chemical coexposures (4.90, 95% CI: 1.23, 19.6).
In Charbotel et al. (2006), exposure to TCE was strongly associated with exposure to
cutting fluids and petroleum oils (22 of the 37 TCE-exposed cases were exposed to both).
Statistical modeling of all factors significant at 10% threshold showed the OR for cutting fluids
to be almost equal to one, whereas the OR for the highest level of TCE exposure was close to
two (Charbotel et al., 2006). Moreover, when exposure to cutting oils was divided into
three levels, a decrease in OR with level of exposure was found. In conditional logistic
regression adjusted for cutting oil exposure, the relative risk (OR) for renal cell carcinoma and
TCE was similar to relative risks unadjusted for cutting fluid exposures (high cumulative TCE
exposure: 1.96 [95% CI: 0.71-5.37] compared to 2.16 [95% CI: 1.02-4.60]; high cumulative and
peak: 2.63 [95% CI: 0.79-8.83] compared to 2.73 [95% CI: 1.06-7.07] (Charbotel et al., 2006).
Charbotel et al. (2009) further examined TCE exposure defined as the highest TWA in any job
held, inferior to cumulative exposure given its lack of consideration of TCE exposure potential in
other jobs, either as exposure to TCE alone, cutting fluids alone, or to both after adjusting for
smoking, body mass index, age, sex, and exposure to other oils (TCE alone: 1.62 [95% CI: 0.75,
3.44]); cutting fluids alone: 2.39 (95% CI: 0.52, 11.03); TCE >50-ppm TWA + cutting fluids:
2.70 (95%) CI: 1.02, 7.17). There were few cases exposed to cutting fluids alone (n = 3) or to
This document is a draft for review purposes only and does not constitute Agency policy.
201 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
TCE alone (n = 15), all of whom had TCE exposure (in the highest exposed job held) of
<35-ppm TWA, and the subgroup analyses were of limited statistical power. A finding of higher
risk for both cutting oil and TCE exposure >50 ppm compared to cutting oil alone supports a
TCE effect for kidney cancer. Adjustment for cutting oil exposures, furthermore, did not greatly
affect the magnitude of TCE effect measures in the many analyses presented by Charbotel et al.
(2006; 2009) suggesting cutting fluid exposure as not greatly confounding TCE effect measures.
Two other kidney case-control studies of TCE exposure examined the effect of cutting oil as a
single occupational exposure on kidney cancer risk (Briining et al., 2003; Karami et al., 2011).
Although Briining et al. (2003) reported an odds ratio of 2.11 (95% CI: 0.66, 6.70) for
self-reported cutting oil exposure and kidney cancer, cutting oil exposure did not appear highly
correlated with TCE exposure as only 5 cases reported exposure to cutting oils compared to 25
cases reporting TCE exposure,. Karami et al. (2011), who examined mineral oil or cutting fluid
exposure among cases and controls in Moore et al. (2010), reported an odds ratio of 0.8 (95% CI:
0.6, 1,1) and 1.1 (95% CI: 0.8, 1.4), for cutting oil mists or other mineral oil mists respectively,
and provides evidence that the reported association with TCE exposure in Moore et al. (2010) is
not likely confounded by cutting or mineral oil exposures. Moreover, cutting oils and mineral
oils have not been associated with kidney cancer in other cohort or case-control studies (Mirer,
2010; NIOSH, 1998), and provides additional support for potential confounding by cutting oils
as of minimal concern.
Boice et al. (2006b) was unable to directly examine hydrazine exposure on TCE effect
measures because of a lack of model convergence in statistical analyses. Three of
7 TCE-exposed kidney cancer cases were identified with hydrazine exposure of 1.5 years or less
and the absence of exposure to the other four cases suggested confounding related to hydrazine
was unlikely to greatly modify observed association between TCE and kidney cancer.
4.4.2.1.5. Susceptible Populations—Kidney Cancer and Trichloroethylene (TCE) Exposure
Two studies of kidney cancer cases from the Arnsberg region in Germany and the study
of kidney cancer cases from three Central and Eastern European countries have examined the
influence of polymorphisms of the glutathione-S-transferase metabolic pathway on renal cell
carcinoma risk and TCE exposure (Briining et al., 1997a; Moore et al., 2010; Wiesenhiitter et al.,
2007). In their study of 45 TCE-exposed male and female renal cell carcinoma cases pending
legal compensation and 48 unmatched male TCE-exposed controls, Briining et al. (1997a)
observed a higher prevalence of exposed cases homozygous and heterozygous for GSTM1
positive, 60%), than the prevalence for this genotype among exposed controls, 35%>. The
This document is a draft for review purposes only and does not constitute Agency policy.
202 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
frequency of GSTM1 positive was lower among this control series than the frequency found in
other European population studies, 50% (Briining et al., 1997a). The prevalence of the GSTT1
positive genotype was 93% among exposed cases and 77% among exposed controls. The
prevalence of GSTT1 positive genotype in the European population is 75% (Briining et al.,
1997a).
Wiesenhiitter et al. (2007) compares the frequency of genetic polymorphism among
subjects from the renal cancer case-control study of Briining et al. (2003) and to the frequencies
of genetic polymorphisms in the areas of Dormund and Lutherstadt Wittenberg, Germany.
Wiesenhiitter et al. (2007) identified the genetic frequencies of GSTM1 and GSTT1 phenotypes
for 98 of the original 134 cases (73%) and 324 of the 401 controls (81%). The prevalence of
GSTM1 positive genotype was 48% among all renal cell carcinoma cases, 40% among
TCE-exposed cases, and 52% among all controls. The prevalence of GSTT1 positive genotypes
was 8P/o among all cases and 81% among all controls. The prevalence of GSTT1 positive
genotypes reported in this paper for all TCE-exposed cases was 20%. Wiesenhiitter et al. (2007)
noted background frequencies in the German population in the expanded control group were
50%) for GSTM1 positive and 81% for GSTT1 positive genotypes. The observations are limited
as the paper is sparsely reported and numbers of exposed (n = 4) and unexposed (n= 15) GSTT1
positive cases does not sum to the 79 cases with the GSTT1 positive genotype identified in the
table's first row.
Moore et al. (2010) presents associations between TCE exposure and renal cell
carcinoma risk stratified by GSTT genotype and for single nucleotide polymorphisms (SNPs) of
the renal cysteine conjugate P-lyase gene. Genotyping was available for 925 of the 1,097 cases
and 1,192 of the 1,476 controls. The percentage of cases and controls genotyped did not
significantly differ among TCE-exposed and unexposed subjects nor was the active GSTT1
genotype association with kidney cancer risk (0.94, 95% CI: 0.75, 1.19). However, adopting
statistical analysis examining TCE exposure and kidney cancer that stratified on GSTT1
polymorphism as null (deleted allele) or active (>1 intact allele), Moore et al. (2010) reported
significant associations for GSTT1 active genotype and no association suggested for subjects
with GSTT1 null genotype. The risk estimate for the association for TCE exposure and kidney
cancer among subjects with an active GSTT1 genotype, was 1.88 (95% CI: 1.06, 3.33), with
higher risk estimates for long exposure duration, cumulative exposure and average exposure
intensity [>13.5 years, 2.13 (95% CI: 1.04, 4.39); >1.58 ppm-years, 2.59 (95% CI: 1.25, 5.35);
>0.076 ppm, 2.77 (95% CI: 1.01, 7.58)] and a positive trend with increasing exposure duration,
cumulative exposure or average intensity categories (p < 0.03) (Moore et al., 2010). The
associations between TCE exposure and kidney cancer was stronger for subjects with a
This document is a draft for review purposes only and does not constitute Agency policy.
203 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
functionally active GSTT1 than those for all subjects (both genotypes combined) (see
Table 4-39). Moore et al. (2010) tested but did not find statistical interaction between GSTT1
genotype and TCE exposure (p > 0.17). Moore et al. (2010) also examined the effect of
polymorphisms of the cysteine conjugate P-lyase gene on TCE risk and reported interaction
between TCE exposure and four minor alleles (SNPs rs2293968, rs2280841, rs2259043, and
rs941960) (p < 0.05). Associations with TCE exposure and kidney cancer were threefold higher
compared to unexposed subjects with these SNPs.
Observations in Briining et al. (1997a) and Wiesenhiitter et al. (2007) must be interpreted
cautiously. Few details are provided in these studies on selection criteria and not all subjects
from the Briining et al. (2003) case-control study are included. For GSTM1 positive, the higher
prevalence among exposed cases in Briining et al. (1997a) compared Wiesenhiitter et al. (2007)
and the lower prevalence among controls compared to background frequency in the European
population may reflect possible selection biases. On the other hand, the broader base population
included in Briining et al. (2003) may explain the observed lower frequency of GSTM1 positive
cases in Wiesenhiitter et al. (2007). Moreover, Wiesenhiitter et al. (2007) does not report
genotype frequencies for controls by exposure status and this information is essential to an
examination of whether renal cell carcinoma risk and TCE exposure may be modified by
polymorphism status. The statistical analyses in both studies was a simple comparison of
exposure prevalence between cases and controls and did not include analyses that stratified on
exposure status. An examination of exposure prevalence is limited as Moore et al. (2010), too,
reported TCE exposure prevalence as similar between exposed cases and controls. Associations
between TCE exposure and kidney cancer for GSTT1 active genotype, however, were reported
in stratified analyses. The more rigorous study design and statistical methods in Moore et al.
(2010) affords more weight to their reported observations than for Briining et al. (1997a) and
Wiesenhiitter et al. (2007). Moore et al. (2010) provides evidence of greater susceptibility to
TCE exposure and kidney cancer among subjects with a functionally active GSTT
polymorphism, particularly among those with certain alleles in single nucleotide polymorphisms
of the cysteine conjugation P-lyase gene region.
Of the three larger (in terms of number of cases) studies that did provide results
separately by sex, Dosemeci et al. (1999) suggest that there may be a sex difference for TCE
exposure and renal cell carcinoma (OR: 1.04, [95% CI: 0.6, 1.7]) in males and 1.96 (95% CI:
1.0, 4.0 in females), while Raaschou-Nielsen et al. (2003) report the same standardized incidence
ratio (SIR = 1.2) for both sexes and crude ORs calculated from data from the Pesch et al. (2000a)
study (provided in a personal communication from Beate Pesch, Forschungsinstitut fiir
Arbeitsmedizin, to Cheryl Scott, EPA, 21 February 2008) are 1.28 for males and 1.23 for
This document is a draft for review purposes only and does not constitute Agency policy.
204 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
females. Whether the Dosemeci et al. (1999) observations are due to susceptibility differences or
to exposure differences between males and females cannot be evaluated. Blair et al. (1998) and
Hansen et al. (2001) also present some results by sex, but these two studies have too few cases to
be informative about a sex difference for kidney cancer.
4.4.2.1.6. Meta-Analysis for Kidney Cancer
Meta-analysis (detailed methodology in Appendix C) was adopted as a tool for
examining the body of epidemiologic evidence on kidney cancer and TCE exposure and to
identify possible sources of heterogeneity. The meta-analyses of the overall effect of TCE
exposure on kidney cancer suggest a small, statistically significant increase in risk that was
stronger in a meta-analysis of the highest exposure group. There was no observable
heterogeneity for any of the meta-analyses of the 15 studies and no indication of publication bias.
Thus, these findings of increased risks of kidney cancer associated with TCE exposure are
robust.
The meta-analysis of kidney cancer examines 15 cohort and case-control studies
identified through a systematic review and evaluation of the epidemiologic literature on TCE
exposure (Anttila et al., 1995; Axelson et al., 1994; Blair et al., 1998; Boice et al., 1999; Briining
et al., 2003; Charbotel et al., 2006; Dosemeci et al., 1999; Greenland et al., 1994; Hansen et al.,
2001; Moore et al., 2010; Morgan et al., 1998; Pesch et al., 2000a; Raaschou-Nielsen et al.,
2003; Siemiatycki, 1991; Zhao et al., 2005). Details of the systematic review and meta-analysis
of the TCE studies are fully discussed in Appendix B and C.
The summary relative risk (RRm) estimate from the primary random effects
meta-analysis of the 15 studies was 1.27(95% CI: 1.13, 1.43). The analysis was dominated by
two (contributing almost 70% of the weight) or three (almost 80% of the weight) large studies
(Dosemeci et al., 1999; Pesch et al., 2000a; Raaschou-Nielsen et al., 2003). Figure 4-1 arrays
individual studies by their weight. No single study was overly influential; removal of individual
studies resulted in RRm estimates that were all statistically significant (p < 0.005) and that
ranged from 1.24 (with the removal of Briining et al. (2003)) to 1.30 (with the removal of
Raaschou-Nielsen et al. (2003)). Similarly, the overall RRm estimate was not highly sensitive to
alternate RR estimate selections nor was publication bias apparent. There was no apparent
heterogeneity across the 15 studies, i.e., the random effects model and the fixed effect model
gave the same results (phetero = 0,67; / = 0%). Nonetheless, subgroup analyses were done
examining the cohort and case-control studies separately with the random effects model; the
resulting RRm estimates were 1.16 (95% CI: 0.96, 1.40) for the cohort studies and 1.48 (1.15,
This document is a draft for review purposes only and does not constitute Agency policy.
205 DRAFT—DO NOT CITE OR QUOTE
-------
1 1.91) for the case-control studies. There was no heterogeneity in the cohort subgroup (p = 0.998;
2 / = 0%). There was heterogeneity in the case-control subgroup, but it was not statistically
This document is a draft for review purposes only and does not constitute Agency policy.
206 DRAFT—DO NOT CITE OR QUOTE
-------
L/l
K>
s?
>!
§•
cs
s;
>1
£
J
° ^
-J §
>!
o
a
a
a
§•
>!
5
o
o
fe
H
81
w
2; ^
o^
H J
1—1 s
H 2.
W|
O •
o
c
o
H
W
Study
Anttila (1Q95)
Avelson 11904)
Beiee (1999)
Greenland (1094)
Hansen (2001)
Morgan (1068)
Raaschou-Nielsen (2003)
Radican (2008)
Zhao (2005)
B runing (2003)
Charbotel (2008)
Dosemeci (199B)
Moore (2010)
Pesch (2000)
Siemiat/cki (1991)
OVERALL
0.1
TCE Exposuie anil Kidney Cancel
Relative Risk and 95% CI
RR
LCL
UCL
0.87
0.32
1.89
1.18
0.42
2 52
0.99
0.40
2.04
0.09
0.30
3.32
1.10
0.30
2.80
1.14
0.51
2.58
1.20
0.94
1.50
1.18
0.47
2.94
1.70
0.38
7.90
2.47
1.36
4,49
1.88
0.80
3.98
1.30
0.00
1.90
2.05
1.13
3.73
1.24
1.00
1.50
0.30
0.30
2.20
1.27
1.13
1.43
10
Figure 4-1. Meta-analysis of kidney cancer and overall TCE exposure (the summary estimate is in the bottom
row, represented by the diamond). Random effects model; fixed effect model same. Symbol sizes reflect relative
weights of the studies.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
2
significant (p = 0.14) and the / value of 41% suggests that the extent of the heterogeneity in this
subgroup was low-to-moderate.
Ten studies reported risks for higher exposure groups (Blair et al., 1998; Boice et al.,
1999; Briining et al., 2003; Charbotel et al., 2006; Dosemeci et al., 1999; Moore et al., 2010;
Morgan et al., 1998; Parent et al., 2000a; Pesch et al., 2000a; Raaschou-Nielsen et al., 2003;
Siemiatycki, 1991; Zhao et al., 2005). Different exposure metrics were used in the various
studies, and the purpose of combining results across the different highest exposure groups was
not to estimate an RRm associated with some level of exposure. Instead, the focus on the highest
exposure category was meant to result in an estimate less affected by exposure misclassification.
In other words, it is more likely to represent a greater differential TCE exposure compared to
people in the referent group than the exposure differential for the overall (typically any vs. none)
exposure comparison. Thus, if TCE exposure increases the risk of kidney cancer, the effects
should be more apparent in the highest exposure groups.
The RRm estimate from the random effects meta-analysis of the studies with results
presented for higher exposure groups was 1.64 (95% CI: 1.31, 2.04), higher than the RRm from
the overall kidney cancer meta-analysis. As with the overall analyses, the meta-analyses of the
highest-exposure groups were dominated by Pesch et al. (2000a) and Raaschou-Nielsen et al.
(2003), which provided about 60% of the weight. Axelson et al. (1994), Anttila et al. (1995) and
Hansen et al. (2001) do not report risk ratios for kidney cancer by higher exposure and a
sensitivity analysis was carried out to address reporting bias. The RRm estimate from the
primary random effects meta-analysis with null RR estimates (i.e., RR = 1.0) included for
Axelson et al. (1994), Anttila et al. (1995) and Hansen et al. (2001) to address reporting bias
associated with ever exposed was 1.58 (95% CI: 1.28, 1.96). Figure 4-2 arrays individual studies
by their weight. The inclusion of these three additional studies contributed less than 7% of the
total weight. No single study was overly influential; removal of individual studies resulted in
RRm estimates that were all statistically significant (p < 0.005) and that ranged from 1.52 (with
the removal of Raaschou-Nielsen et al. (2003)) to 1.64 (with the removal of Pesch et al. (2000a)).
Similarly, the RRm estimate was not highly sensitive to alternate RR estimate selections (all with
p < 0.0005) and other than a negligible amount of heterogeneity observed in the sensitivity
analysis with the Pesch job-exposure matrix (JEM) alternate (/ = 0.64%), there was no
observable heterogeneity across the studies for any of the meta-analyses conducted with the
highest-exposure groups, including those in which RR values for Anttila, Axelson, and Hansen
were assumed (/ = 0%), For Pesch, the job-task exposure matrix (JTEM) approach is preferred
because it seemed to be a more comprehensive and discriminating approach, taking actual job
This document is a draft for review purposes only and does not constitute Agency policy.
208 DRAFT—DO NOT CITE OR QUOTE
-------
Study
Boice !;19S0i
Morgan (1 £'£'8)
Raaschou-Nielsen (2003)
Radican (2008)
Zhao (2005)
Burning 1:2003)
Chart)otel (2006)
Moore (2010)
P esc h (2000)
Siemiatycki (19£'1)
Anttila f19&5)
A*elson (1994)
Hansen (2001)
OVERALL
r
TCE Exposwe
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
tasks into account, rather than just larger job categories. No subgroup analyses (e.g., cohort vs.
case-control studies) were done with the highest exposure group results.
NRC (2006) deliberations on trichloroethylene commented on two prominent evaluations
of the then-current TCE epidemiologic literature using meta-analysis techniques, Wartenberg
et al. (2000) and Kelsh et al. (2005), submitted by Exponent-Health Sciences to NRC during
their deliberations and who updated their analysis by including subsequently published studies of
Boice et al. (2006b) and Charbotel et al. (2006), but not Radican et al. (2008), and presented
summary relative risk estimates for cohort and case-control studies, separately, and combined
(Kelsh et al., 2010). Wartenberg et al. (2000) reported an RRm of 1.7 (95% CI: 1.1, 2.7) for
kidney cancer incidence in the TCE subcohorts (Anttila et al., 1995; Axelson et al., 1994; Blair et
al., 1998; Henschler et al., 1995). For kidney cancer mortality in TCE subcohorts (Blair et al.,
1998; Boice et al., 1999; Henschler et al., 1995; Morgan et al., 1998; Ritz, 1999a), Wartenberg
et al. (2000) reported an RRm of 1.2 (95% CI: 0.8, 1.7). Kelsh et al. (2010) examined a slightly
different grouping of cohort studies as did Wartenberg et al. (2000), presenting a summary
relative risk estimate for kidney cancer incidence and mortality combined. The RRm for kidney
cancer in Group I cohort studies (Anttila et al., 1995; Axelson et al., 1994; Blair et al., 1998;
Boice et al., 1999; Hansen et al., 2001; Morgan et al., 1998; Raaschou-Nielsen et al., 2003) was
1.34 (95%) CI: 1.07-1.67) with no evidence of heterogeneity and, in Group II cohort studies,
studies lacking documented TCE exposure (Blair et al., 1989; Chang et al., 2003; Costa et al.,
1989; Garabrant et al., 1988; Henschler et al., 1995; Selden and Ahlborg, 1991; Sinks et al.,
1992), was 1.58 (95%> CI: 0.75, 3.32) with evidence of heterogeneity. Removing both Henschler
et al. (1995) and Sinks et al. (1992), considered by Kelsh et al. (2010) as outliers, eliminated
observed heterogeneity and the summary risk estimate was 0.88 (95%> CI: 0.8, 1.33). Kelsh et al.
(2010), also, presented separately a summary relative risk for renal cancer case-control studies
and TCE. For case-control studies Charbotel et al, 2006 (Briining et al., 2003; Dosemeci et al.,
1999; Greenland et al., 1994; Pesch et al., 2000a; Siemiatycki, 1991; Vamvakas et al., 1998), the
RRm for renal cell carcinoma was 1.57 (95%> CI: 1.06, 2.30) with evidence of heterogeneity, and
RRm of 1.33 (95%> CI: 1.02, 1.73) and no evidence of heterogeneity in a sensitivity analysis
removing Vamvakas et al. (1998), a study Kelsh et al. (2010) considered as an outlier. Last,
Kelsh et al. (2010) presented three summary relative risk estimates for renal cell cancer Groups I
and II cohort and case-control studies combined: 1.30 (95%> CI: 1.04, 1.61) with evidence of
heterogeneity and included 23 studies with kidney cancer risk estimates for all subjects, those
with documented TCE exposure and those unexposed to TCE, and Ritz (1999a) in Group I
studies; 1.42 (95%> CI 1.13, 1.77) with evidence of heterogeneity and included 23 studies, with
TCE subcohort kidney cancer risk estimates replacing the total cohort estimate for Group I
This document is a draft for review purposes only and does not constitute Agency policy.
210 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
studies; and, 1.24 (95% CI; 1.06, 1.45) with no evidence of heterogeneity and included
20 studies, counting TCE subcohort kidney cancer risk estimates in Group I studies and
removing the three studies Kelsh et al. (2010) considered as outliers.
The present analysis was conducted according to NRC (2006) suggestions for
transparency, systematic review criteria, and examination of both cohort and case-control
studies. EPA's meta-analysis has several advantages to previous ones of TCE exposure and
cancer. The selection criteria adopted in this meta-analysis were intended to identify informative
studies for the evaluation of TCE exposure and cancer, studies with reduced systematic errors.
Neither Henschler et al. (1995) nor Vamvakas et al. (1998), two studies with incomplete cohort
identification or potential selection bias of study controls, met our inclusion criteria and their
inclusion in other meta-analysis may have contributed to the observed heterogeneity in kidney
cancer RRm (Kelsh et al., 2010). Studies with background or low TCE exposure potential also
did not meet another selection criterion as our analysis focused on TCE exposure potential
inferred to each subject by reference to industrial hygiene records, individual biomarkers,
job-exposure matrices, water distribution models, or questionnaire responses that likely had
fewer biases associated with exposure misclassification, although this bias would not have been
completely minimized. Inclusion of studies of lower exposure potential in meta-analyses can
have important implications for identifying a cancer hazard (Steinmaus et al., 2008; Zhang et al.,
2008; Vlaanderen et al., 2011). The present analysis includes the recently published studies of
Charbotel et al. (2006), Moore et al. (2010), and updated mortality of the Blair et al. (1998)
cohort by Radican et al. (2008). As discussed above, the summary estimate from the primary
random effects meta-analysis of the 15 studies was 1.27 (95% CI: 1.13, 1.43). Additionally,
EPA examined kidney cancer risk for higher exposure group. The RRm estimate from the
random effects meta-analysis of the studies with results presented for higher exposure groups
was 1.64 (95% CI: 1.31, 2.04), higher than the RRm from the overall kidney cancer
meta-analysis, and 1.58 (95% CI: 1.28, 1.96) in the meta-analysis with null RR estimates (i.e.,
RR = 1.0) to address possible reporting bias for three studies.
4.4.3. Human Studies of Somatic Mutation of von Hippel-Lindau (VHL) Gene
Studies have been conducted to identify mutations in the VHL gene in renal cell
carcinoma patients, with and without TCE exposures (Brauch et al., 1999; Charbotel et al., 2007;
Furge et al., 2007; Kenck et al., 1996; Schraml et al., 1999; Toma et al., 2008; Wells et al.,
2009). Inactivation of the VHL gene through mutations, LOH and imprinting has been observed
in about 70% of sporadic renal clear cell carcinomas, the most common renal cell carcinoma
This document is a draft for review purposes only and does not constitute Agency policy.
211 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
subtype (Kenck et al., 1996). Other genes or pathways, including c-myc activation and VEGF,
have also been examined as to their role in various renal cell carcinoma subtypes (Furge et al.,
2007; Toma et al., 2008). Furge et al. (2007) reported that there are molecularly distinct forms
of RCC and possibly molecular differences between clear-cell renal cell carcinoma subtypes.
This study was performed using tissues obtained from paraffin blocks. These results are
supported by a more recent study which examined the genetic abnormalities of clear cell renal
cell carcinoma using frozen tissues from 22 cc-RCC patients and paired normal tissues (Toma et
al., 2008). This study found that 20 (91%) of the 22 cases had LOH on chromosome 3p
(harboring the VHL gene). Alterations in copy number were also found on chromosome 9 (32%
of cases), chromosome arm 14q (36% of cases), chromosome arm 5q (45% of cases) and
chromosome 7 (32% of cases), suggesting roles for multiple genetic changes in RCC, and is also
supported by genomes-wide single-nucleotide polymorphism analysis (Toma et al., 2008).
Several papers link mutation of the VHL gene in renal cell carcinoma patients to TCE
exposure. These reports are based on comparisons of VHL mutation frequencies in TCE exposed
cases from renal cell carcinoma case-control studies or from comparison to background mutation
rates among renal cell carcinoma case series (see Table 4-42). Briining et al. (1997b) first
reported a high somatic mutation frequency (100%) in a series of 23 renal cell carcinomas cases
with medium to high intensity TCE exposure as determined by an abnormal single stand
conformation polymorphism (SSCP) pattern, with most variations found in exon two. Only four
samples were sequenced at the time of publication and showed mutations in exon one, two and
three (see Table 4-42). Some of the cases in this study were from the case-control study of
Vamvakas et al. (1998) (see Section 4.4.3 and Appendix C).
Brauch et al. (1999; 2004) analyzed renal cancer cell tissues for mutations of the VHL
gene and reported increased occurrence of mutations in patients exposed to high concentrations
of TCE. In the first study (Brauch et al., 1999), an employer's liability or worker's
compensation registry was used to identify 44 renal cell carcinoma cases, 18 of whom were also
included in Briining et al. (1997b). Brauch et al. (1999) found multiple mutations in 42% of the
exposed patients who experienced any mutation and 57% showed loss of heterozygosity. A hot
spot mutation of cytosine to thymine at nucleotide 454 (C454T) was found in 39% of samples
that had a VHL mutation and was not found in renal cell cancers from nonexposed patients or in
lymphocyte DNA from either exposed or nonexposed cases or controls. As discussed above,
little information was given on how subjects were selected and whether there was blinding of
from the renal cell carcinoma case-control study of Vamvakas et al. (1998). Brauch et al. (2004)
compared age at diagnosis and histopathologic parameters of tumors as well as somatic mutation
This document is a draft for review purposes only and does not constitute Agency policy.
212 DRAFT—DO NOT CITE OR QUOTE
-------
1 characteristics in the VHL tumor suppressor gene between the TCE-exposed and non-TCE
2 exposed renal cell carcinoma patient groups (TCE-exposed from their previous 1999 publication
This document is a draft for review purposes only and does not constitute Agency policy.
213 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-42. Summary of human studies on somatic mutations of the VHL genea
TCE exposure
status
Briining et al.
(1997b)
Brauch et al. (1999)
Schraml et al. (1999)
Brauch et al. (2004)
Charbotel et al. (2007)
Exposed
Exposed
Unexposed
Exposed
Unexposed
Exposed
Unexposed
Exposed
Unexposed
Number of
subjects/
Number with
mutations (%)
23/23 (100%)
44/33 (75%)
73/42 (58%)
9/3 (33%)
113/38
(34%)
17/14 (82%)
21/2 (10%)
25/2 (9%)
23/2 (8%)
Renal cell
carcinoma
subtype
Unknown
Unknown
Clear cell 9 (75%)
Papillary 2 (18%)
Oncocytomas 1 (8%)
Unknown
Clear cell 37 (%)
Oncocytic adenoma 1 (%)
Bilateral metachronous
1 (%)
Clear cell 51 (75%)
Papillary 10 (10-15%)
Chromophobe 4 (5%)
Oncocytomas 4 (5%)
Tissue type
analyzed
Paraffin
Paraffin, fresh
(lymphocyte)
Paraffin
Paraffin
Paraffin, frozen tissues,
Bouin's fixative
Assay
SSCP,b sequencing15
SSCP, sequencing,
restriction enzyme
digestion
CGH, sequencing
Sequencing
Sequencing
Number of
mutations
23
50
42
4
50
24
2
2
2
Type of mutation
Missense
Nonmissense0
1
3
27
23
NA
NA
1
3
Unknown
Unknown
17
7
2
0
1
1
1
1
a Adapted from NRC (2006) with addition of Schraml et al. (1999) and Charbotel et al. (2007).
bBy SSCP. Four (4) sequences confirmed by comparative genomic hybridization.
0 Includes insertions, frameshifts, and deletions.
CGH = comparative genomic hybridization.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
to the non TCE-exposed cases newly sequenced in this study). Renal cell carcinoma did not
differ with respect to histopathologic characteristics in either patient group. Comparing results
from TCE-exposed and nonexposed patients revealed clear differences with respect to
(1) frequency of somatic VHL mutations, (2) incidence of C454T transition, and (3) incidence of
multiple mutations. The C454T hot spot mutation at codon 81 was exclusively detected in
tumors from TCE-exposed patients, as were multiple mutations. Also, the incidence of VHL
mutations in the TCE-exposed group was at least twofold higher than in the nonexposed group.
Overall, these finding support the view that the effect of TCE is not limited to clonal expansion
of cells mutated spontaneously or by some other agent.
Brauch et al. (2004) were not able to analyze all RCCs from the Vamvakas study
(Vamvakas et al., 1998), in part because samples were no longer available. Using the data
described by Brauch et al. (2004) (VHL mutation found in 15 exposed and 2 nonexposed
individuals, and VHL mutation not found in 2 exposed and 19 unexposed individuals), the
calculated OR is 71.3. The lower bound of the OR including the excluded RCCs is derived from
the assumption that all 20 cases that were excluded were exposed but did not have mutations in
VHL (VHL mutations were found in 15 exposed and 2 unexposed individuals and VHL was not
found in 22 exposed and 18 unexposed individuals), leading to an OR of 6.5 that remains
statistically significant.
Charbotel et al. (2007) examines somatic mutations in the three VHL coding exons in
RCC cases from their case-control study (Charbotel et al., 2006). Of the 87 RCCs in the
case-control study, tissue specimens were available for 69 cases (79%) of which 48 were
cc-RCC. VHL sequencing was carried out for only the cc-RCC cases, 66% of the 73 cc-RCC
cases in Charbotel et al. (2006). Of the 48 cc-RCC cases available for VHL sequencing,
15 subjects were identified with TCE exposure (31%), an exposure prevalence lower than 43%
observed in the case-control study. Partial to full sequencing of the VHL gene was carried out
using polymerase chain reaction (PCR) amplification and VHL mutation pattern recognition
software of Beroud et al. (1998). Full sequencing of the VHL gene was possible for only
26 RCC cases (36% of all RCC cases). Single point mutations were identified in four cases
(8%> prevalence): two unexposed cases, a G > C mutation in exon 2 splice site and a G > A in
ex on 1; one case identified with low/medium exposure, T > C mutation in exon 2, and, one case
identified with high TCE exposure, T > C in exon 3. It should be noted that the two cases with
T > C mutations were smokers unlike the cases with G > A or G > C mutations. The prevalence
of somatic VHL mutation in this study is quite low compared to that observed in other RCC case
series from this region; around 50% (Bailly et al., 1995; Gallou et al., 2001). To address possible
bias from misclassification of TCE exposure, Charbotel et al. (2006) examined renal cancer risk
This document is a draft for review purposes only and does not constitute Agency policy.
215 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
for jobs associated with a high level of confidence for TCE exposure. As would be expected if
bias was a result of misclassification, they observed a stronger association between higher
confidence TCE exposure and RCC, suggesting that some degree of misclassification bias is
associated with their broader exposure assessment approach. Charbotel et al. (2007) do not
present findings on VHL mutations for those subjects with higher level of confidence TCE
exposure assignment.
Schraml et al. (1999) did not observe statistically significant differences in DNA
sequence or mutation type in a series of 12 renal cell carcinomas from subjects exposed to
solvents including varying TCE intensity and a parallel series of 113 clear cell carcinomas from
non-TCE exposed patients. Only nine of the RCC were cc-RCC and were sequenced for
mutations. VHL mutations were observed in clear cell tumors only; four mutations in three
TCE-exposed subjects compared to 50 mutations in tumors of 38 nonexposed cases. Details as
to exposure conditions are limited to a statement that subjects had been exposed to high doses of
solvents, potential for mixed solvent exposures, and that exposure included a range of TCE
concentrations. Limitations of this study include having a wider range of TCE exposure
intensities as compared to the studies described above (Brauch et al., 1999; Briining et al.,
1997b), which focused on patients exposed to higher levels of TCE, and the limited number of
TCE-exposed subjects analyzed, being the smallest of all available studies on RCC, TCE and
VHL mutation. For these reasons, Schraml et al. (1999) is quite limited for examining the
question of VHL mutations and TCE exposure.
Szymanska et al. (2010) examines somatic mutations in three VHL coding exons in 359
RCC cases, 334 of whom with clear-cell carcinomas, from the case-control study of Moore et al.
(2010) as part of a pilot examination of mutation in three other genes, TP53, EGFR, and KRAS.
The prevalence of VHL mutations was high in the RCC series, 72% of the tumors carried at least
one function mutation, Although occupational exposures were not examined and data was not
presented, Szymanska et al. (2010) reported that VHL mutations were not associated with TCE
exposure.
A number of additional methodological issues need to be considered in interpreting these
studies. Isolation of DNA for mutation detection has been performed using various tissue
preparations, including frozen tissues, formalin fixed tissues and tissue sections fixed in Bouin's
solution. Ideally, studies would be performed using fresh or freshly frozen tissue samples to
limit technical issues with the DNA extraction. When derived from other sources, the quality
and quantity of the DNA isolated can vary, as the formic acid contained in the formalin solution,
fixation time and period of storage of the tissue blocks often affect the quality of DNA. Picric
acid contained in Bouin's solution is also known to degrade nucleic acids resulting in either low
This document is a draft for review purposes only and does not constitute Agency policy.
216 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
yield or poor quality of DNA. In addition, during collection of tumor tissues, contamination of
neighboring normal tissue can easily occur if proper care is not exercised. This could lead to the
'dilution effect' of the results—i.e., because of the presence of some normal tissue, frequency of
mutations detected in the tumor tissue can be lower than expected. These technical difficulties
are discussed in these papers, and should be considered when interpreting the results.
Additionally, selection bias is possible given tissue specimens were not available for all RCC
cases in Vamvakas et al. (1998) or in Charbotel et al. (2006). Some uncertainty associated with
misclassification bias is possible given the lack of TCE exposure information to individual
subjects in Schraml et al. (1999) and in Charbotel et al. (2007) from their use of broader
exposure assessment approach compared to that associated with the higher confident exposure
assignment approach. A recent study by Nickerson et al. (2008) addresses many of these
concerns by utilizing more sensitive methods to look at both the genetic and epigenetic issues
related to VHL inactivation. This study was performed on DNA from frozen tissue samples and
used a more sensitive technique for analysis for mutations (endonuclease scanning) as well as
analyzing for methylation changes that may lead to inactivation of the VHL gene. This method
of analysis was validated on tissue samples with known mutations. Of the 205 cc-RCC samples
analyzed, 169 showed mutations in the VHL gene (82.4%). Of those 36 without mutation, 11
were hypermethylated in the promoter region, which will also lead to inactivation of the VHL
gene. Therefore, this study showed inactivating alterations in the VHL gene (either by mutation
or hypermethylation) in 91% tumor samples analyzed.
The limited animal studies examining the role of VHL mutation following exposure to
chemicals including TCE are described below in Section 4.4.6.1.1. Conclusions as to the role of
VHL mutation in TCE-induced kidney cancer, taking into account both human and experimental
data, are presented below in Section 4.4.7.
4.4.4. Kidney Noncancer Toxicity in Laboratory Animals
Acute, subchronic, and chronic exposures to TCE cause toxicity to the renal tubules in
rats and mice of both sexes, via both inhalation (see Table 4-43) and oral (see Table 4-44)
exposures. Nephrotoxicity from acute exposures to TCE has only been reported at relatively
high doses, although histopathological changes have not been investigated in these experiments.
Information about specific location of lesions is presented where available. TCE exposure for
13-weeks (corn oil gavage) led to increased nephrotoxicity but no significant increases in
preneoplastic or neoplastic lesions as compared to controls (Mally et al., 2006). Chronic
nephropathy was also observed in both sexes of Osborne-Mendel rats following exposure to TCE
This document is a draft for review purposes only and does not constitute Agency policy.
217 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
(549 and 1,097 mg/kg-day, 78 week). Chakrabarti and Tuchweber (1988) found that TCE
administered to male F344 rats by intraperitoneal injection (723-2,890 mg/kg) or by inhalation
(1,000-2,000 ppm for 6 hours) produced elevated urinary NAG, y-glutamyl transpeptidase
(GGT), glucose excretion, blood urea nitrogen (BUN), and high molecular weight protein
excretion, characteristic signs of proximal tubular, and possibly glomerular injury, as soon as
24 hours postexposure. In the intraperitoneal injection experiments, inflammation was observed,
although some inflammation is expected due to the route of exposure, and nephrotoxicity effects
were only statistically significantly elevated at the highest dose (2,890 mg/kg). In the inhalation
experiments, the majority of the effects were statistically significant at both 1,000 and
2,000 ppm. Similarly, at these exposures, renal cortical slice uptake of />aminohippurate was
inhibited, indicating reduced proximal tubular function. Cojocel et al. (1989) found similar
effects in mice administered TCE by intraperitoneal injection (120-1,000 mg/kg) at 6 hours
postexposure, such as the dose-dependent increase in plasma BUN concentrations and decrease
in /;-aminohippurate accumulation in renal cortical slices. In addition, malondialdehyde (MDA)
and ethane production were increased, indicating lipid peroxidation.
Kidney weight increases have been observed following inhalation exposure to TCE in
both mice (Kjellstrand et al., 1983b) and rats (Woolhiser et al., 2006) and following lifetime
drinking water exposure in a genetically-prone murine model (Peden-Adams et al., 2008).
Kjellstrand et al. (1983b) demonstrated an increase in kidney weights in both male (20%
compared to control) and female (10% compared to control) mice following intermittent and
continuous TCE whole-body inhalation exposure (up to 120 days). This increase was significant
in males as low as 75 ppm exposure and in females starting at 150-ppm exposure. The latter
inhalation study, an unpublished report by Woolhiser et al. (2006), was designed to examine
immunotoxicity of TCE but also contains information regarding kidney weight increases in
female Sprague Dawley (S-D) rats exposed to 0-, 100-, 300-, and 1,000-ppm TCE for
6 hours/day, 5 days/week, for 4 weeks. Relative kidney weights were significantly elevated
This document is a draft for review purposes only and does not constitute Agency policy.
218 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-43. Inhalation studies of kidney noncancer toxicity in laboratory animals
Reference
Animals (sex)
Exposure route
Dose/exposure concentration
Exposed
Kidney effect(s) discussed in Section 4.4.4
Chakrabarti and
Tuchweber(1988)
Fischer 344 rats
(M)
Inhalation
0-20,00 ppm, 6 h
6/group
Increased signs of proximal tubular damage.
Green et al. (1998)
Fischer 344 rats
(M)
Inhalation
0, 250, 500 ppm, 6 h/d for 1, 7, 15,
21, 28 d
3-5/group
Increased formic acid excretion; plasma and
urinary markers of nephrotoxicity
unchanged.
Kjellstrand et al.
(1983b)
NMRI mice (M
and F)
Inhalation
0-3.600 ppm, variable time periods
of 1-24 h/d, for 30 or 120 d
10-20/group
Increased kidney weight.
Maltoni et al.
(1986)
Sprague-Dawley
rats, (M and F)
B6C3F1 mice (M
andF)
Inhalation
0, 100, 300, 600 ppm, 7 h/d, 5 d/wk,
104 wk exposure, observed for
lifespan
116-141/group
Meganucleocytosis in male rats (Details in
Table 4.49).
Mensing et al.
(2002)
Long-Evans rats
(M)
Inhalation
0-500 ppm, 6 h/5 d/wk, 6 mo
5/group
Increased signs of nephrotoxicity.
Woolhiser et al.
(2006)
Sprague-Dawley
rats(F)
Inhalation
0,100,300, and 1,000 ppm, 6 h/d, 5
d/wk, 4 wk
16/group
Increased kidney weight.
Bolded study(ies) carried forward for consideration in dose-response assessment (see Section 5).
-------
Table 4-44. Oral and i.p. studies of kidney noncancer toxicity in laboratory animals
Reference
Animals (sex)
Exposure route
Dose/exposure concentration
Exposed
Kidney effect(s) discussed in Section 4.4.4
Chakrabarti and
Tuchweber(1988)
Fischer 344 rats
(M)
Intraperitoneal
injection
0-2,890 mg/kg-day
6/group
Increased signs of proximal tubular damage.
Cojocel et al.
(1989)
NMRI mice (M)
Intraperitoneal
injection (sesame
oil)
0-1,000 mg/kg
4/group
Increased signs of nephrotoxicity.
Green et al.
(1997a)
Fischer 344 rats
(M)
B6C3F1 mice (M)
Gavage (corn oil)
0, 500, 2,000 mg/kg-day, 1 or 10 d
5 or
10/group
Increases in biochemical markers of kidney
damage.
Green et al. (2003)
Fischer 344 rats
(M)
Drinking water
0-54.3 mg/kg-day, 52 wk
60/group
Increased kidney weights and tubular
degeneration.
Mally et al. (2006)
Eker rat (M)
Gavage (corn oil)
0-1,000 mg/kg BW, 5 d/wk, 13 wk
10/group
Increased nephrotoxicity.
Maltoni et al.
(1986)
Sprague-Dawley
rats (M and F)
Gavage (olive
oil)
0, 50, 250 mg/kg-day 4-5 d/wk, 52
wk
30/group
Megakaryocytosis in male rats (Details in
Table 4.47).
NCI (1976)
Osborne-Mendel
rats (M and F)
B6C3F1 mice (M
and F)
Gavage (corn
oil)
0-2,339 mg/kg-day, variable doses,
5 d/wk, 78 wk
50/group
Toxic nephrosis in all exposed animals
(Details in Table 4.46).
NTP (1988)
ACI, August,
Marshall, and
Osborne-Mendel
rats (M and F)
Gavage (corn
oil)
0,500,1,000 mg/kg-day, 5 d/wk,
103 wk
50/group
Cytomegaly and toxic nephropathy observed
in all exposed rats (Details in Table 4-48).
NTP (1990)
Fischer 344 rats
(M and F)
B6C3F1 mice (M
andF)
Gavage (corn oil)
Rats: 0-2,000 mg/kg-day, Mice:
0-6,000 mg/kg-day, 5d/wk, 13 wk
10/group
Cytomegaly and karyomegaly of renal tubular
epithelium in mice and rats (Details in
Table 4.45).
Peden-Adams
et al. (2008)
MRL mice (M and
F)
Drinking water
0, 1.400, 14.000 ppb, lifetime
6/group
Increased kidney weight in male mice.
Bolded study(ies) carried forward for consideration in dose-response assessment (see Section 5).
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
(17.4% relative to controls) at 1,000-ppm TCE exposure. However, the small number of animals
and the variation in initial animal weight limit the ability of this study to determine statistically
significant increases. The Peden-Adams et al. (2008) study was designed to assess the effects of
TCE exposure in a genetically-prone murine lupus model. Although the study did not
demonstrate an increase in the development of autoimmune disease markers (for details see
Section 4.6.2), changes in body weight and organ weights in males were observed. Following
lifetime exposure to TCE (14.000 ppb) in drinking water, males exhibited a decreasing trend in
body mass of 12% from controls (female body weights not altered). Spleen, thymic and kidney
mass in females were not altered following exposure to TCE, while an 18% increase in kidney
mass was observed in the high dose treatment group (14.000 ppb) in males.
Similarly, overt signs of subchronic nephrotoxicity, such as changes in blood or urinary
biomarkers, are also primarily a high dose phenomenon, although histopathological changes are
evident at lower exposures. Green et al. (1997a) reported administration of 2,000 mg/kg-day
TCE by corn oil gavage for 42 days in F344 rats caused increases of around twofold of control
results in urinary markers of nephrotoxicity such as urine volume and protein (both 1.8x), NAG
(1.6x), glucose (2.2x) and alkaline phosphatase (ALP; 2.Ox), similar to the results of the acute
study of Chakrabarti and Tuchweber (1988), above. No morphological changes were observed
in kidneys from any animals (Green et al., 1997a). At lower dose levels, Green et al. (1998)
reported that plasma and urinary markers of nephrotoxicity were unchanged. In particular, after
1-28 day exposures to 250 or 500 ppm TCE for 6 hours/day, there were no statistically
significant differences in plasma levels of BUN or in urinary levels of creatinine, protein, ALP,
NAG, or GGT. However, increased urinary excretion of formic acid, accompanied by changes
in urinary pH and increased ammonia, was found at these exposures. Interestingly, at the same
exposure level of 500 ppm (6 hours/day, 5 days/week, for 6 months), Mensing et al. (2002)
reported elevated excretion of low molecular weight proteins and NAG, biomarkers of
nephrotoxicity, but after the longer exposure duration of 6 months.
Numerous studies have reported histological changes from TCE exposure for subchronic
and chronic durations (Maltoni et al., 1988, 1986; Mensing et al., 2002; NTP, 1990, 1988). As
summarized in Table 4-45, in 13-week studies in F344 rats and B6C3F1 mice, NTP (1990)
reported relatively mild cytomegaly and karyomegaly of the renal tubular epithelial cells at the
doses 1,000-6,000 mg/kg-day (at the other doses, tissues were not examined). The NTP report
noted that "these renal effects were so minimal that they were diagnosed only during a
reevaluation of the tissues...prompted by the production of definite renal toxicity in the 2-year
study." In the 6 month, 500-ppm inhalation exposure experiments of Mensing et al. (2002),
some histological changes were noted in the glomeruli and tubuli of exposed rats, but they
This document is a draft for review purposes only and does not constitute Agency policy.
221 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-45. Summary of renal toxicity and tumor findings in gavage studies
2 of trichloroethylene by NTP (1990)
3
Sex
Dose (mg/kg)a
Cytomegaly and karyomegaly
incidence (severity1^
Adenoma
(overall;
terminal)
Adenocarcinoma
(overall; terminal)
1/d, 5 d/wk, 13-wk study, F344/N rats
Male
0, 125, 250, 500, 100
Tissues not evaluated
None reported
2,000
8/9 (Minimal/mild)
Female
0, 62.5, 125, 250, 500
Tissues not evaluated
1,000
5/10 (Equivocal/minimal)
1/d, 5 d/wk, 13-wk study, B6C3F1 mice
Male
0, 375, 750, 1,500
Tissues not evaluated
None reported
3,000
7/10° (Mild/moderate)
6,000
d
Female
0, 375, 750, 1,500
Tissues not evaluated
3,000
9/10 (Mild/moderate)
6,000
1/10 (Mild/moderate)
1/d, 5 d/wk, 103-wk study, F344/N rats
Male
0
0% (0)
0/48; 0/33
0/48; 0/33
500
98% (2.8)
2/49; 0/20
0/49; 0/20
1,000
98% (3.1)
0/49; 0/16
3/49; 3/16e
Female
0
0% (0)
0/50; 0/37
0/50; 0/37
500
100% (1.9)
0/49; 0/33
0/49; 0/33
1,000
100% (2.7)
0/48; 0/26
1/48; 1/26
1/d, 5 d/wk, 103-wk study, B6C3F1 mice
Male
0
0% (0)
1/49; 1/33
0/49; 0/33
1,000
90% (1.5)
0/50; 0/16
1/50; 0/16
Female
0
0% (0)
0/48; 0/32
0/48; 0/32
1,000
98% (1.8)
0/49; 0/23
0/49; 0/23
4
5 Study carried forward for consideration in dose-response assessment (see Section 5).
6 a Corn oil vehicle.
7 b Numerical scores reflect the average grade of the lesion in each group (1, slight; 2, moderate; 3, well marked; and
8 4, severe).
9 Observed in four mice that died after 7-13 wk and in three that survived the study.
10 d All mice died during the first wk.
11 sp = 0.028.
12
13
This document is a draft for review purposes only and does not constitute Agency policy.
222 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
provided no detailed descriptions beyond the statement that "perivascular, interstitial infections
and glomerulonephritis could well be detected in kidneys of exposed rats."
After 1-2 years of chronic TCE exposure by gavage (NCI, 1976; NTP, 1990, 1988) or
inhalation (Maltoni et al., 1988) (see Tables 4-45 to 4-49), both the incidence and severity of
these effects increases, with mice and rats exhibiting lesions in the tubular epithelial cells of the
inner renal cortex that are characterized by cytomegaly, karyomegaly, and toxic nephrosis. As
with the studies at shorter duration, these chronic studies reported cytomegaly and karyomegaly
of tubular cells. NTP (1990) specified the area of damage as the pars recta, located in the
corticomedullary region. It is important to note that these effects are distinct from the chronic
nephropathy and inflammation observed in control mice and rats (Lash et al., 2000b; Maltoni et
al., 1988; NCI, 1976).
These effects of TCE on the kidney appear to be progressive. Maltoni et al. (1988) noted
that the incidence and degree of renal toxicity increased with increased exposure time and
increased time from the start of treatment. As mentioned above, signs of toxicity were present in
the 13 week study (NTP, 1988), and NTP (1990) noted cytomegaly at 26 weeks. NTP (1990)
noted that as "exposure time increased, affected tubular cells continued to enlarge and additional
tubules and tubular cells were affected," with toxicity extending to the cortical area as kidneys
became more extensively damaged. NTP (1988, 1990) noted additional lesions that increased in
frequency and severity with longer exposure, such as dilation of tubules and loss of tubular cells
lining the basement membrane ("stripped appearance" (NTP, 1988) or flattening of these cells
(NTP, 1990)). NTP (1990) also commented on the intratubular material and noted that the
tubules were empty or "contained wisps of eosinophilic material."
With gavage exposure, these lesions were present in both mice and rats of both sexes, but
were on average more severe in rats than in mice, and in male rats than in female rats (NTP,
1990). Thus, it appears that male rats are most sensitive to these effects, followed by female rats
and then mice. This is consistent with the experiments of Maltoni et al. (1988), which only
reported these effects in male rats. The limited response in female rats or mice of either sex in
these experiments may be related to dose or strain. The lowest chronic gavage doses in the
National Cancer Institute (NCI, 1976) and NTP (1988, 1990) F344 rat experiments was
500 mg/kg-day, and in all these cases at least 80% (and frequently 100%) of the animals showed
cytomegaly or related toxicity. By comparison, the highest gavage dose in the Maltoni et al.
(1988) experiments (250 mg/kg-day) showed lower incidences of renal cytomegaly and
karyomegaly in male Sprague-Dawley rats (47 and 67%, overall and corrected incidences) and
none in female rats. The B6C3F1 mouse strain was used in the NCI (1976), NTP (1990), and
This document is a draft for review purposes only and does not constitute Agency policy.
223 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Table 4-46. Summary of renal toxicity and tumor findings in gavage studies
of trichloroethylene by NCI (1976)
Sex
Dose (mg/kg)a
Toxic nephrosis
(overall; terminal)
Adenoma or adenocarcinoma
(overall; terminal)b
1/d, 5 d/wk, 2-yr study, Osborn-Mendel rats
Males
0
0/20; 0/2
0/20; 0/2
549
46/50; 7/7
l/50;c 0/7
1,097
46/50; 3/3
0/50; 0/3
Females
0
0/20; 0/8
0/20; 0/8
549
39/48; 12/12
0/48; 0/12
1,097
48/50; 13/13
0/50; 0/13
1/d, 5 d/wk, 2-yr study, B6C3F1 mice
Males
0
0/20; 0/8
0/20; 0/8
1,169
48/50; 35/35
0/50; 0/35
2,339
45/50; 20/20
l/50;d 1/20
Females
0
0/20; 0/17
0/20; 0/17
869
46/50; 40/40
0/50; 0/40
1,739
46/47;e 39/39
0/47; 0/39
Study carried forward for consideration in dose-response assessment (see Section 5).
a Treatment period was 48 wk for rats, 66 wk for mice. Doses were changed several times during the study based on
monitoring of body weight changes and survival. Dose listed here is the time-weighted average dose over the days
on which animals received a dose.
b A few malignant mixed tumors and hamartomas of the kidney were observed in control and low dose male rats, but
are not counted here.
0 Tubular adenocarcinoma.
d Tubular adenoma.
e One mouse was reported with "nephrosis," but not "nephrosis toxic," and so was not counted here.
This document is a draft for review purposes only and does not constitute Agency policy.
224 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-47. Summary of renal toxicity findings in gavage studies of
2 trichloroethylene by Maltoni et al. (1988)
3
Sex
Dose (mg/kg)a
Megalonucleocytosisb (overall;
corrected0)
1/d, 4-5 d/wk, 52-wk exposure, observed for lifespan, Sprague-Dawley rats
Males
0
0/20; 0/22
50
0/30; 0/24
250
14/30; 14/21
Females
0
0/30; 0/30
50
0/30; 0/29
250
0/30; 0/26
4
5 a Olive oil vehicle.
6 b Renal tubuli megalonucleocytosis is the same as cytomegaly and karyomegaly of renal tubuli cells (Maltoni et al.,
7 1988).
8 0 Denominator for "corrected" incidences is the number of animals alive at the time of the first kidney lesion in this
9 experiment (39 wk).
10
11
This document is a draft for review purposes only and does not constitute Agency policy.
225 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-48. Summary of renal toxicity and tumor incidence in gavage studies
2 of trichloroethylene by NTP (1988)
3
Sex
Dose (mg/kg)a
Cytomegaly
Toxic
nephropathy
Adenoma
(overall; terminal)
Adenocarcinoma
(overall; terminal)
1/d, 5 d/wk, 2-yr study, ACI rats
Male
0
0/50
0/50
0/50; 0/38
0/50; 0/38
500
40/49
18/49
0/49; 0/19
1/49; 0/19
1,000
48/49
18/49
0/49; 0/11
0/49; 0/11
Female
0
0/48
0/48
0/48; 0/34
0/48; 0/34
500
43/47
21/47
2/47; 1/20
1/47; 1/20
1,000
42/43
19/43
0/43; 0/19
1/43; 0/19
1/d, 5 d/wk, 2-yr study, August rats
Male
0
0/50
0/50
0/50; 0/21
0/50; 0/21
500
46/50
10/50
1/50; 0/13
1/50; 1/13
1,000
46/49
31/49
1/49; 1/16
0/49; 0/16
Female
0
0/49
0/49
1/49; 1/23
0/49; 0/23
500
46/48
8/48
2/48; 1/26
2/48; 2/26
1,000
50/50
29/50
0/50; 0/25
0/50; 0/25
1/d, 5 d/wk, 2-yr study, Marshall rats
Male
0
0/49
0/49
0/49; 0/26
0/49; 0/26
500
48/50
18/50
1/50; 0/12
0/50; 0/12
1,000
47/47
23/47
0/47; 0/6
1/47; 0/6
Female
0
0/50
0/50
1/50; 0/30
0/50; 0/30
500
46/48
30/48
1/48; 1/12
1/48; 0/12
1,000
43/44
30/44
0/44; 0/10
1/44; 1/10
1/d, 5 d/wk, 2-yr study, Osborne-Mendel rats
Male
0
0/50
0/50
0/50; 0/22
0/50; 0/22
500
48/50
39/50
6/50; 5/17
0/50; 0/17
1,000
49/50
35/50
1/50; 1/15
1/50; 0/15
Female
0
0/50
0/50
0/50; 0/20
0/50; 0/20
500
48/50
30/50
0/50; 0/11
0/50; 0/11
1,000
49/49
39/49
1/49; 0/7
0/49; 0/7
4
5 Study carried forward for consideration in dose-response assessment (see Section 5).
6 a Corn oil vehicle.
7
This document is a draft for review purposes only and does not constitute Agency policy.
226 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Table 4-49. Summary of renal toxicity and tumor findings in inhalation
studies of trichloroethylene by Maltoni et al. (1988)a
Sex
Concentration
(ppm)
Meganucleocytosisb
(overall; corrected)
Adenoma
(overall;
corrected)
Adenocarcinoma
(overall;
corrected)
7 h/d, 5 d/wk, 2-yr exposure, observed for lifespan, Sprague-Dawley rats0
Male
0
0/135; 0/122
0/135; 0/122
0/135; 0/122
100
0/130; 0/121
1/130; 1/121
0/130; 0/121
300
22/130; 22/116
0/130; 0/116
0/130; 0/116
600
101/130; 101/124
1/130; 1/124
4/130; 4/124
Female
0
0/145; 0/141
0/145; 0/141
0/145; 0/141
100
0/130; 0/128
1/130; 1/128
0/130; 0/128
300
0/130; 0/127
0/130; 0/127
0/130; 0/127
600
0/130; 0/127
0/130; 0/127
1/130; 1/127
7 h/d, 5 d/wk, 78-wk exposure, observed for lifespan, B6C3F1 miced
Male
0
0/90
0/90
0/90
100
0/90
0/90
1/90
300
0/90
0/90
0/90
600
0/90
0/90
0/90
Female
0
0/90
0/90
1/90
100
0/90
0/90
0/90
300
0/90
0/90
0/90
600
0/90
0/90
0/90
Study carried forward for consideration in dose-response assessment (see Section 5).
a Three inhalation experiments in this study found no renal megalonucleocytosis, adenomas, or adenocarcinomas:
BT302 (8-wk exposure to 0, 100, 600 ppm in Sprague-Dawley rats); BT303 (8-wk exposure to 0, 100, or 600 ppm
in Swiss mice); and BT305 (78-wk exposure to 0, 100, 300, or 600 ppm in Swiss mice).
b Renal tubuli meganucleocytosis is the same as cytomegaly and karyomegaly of renal tubuli cells (Maltoni et al.,
1988).
0 Combined incidences from experiments BT304 and BT304bis. Corrected incidences reflect number of rats alive at
47 wk, when the first renal tubular megalonucleocytosis in these experiments appeared.
dFemale incidences are from experiment BT306, while male incidences are from experiment BT306bis, which was
added to the study because of high, early mortality due to aggressiveness and fighting in males in experiment
BT306. Corrected incidences not show, because only the renal adenocarcinomas appeared at 107 wk in the male
and 136 in the female, when the most of the mice were already deceased.
This document is a draft for review purposes only and does not constitute Agency policy.
227 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Maltoni et al. (1988) studies (see Tables 4-45-4-49). While the two gavage studies (NCI, 1976;
NTP, 1990) were consistent, reporting at least 90% incidence of cytomegaly and karyomegaly at
all studied doses, whether dose accounts for the lack of kidney effects in Maltoni et al. (1988)
requires comparing inhalation and gavage dosing. Such comparisons depend substantially on the
internal dose-metric, so conclusions as to whether dose can explain differences across studies
cannot be addressed without dose-response analysis using PBPK modeling. Some minor
differences were found in the multistrain NTP study (1988), but the high rate of response makes
distinguishing among them difficult. Soffritti (personal communication with JC Caldwell,
February 14, 2006) did note that the colony from which the rats in Maltoni et al. (1988; 1986)
experiments were derived had historically low incidences of chronic progressive nephropathy
and renal cancer.
4.4.5. Kidney Cancer in Laboratory Animals
4.4.5.1.1. Inhalation Studies of Trichloroethylene (TCE)
A limited number of inhalation studies examined the carcinogenicity of TCE, with no
statistically-significantly increases in kidney tumor incidence reported in mice or hamsters
(Fukuda et al., 1983; Henschler et al., 1980; Maltoni et al., 1988; 1986). The cancer bioassay by
Maltoni et al. (1988; 1986) reported no statistically significant increase in kidney tumors in mice
or hamsters, but renal adenocarcinomas were found in male (4/130) and female (1/130) rats at
the high dose (600 ppm) after 2 years exposure and observation at natural death. In males, these
tumors seemed to have originated in the tubular cells, and were reported to have never been
observed in over 50,000 Sprague-Dawley rats (untreated, vehicle-treated, or treated with
different chemicals) examined in previous experiments in the same laboratory (Maltoni et al.,
1986). The renal adenocarcinoma in the female rat was cortical and reported to be similar to that
seen infrequently in historical controls. This study also demonstrated the appearance of
increased cytokaryomegaly or megalonucleocytosis in the tubular cells, a lesion that was
significantly and dose-dependently increased in male rats only (see Table 4-49). Maltoni et al.
(1986) noted that some considerations supported either the hypothesis that these were precursor
lesions of renal adenocarcinomas cancer or the hypothesis that these are not precursors but rather
the morphological expression of TCE-induced regressive changes. The inhalation studies by
Fukuda et al. (1983) in Sprague-Dawley rats and female ICR mice, reported one clear cell
carcinoma in rats exposed to the highest concentration (450 ppm) but saw no increase in kidney
tumors in mice. This result was not statistically significant (see Table 4-50) and no details are
This document is a draft for review purposes only and does not constitute Agency policy.
228 DRAFT—DO NOT CITE OR QUOTE
-------
1 given about the specific location of the tumors. One negative study (Henschler et al., 1980)
2 tested NMRI mice, Wistar rats, and Syrian hamsters of both sexes (60 animals per strain), and
3
This document is a draft for review purposes only and does not constitute Agency policy.
229 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-50. Summary of renal tumor findings in inhalation studies of
2 trichloroethylene by Henschler et al. (1980)a and Fukuda et al. (1983)b
3
Sex
Concentration (ppm)
Adenomas
Adenocarcinomas
6 h/d, 5 d/wk, 18-mo exposure, 30-mo observation, Han:NMRI mice (Henschler et al., 1980)
Males
0
4/30
1/30
100
1/29
0/30
500
1/29
0/30
Females
0
0/29
0/29
100
0/30
0/30
500
0/28
0/28
6 h/d, 5 d/wk, 18-mo exposure, 36-mo observation, Han:WIST rats (Henschler et al., 1980)
Males
0
2/29
0/29
100
1/30
0/30
500
2/30
1/30
Females
0
0/28
0/28
100
0/30
0/30
500
1/30
0/30
7 h/d, 5 d/wk, 2-yr study, Crj:CD (S-D) rats (Fukuda et al., 1983)
Females
0
0/50
0/50
50
0/50
0/50
150
0/47
0/47
450
0/51
1/50
4
5 a Henschler et al. (1980) observed no renal tumors in control or exposed Syrian hamsters.
6 b Fukuda et al. (1983) observed no renal tumors in control or exposed Crj :CD-1 (ICR) mice.
7
8
9 observed no significant increase in renal tubule tumors any of the species tested. Benign
10 adenomas were observed in male mice and rats, a single adenocarcinoma was reported in male
11 rats at the highest dose, and no renal adenocarcinomas reported in females of either species (see
12 Table 4-50). Renal cell carcinomas appear to be very rare in Wistar rats, with historical control
13 rates reported to be about 0.4% in males and 0.2% in females (Poteracki and Walsh, 1998), so
14 these data are very limited in power to detect small increases in their incidence.
This document is a draft for review purposes only and does not constitute Agency policy.
230 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
4.4.5.1.2. Gavage and Drinking Water Studies of Trichloroethylene (TCE)
Several chronic gavage studies exposing multiple strains of rats and mice to
0-3,000 mg/kg TCE for at least 52 weeks have been conducted (see Tables 4-45 to 4-48, 4-51)
(Henschler et al., 1984; Maltoni et al., 1986; NCI, 1976; NTP, 1988, 1990; Van Duuren et al.,
1979). Van Duuren et al. (1979) examined TCE and 14 other halogenated compounds for
carcinogenicity in both sexes of Swiss mice. While no excess tumors were observed, the dose
rate (0.5 mg once per week, or an average dose rate of approximately 2.4 mg/kg-day for a 30 g
mouse) is about 400-fold lower than that in the other gavage studies. Inadequate design and
reporting of this study limit the ability to use the results as an indicator of TCE carcinogenicity.
In the NCI (1976) study, the results for Osborne-Mendel rats were considered by the authors to
be inconclusive due to significant early mortality. Two male rats demonstrated kidney lesions
(dilated renal pelvis and dark red renal medulla), but in rats of both sexes, no increase was seen
in primary tumor induction over that observed in controls. While both sexes of B6C3F1 mice
showed a compound-related increase in nephropathy, no increase in tumors over controls was
observed. The NCI study (1976) used technical grade TCE which contained two known
carcinogenic compounds as stabilizers (epichlorohydrin and 1,2-epoxybutane). However, a
subsequent study by Henschler et al. (1984) in mice reported no significant differences in
systemic tumorigenesis between pure, industrial, and stabilized TCE, suggesting that
concentrations of these stabilizers are too low to be the cause of tumors. A later gavage study by
NTP (1988), using TCE stabilized with diisopropylamine, observed an increased incidence of
renal tumors in all four strains of rats (ACI, August, Marshall, and Osborne-Mendel). All
animals exposed for up to 2 years (rats and mice) had non-neoplastic kidney lesions (tubular cell
cytomegaly), even if they did not later develop kidney cancer (see Table 4-48). This study was
also considered inadequate by the authors because of chemically induced toxicity, reduced
survival, and incomplete documentation of experimental data. The final NTP study (1990) in
male and female F344 rats and B6C3F1 mice used epichlorohydrin-free TCE. Only in the
highest-dose group (1,000 mg/kg) of male F344 rats was renal carcinoma statistically significant
increased. The results for detecting a carcinogenic response in rats were considered by the
authors to be equivocal because both groups receiving TCE showed significantly reduced
survival compared to vehicle controls and because of a high rate (e.g., 20% of the animals in the
high-dose group) of death by gavage error. However, historical control incidences at NTP of
This document is a draft for review purposes only and does not constitute Agency policy.
231 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-51. Summary of renal tumor findings in gavage studies of
2 trichloroethylene by Henschler et al. (1984)a and Van Duuren et al. (1979)b
3
Sex
(TCE dose)
Control or TCE exposed
(stabilizers if present)
Adenomas
Adenocarcinomas
5 d/wk, 18-mo exposure, 24-mo observation, Swiss mice (Henschler et al., 1984)
Males
(2.4g/kg BW)
Control (none)
1/50
1/50
TCE (triethanolamine)
1/50
1/50
TCE (industrial)
0/50
0/50
TCE (epichlorohydrin (0.8%))
0/50
0/50
TCE (1,2-epoxybutane (0.8%))
2/50
2/50
TCE (both epichlorohydrin (0.25%)
and 1,2-epoxybutane (0.25%))
0/50
0/50
Females
(1.8 g/kg BW)
Control (none)
0/50
1/50
TCE (triethanolamine)
4/50
0/50
TCE (industrial)
0/50
0/50
TCE (epichlorohydrin (0.8%))
0/50
0/50
TCE (1,2-epoxybutane (0.8%))
0/50
0/50
TCE (both epichlorohydrin (0.25%)
and 1,2-epoxybutane (0.25%))
0/50
0/50
1 d/wk, 89-wk exposure, Swiss rats (Van Duuren et al., 1979)
Males
(0.5mg)
Control
0/30
0/30
TCE (unknown)
0/30
0/30
Females
(0.5mg)
Control
0/30
0/30
TCE(unknown)
0/30
0/30
4
5 "Henschler et al. (1984). Due to poor condition of the animals resulting from the nonspecific toxicity of high doses
6 of TCE and/or the additives, gavage was stopped for all groups during wk 35-40, 65, and 69-78, and all doses
7 were reduced by a factor of 2 from the 40th wk on.
8 b Van Duuren et al. (1979) observed no renal tumors in control or exposed Swiss mice.
9
This document is a draft for review purposes only and does not constitute Agency policy.
232 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
kidney tumors in F344 rats is very low,5 lending biological significance to their occurrence in
this study, despite the study's limitations. Cytomegaly and karyomegaly were also increased,
particularly in male rats. The toxic nephropathy (specific location in kidney not stated) observed
in both rats and mice and contributed to the poor survival rate (see Table 4-45). As discussed
previously, this toxic nephropathy was clearly distinguishable from the spontaneous chronic
progression nephropathy commonly observed in aged rats.
4.4.5.1.3. Conclusions: Kidney Cancer in Laboratory Animals
Chronic TCE carcinogenicity bioassays have shown evidence of neoplastic lesions in the
kidney in rats (mainly in males, with less evidence in females), treated via inhalation and gavage.
As discussed above, individual studies have a number of limitations and have shown limited
increases in kidney tumors. However, given the rarity of these tumors as assessed by historical
controls and the repeatability of this result, these are considered biologically significant.
4.4.6. Role of Metabolism in Trichloroethylene (TCE) Kidney Toxicity
It is generally thought that one or more TCE metabolites rather than the parent compound
are the active moieties for TCE nephrotoxicity. As reviewed in Section 3.3, oxidation by CYPs,
of which CYP2EI is thought to be the most active isoform, results in the production of chloral
hydrate, trichloroacetic acid, dichloroacetic acid and trichloroethanol. The glutathione
conjugation pathway produces metabolites such as DCVG, DCVC, dichlorovinylthiol, and
NAcDCVC, although, as discussed in Section 3.3.3.2, the quantitative estimates of the amount
systemically produced following TCE exposure remains uncertain. Because several of the steps
for generating these reactive metabolites occur in the kidney, the GSH conjugation pathway has
been thought to be responsible for producing the active moiety or moieties of TCE
nephrotoxicity. A comparison of TCE's nephrotoxic effects with the effects of TCE metabolites,
both in vivo and in vitro, thus, provides a basis for assessing the relative roles of different
metabolites. While most of the available data have been on metabolites from GSH conjugation,
such as DCVC, limited information is also available on the major oxidative metabolites TCOH
and TCA.
5 NTP (19901 reported a historical control incidence of 0.4% in males. The NTP web site reports historical control
rates of renal carcinomas for rats dosed via corn oil gavage on the NIH-07 diet (used before 1995, when the TCE
studies were conducted) to be 0.5% (2/400) for males and 0% (0/400) for females
(http://ntp-server.niehs.nih.gov/ntp/research/database_searches/historical_controls/path/r_gavco.txt). In addition,
the two occurences in males came from the same study, with all other studies reporting 0/50 carcinomas.
This document is a draft for review purposes only and does not constitute Agency policy.
233 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
4.4.6.1.1. In Vivo Studies of the Kidney Toxicity of Trichloroethylene (TCE) Metabolites
Studies of kidney toxicity of TCE metabolites discussed in this section are shown in
Table 4-52.
4.4.6.1.2. Role of GSH conjugation metabolites of trichloroethylene (TCE)
In numerous studies, DCVC has been shown to be acutely nephrotoxic in rats and mice.
Mice receiving a single dose of 1 mg/kg DCVC (the lowest dose tested in this species) exhibited
karyolytic proximal tubular cells in the outer stripe of the outer medulla, with some sloughing of
cells into the lumen and moderate desquamation of the tubular epithelium (Eyre et al., 1995b).
Higher doses in mice were associated with more severe histological changes similar to those
induced by TCE, such as desquamation and necrosis of the tubular epithelium (Darnerud et al.,
1989; TERRACINI and PARKER, 1965; Vaidya et al., 2003a, b). In rats, no histological
changes in the kidney were reported after single doses of 1, 5, and 10 mg/kg DCVC (Eyre et al.,
1995a; Green et al., 1997a), but cellular debris in the tubular lumen was reported at 25 mg/kg
(Eyre et al., 1995b) and slight degeneration and necrosis were seen at 50 mg/kg (Green et al.,
1997a). Green et al. (1997a) reported no histological changes were noted in rats after 10 doses
of 0.1-5.0 mg/kg DCVC (although increases in urinary protein and GGT were found), but some
karyomegaly was noted in mice after 10 daily doses of 1 mg/kg. Therefore, mice appear more
sensitive than rats to the nephrotoxic effects of acute exposure to DCVC, although the number of
animals used at each dose in these studies was limited (10 or less). Although the data are not
sufficient to assess the relatively sensitivity of other species, it is clear that multiple species,
including rabbits, guinea pigs, cats, and dogs, are responsive to DCVC's acute nephrotoxic
effects (Jaffe et al., 1984; Krejci et al., 1991; TERRACINI and PARKER, 1965; Wolfgang et al.,
1989b).
Very few studies are available at longer durations. Terracini and Parker (1965) gave
DCVC in drinking water to rats at a concentration of 0.01% for 12 weeks (approximately
10 mg/kg-day), and reported consistent pathological and histological changes in the kidney. The
progression of these effects was as follows: (1) during the first few days, completely necrotic
tubules, with isolated pyknotic cells being shed into the lumen; (2) after 1 week, dilated tubules
in the inner part of the cortex, lined with flat epithelial cells that showed thick basal membranes,
some with big hyperchromatic nuclei; (3) in the following weeks, increased prominence of
tubular cells exhibiting karyomegaly, seen in almost all animals, less pronounced tubular
This document is a draft for review purposes only and does not constitute Agency policy.
234 DRAFT—DO NOT CITE OR QUOTE
-------
1 dilation, and cytomegaly in the same cells showing karyomegaly. In addition, increased mitotic
2 activity was reported the first few days, but was not evident for the rest of the experiment.
3 Terracini and Parker (1965) also reported the results of a small experiment (13 male and
This document is a draft for review purposes only and does not constitute Agency policy.
23 5 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-52. Laboratory animal studies of kidney noncancer toxicity of TCE metabolites
Reference
Animals (sex)
Exposure route
Dose/exposure concentration
Exposed
Kidney effect(s) discussed in Section 4.4.4
Dow and Green
(2000)
Fischer 344 rats
(M)
Drinking water
0, 0.5, 1 g/L Trichloroethanol, 12 wk
3/group
Increased formic acid in urine.
Jaffe et al. (1984)
Swiss-Webster
mice (M)
Drinking water
0-22 mg/kg-day DCVC, 37 wk
5/group
Cytomegaly and tubular degeneration.
Mather et al.
(1990)
Sprague-Dawley
rats (M)
Drinking water
0-355 mg/kg-day TCA, 90 d
10/group
Increased kidney weight.
Terracini and
Parker (1965)
Wistar rats
(Gender not
specified)
Grey mice
(Gender not
specified)
Drinking water
0,0.01% DCVC, 12 wk
3 5/group
Necrosis of tubular epithelium in mice and rats.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
5 female rats) given the same concentration of DCVC in drinking water for 46 weeks, and
observed for 87 weeks. They noted renal tubular cells exhibiting karyomegaly and cytomegaly
consistently throughout the experiment. Moreover, a further group of eight female rats given
DCVC in drinking water at a concentration of 0.001% (approximately 1 mg/kg-day) also
exhibited similar, though less severe, changes in the renal tubules. In mice, Jaffe et al. (1984)
gave DCVC in drinking water at concentrations of 0.001, 0.005, and 0.01% (estimated daily dose
of 1-2, 7-13, and 17-22 mg/kg-day), and reported similar effects in all dose groups, including
cytomegaly, nuclear hyperchromatism, and multiple nucleoli, particularly in the pars recta
section of the kidney. Thus, effects were noted in both mice and rats under chronic exposures at
doses as low as 1-2 mg/kg-day (the lowest dose tested). Therefore, while limited, the available
data do not suggest differences between mice and rats to the nephrotoxic effects of DCVC under
chronic exposure conditions, in contrast to the greater sensitivity of mice to acute and subchronic
DCVC-induced nephrotoxicity.
Importantly, as summarized in Table 4-53, the histological changes and their location in
these subchronic and chronic experiments with DCVC are quite similar to those reported in
chronic studies of TCE, described above, particularly the prominence of karyomegaly and
cytomegaly in the pars recta section of the kidney. Moreover, the morphological changes in the
tubular cells, such as flattening and dilation, are quite similar. Similar pathology is not observed
with the oxidative metabolites alone (see Section 4.4.6.1.2).
Additionally, it is important to consider whether sufficient DCVC may be formed from
TCE exposure to account for TCE nephrotoxicity. While direct pharmacokinetic measurements,
such as the excretion of NAcDCVC, have been used to argue that insufficient DCVC would be
formed to be the active moiety for nephrotoxicity (Green et al., 1997a), as discussed in Section 3,
urinary NAcDCVC is a poor marker of the flux through the GSH conjugation pathway because
of the many other possible fates of metabolites in that pathway. In another approach, Eyre et al.
(1995b) using acid-labile adducts as a common internal dosimeter between TCE and DCVC, and
reported that a single TCE dose of 400 mg/kg in rats (similar to the lowest daily doses in the NCI
and NTP rat bioassays) and 1,000 mg/kg (similar to the lowest daily doses in the NCI and NTP
mouse bioassays) corresponded to a single equivalent DCVC dose of 6 and 1 mg/kg-day in rats
and mice, respectively. These equivalent doses of DCVC are greater or equal to those in which
nephrotoxicity has been reported in these species under chronic conditions. Therefore, assuming
that this dose correspondence is accurate under chronic conditions, sufficient DCVC would be
formed from TCE exposure to explain the observed histological changes in the renal tubules.
Nevertheless, direct estimates of how much DCVC is formed after TCE exposure are lacking.
This document is a draft for review purposes only and does not constitute Agency policy.
237 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-53. Summary of histological changes in renal proximal tubular cells induced by chronic exposure to
TCE, DCVC, and TCOH
Effects
TCE
DCVC
TCOH
Karyomegal
y
Enlarged, hyperchromatic nuclei, irregular to
oblong in shape. Vesicular nuclei containing
prominent nucleoli.
Enlarged, hyperchromatic nuclei with
and multiple nucleoli. Nuclear
pyknosis and karyorrhexis.
None reported.
Cytomegaly
Epithelial cells were large, elongated and
flattened.
Epithelial cells were large, elongated
and flattened cells.
No report of enlarged cells.
Cell
necrosis/
hyperplasia
Stratified epithelium that partially or
completely filled the tubular lumens. Cells in
mitosis were variable in number or absent.
Cells had abundant eosinophilic or basophilic
cytoplasm.
Thinning of tubular epithelium, frank
tubular necrosis, re-epitheliation.
Tubular atrophy, interstitial fibrosis
and destruction of renal parenchyma.
More basophilic and finely
vacuolated.
No flattening or loss of
epithelium reported.
Increased tubular cell
basophilia, followed by
increased cellular
eosinophilia, tubular cell
vacuolation.
Morphology
/ content of
tubules
Some tubules enlarged/dilated to the extent that
they were difficult to identify. Portions of
basement membrane had a stripped appearance.
Tubules were empty or contained "wisps of
eosinophilic material."
Tubular dilation, denuded tubules.
Thick basal membrane. Focal areas
of dysplasia, intraluminal casts.
No tubular dilation reported.
Intratubular cast formation.
Sources: NCI (1976); NTP (1988, 1990); Maltoni et al. (1988); Terracini and Parker (1965); Jaffe et al. (1985); Green et al. (2003).
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
The Eker rat model (Tsc-2is at increased risk for the development of spontaneous renal
cell carcinoma and as such has been used to understand the mechanisms of renal carcinogenesis
(Stemmer et al., 2007; Wolf et al., 2000). One study has demonstrated similar pathway
activation in Eker rats as that seen in humans with VHL mutations leading to renal cell
carcinoma, suggesting Tsc-2 inactivation is analogous to inactivation of VHL in human renal cell
carcinoma (Liu et al., 2003). Although the Eker rat model is a useful tool for analyzing
progression of renal carcinogenesis, it has some limitations in analysis of specific genetic
changes, particularly given the potential for different genetic changes depending on type of
exposure and tumor. The results of short-term assays to genotoxic carcinogens in the Eker rat
model (Morton et al., 2002; Stemmer et al., 2007) reported limited preneoplastic and neoplastic
lesions which may be related to the increased background rate of renal carcinomas in this animal
model.
Recently, Mally et al. (2006) exposed male rats carrying the Eker mutation to TCE
(0-1,000 mg/kg BW) by corn oil gavage and demonstrated no increase in renal preneoplastic
lesions or tumors. Primary Eker rat kidney cells exposed to DCVC in this study did induce an
increase in transformants in vitro but no DCVC-induced vhl or Tsc-2 mutations were observed.
In vivo exposure to TCE (5 days/week for 13 weeks), decreased body weight gain and increased
urinary excretion at the two highest TCE concentrations analyzed (500 and 1,000 mg/kg BW)
but did not change standard nephrotoxicity markers (GGT, creatinine and urinary protein).
Renal tubular epithelial cellular proliferation as measured by BrdU incorporation was
demonstrated at the three highest concentrations of TCE (250, 500 and 1,000 mg/kg-day). A
minority of these cells also showed karyomegaly at the two higher TCE concentrations.
Although renal cortical tumors were demonstrated in all TCE exposed groups, these were not
significantly different from controls (13 weeks). These studies were complemented with in vitro
studies of DCVC (10-50 |iM) in rat kidney epithelial (RKE) cells examining proliferation at 8,
24, and 72 hours and cellular transformation at 6-7 weeks. Treatment of RKE cells from
susceptible rats with DCVC gave rise to morphologically transformed colonies consistently
higher than background (Mally et al., 2006). Analyzing ten of the renal tumors from the TCE
exposed rats and nine of the DCVC transformants from these studies for alterations to the VHL
gene that might lead to inactivation found no alterations to VHL gene expression or mutations.
One paper has linked the VHL gene to chemical-induced carcinogenesis. Shiao et al.
(1998) demonstrated VHL gene somatic mutations in A'-nitrosodi methyl am ine-induced rat kidney
cancers that were of the clear cell type. The clear cell phenotype is rare in rat kidney cancers,
but it was only the clear cell cancers that showed VHL somatic mutation (three of eight tumors
analyzed). This provided an additional link between VHL inactivation and clear cell kidney
This document is a draft for review purposes only and does not constitute Agency policy.
239 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
cancer. However, this study examined archived formalin fixed paraffin embedded tissues from
previous experiments. As described previously (see Section 4.4.2), DNA extraction from this
type of preparation creates some technical issues. Similarly, archived formalin-fixed paraffin
embedded tissues from rats exposed to potassium bromide were analyzed in a later study by
Shiao et al. (2002). This later study examined the VHL gene mutations following exposure to
potassium bromide, a rat renal carcinogen known to induce clear cell renal tumors. Clear cell
renal tumors are the most common form of human renal epithelial neoplasms, but are extremely
rare in animals. Although F344 rats exposed to potassium bromide in this study did develop
renal clear cell carcinomas, only two of nine carried the same C to T mutation at the core region
of the Spl transcription-factor binding motif in the VHL promoter region, and one of
four untreated animals had a C to T mutation outside the conserved core region. Mutation in the
VHL coding region was only detected in one tumor, so although the tumors developed following
exposure to potassium bromide were morphologically similar to those found in humans; no
similarities were found in the genetic changes.
Elfarra et al. (1984) found that both DCVG and DCVC administered to male F344 rats by
intraperitoneal injections in isotonic saline resulted in elevations in BUN and urinary glucose
excretion. Furthermore, inhibition of renal GGT activity with acivicin protected rats from
DCVG-induced nephrotoxicity. In addition, both the B-lyase inhibitor AOAA and the renal
organic anion transport inhibitor probenecid provided protection from DCVC, demonstrating a
requirement for metabolism of DCVG to the cysteine conjugate by the action of renal GGT and
dipeptidase, uptake into the renal cell by the organic anion transporter, and subsequent activation
by the B-lyase. This conclusion was supported further by showing that the methyl analog of
DCVC, which cannot undergo a B-elimination reaction due to the presence of the methyl group,
was not nephrotoxic.
Korrapati et al. (2005) builds upon a series of investigations of hetero- (by mercuric
chloride [HgCh]) and homo-(by DCVC, 15 mg/kg) protection against a lethal dose of DCVC
(75 mg/kg). Priming, or preconditioning, with pre-exposure to either HgCb or DCVC of male
Swiss-Webster mice was said to augment and sustain cell division and tissue repair, hence
protecting against the subsequent lethal DCVC dose (Vaidya et al., 2003a, b; 2003c). Korrapati
et al. (2005) showed that a lethal dose of DCVC downregulates phosphorylation of endogenous
retinoblastoma protein (pRb), which is considered critical in renal proximal tubular and
mesangial cells for the passage of cells from G1 to S-phase, thereby leading to a block of renal
tubule repair. Priming, in contrast, upregulated P-pRB which was sustained even after the
administration of a lethal dose of DCVC, thereby stimulating S-phase DNA synthesis, which was
concluded to result in tissue repair and recovery from acute renal failure and death. These
This document is a draft for review purposes only and does not constitute Agency policy.
240 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
studies are more informative about the mechanism of autoprotection than on the mechanism of
initial injury caused by DCVC. In addition, the priming injury (not innocuous, as it caused
25-50% necrosis and elevated blood urea nitrogen) may have influenced the toxicokinetics of
the second DCVC injection.
4.4.6.1.3. Role of oxidative metabolites of Trichloroethylene (TCE)
Some investigators (Dow and Green, 2000; Green et al., 2003; Green et al., 1998) have
proposed that TCE nephrotoxicity is related to formic acid formation. They demonstrated that
exposure to either trichloroethanol or trichloroacetic acid causes increased formation and urinary
excretion of formic acid (Green et al., 1998). The formic acid does not come from
trichloroethylene. Rather, trichloroethylene (or a metabolite) has been proposed to cause a
functional depletion of vitamin B12, which is required for the methionine salvage pathway of
folate metabolism. Vitamin Bi2 depletion results in folate depletion. Folate is a cofactor in
one-carbon metabolism and depletion of folate allows formic acid to accumulate, and then to be
excreted in the urine (Dow and Green, 2000).
TCE (1 and 5 g/L), TCA (0.25, 0.5 and 1 g/L) and TCOH (0.5 and 1.0 g/L) exposure in
male Fischer rats substantially increased excretion of formic acid in urine, an effect suggested as
a possible explanation for TCE-induced renal toxicity in rats (Green et al., 1998). Green et al.
(2003) reported tubular toxicity as a result of chronic (1 year) exposure to TCOH (0, 0.5, and
1.0 g/L). Although TCOH causes tubular degeneration in a similar region of the kidney as TCE,
there are several dissimilarities between the characteristics of nephrotoxicity between the
two compounds, as summarized in Table 4-53. In particular, Green et al. (1998) did not observe
TCOH causing karyomegaly and cytomegaly. These effects were seen as early as 13 weeks after
the commencement of TCE exposure (NTP, 1990), with 300 ppm inhalation exposures to TCE
(Maltoni et al., 1988), as well as at very low chronic exposures to DCVC (Jaffe et al., 1984;
TERRACINI and PARKER, 1965). In addition, Green et al. (2003) reported neither flattening
nor loss of the tubular epithelium nor hyperplasia, but suggested that the increased early
basophilia was due to newly divided cells, and therefore, represented tubular regeneration in
response to damage. Furthermore, they noted that such changes were seen with the spontaneous
damage that occurs in aging rats. However, several of the chronic studies of TCE noted that the
TCE-induced damage observed was distinct from the spontaneous nephropathy observed in rats.
A recent in vitro study of rat hepatocytes and primary human renal proximal tubule cells from
two donors measured formic acid production following exposure to CH (0.3-3 mM, 3-10 days)
(Lock et al., 2007). This study observed increased formic acid production at day 10 in both
This document is a draft for review purposes only and does not constitute Agency policy.
241 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
human renal proximal tubule cell strains, but a similar level of formic acid was measured when
CH was added to media alone. The results of this study are limited by the use of only two
primary human cell strains, but suggest exposure to CH does not lead to significant increases in
formic acid production in vivo.
Interestingly, it appears that the amount of formic acid excreted reaches a plateau at a
relatively low dose. Green et al. (2003) added folic acid to the drinking water of the group of
rats receiving the lower dose of TCOH (18.3 mg/kg-day) in order to modulate the excretion of
formic acid in that dose group, and retain the dose-response in formic acid excretion relative to
the higher-dose group (54.3 mg/kg-day). These doses of TCOH are much lower than what
would be expected to be formed in vivo at chronic gavage doses. For instance, after a single
500-mg/kg dose of TCE (the lower daily dose in the NTP rat chronic bioassays), Green and
Prout (1985) reported excretion of about 41% of the TCE gavage dose in urine as TCOH or
trichloroethanol-glucuronide conjugate (TCOG) in 24 hours. Thus, using the measure of
additional excretion after 24 hours and the TCOH converted to TCA as a lower bound as to the
amount of TCOH formed by a single 500 mg/kg dose of TCE, the amount of TCOH would be
about 205 mg/kg, almost fourfold greater than the high dose in the Green et al. (2003) study. By
contrast, these TCOH doses are somewhat smaller than those expected from the inhalation
exposures of TCE. For instance, after 6 hour exposure to 100 and 500 ppm TCE (similar to the
daily inhalation exposures in Maltoni et al. (1988)), male rats excreted 1.5 and 4.4 mg of TCOH
over 48 hours, corresponding to 5 and 15 mg/kg for a rat weighing 0.3 kg (Kaneko et al., 1994a).
The higher equivalent TCOH dose is similar to the lower TCOH dose used in Green et al.
(2003), so it is notable that while Maltoni et al. (1988) reported a substantial incidence of
cytomegaly and karyomegaly after TCE exposure (300 and 600 ppm), none was reported in
Green et al. (2003).
TCOH alone does not appear sufficient to explain the range of renal effects observed
after TCE exposure, particularly cytomegaly, karyomegaly, and flattening and dilation of the
tubular epithelium. However, given the studies described above, it is reasonable to conclude that
TCOH may contribute to the nephrotoxicity of TCE, possibly due to excess formic acid
production, because (1) there are some similarities between the effects observed with TCE and
TCOH and (2) the dose at which effects with TCOH are observed overlap with the approximate
equivalent TCOH dose from TCE exposure in the chronic studies.
Dow and Green (2000) noted that TCA also induced formic acid accumulation in rats,
and suggested that TCA may therefore, contribute to TCE-induced nephrotoxicity. However,
TCA has not been reported to cause any similar histologic changes in the kidney. Mather et al.
(1990) reported an increase of kidney-weight to body-weight ratio in rats after 90 days of
This document is a draft for review purposes only and does not constitute Agency policy.
242 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
exposure to trichloroacetic acid in drinking water at 5,000 ppm (5 g/L) but reported no
histopathologic changes in the kidney. DeAngelo et al. (1997) reported no effects of
trichloroacetic acid on kidney weight or histopathology in rats in a 2-year cancer bioassay.
Dow and Green (2000) administered TCA at quite high doses (1 and 5 g/L in drinking water),
greater than the subsequent experiments of Green et al. (2003) with TCOH (0.5 and 1 g/L in
drinking water), and reported similar amounts of formic acid produced (about 20 mg/day for
each compound). However, cytotoxicity or karyomegaly did not appear to be analyzed.
Furthermore, much more TCOH is formed from TCE exposure than TCA. Therefore, if TCA
contributes substantially to the nephrotoxicity of TCE, its contribution would be substantially
less than that of TCOH. Lock et al. (2007) also measured formic acid production in human renal
proximal tubule cells exposed to 0.3-3 mM CH for 10 days CH. This study measured
metabolism of CH to TCOH and TCA as well as formic acid production and subsequent
cytotoxicity. Increased formic acid was not observed in this study, and limited cytotoxicity was
observed. However, this study was performed in human renal proximal tubular cells from only
two donors, and there is potential for large interindividual variability in response, particularly
with CYP enzymes.
In order to determine the ability of various chlorinated hydrocarbons to induce
peroxisomal enzymes, Goldsworthy and Popp (1987) exposed male Fischer 344 rats and male
B6C3F1 mice to TCE (1,000 mg/kg BW) and TCA (500 mg/kg BW) by corn oil gavage for
10 consecutive days. Peroxisomal activation was measured by palmitoyl coenzyme A (CoA)
oxidase activity levels. TCE led to increased peroxisomal activation in the kidneys of both rats
(300% of control) and mice (625% of control), while TCA led to an increase only in mice (280%
of control). A study by Zanelli et al. (1996) exposed Sprague-Dawley rats to TCA for 4 days and
measured both renal and hepatic peroxisomal and cytochrome P450 enzyme activities.
TCA-treated rats had increased activity in CYP 4A subfamily enzymes and peroxisomal
palmitoyl-CoA oxidase. Both of these acute studies focused on enzyme activities and did not
further analyze resulting histopathology.
4.4.6.1.4. In Vitro Studies of Kidney Toxicity of Trichloroethylene (TCE) and Metabolites
Generally, it is believed that TCE metabolites are responsible for the bulk of kidney
toxicity observed following exposure. In particular, studies have demonstrated a role for DCVG
and DCVC in kidney toxicity, though, as discussed in Section 3.3.3.2, the precise metabolic yield
of these metabolites following TCE exposure remains uncertain. The work by Lash and
colleagues (Cummings and Lash, 2000; Cummings et al., 2000b; Cummings et al., 2000c; Lash
This document is a draft for review purposes only and does not constitute Agency policy.
243 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
et al., 2000b) examined the effect of trichloroethylene and its metabolites in vitro.
Trichloroethylene and DCVC are toxic to primary cultures of rat proximal and distal tubular cells
(Cummings et al., 2000a) while the TCE metabolites DCVG and DCVC have been demonstrated
to be cytotoxic to rat and rabbit kidney cells in vitro (Groves et al., 1993; Hassall et al., 1983;
Wolfgang et al., 1989a) Lash et al., 2000b, 2001b. Glutathione-related enzyme activities were
well maintained in the cells, whereas CYP activities were not. The enzyme activity response to
DCVC was greater than the response to trichloroethylene; however, the proximal and distal
tubule cells had similar responses even though the proximal tubule is the target in vivo. The
authors attributed this to the fact that the proximal tubule is exposed before the distal tubule in
vivo and to possible differences in uptake transporters. They did not address the extent to which
transporters were maintained in the cultured cells.
In further studies, Lash et al. (2001b) assessed the toxicity of trichloroethylene and its
metabolites DCVC and DCVG using in vitro techniques (Lash et al., 2001b) as compared to in
vivo studies. Experiments using isolated cells were performed only with tissues from
Fischer 344 rats, and lactate dehydrogenase release was used as the measure of cellular toxicity.
The effects were greater in males. DCVC and trichloroethylene had similar effects, but DCVG
exhibited increased efficacy compared with trichloroethylene and DCVC.
In vitro mitochondrial toxicity was assessed in renal cells from both Fischer 344 rats and
B6C3F1 mice following exposure to both DCVC and DCVG (Lash et al., 2001b). Renal
mitochondria from male rats and mice responded similarly; a greater effect was seen in cells
from the female mice. These studies show DCVC to be slightly more toxic than
trichloroethylene and DCVG, but species differences are not consistent with the effects observed
in long-term bioassays. This suggests that in vitro data be used with caution in risk assessment,
being mindful that in vitro experiments do not account for in vivo pharmacokinetic and
metabolic processes.
In LLC-PK1 cells, DCVC causes loss of mitochondrial membrane potential,
mitochondrial swelling, release of cytochrome c, caspase activation, and apoptosis (Chen et al.,
2001). Thus, DCVC is toxic to mitochondria, resulting in either apoptosis or necrosis.
DCVC-induced apoptosis also has been reported in primary cultures of human proximal tubule
cells (Lash et al., 2001a).
DCVC was further studied in human renal proximal tubule cells for alterations in gene
expression patterns related to proposed modes of action in nephrotoxicity (Lock et al., 2006). In
cells exposed to subtoxic levels of DCVC to better mimic workplace exposures, the expression
of genes involved with apoptosis (caspase 8, FADD-like regulator) was increased at the higher
dose (1 |iM) but not at the lower dose (0.1 (xM) of DCVC exposure. Genes related to oxidative
This document is a draft for review purposes only and does not constitute Agency policy.
244 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
stress response (SOD, NF-kB, p53, c-Jun) were altered at both subtoxic doses, with genes
generally upregulated at 0.1 |iM DCVC being downregulated at 1 [xM DCVC. The results of this
study support the need for further study, and highlight the involvement of multiple pathways and
variability of response based on different concentrations.
Lash et al. (2007) examined the effect of modulation of renal metabolism on toxicity of
TCE in isolated rat cells and microsomes from kidney and liver. Following exposure to
modulating chemicals, lactate dehydrogenase (LDH) was measured as a marker of cytotoxicity,
and the presence of specific metabolites was documented (DCVG, TCA, TCOH, and CH).
Inhibition of the CYP stimulated an increase of GSH conjugation of TCE and increased
cytotoxicity in kidney cells. This modulation of CYP had a greater effect on TCE-induced
cytotoxicity in liver cells than in kidney cells. Increases in GSH concentrations in the kidney
cells led to increased cytotoxicity following exposure to TCE. Depletion of GSH in hepatocytes
exposed to TCE, however, led to an increase in hepatic cytotoxicity. The results of this study
highlight the role of different bioactivation pathways needed in both the kidney and the liver,
with the kidney effects being more affected by the GSH conjugation pathways metabolic
products.
In addition to the higher susceptibility of male rats to TCE-induced
nephrocarcinogenicity and nephrotoxicity, isolated renal cortical cells from male F344 rats are
more susceptible to acute cytotoxicity from TCE than cells from female rats. TCE caused a
modest increase in LDH release from male rat kidney cells but had no significant effect on LDH
release from female rat kidney cells. These results on male susceptibility to TCE agree with the
in vivo data.
4.4.6.1.5. Conclusions as to the Active Agents of Trichloroethylene (TCE)-Induced
Nephrotoxicity
In summary, the TCE metabolites DCVC, TCOH, and TCA have all been proposed as
possible contributors to the nephrotoxicity of TCE. Both in vivo and in vitro data strongly
support the conclusion that DCVC and related GSH conjugation metabolites are the active agents
of TCE-induced nephrotoxicity. Of these, DCVC induces effects in renal tissues, both in vivo
and in vitro, that are most similar to those of TCE, and formed in sufficient amounts after TCE
exposure to account for those effects. A role for formic acid due to TCOH or TCA formation
from TCE cannot be ruled out, as it is known that substantial TCOH and TCA are formed from
TCE exposure, that formic acid is produced from all three compounds, and that TCOH exposure
leads to toxicity in the renal tubules. However, the characteristics of TCOH-induced
This document is a draft for review purposes only and does not constitute Agency policy.
245 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
nephrotoxicity do not account for the range of effects observed after TCE exposure while those
of DCVC-induced nephrotoxicity do. Also, TCOH does not induce the same pathology as TCE
or DCVC. TCA has also been demonstrated to induce peroxisomal proliferation in the kidney
(Goldsworthy and Popp, 1987), but this has not been associated with kidney cancer. Therefore,
although TCOH and possibly TCA may contribute to TCE-induced nephrotoxicity, their
contribution is likely to be small compared to that of DCVC. However, as discussed in
Section 3.3.3.2, the precise metabolic yield of these DCVC following TCE exposure remains
uncertain.
4.4.7. Mode(s) of Action for Kidney Carcinogenicity
This section will discuss the evidentiary support for several hypothesized modes of action
for kidney carcinogenicity, including mutagenicity, cytotoxicity and regenerative proliferation,
peroxisome proliferation, o2\i-related nephropathy and formic acid-related nephropathy,
following the framework outlined in the Cancer Guidelines (U.S. EPA, 2005c, d).6 The data and
conclusions for the MOAs with the greatest experimental support are summarized in Table 4-54.
4.4.7.1.1. Hypothesized Mode of Action: Mutagenicity
One hypothesis is that a mutagenic mode of action is operative in TCE-induced renal
carcinogenesis. According to this hypothesis, the key events leading to TCE-induced kidney
tumor formation constitute the following: TCE GSH conjugation metabolites (e.g., DCVG,
DCVC, NAcDCVC, and/or other reactive metabolites derived from subsequent beta-lyase, flavin
monooxygenases [FMO], or CYP metabolism) derived from the GSH-conjugation pathway, after
being either produced in situ in or delivered systemically to the kidney, cause direct alterations to
DNA (e.g., mutation, DNA damage, and/or micronuclei induction). Mutagenicity is a
well-established cause of carcinogenicity.
6 As recently reviewed (Guvton et al.. 20081 the approach to evaluating mode of action information described in
EPA's Cancer Guidelines (2005c. d) considers the issue of human relevance of a hypothesized mode of action in the
context of hazard evaluation. This excludes, for example, consideration of toxicokinetic differences across species;
specifically, the Cancer Guidelines state, "the toxicokinetic processes that lead to formation or distribution of the
active agent to the target tissue are considered in estimating dose but are not part of the mode of action." In
addition, information suggesting quantitative differences in the occurrence of a key event between test species and
humans are noted for consideration in the dose-response assessment, but is not considered in human relevance
determination. In keeping with these principles, a formal analysis of the dose-response of key events in the
hypothesized modes of action is not presented unless it would aid in the overall weight of evidence analysis for
carcinogenicity, as presented in Section 4.11.
This document is a draft for review purposes only and does not constitute Agency policy.
246 DRAFT—DO NOT CITE OR QUOTE
-------
1
This document is a draft for review purposes only and does not constitute Agency policy.
247 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-54. Summary of major mode of action conclusions for TCE kidney carcinogenesis
Hypothesized MOA
postulated key events
Experimental support
Human relevance
Weight-of-evidence
conclusion
Mutagenicity
GSH conjugation-derived metabolites
produced in situ or delivered
systemically to kidney
Multiple in vitro and in vivo studies demonstrate GSH
conjugation of TCE, and availability to the kidney (see
Section 3.3.3).
Uncertainties are quantitative (precise amount of flux), not
qualitative.
Yes: demonstrated in
humans in vivo and in
human cells in vitro.
Highly likely that both
human and rodent kidneys
are exposed to the GSH-
conjugation derived
metabolites.
GSH conjugation-derived metabolites
cause direct alterations to DNA,
resulting in mutations
GSH conjugation derived metabolites (DCVG, DCVC,
NAcDCVC) demonstrated to be genotoxic in most in vitro assays
in which they have been tested, including Ames test (see
Section 4.2.1.4.1).
Kidney-specific genotoxicity in rats and rabbits after in vivo
administration of TCE or DCVC. Not seen in mice, but may be
due to species differences in metabolism and in sensitivity to renal
carcinogenesis.
Yes: no basis for
discounting in vitro
genotoxicity results.
Inconsistent results with
respect to VHL
mutation and
TCE-induced kidney
tumors.
Predominance of positive
genotoxicity data
consistent with
GSH-conjugation derived
metabolites causing
mutations in the kidney.
Mutations cause cancer
Well established.
Yes: well established.
Mutagenicity is a well-
established cause of
cancer.
Cytotoxicity and regenerative proliferation
GSH conjugation-derived metabolites
produced in situ or delivered
systemically to kidney
Multiple in vitro and in vivo studies demonstrate GSH
conjugation of TCE, and availability to the kidney (see
Section 3.3.3).
Uncertainties are quantitative (precise amount of flux), not
qualitative.
Yes: demonstrated in
humans in vivo and in
human cells in vitro.
Highly likely that both
human and rodent kidneys
are exposed to the GSH-
conjugation derived
metabolites.
-------
Table 4-54. Summary of major mode of action conclusions for TCE kidney carcinogenesis (continued)
Hypothesized MOA
postulated key events
Experimental support
Human relevance
Weight-of-evidence
conclusion
Mutagenicity
GSH conjugation-derived metabolites
cause cell death (cytotoxicity) in the
kidney.
Multiple human and laboratory animal studies demonstrating TCE
to be nephrotoxic, including chronic studies (see Sections 4.4.1
and 4.4.4).
Multiple laboratory animal studies and in vitro studies in rat and
human kidney cells demonstrating DCVC to be nephrotoxic (see
Sections 4.4.6.1.1 and 4.4.6.2).
Some evidence that TCOH is nephrotoxic, but histological
changes caused by TCE more similar to those caused by DCVC
(see Section 4.4.6.1 and 4.4.6.3).
Yes: demonstrated
human nephrotoxicity
of TCE in vivo and
DCVC in vitro.
TCE is nephrotoxic in
humans, and DCVC is
likely the predominant
moeity responsible.
Cell proliferation increases in the
kidney to repair damage.
No data specific to TCE or GSH conjugation-derived metabolites.
Yes
Increased cell turnover increases the
rate of mutations.
No data specific to TCE or GSH conjugation-derived metabolites.
Yes
Increased proliferation cause clonal
expansion of initiated (premalignant)
cells.
No increase in preneoplastic or neoplastic lesions in Eker rats
exposed to TCE for 13 wk, but no data on longer durations or
from other rat strains sensitive to TCE renal carcinogenesis.
Yes
Increased number of mutations and/or
initiated cells causes cancer.
Well established.
Yes: well established.
Increases in mutations and
in initiated cells are well-
established causes of
cancer.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
4.4.7.1.2. Experimental support for the hypothesized mode of action
Evidence for the hypothesized mode of action for TCE includes (1) the formation of
GSH-conjugation pathway metabolites in the kidney demonstrated in TCE toxicokinetics studies;
and (2) the genotoxicity of these GSH-conjugation pathway metabolites demonstrated in most
existing in vitro and in vivo assays of gene mutations (i.e., Ames test) and in assays of
unscheduled DNA synthesis, DNA strand breaks, and micronuclei using both "standard" systems
and renal cells/tissues.7 Additional relevant data come from analyses of VHL mutations in
human kidney tumors and studies using the Eker rat model. These lines of evidence are
elaborated below.
Toxicokinetic data are consistent with these genotoxic metabolites either being delivered
to or produced in the kidney. As discussed in Section 3, following in vivo exposure to TCE, the
metabolites DCVG, DCVC, and NAcDCVC have all been detected in the blood, kidney, or urine
of rats, and DCVG in blood and NAcDCVC in urine have been detected in humans (Bernauer et
al., 1996; Birner et al., 1993; Lash et al., 1999a; 2006). In addition, in vitro data have shown
DCVG formation from TCE in cellular and subcellular fractions from the liver, from which it
would be delivered to the kidney via systemic circulation, and from the kidney (see
Tables 3-23-3-24, and references therein). Furthermore, in vitro data in both humans and
rodents support the conclusion that DCVC is primarily formed from DCVG in the kidney itself,
with subsequent in situ transformation to NAcDCVC by iV-Acetyl transferase or to reactive
metabolites by beta-lyase, FMO, or CYPs (see Sections 3.3.3.2.2-3.3.3.2.5). Therefore, it is
highly likely that both human and rodent kidneys are exposed to these TCE metabolites.
As discussed in Section 4.2.1.4.2, DCVG, DCVC, and NAcDCVC have been
demonstrated to be genotoxic in most available in vitro assays.8 In particular, DCVC was
7 The EPA Cancer Guidelines (2005c. d) note reliance on "evaluation of in vivo or in vitro short-term testing results
for genetic endpoints" and evidence that "the carcinogen or a metabolite is DNA-reactive and/or has the ability to
bind to DNA"as part of this weight of evidence supporting a mutagenic mode of action. While evidence from
hypothesis-testing experiments that mutation is an early step in the carcinogenic process is considered if available, it
is not required for determination of a mutagenic mode of action; rather, reliance on short-term genotoxicity tests is
emphasized. Thus, such tests are the focus of this analysis, which also includes an analysis of other available data
from humans and animals. In keeping with these principles, a formal analysis of the temporal concordance of key
events in the hypothesized modes of action is not presented unless it would aid in the overall weight of evidence
analysis for carcinogenicity, as presented in Section 4.11.
8 Evaluation of genotoxicity data entails a weight of evidence approach that includes consideration of the various
types of genetic damage that can occur. In acknowledging that genotoxicity tests are by design complementary
evaluations of different mechanisms of genotoxicity, a recent IPCS publication (Eastmond et al.. 20091 notes that
"multiple negative eresults may not be sufficient to remove concern for mutagenicity raised by a clear positive result
in a single mutagenicity assay.". These considerations inform the present approach. In addition, consistent with
EPA's Cancer Guidelines (2005c. d), the approach does not address relative potency (e.g., among TCE metabolites,
or of such metabolites with other known genotoxic carcinogens) per se. nor does it consider quantitative issues
This document is a draft for review purposes only and does not constitute Agency policy.
250 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
mutagenic in the Ames test in three of the tested strains of S. typhimurium (TA100, TA2638,
TA98) (Dekant et al., 1986c; Vamvakas et al., 1988b) and caused dose-dependent increases in
unscheduled DNA synthesis in the two available assays: porcine kidney tubular epithelial cell
line (Vamvakas et al., 1996) and Syrian hamster embryo fibroblasts (Vamvakas et al., 1988a).
DCVC has also been shown to induce DNA strand breaks in both available studies (Jaffe et al.,
1985; Robbiano et al., 2004), and induce micronucleus formation in primary kidney cells from
rats and humans (Robbiano et al., 2004) but not in Syrian hamster embryo fibroblasts (Vamvakas
et al., 1988a). Only one study each is available for DCVG and A-AcDCVC, but notably both
were positive in the Ames test (Vamvakas et al., 1987; 1988b). Although the number of test
systems was limited, these results are consistent.
These in vitro results are further supported by studies reporting kidney-specific
genotoxicity after in vivo administration of TCE or DCVC. In particular, Robbiano et al. (1998)
reported increased numbers of micronucleated cells in the rat kidney following oral TCE
exposure. Oral exposure to DCVC in both rabbits (Jaffe et al., 1985) and rats (Clay, 2008)
increased DNA strand breaks in the kidney. However, in one inhalation exposure study in rats,
TCE did not increase DNA breakage in the rat kidney, possibly due to study limitations (limited
exposure time [6 hours/day for only 5 d] and small number of animals exposed [n = 5] ; Clay,
(2008). One study of TCE exposure in the Eker rat, a rat model heterozygous for the tumor
suppressor gene Tsc-2, reported no significant increase in kidney tumors as compared to controls
(Mally et al., 2006). Inactivation of Tsc-2 in this rat model is associated with spontaneous renal
cell carcinoma with activation of pathways similar to that of VHL inactivation in humans (Liu et
al., 2003). TCE exposure for 13-weeks (corn oil gavage) led to increased nephrotoxicity but no
significant increases in preneoplastic or neoplastic lesions as compared to controls (Mally et al.,
2006). This lack of increased incidence of neoplastic or preneoplastic lesions reported by Mally
et al. (2006) in the tumor-prone Eker rat is similar to lack of significant short-term response
exhibited by other genotoxic carcinogens in the Eker rat (Morton et al., 2002; Stemmer et al.,
2007) and may be related to the increased background rate of renal carcinomas in this animal
model. Mally et al. (2006) also exposed primary kidney epithelial cells from the Eker rat to
DCVC in vitro and demonstrated increased transformation similar to that of other renal
carcinogens (Horesovsky et al., 1994).
As discussed in Section 4.2.1.4.1, although Douglas et al. (1999) did not detect increased
mutations in the kidney of lacZ transgenic mice exposed to TCE for 12 days, these results are not
related to the probable production of these metabolites in vivo. Instead, the analysis of genetic toxicity data
presented in Section 4.2 and summarized here focuses on the identification of a genotoxic hazard of these
metabolites; a quantitative analysis of TCE metabolism to reactive intermediates, via PBPK modeling, is presented
in Section 3.5.
This document is a draft for review purposes only and does not constitute Agency policy.
251 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
highly informative as to the role of mutagenicity in TCE-induced kidney tumors, given the
uncertainties in the production in genotoxic GSH conjugation metabolites in mice and the low
carcinogenic potency of TCE for kidney tumors in rodents relative to what is detectable in
experimental bioassays. Limited, mostly in vitro, toxicokinetic data do not suggest mice have
less GSH conjugation or subsequent renal metabolism/bioactivation (see Section 3.3.3.2.7), but
quantitatively, the uncertainties in the flux through these pathways remain significant (see
Section 3.5). In additional, similar to other genotoxic renal carcinogens analyzed by NTP, there
is limited evidence of mouse kidney tumors following TCE exposure. However, given the
already low incidences of kidney tumors observed in rats, a relatively small difference in potency
in mice would be undetectable in available chronic bioassays. Notably, of seven chemicals
categorized as direct-acting genotoxic carcinogens that induced rat renal tumors in NTP studies,
only two also led to renal tumors in the mouse (tris[2,3-dibromopropyl]phosphate and
ochratoxin A) (Kanisawa and Suzuki, 1978; Reznik et al., 1979), so the lack of detectable
response in mouse bioassays does not preclude a genotoxic MOA.
VHL inactivation (via mechanisms such as deletion, silencing or mutation) observed in
human renal clear cell carcinomas, is the basis of a hereditary syndrome of kidney cancer
predisposition, and is hypothesized to be an early and causative event in this disease (e.g., 2008).
Therefore, specific actions of TCE metabolites that produce or select for mutations of the VHL
suppressor gene could lead to kidney tumorigenesis. Several studies have compared VHL
mutation frequencies in cases with TCE exposures with those from control or background
populations. Briining et al. (1997b) and Brauch et al. (1999; 2004) reported differences between
TCE-exposed and nonexposed renal cell carcinoma patients in the frequency of somatic VHL
mutations, the incidence of a hot spot mutation of cytosine to thymine at nucleotide 454, and the
incidence of multiple mutations. These data suggest that kidney tumor genotype data in the form
of a specific mutation pattern may potentially serve to discriminate TCE-induced tumors from
other types of kidney tumors in humans. If validated, this would also suggest that TCE-induced
kidney tumors are dissimilar from those occurring in unexposed individuals. Thus, while not
confirming a mutation MOA, these data suggest that TCE-induced tumors may be distinct from
those induced spontaneously in humans. However, it has not been examined whether a possible
linkage exists between VHL loss or silencing and mutagenic TCE metabolites.
By contrast, Schraml et al. (1999) and Charbotel et al. (2007) reported that TCE-exposed
renal cell carcinoma patients did not have significantly higher incidences of VHL mutations
compared to nonexposed patients. However, details as to the exposure conditions were lacking
in Schraml et al. (1999). In addition, the sample preparation methodology employed by
Charbotel et al. (2007) and others (Brauch et al., 1999; Briining et al., 1997b) often results in
This document is a draft for review purposes only and does not constitute Agency policy.
252 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
poor quality and/or low quantity DNA, leading to study limitations (less than 100% of samples
were able to be analyzed). Therefore, further investigations are necessary to either confirm or
contradict the validity of the genetic biomarkers for TCE-related renal tumors reported by
Briining et al. (1997b) and Brauch et al. (1999; 2004).
In addition, while exposure to mutagens is certainly associated with cancer induction (as
discussed with respect to the liver in Appendix E, Sections E.3.1 and E.3.2), examination of
end-stage tumor phenotype or genotype has limitations concerning determination of early key
events. The mutations that are observed with the progression of neoplasia are associated with
increased genetic instability and an increase in mutation rate. Further, inactivation of the VHL
gene also occurs through other mechanisms in addition to point mutations, such as loss of
heterozygosity or hypermethylation (Kenck et al., 1996; Nickerson et al., 2008) not addressed in
these studies. Recent studies examining the role of other genes or pathways suggest roles for
multiple genes in renal cell carcinoma development (Furge et al., 2007; Toma et al., 2008).
Therefore, the inconsistent results with respect to VHL mutation status do not constitute negative
evidence for a mutational MOA and the positive studies are suggestive of a TCE-induced kidney
tumor genotype.
In sum, the predominance of positive genotoxicity data in the database of available
studies of TCE metabolites derived from GSH conjugation (in particular the evidence of
kidney-specific genotoxicity following in vivo exposure to TCE or DCVC), coupled with the
toxicokinetic data consistent with the in situ formation of these GSH-conjugation metabolites of
TCE in the kidney, is consistent with the hypothesis that a mutagenic MOA is operative in
TCE-induced kidney tumors. Mutagenicity is a well-established cause of carcinogenicity.
Available data on the VHL gene in humans add biological plausibility to these conclusions.
Quantitatively, however, as discussed in Section 3.3.3.2, the precise metabolic yield of the GSH
conjugation metabolites following TCE exposure remains uncertain.
4.4.7.1.3. Hypothesized Mode of Action: Cytotoxicity and Regenerative Proliferation
Another hypothesis is that TCE acts by a cytotoxicity mode of action in TCE-induced
renal carcinogenesis. According to this hypothesis, the key events leading to TCE-induced
kidney tumor formation comprise the following: the TCE GSH-conjugation metabolite DCVC,
after being either produced in situ in or delivered systemically to the kidney, causes cytotoxicity,
leading to compensatory cellular proliferation and subsequently increased mutations and clonal
expansion of initiated cells.
This document is a draft for review purposes only and does not constitute Agency policy.
253 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
4.4.7.1.4. Experimental support for the hypothesized mode of action
Evidence for the hypothesized MOA consist primarily of (1) the demonstration of
nephrotoxicity following TCE exposure at current occupational limits in human studies and
chronic TCE exposure in animal studies; (2) the relatively high potential of the TCE metabolite
DCVC to cause nephrotoxicity; and (3) toxicokinetic data demonstrating that DCVC is formed in
the kidney following TCE exposure. Data on nephrotoxicity of TCE and DCVC are discussed in
more detail below, while the toxicokinetic data were summarized previously in the discussion of
mutagenicity. Thus, the data are consistent with the hypothesized MOA, and therefore do not
rule out a contribution from cytotoxicity and regenerative proliferation to TCE-induced kidney
carcinogenesis. However, there are a lack of experimental data supporting a causal link between
TCE nephrotoxicity combined with sustained cellular proliferation and TCE-induced
nephrocarcinogeni city.
There is substantial evidence that TCE is nephrotoxic in humans and laboratory animals
and that its metabolite DCVC is nephrotoxic in laboratory animals. Epidemiological studies
have consistently demonstrated increased excretion of nephrotoxicity markers (NAG, protein,
albumin) at occupational (Green et al., 2004) and higher (Bolt et al., 2004; Briining et al., 1999a;
1999b) levels of TCE exposure. However, direct evidence of tubular toxicity, particularly in
renal cell carcinoma cases, is not available. These studies are supported by the results of
multiple laboratory animal studies. Chronic bioassays have reported very high (nearly 100%)
incidences of nephrotoxicity of the proximal tubule in rats (NTP, 1988, 1990) and mice (NCI,
1976; NTP, 1990) at the highest doses tested. In vivo studies examining the effect of TCE
exposure on nephrotoxicity showed increased proximal tubule damage following intraperitoneal
injection and inhalation of TCE in rats (Chakrabarti and Tuchweber, 1988) and intraperitoneal
injection in mice (Cojocel et al., 1989). Studies examining DCVC exposure in rats (Elfarra et
al., 1986; TERRACINI and PARKER, 1965) and mice (Darnerud et al., 1989; Jaffe et al., 1984)
have also shown increases in kidney toxicity. The greater potency for kidney cytotoxicity for
DCVC compared to TCE was shown by in vitro studies (Lash et al., 1986; 1995; Stevens et al.,
1986). These studies also further confirmed the higher susceptibility of male rats or mice to
DCVC-induced cytotoxicity. Cytokaryomegaly (an effect specific to TCE and not part of the
chronic progressive nephropathy or the pathology that occurs in aging rat kidneys) was observed
in the majority of rodent studies and may or may not progress to carcinogenesis. Finally, as
discussed extensively in Section 4.4.6.1, a detailed comparison of the histological changes in the
kidney caused by TCE and its metabolites supports the conclusion that DCVC is the predominant
moiety responsible for TCE-induced nephrotoxicity.
This document is a draft for review purposes only and does not constitute Agency policy.
254 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Because it is known that not all cytotoxins are carcinogens (i.e., cytotoxicity is not a
specific predictor of carcinogenicity), additional experimental support is required to causally link
nephrotoxicity to nephrocarcinogenicity. For chemicals that bind to a2|i-globulin, a MOA
involving cell necrosis followed by subsequent regenerative proliferation has been hypothesized
to cause kidney tumors in the absence of genotoxicity (Short, 1993). However, for other
chemicals, toxicity and increased cell proliferation have been observed in the kidney without
inducing tumors even after chronic exposure (Tennant et al., 1991). Similarly, in the liver,
partial hepatectomy leading to regenerative hyperplasia does not by itself lead to increased
hepatocarcinogenicity, and requires administration of a mutagen to exhibit enhanced
carcinogenic effects. By analogy, a biologically plausible MOA may involve a combination of
mutagenicity and cytotoxicity, with mutagenicity increasing the rate of mutation and
regenerative proliferation induced by cytotoxicity enhancing the selection, survival or clonal
expansion of mutated cells.
For TCE and kidney cancer, clearly, cytotoxicity occurs at doses below those causing
carcinogenicity, as the incidence of nephrotoxicity in chronic bioassays is an order of magnitude
higher than that of renal tumors. Thus, these data are consistent with cytotoxicity being a
precursor to carcinogenicity (i.e., if the opposite were the case—carcinogenicity without
cytoxicity—then the hypothesis would be falsified). While chronic nephrotoxicity was reported
in the same bioassays showing increased kidney tumor incidences, the use of such data to inform
MOA is indirect and associative, and do not offer a test of the hypothesis (Short, 1993).
Nephrotoxicity is observed in both mice and rats, in some cases with nearly 100% incidence in
all dose groups, but kidney tumors are only observed at low incidences in rats at the highest
tested doses (NCI, 1976; NTP, 1990). Therefore, nephrotoxicity alone appears to be insufficient,
or at least not rate-limiting, for rodent renal carcinogenesis, since maximal levels of toxicity are
reached before the onset of tumors. Furthermore, there are multiple mechanisms by which TCE
has been hypothesized to induce cytotoxicity, including oxidative stress, disturbances in calcium
ion homeostasis, mitochondrial dysfunction, and protein alkylation (Lash et al., 2000b). Some of
these effects may therefore, have ancillary consequences related to tumor induction which are
independent of cytotoxicity per se. Therefore, data currently cannot distinguish as to whether
cytoxicity is causally related to tumorigenesis or merely associated by virtue of being a marker
for a different, key causal event.
Under the hypothesized MOA, cytotoxicity leads to the induction of repair processes and
compensatory proliferation that could lead to an increased production or clonal expansion of
cells previously initiated by mutations occurred spontaneously, from coexposures, or from TCE
or its metabolites. Data on compensatory cellular proliferation and the subsequent hypothesized
This document is a draft for review purposes only and does not constitute Agency policy.
255 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
key events in the kidney are few, with no data from rat strains used in chronic bioassays. In rats
carrying the Eker mutation, Mally et al. (2006) reported increased DNA synthesis as measured
by BrdU incorporation in animals exposed to the high dose of TCE (1,000 mg/kg-day) for
13 weeks, but there was no evidence of clonal expansion or tumorigenesis in the form of
increased preneoplastic or neoplastic lesions as compared to controls. Therefore, in both rodent
and human studies of TCE, data demonstrating a causal link between compensatory proliferation
and the induction of kidney tumors are lacking.
In sum, the predominance of positive nephrotoxicity data in the database of available
studies of TCE metabolites derived from GSH conjugation (in particular the evidence of
kidney-specific cytotoxicity following in vivo exposure to TCE or DCVC), coupled with the
toxicokinetic data consistent with the in situ formation of these GSH-conjugation metabolites of
TCE in the kidney, is consistent with the hypothesis that a MOA involving cytotoxicity and
regenerative proliferation contributes TCE-induced kidney tumors, either independently or in
combination with a mutagenic MOA. However, nephrotoxicity is not in itself predictive of
tumorigenesis, and experimental data supporting for a causal link between TCE nephrotoxicity
combined with sustained cellular proliferation and TCE-induced nephrocarcinogenicity are
lacking. A more biologically plausible MOA may involve a combination of mutagenicity and
cytotoxicity, with mutagenicity increasing the rate of mutation and regenerative proliferation
enhancing the selection, survival or clonal expansion of mutated cells. However, this hypothesis
has yet to be tested experimentally.
4.4.7.1.5. Additional Hypothesized Modes of Action with Limited Evidence or Inadequate
Experimental Support
Along with metabolites derived from GSH conjugation of TCE, oxidative metabolites are
also present and could induce toxicity in the kidney. After TCE exposure, the oxidative
metabolite and peroxisome proliferator TCA is present in the kidney and excreted in the urine as
a biomarker of exposure. Hypotheses have also been generated regarding the roles of
a.2|i-globulin or formic acid in nephrotoxicity induced by TCE oxidative metabolites TCA or
TCOH. However, the available data are limited or inadequate for supporting these hypothesized
MOAs.
This document is a draft for review purposes only and does not constitute Agency policy.
256 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
4.4.7.1.6. Peroxisome proliferation
Although not as well studied as the effects of glutathione metabolites in the kidney, there
is evidence that oxidative metabolites affect the kidney after TCE exposure. Both TCA and
DCA are peroxisome proliferator activated receptor alpha (PPARa) agonists although most
activity has been associated with TCA production after TCE exposure. Exposure to TCE has
been found to induce peroxisome proliferation not only in the liver but also the kidney.
Peroxisome proliferation in the kidney has been evaluated by only one study of TCE
(Goldsworthy and Popp, 1987), using increases in cyanide-insensitive palmitoyl-CoA oxidation
(PCO) activity as a marker. Increases in renal PCO activity were observed in rats (3.0-fold) and
mice (3.6-fold) treated with TCE at 1,000 mg/kg-day for 10 days, with smaller increases in both
species from TCA treatment at 500 mg/kg-day for 10 days. However, no significant increases in
kidney/body weight ratios were observed in either species. There was no relationship between
induction of renal peroxisome proliferation and renal tumors (i.e., a similar extent of peroxisome
proliferation-associated enzyme activity occurred in species with and without TCE-induced renal
tumors). However, the increased peroxisomal enzyme activities due to TCE exposure are
indicative of oxidative metabolites being present and affecting the kidney. Such metabolites
have been associated with other tumor types, especially liver, and whether coexposures to
oxidative metabolites and glutathione metabolites contribute to kidney tumorigenicity has not
been examined.
4.4.7.1.7. a2ji-Globulin-related nephropathy
Induction of a.2|i-globulin nephropathy by TCE has been investigated by Goldsworthy
et al. (1988), who reported that TCE did not induce increases in this urinary protein, nor did it
stimulate cellular proliferation in rats. In addition, whereas kidney tumors associated with
a.2|i-globulin nephropathy are specific to the male rat, as discussed above, nephrotoxicity is
observed in both rats and mice and kidney tumor incidence is elevated (though not always
statistically significant) in both male and female rats. TCOH was recently reported to cause
hyaline droplet accumulation and an increase in a2|i-globulin, but these levels were insufficient
to account for the observed nephropathy as compared to other exposures (Green et al., 2003).
Therefore, it is unlikely that a2|i-globulin nephropathy contributes significantly to TCE-induced
renal carcinogenesis.
This document is a draft for review purposes only and does not constitute Agency policy.
257 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
4.4.7.1.8. Formic acid-related nephrotoxicity
Another MOA hypothesis proposes that TCE nephrotoxicity is mediated by increased
formation and urinary excretion of formic acid mediated by the oxidative metabolites TCA or
TCOH (Dow and Green, 2000; Green et al., 2003; 1998). The subsequent hypothesized key
events are the same as those for DCVC-induced cytotoxicity, discussed above (see
Section 4.4.7.2). As discussed extensively in Section 4.4.6.1.2, these oxidative metabolites do
not appear sufficient to explain the range of renal effects observed after TCE exposure,
particularly cytomegaly, karyomegaly, and flattening and dilation of the tubular epithelium.
Although TCOH and possibly TCA may contribute to the nephrotoxicity of TCE, perhaps due to
excess formic acid production, these metabolites do not show the same range of cytotoxic effects
observed following TCE exposure (see Table 4-53). Therefore, without specific evidence
linking the specific nephrotoxic effects caused by TCOH or TCA to carcinogenesis, and in light
of the substantial evidence that DCVC itself can adequately account for the nephrotoxic effects
of TCE, the weight of evidence supports a conclusion that cytotoxicity mediated by increased
formic acid production induced by oxidative metabolites TCOH and possibly TCA is not
responsible for the majority of the TCE-induced cytotoxicity in the kidneys, and therefore, would
not be the major contributor to the other hypothesized key events in this MOA, such as
subsequent regenerative proliferation.
4.4.7.1.9. Conclusions About the Hypothesized Modes of Action
4.4.7.1.10. Is the hypothesized mode of action sufficiently supported in the test animals
4.4.7.1.10.1. Mutagenicity
The predominance of positive genotoxicity data in the database of available studies of
TCE metabolites derived from GSH conjugation (in particular the evidence of kidney-specific
genotoxicity following in vivo exposure to TCE or DCVC), coupled with the toxicokinetic data
consistent with the in situ formation of these GSH-conjugation metabolites of TCE in the kidney,
supports the conclusion that a mutagenic MOA is operative in TCE-induced kidney tumors.
This document is a draft for review purposes only and does not constitute Agency policy.
258 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
4.4.7.1.10.2. Cytotoxicity
As reviewed above, in vivo and in vitro studies have shown a consistent nephrotoxic
response to TCE and its metabolites in proximal tubule cells from male rats. Therefore, it has
been proposed that cytotoxicity seen in this region of the kidney is a precursor to carcinogenicity.
Available data are consistent with the hypothesis that a MOA involving cytotoxicity and
regenerative proliferation contributes to TCE-induced kidney tumors, either independently or in
combination with a mutagenic MOA. However, it has not been determined whether tubular
toxicity is a necessary precursor of carcinogenesis, and there is a lack of experimental support for
causal links, such as compensatory cellular proliferation or clonal expansion of initiated cells,
between nephrotoxicity and kidney tumors induced by TCE. Nephrotoxicity alone appears to be
insufficient, or at least not rate-limiting, for rodent renal carcinogenesis, since maximal levels of
toxicity are reached before the onset of tumors. A more biologically plausible MOA may
involve a combination of mutagenicity and cytotoxicity, with mutagenicity increasing the rate of
mutation and regenerative proliferation enhancing the survival or clonal expansion of mutated
cells. However, this hypothesis has yet to be tested experimentally.
4.4.7.1.10.3. Additional hypothese
The kidney is also exposed to oxidative metabolites that have been shown to be
carcinogenic in other target organs. TCA is excreted in kidney after its metabolism from TCE
and also can cause peroxisome proliferation in the kidney, but there are inadequate data to define
a MOA for kidney tumor induction based on peroxisome proliferation. TCE induced little or no
a.2|i-globulin and hyaline droplet accumulation to account for the observed nephropathy, so
available data do not support this hypothesized MOA. The production of formic acid following
exposure to TCE and its oxidative metabolites TCOH and TCA may also contribute to
nephrotoxicity; however, the available data indicate that TCOH and TCA are minor contributors
to TCE-induced nephrotoxicity, and therefore, do not support this hypothesized MOA. Because
these additional MOA hypotheses are either inadequately defined or are not supported by the
available data, they are not considered further in the conclusions below.
This document is a draft for review purposes only and does not constitute Agency policy.
259 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
4.4.7.1.11. 2. Is the hypothesized mode of action relevant to humans
4.4.7.1.11.1. Mutagenicity
The evidence discussed above demonstrates that TCE GSH-conjugation metabolites are
mutagens in microbial as well as test animal species. Therefore, the presumption that they would
be mutagenic in humans. Available data on the VHL gene in humans add biological plausibility
to this hypothesis. The few available data from human studies concerning the mutagenicity of
TCE and its metabolites suggest consistency with this MO A, but are not sufficiently conclusive
to provide direct supporting evidence for a mutagenic MOA. Therefore, this MOA is considered
relevant to humans.
4.4.7.1.11.2. Cytotoxicity
Although data are inadequate to determine that the MOA is operative, none of the
available data suggest that this MOA is biologically precluded in humans. Furthermore, both
animal and human studies suggest that TCE causes nephrotoxicity at exposures that also induce
renal cancer, constituting positive evidence of the human relevance of this hypothesized MOA.
4.4.7.1.12. 3. Which populations or lifestages can be particularly susceptible to the
hypothesized mode of action
4.4.7.1.12.1. Mutagenicity
The mutagenic MOA is considered relevant to all populations and lifestages. According
to EPA's Cancer Guidelines (U.S. EPA, 2005c) and Supplemental Guidance (U.S. EPA, 2005d),
there may be increased susceptibility to early-life exposures for carcinogens with a mutagenic
mode of action. Therefore, because the weight of evidence supports a mutagenic mode of action
for TCE carcinogenicity and in the absence of chemical-specific data to evaluate differences in
susceptibility, early-life susceptibility should be assumed and the age-dependent adjustment
factors (ADAFs) should be applied, in accordance with the Supplemental Guidance.
In addition, because the MOA begins with GSH-conjugation metabolites being delivered
systemically or produced in situ in the kidney, toxicokinetic differences—i.e., increased
production or bioactivation of these metabolites—may render some individuals more susceptible
to this MOA. However, as discussed in Section 3.3.3.2, quantitative estimates of the amount of
This document is a draft for review purposes only and does not constitute Agency policy.
260 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
GSH conjugation following TCE exposure remain uncertain. Toxicokinetic-based susceptibility
is discussed further in Section 4.10.
In rat chronic bioassays, TCE-treated males have higher incidence of kidney tumors than
similarly treated females. However, the basis for this sex-difference is unknown, and whether it
is indicative of a sex difference in human susceptibility to TCE-induced kidney tumors is
likewise unknown. The epidemiologic studies generally do not show sex differences in kidney
cancer risk. Lacking exposure-response information, it is not known if the sex-difference in one
renal cell carcinoma case-control study (Dosemeci et al., 1999) may reflect exposure differences
or susceptibility differences.
4.4.7.1.12.2. Cytotoxicity
Populations which may be more susceptible based on the toxicokinetics of the production
of GSH conjugation metabolites and the sex differences observed in rat chronic bioassays are the
same as for a mutagenic MOA. No data are available as to whether other factors may lead to
different populations or lifestages being more susceptible to a cytotoxic MOA for TCE-induced
kidney tumors. For instance, it is not known how the hypothesized key events in this MOA
interact with known risk factors for human renal cell carcinoma.
The weight of evidence sufficiently supports a mutagenic MOA for TCE in the kidney,
based on supporting data that GSH-metabolites are genotoxic and produced in sufficient
quantities in the kidney to lead to tumorigenesis. Cytotoxicity and regenerative proliferation
were considered as an alternate MOA, however, there are inadequate data to support a causal
association between cytotoxicity and kidney tumors. Further, hypothesized MO As relating to
peroxisomal proliferation, a2[j,-globulin nephropathy and formic acid-related nephrotoxicity
were considered and rejected due to limited evidence and/or inadequate experimental support.
4.4.8. Summary: Trichloroethylene (TCE) Kidney Toxicity, Carcinogenicity, and Mode-
of-Action
Human studies have shown increased levels of proximal tubule damage in workers
exposed to high levels of TCE (NRC, 2006). These studies analyzed workers exposed to TCE
alone or in mixtures and reported increases in various urinary biomarkers of kidney toxicity or
end-stage renal disease (P2-microglublin, total protein, NAG, al-microglobulin) (Bolt et al.,
2004; Briining et al., 1999a; 1999b; Green et al., 2004; Jacob et al., 2007; Nagaya et al., 1989b;
Radican et al., 2006; Selden et al., 1993). Laboratory animal studies examining TCE exposure
This document is a draft for review purposes only and does not constitute Agency policy.
261 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
provide additional support, as multiple studies by both gavage and inhalation exposure show that
TCE causes renal toxicity in the form of cytomegaly and karyomegaly of the renal tubules in
male and female rats and mice. By gavage, incidences of these effects under chronic bioassay
conditions approach 100%, with male rats appearing to be more sensitive than either female rats
or mice of either sex based on the severity of effects. Under chronic inhalation exposures, only
male rats exhibited these effects. Further studies with TCE metabolites have demonstrated a
potential role for DCVC, TCOH, and TCA in TCE-induced nephrotoxicity. Of these, DCVC
induces the renal effects that are most like TCE, and it is formed in sufficient amounts following
TCE exposure to account for these effects.
Kidney cancer risk from TCE exposure has been studied related to TCE exposure in
cohort, case-control and geographical studies. These studies have examined TCE in mixed
exposures as well as alone. Elevated risks are observed in many of the cohort and case-control
studies examining kidney cancer incidence in industries or job titles with historical use of TCE
(see Table 4-39 and 4-40), particularly among subjects ever exposed to TCE (Briining et al.,
2003; Dosemeci et al., 1999; Moore et al., 2010; Raaschou-Nielsen et al., 2003) or subjects with
TCE surrogate for high exposure (Briining et al., 2003; Charbotel et al., 2006; Moore et al.,
2010; Raaschou-Nielsen et al., 2003; Zhao et al., 2005). Greater susceptibility to TCE exposure
and kidney cancer is observed among subjects with a functionally active GSTT polymorphism,
particularly among those with certain alleles in single nucleotide polymorphisms of the cysteine
conjugation P-lyase gene region (Moore et al., 2010). Although there are some controversies
related to deficiencies of the epidemiological studies (Henschler et al., 1995; Vamvakas et al.,
1998), many of these are overcome in later studies (Briining et al., 2003; Charbotel et al., 2006;
Moore et al., 2010). A meta-analysis of the overall effect of TCE exposure on kidney cancer,
additionally, suggests a small, statistically significant increase in risk (RRm = 1.27 95% CI: 1.13,
1.43) with a summary relative risk estimate in the higher exposure group of 1.64, (95% CI: 1. 31,
2.04), robust in sensitivity to alternatives and lacking observed statistical heterogeneity among
17 studies meeting explicitly-defined inclusion criteria.
In vivo laboratory animal studies to date suggest a small increase in renal tubule tumors
in male rats and, to a lesser extent, in female rats, with no increases seen in mice or hamsters.
These results are based on limited studies of both oral and inhalation routes, some of which were
deemed insufficient to determine carcinogenicity based on various experimental issues.
However, because of the rarity of kidney tumors in rodents, the repeatability of this finding
across strains and studies supports their biological significance despite the limitations of
individual studies and relatively small increases in reported tumor incidence.
This document is a draft for review purposes only and does not constitute Agency policy.
262 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Some but not all human studies have suggested a role for VHL mutations in TCE-induced
kidney cancer (Brauch et al., 1999; 2004; Briining et al., 1997b; Charbotel et al., 2007; Schraml
et al., 1999). Certain aspects of these studies may explain some of these discrepant results. The
majority of these studies have examined paraffinized tissue that may lead to technical difficulties
in analysis, as paraffin extractions yield small quantities of often low-quality DNA. The
chemicals used in the extraction process itself may also interfere with enzymes required for
further analysis (PCR, sequencing). Although these studies do not clearly show mutations in all
TCE-exposed individuals, or in fact in all kidney tumors examined, this does not take into
account other possible means of VHL inactivation, including silencing or loss, and other potential
targets of TCE mutagenesis were not systematically examined. A recent study by Nickerson
et al. (2008) analyzed both somatic mutation and promoter hypermethylation of the VHL gene in
cc-RCC frozen tissue samples using more sensitive methods. The results of this study support
the hypothesis that VHL alterations are an early event in clear cell RCC carcinogenesis, but these
alterations may not be gene mutations. No experimental animal studies have been performed
examining vhl inactivation following exposure to TCE, although one in vitro study examined vhl
mutation status following exposure to the TCE-metabolite DCVC (Mally et al., 2006). This
study found no mutations following DCVC exposure, although this does not rule out a role for
DCVC in vhl inactivation by some other method or vhl alterations caused by other TCE
metabolites.
Although not encompassing all of the actions of TCE and its metabolites that may be
involved in the formation and progression of neoplasia, available evidence supports the
conclusion that a mutagenic MOA mediated by the TCE GSH-conjugation metabolites
(predominantly DCVC) is operative in TCE-induced kidney cancer. This conclusion is based on
substantial evidence that these metabolites are genotoxic and are delivered to or produced in the
kidney, including evidence of kidney-specific genotoxicity following in vivo exposure to TCE or
DCVC. Cytotoxicity caused by DCVC leading to compensatory cellular proliferation is also a
potential MOA in renal carcinogenesis. A combination of mutagenicity and cytotoxicity, with
mutagenicity increasing the rate of mutation and regenerative proliferation enhancing the
survival or clonal expansion of mutated cells, while biologically plausible, has yet to be tested
experimentally. The additional MOA hypotheses of peroxisome proliferation, accumulation of
a2[j,-globulin, and cytotoxicity mediated by TCE-induced excess formic acid production are not
supported by the available data.
This document is a draft for review purposes only and does not constitute Agency policy.
263 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
4.5. LIVER TOXICITY AND CANCER
4.5.1. Liver Noncancer Toxicity in Humans
The complex of chronic liver disease is a spectrum of effects and comprises nonalcoholic
fatty liver disease (nonalcoholic steatohepatitis) and cirrhosis, more rare anomalies ones such as
autoimmune hepatitis, primary biliary cirrhosis, and primary sclerosing cholangitis, and
hepatocellular and cholangiocarcinoma (intrahepatic bile duct cancer) (Juran and Lazaridis,
2006). Chronic liver disease and cirrhosis, excluding neoplasia, is the 12th leading cause of death
in the United States in 2005 with 27,530 deaths (Kung et al., 2008) with a morality rate of 9.0
per 100,000 (Jemal et al., 2008).
Eight studies reported on liver outcomes and TCE exposure and are identified in
Table 4-55. Three studies are suggestive of effects on liver function tests in metal degreasers
occupationally exposed to trichloroethylene (Nagaya et al., 1993; Rasmussen et al., 1993b; Xu et
al., 2009). Nagaya et al. (1993) in their study of 148 degreasers in metal parts factories,
semiconductor factors, or other factories, observed total mean serum cholesterol concentration,
mean serum high density lipoprotein-cholesterol (HDL-C) concentrations to increase with
increasing TCE exposure, as defined by U-TTC), although a statistically significant linear trend
was not found. Nagaya et al. (1993) estimated subjects in the low exposure group had TCE
exposure to 1 ppm-, 6-ppm TCE in the moderate exposure group, and 210-ppm TCE in the high
exposure group. No association was noted between serum liver function tests and U-TTC, a
finding not surprising given individuals with a history of hepatobiliary disease were excluded
from this study. Nagaya et al. (1993) follows 13 workers with higher U-TTC concentrations
over a 2-year period; serum HDL-C and two hepatic function enzymes, GGT and aspartate
aminotrasferase (AST) concentrations were highest during periods of high level exposure, as
indicated from U-TTC concentrations. Similarly, in a study of 95 degreasers, 70 exposed to
trichloroethylene exposure and 25 to CFC113 (Rasmussen et al., 1993b), mean serum GGT
concentration for subjects with the highest TCE exposure duration was above normal reference
values and were about threefold higher compared to the lowest exposure group. Rasmussen
et al. (1993b) estimated mean urinary TCE concentration in the highest exposure group as
7.7 mg/L with past exposures estimated as equivalent to 40-60 mg/L. Multivariate regression
analysis showed a small statistically nonsignificant association due to age and a larger effect due
to alcohol abuse that reduced and changed direction of a TCE exposure affect. The inclusion of
CFC113 exposed subjects introduces a downward bias since liver toxicity is not associated with
CFC113 exposure (U.S. EPA, 2008a) and would underestimate any possible TCE effect. Xu
et al. (2009) reported symptoms and liver function tests of 21 metal degreasers with severe
This document is a draft for review purposes only and does not constitute Agency policy.
264 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
hypersensitivity dermatitis (see last paragraph in this section for discussion of other liver effects
in hypersensitivity dermatitis cases). TCE concentration of agent used to clean metal parts
ranged from 10.2-63.5% with workplace ambient monitoring time-weighted-average TCE
"3
concentrations of 18-683 mg/m (3-127 ppm). Exposure was further documented by urinary
TCA levels in 14 of 21 cases above the recommended occupation level of 50 mg/L. The
prevalence of elevated liver enzymes among these subjects was 90% (19 cases) for alanine
aminotrasferase, 86% (18 cases) for asparatate aminotrasferase, and 76% (16 cases) for total
bilirubin (Xu et al., 2009). Two studies provide evidence of plasma or serum bile acids changes
among TCE-exposed degreasers. Neghab et al. (1997) in a small prevalence study of 10 healthy
workers (5 unexposed controls and 5 exposed) observed statistically significantly elevated total
serum bile acids, particularly deoxycholic acid and the subtotal of free bile acids, among TCE
subjects at postexposure compared to their pre-exposure concentrations and serum bile acid
levels correlated well with TCE exposure (r = 0.94). Total serum bile acid concentration did not
change in control subjects between pre- and postexposure, nor did enzyme markers of liver
function in either unexposed or exposed subjects differ between pre and postexposure period.
However, the statistical power of this study is quite limited and the prevalence design does not
include subjects who may have left employment because of possible liver problems. The paper
provides minimal details of subject selection and workplace exposure conditions, except that
pre-exposure testing was carried out on the 1st work day of the week (pre-exposure), repeated
sampling after 2 days (postexposure), and a postexposure 8-hour time-weighted-average TCE
concentration of 9 ppm for exposed subjects; no exposure information is provided for control
subjects. Driscoll et al. (1992) in a study of 22 subjects (6 unexposed and 16 exposed) employed
at a factory manufacturing small appliances reported statistically significant group differences in
logistic regression analyses controlling for age and alcohol consumption in mean fasting plasma
bile acid concentrations. Other indicators of liver function such as plasma enzyme levels were
statistically significant different between exposed and unexposed subjects. Laboratory samples
were obtained at the start of subject's work shift. Exposure data are not available on the
22 subjects and assignment of exposed and unexposed was based on work duties. Limited
Table 4-55. Summary of human liver toxicity studies
Subjects
Effect
Exposure
Reference
This document is a draft for review purposes only and does not constitute Agency policy.
265 DRAFT—DO NOT CITE OR QUOTE
-------
148 male metal degreasers in
metal parts, semiconductor and
other factories
Serum liver function enzyme
(HDL-C, AST, and GGT)
concentrations did not
correlated with TCE exposure
assesses in a prevalence study
but did correlate with TCE
concentration over a 2-yr
follow-up period
U-TTC levels obtained from spot
urine sample obtained during
working h used to assign exposure
category included the following:
High: 209 + 99 mg/g Cr
Medium: 35 +27 mg/g Cr
Low: 5+2 mg/g Cr
Note: this study does not include
an unexposed referent group
Nagaya et al.
(1993)
95 workers (70 TCE exposed,
25 CFC113 exposed) selected
from a cohort of 240 workers at
72 factors engaged in metal
degreasing with chlorinated
solvents
Increased serum GGT
concentration with increasing
cumulative exposure
4 groups (cumulative number of yr
exposed over a working life):
I: 0.6 (0-0.99)
II: 1.9 (1-2.8)
III: 4.4 (2.9-6.7)
IV: 14.4 (6.8-35.6)
Rasmussen
et al. (1993b)
21 metal degreasers with severe
hypersensitivity dermatitis
High prevalence of serum
liver function enzymes above
normal levels: ALT, 19 or
21 cases; AST, 18 of
21 cases, and T-Bili, 16 of
21 cases
TWA mean ambient TCE
concentration occupational setting
of cases, 18 mg/m3-683 mg/m3
14 of 21 cases withU-TCE above
recommended occupational level
of 50 mg/L
Xu et al.
(2009)
Five healthy workers engaged
in decreasing activities in steel
industry and five healthy
workers from clerical section of
same company
Total serum bile acid
concentration increased
between pre- and
postexposure (2-d period)
8-h TWA mean personal air:
8.9 + 3.2 ppmpostexposure
Neghab et al.
(1997)
22 workers at a factory
manufacturing small appliances
Increased in several bile acids
Regular exposure to <5 ppm TCE;
peak exposure for two workers to
>250 00m
Driscoll et al.
(1992)
4,489 males and female
residents from 15 Superfund
site and identified from
ATSDR Trichloroethylene
Exposure Subregistry
Liver problems diagnosed
with past yr
Residency in community with
Superfund site identified with TCE
and other chemicals
Davis et al.,
2005
Case reports from eight
countries of individuals with
idiosyncratic generalized skin
disorders
Hepatitis in 46-94% of cases;
other liver effects includes
hepatomegaly and elevated
liver function enzymes; and
in rare cases, acute liver
failure
If reported, TCE, from <50 mg/m3
to more than 4,000 mg/m3.
Symptoms developed within
2-5 wk of initial exposure, with
some intervals up to 3 mo
Kamijima
et al. (2007)
1
This document is a draft for review purposes only and does not constitute Agency policy.
266 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Table 4-55. Summary of human liver toxicity studies (continued)
Subjects
Effect
Exposure
Reference
Deaths in California between
1979-1981 due to cirrhosis
SMR of 211 (95% CI: 136,
287) for white male sheet
metal workers and
SMR = 174 (95% CI:
150-197) for metal workers
Occupational title on death
certificate
Leigh and
Jiang (1993)
ALT = alanine aminotrasferase.
personal monitoring from other nonparticipating workers at this facility indicated TCE exposure
as low, less than 5 ppm, with occasional peaks over 250 ppm although details are lacking
whether these data represent exposures of study subjects.
Davis et al. 2005 in their analysis of subjects from the TCE subregistry of ATSDR's
National Exposure Registry examined the prevalence of subjects reporting liver problems
(defined as seeking treatment for the problem from a physician within the past year) using rates
for the equivalent health condition from the National Health Interview Survey (a nationwide
multipurpose health survey conducted by the National Center for Health Statistics, Centers for
Disease Control and Prevention). The TCE subregistry is a cohort of exposed persons from
15 sites in 5 states. The shortest time interval from inclusion in the exposure registry and last
follow-up was 5 years for one site and 10 years for seven sites. Excess in past-year liver
disorders relative to the general population persisted for much of the lifetime of follow-up.
SMRs for liver problems were 3rd follow-up, SMR = 2.23 (99% CI: 1.13, 3.92); 4th follow-up,
SMR = 3.25 (99% CI: 1.82, 5.32); and, 5th follow-up, SMR = 2.82 (99% CI: 1.46, 4.89).
Examination by TCE exposure, duration or cumulative exposure to multiple organic solvents did
not show exposure-response patterns. Overall, these observations are suggestive of liver
disorders as associated with potential TCE exposure, but whether TCE caused these conditions is
not possible to determine given the study's limitations. These limitations include a potential for
misclassification bias, the direction of which could dampen observations in a negative direction,
and lack of adjustment in statistical analyses for alcohol consumption, which could bias
observations in a positive direction.
Evaluation in epidemiologic studies of risk factors for cirrhosis other than alcohol
consumption and Hepatitis A, B, and C is quite limited. NRC (2006) cited a case report of
cirrhosis developing in an individual exposed occupationally to TCE for 5 years from a
hot-process degreaser and to 1,1,1-trichloroethane for 3 months thereafter (Thiele et al., 1982).
One cohort study on cirrhosis deaths in California between 1979 and 1981 and occupational risk
This document is a draft for review purposes only and does not constitute Agency policy.
267 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
factors as assessed using job title observed elevated risks with occupational titles of sheet metal
workers and metalworkers and cirrhosis among white males who comprised the majority of
deaths (Leigh and Jiang, 1993). This analysis lacks information on alcohol patterns by
occupational title in addition to specific chemical exposures. Few deaths attributable to cirrhosis
are reported for nonwhite male and for both white and nonwhite female metalworkers with
analyses examining these individuals limited by low statistical power. Some but not all
trichloroethylene mortality studies report risk ratios for cirrhosis (see Table 4-56). A statistically
significant deficit in cirrhosis mortality is observed in three studies (Boice et al., 1999; Boice et
al., 2006b; Morgan et al., 1998) and with risk ratios including a risk of 1.0 in the remaining
studies (ATSDR, 2004; Blair et al., 1989; 1998; Garabrant et al., 1988; Ritz, 1999a). These
results do not rule out an effect of TCE on liver cirrhosis since disease misclassification may
partly explain observations. Available studies are based on death certificates where a high
degree of underreporting, up to 50%, is known to occur (Blake et al., 1988).
A number of case reports exist of liver toxicity including hepatitis accompanying
immune-related generalized skin diseases described as a variation of erythema multiforme,
Stevens-Johnson syndrome, toxic epiderma necrolysis patients, and hypersensitivity syndrome
(Section 4.6.1.2 describes these disorders and evidence on TCE) (Kamijima et al., 2007).
Kamijima et al. (2007) reported hepatitis was seen in 92-94% of cases presenting with an
immune-related generalized skin diseases of variation of erythema multiforme, Stevens-Johnson
syndrome, and toxic epiderma necrolysis patients, but the estimates within the hypersensitivity
syndrome group were more variable (46-94%). Many cases developed with a short time after
initial exposure and presented with jaundice, hepatomegaly or hepatosplenomegaly, in addition,
to hepatitis. Hepatitis development was of a nonviral etiology, as antibody titers for Hepatitis A,
B, and C viruses were not detectable, and not associated with alcohol consumption (Huang et al.,
2002; Kamijima et al., 2007). Liver failure was moreover a leading cause of death among these
subjects. Kamijima et al. (2007) note the similarities between specific skin manifestations and
accompanying hepatic toxicity and case presentations of TCE-related generalized skin diseases
and conditions that have been linked to specific medications (e.g., carbamezepine, allupurinol,
antibacterial sulfonamides), possibly in conjunction with reactivation of specific latent viruses.
However, neither cytomegalovirus or Epstein-Barr viruses are implicated in the few reports
which did include examination of viral antibodies.
This document is a draft for review purposes only and does not constitute Agency policy.
268 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
4.5.2. Liver Cancer in Humans
Primary hepatocellular carcinoma and cholangiocarcinoma (intrahepatic and extrahepatic
bile ducts) are the most common primary hepatic neoplasms (Blechacz and Gores, 2008; El-
Serag, 2007). Primary hepatocellular carcinoma is the 5th most common of cancer deaths in
males and 9th in females (Jemal et al., 2008). Age-adjusted incidence rates of hepatocellular
carcinoma (HCC) and intrahepatic cholangiocarcinoma (ICC) are increasing, with a twofold
increase in HCC over the past 20 years. This increase is higher than expected from an expanded
definition of liver cancer to include primary or secondary neoplasms since International
Classification of Disease (ICD)-9, incorrect classification of hilar cholangiocarcinomas in ICD-0
as ICC, or to improved detection methods (El-Serag, 2007). It is estimated that
21,370 Americans will be diagnosed in 2008 with liver and intrahepatic bile cancer; age-adjusted
incidence rates for liver and intrahepatic bile duct cancer for all races are 9.9 per 100,000 for
males and 3.5 per 100,000 for females Ries et al., 2008. Survival for liver and biliary tract
cancers remains poor and age-adjusted mortality rates are just slightly lower than incidence rates.
While hepatitis B and C viruses and heavy alcohol consumption are believed major risk factors
for HCC and intrahepatic cholangiocarcinoma, these risk factors cannot fully account for roughly
10 and 20% of HCC cases Kulkarni et al., 2004. Cirrhosis is considered a premalignant
condition for HCC, however, cirrhosis is not a sufficient cause for HCC since 10-25% of HCC
cases lack evidence of cirrhosis at time of detection (Chiesa et al., 2000; Fattovich et al., 2004;
Kumar et al., 2007). Nonalcoholic steatohepatitis reflecting obesity and metabolic syndrome is
recently suggested as contributing to liver cancer risk (El-Serag, 2007).
This document is a draft for review purposes only and does not constitute Agency policy.
269 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
This document is a draft for review purposes only and does not constitute Agency policy.
270 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-56. Selected results from epidemiologic studies of TCE exposure and
2 cirrhosis
3
Study
population
Exposure group
Relative risk (95% CI)
No. obs.
events
Reference
Cohort and PMR-mortality
Aerospace workers (Rocketdyne)
Any TCE (utility/eng flush)
0.39 (0.16,0.80)
7
Boice et al. (2006b)
Low cumulative TCE score
Not reported
Zhao et al. (2005)
Medium cumulative TCE score
High TCE score
p for trend
View-master workers
Males
0.76(0.16,2.22)
3
ATSDR (2003a)
Females
1.51 (0.72, 2.78)
10
Electronic workers (Taiwan)
Primary liver, males
Not reported
Chang et al. (2003;
2005)
Primary liver, females
Not reported
Uranium-processing workers
Any TCE exposure
0.91 (0.63, 1.28)
33
Ritz (1999a)
Light TCE exposure, >2 yr duration
Not reported
Mod TCE exposure, >2 yr duration
Not reported
Aerospace workers (Lockheed)
TCE routine exposure
0.61 (0.39,0.91)
23
Boice et al. (1999)
TCE routine-intermittent
Not reported
13
Aerospace workers (Hughes)
TCE subcohort
0.55 (0.30, 0.93)
14
Morgan etal. (1998,
2000b)
Low intensity (<50 ppm)
0.95 (0.43, 1.80)
9
High intensity (>50 ppm)
0.32(0.10,0.74)
5
Aircraft maintenance workers (Hill AFB, Utah)
TCE subcohort
1.1 (0.6, 1.9)a
44
Blair etal. (1998)
Males, cumulative exposure
0
1.0a
<5 ppm-yr
0.6(0.2, 1.3)
10
5-25 ppm-yr
0.8(0.3, 1.9)
9
>25 ppm-yr
1.2(0.6,2.4)
17
4
This document is a draft for review purposes only and does not constitute Agency policy.
271 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-56. Selected results from epidemiologic studies of TCE exposure and
cirrhosis (continued)
Study
population
Exposure group
Relative risk (95% CI)
No. obs.
events
Reference
Aircraft
Females, cumulative exposure
maintenance
workers
(continued)
0
1.0a
Blair etal. (1998)
<5 ppm-yr
2.4 (1.4, 13.7)
6
(continued)
5-25 ppm-yr
1.8 (0.2, 15.0)
1
>25 ppm-yr
0.6(0.1,4.8)
1
TCE subcohort
1.04 (0.56, 1.93) a'b
37
Radican et al. (2008)
Males, cumulative exposure
0.87 (0.43, 1.73)
31
0
1.0 a'b
<5 ppm-yr
0.56 (0.23, 1.40)
8
5-25 ppm-yr
1.07 (0.45, 2.53)
10
>25 ppm-yr
1.06 (0.48, 2.38)
13
Females, cumulative exposure
1.79 (0.54, 5.93)
6
0
1.00a
<5 ppm-yr
3.30 (0.88, 12.41)
4
5-25 ppm-yr
2.20 (0.26, 18.89)
1
>25 ppm-yr
0.59 (0.97, 5.10)
1
Deaths reported to GE pension fund (Pittsfield, MA)
Not reported
Greenland et al. (1994)
U.S. Coast Guard employees
Blair etal. (1989)
Marine inspectors
1.36 (0.79, 2.17)
17
Noninspectors
0.53 (0.23, 1.05)
8
Aircraft manufacturing plant employees (Italy)
Costa et al. (1989)
All subjects
Not reported
Aircraft manufacturing plant employees (San Diego, CA)
Garabrant et al. (1988)
All subjects
0.86 (0.67, 1.11)
63
1
2
3
4
5
6
7
a Referent group are subjects from the same plant or company, or internal referents.
b Numbers of cirrhosis deaths in Radican et al. (2008) are fewer than Blair et al. (1989) because Radican et al. (2008)
excluded cirrhosis deaths due to alcohol.
GE = General Electric.
This document is a draft for review purposes only and does not constitute Agency policy.
272 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
All cohort studies, except Zhao et al. (2005), present risk ratios (SIRs or SMRs) for liver
and biliary tract cancer. More rarely reported in cohort studies are risk ratios for primary liver
cancer (HCC) or for gallbladder and extrahepatic bile duct cancer. Four community studies also
presented risk ratios for liver and biliary tract cancer including a case-control study of primary
liver cancer of residents of Taiwanese community with solvent-contaminated drinking water
wells (ATSDR, 2006a; Lee et al., 2003; Morgan and Cassady, 2002; Vartiainen et al., 1993).
Several population case-control studies examine liver cancer and organic solvents or
occupational job titles with possible TCE usage (Austin et al., 1987; D0ssing et al., 1997;
Hardell et al., 1984; Heinemann et al., 2000; Hernberg et al., 1988; 1984; Ji and Hemminki,
2005; Kvam et al., 2005; Lindbohm et al., 2009; Porru et al., 2001; Stemhagen et al., 1983;
Weiderpass et al., 2003); however, the lack of detailed exposure assessment to TCE, specifically
in the population case-control studies as well as in geographic-based studies, or, too few exposed
cases and controls in those studies that do present some information limits their usefulness for
evaluating hepatobiliary or gall bladder cancer and TCE exposure. Table 4-57 presents
observations from cohort, case-control, and community studies on liver and biliary tract cancer,
primary liver, and gallbladder and extrahepatic bile duct cancer and trichloroethylene.
Excess liver cancer incidence is observed in most studies in which there is a high
likelihood of TCE exposure in individual study subjects (e.g., based on job-exposure matrices or
biomarker monitoring) and which met, to a sufficient degree, the standards of epidemiologic
design and analysis were identified (Anttila et al., 1995; Axelson et al., 1994; Hansen et al.,
2001; Raaschou-Nielsen et al., 2003) as is mortality (ATSDR, 2004; Blair et al., 1998; Boice et
al., 2006b; Morgan et al., 1998; 2008; Ritz, 1999a). Risks for primary liver cancer and for
gallbladder and biliary tract cancers in females were statistically significantly elevated only in
Raaschou-Nielsen et al. (2003), the study with the largest number of observed cases without
suggestion of exposure duration-response patterns. Cohort studies with more uncertain exposure
assessment approaches, e.g., studies of all subjects working at a factory (Blair et al., 1989; Chang
et al., 2003; Chang et al., 2005; Costa et al., 1989; Garabrant et al., 1988), do not show
association but are quite limited given their lacking attribution of who may have higher or lower
exposure potentials. Ritz (1999a), the exception, found evidence of an exposure-response
relationship; mortality from hepatobiliary cancer was found to increase with
This document is a draft for review purposes only and does not constitute Agency policy.
273 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-57. Selected results from epidemiologic studies of TCE exposure and liver cancer
Study
population
Exposure group
Liver and intrahepatic bile
ducts
Primary liver
Gallbladder and extrahepatic bile ducts
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Reference
Cohort and PMR studies—incidence
Aerospace workers (Rocketdyne)
Low cumulative TCE score
Not reported
Zhao et al.
(2005)
Medium cumulative TCE score
Not reported
High TCE score
Not reported
p for trend
Danish blue-collar workers with TCE exposure
Males + females
1.3 (1.0, 1.6f
82
Raaschou-
Nielson et al.
(2003)
Males + females
1.4(1.0, 1.8)b
57
Males, any exposure
1.1 (0.8, 1.5)b
41
1.1 (0.7, 1.6)
27
1.1 (0.6, 1.9)
14
<1 yr employment duration
1.2(0.7, 2.1)b
13
1.3 (0.6,2.5)
9
1.1 (0.3,2.9)
4
1-4.9 yr employment duration
0.9 (0.5, 1.6)b
13
1.0(0.5, 1.9)
9
0.8(0.2,2.1)
4
>5 yr employment duration
1.1 (0.6, 1.7)b
15
1.1 (0.5,2.1)
9
1.4 (0.5, 3.1)
6
Females, any exposure
2.8 (1.6, 4.6)b
16
2.8(1.1,5.8)
7
2.8 (1.3, 5.3)
9
<1 yr employment duration
2.5 (0.7, 6.5)b
4
2.8 (0.3, 10.0)
2
2.3 (0.3, 8.4)
2
1-4.9 yr employment duration
4.5 (2.2, 8.3)b
10
4.1 (1.1, 10.5)
4
4.8(1.7, 10.4)
6
>5 yr employment duration
1.1 (0.1, 3.8)b
2
1.3 (0.0,7.1)
1
0.9 (0.0, 5.2)
1
-------
Table 4-57. Selected results from epidemiologic studies of TCE exposure and liver cancer (continued)
Liver and intrahepatic bile
ducts
Primary liver
Gallbladder and extrahepatic bile ducts
Study
population
Exposure group
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Reference
Biologically-monitored Danish workers
Males + females
2.1 (0.7, 5.0)b
5
1.7(0.2,6.0)
2
2.5 (0.5, 7.3)
3
Hansen et al.
Males
2.6(0.8, 6.0)b
5
1.8(0.2,6.6)
2
3.3 (0.7, 9.7)
3
(2001)
Females
0 (0.4
exp)
0(0.1
exp)
0 (0.3 exp)
Cumulative exposure (Ikeda)
Not reported
<17 ppm-yr
>17 ppm-yr
Mean concentration (Ikeda)
Not reported
<4 ppm
4+ ppm
Employment duration
Not reported
<6.25 yr
>6.25
-------
Table 4-57. Selected results from epidemiologic studies of TCE exposure and liver cancer (continued)
Study
population
Exposure group
Liver and intrahepatic bile
ducts
Primary liver
Gallbladder and extrahepatic bile ducts
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Reference
Aircraft maintenance workers from Hill Air Force Base
TCE subcohort
Not reported
9
Not reported
Blair et al.
(1998)
Males, cumulative exposure
0
1.0C
1.03
<5 ppm-yr
0.6(0.1,3.1)
3
1.2(0.1,2.1)
2
5-25 ppm-yr
0.6(0.1,3.8)
2
1.0(0.1, 16.7)
1
>25 ppm-yr
1.1 (0.2,4.8)
4
2.6(0.3,25.0)
3
Females, cumulative exposure
0
0
Biologically-monitored Finnish workers
All subjects
1.89 (0.86, 3.59)b
9
2.27 (0.74, 5.29)
5
1.56 (0.43, 4.00)
4
Anttila et al.
(1995)
Mean air-TCE (Ikeda extrapolation from U-TCA)
<6 ppm
Not reported
1.64 (0.20, 5.92)
2
6+ ppm
2.74 (0.33, 9.88)
2
Biologically-monitored Swedish workers
Males
1.41 (0.38, 3.60)b
4
Axelson et al.
(1994)
Females
Not reported
Cohort and PMR-mortality
Computer manufacturing workers (IBM), NY
Not reported
1
Clapp and
Hoffman (2008)
-------
Table 4-57. Selected results from epidemiologic studies of TCE exposure and liver cancer (continued)
Study
population
Exposure group
Liver and intrahepatic bile
ducts
Primary liver
Gallbladder and extrahepatic bile ducts
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Reference
Aerospace workers (Rocketdyne)
Any TCE (utility/eng flush)
1.28 (0.35, 3.27)
4
Boice et al.
(2006b)
Low cumulative TCE score
Not reported
Zhao et al.
(2005)
Med cumulative TCE score
High TCE score
p for trend
View-Master workers
Males
2.45 (0.50, 7.12)d
3
1.01 (0.03, 5.63)d
1
8.41 (1.01, 30.4)d
2
ATSDR (2003a)
Females
0
(2.61 exp)
0
(1.66 exp)
0
(0.95 exp)
Electronic workers (Taiwan)
Primary liver, males
Not reported
0
(0.69 exp)
Chang et al.
(2003; 2005)
Primary liver, females
Not reported
0
(0.57 exp)
-------
Table 4-57. Selected results from epidemiologic studies of TCE exposure and liver cancer (continued)
Liver and intrahepatic bile
ducts
Primary liver
Gallbladder and extrahepatic bile ducts
Study
population
Exposure group
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Reference
Uranium-processing workers
Any TCE exposure
Not reported
Ritz (1999a)
Light TCE exposure, >2 yr
ation
0.93 (0.19, 4.53)e
3
Mod TCE exposure, >2 yr
duration
4.97 (0.48, 51.1)e
1
Light TCE exposure, >5 yr
duration
2.86 (0.48, 17.3)f
3
Mod TCE exposure, >5 yr
duration
12.1 (1.03, 144)f
1
Aerospace workers (Lockheed)
TCE routine exposure
0.54 (0.15, 1.38)
4
Boice et al.
TCE routine-intermittent
(1999)
0 yr
1.00c
22
Any exposure
Not reported
13
<1 yr
0.53 (0.18, 1.60)
4
1-4 yr
0.52 (0.15, 1.79)
3
>5 yr
0.94 (0.36, 2.46)
6
p for trend
>0.20
-------
Table 4-57. Selected results from epidemiologic studies of TCE exposure and liver cancer (continued)
Liver and intrahepatic bile
ducts
Primary liver
Gallbladder and extrahepatic bile ducts
Study
population
Exposure group
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Reference
Aerospace workers (Hughes)
TCE subcohort
0.98 (0.36,2.13)
6
Morgan et al.
Low intensity (<50 ppm)e
1.32 (0.27, 3.85)
3
(1998, 2000b)
High intensity (>50 ppm)e
0.78(0.16,2.28)
3
TCE subcohort (Cox analysis)
Never exposed
1.00c
14
Ever exposed
1.48 (0.56, 3.91)8'h
6
Cumulative
Low
2.12(0.59, 7.66)h
3
High
1.19 (0.34, 4.16)h
3
Peak
No/low
1.00c
17
Medium/high
0.98 (0.29, 3.35)h
3
-------
Table 4-57. Selected results from epidemiologic studies of TCE exposure and liver cancer (continued)
Liver and intrahepatic bile
ducts
Primary liver
Gallbladder and extrahepatic bile ducts
Study
population
Exposure group
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Reference
Aircraft maintenance workers (Hill AFB, Utah)
Blair et al.
TCE subcohort
1.3 (0.5, 3.4)c
15
1.7 (0.2, 16.2)3
4
(1998)
Males, cumulative exposure
0
1.0C
<5 ppm-yr
1.1 (0.3,4.1)
6
5-25 ppm-yr
0.9(0.2,4.3)
3
>25 ppm-yr
0.7 (0.2, 3.2)
3
Females, cumulative exposure
0
1.0°
<5 ppm-yr
1.6 (0.2, 18.2)
1
5-25 ppm-yr
0
>25 ppm-yr
2.3 (0.3, 16.7)
2
TCE subcohort
1.12(0.57, 2.19)C1
31
1.25 (0.31, 4.97)c>1
8
Radican et al.
Males, cumulative exposure
1.36 (0.59, 3.11)°
28
2.72 (0.34, 21.88)c
8
(2008)
0
1.0°
1.03
<5 ppm-yr
1.17 (0.45,3.09)
10
3.28 (0.37, 29,45)
4
5-25 ppm-yr
1.16 (0.39,3.46)
6
0
>25 ppm-yr
1.72 (0.68, 4.38)
12
4.05 (0.45, 36.41)
4
-------
Table 4-57. Selected results from epidemiologic studies of TCE exposure and liver cancer (continued)
Study
population
Exposure group
Liver and intrahepatic bile
ducts
Primary liver
Gallbladder and extrahepatic bile ducts
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Reference
Aircraft
maintenance
workers
(continued)
Females, cumulative exposure
0.74 (0.18, 2.97)c
3
0
Radican et al.
(2008)
(continued)
0
1.03
<5 ppm-yr
0.69 (0.08, 5.74)
1
5-25 ppm-yr
0
>25 ppm-yr
0.98 (0.20, 4.90)
2
Deaths reported to GE pension fund
(Pittsfield, MA)
0.54 (0.11, 2.63)J
9
Greenland et al.
(1994)
U.S. Coast Guard employees
Marine inspectors
1.12 (0.23,3.26)
3
Blair et al.
(1989)
Noninspectors
Not reported
0 (2 exp)
Aircraft manufacturing plant employees (Italy)
All subjects
0.70 (0.23, 1.64)
5
Costa et al.
(1989)
Aircraft manufacturing plant employees (San Diego, CA)
All subjects
0.94 (0.40, 1.86)
8
Garabrant et al.
(1988)
Case-control studies
Residents of community with contaminated drinking water (Taiwan)
Village of residency, males
Lee et al. (2003)
Upstream
1.00
Downstream
2.57 (1.21,5.46)
26
-------
L/l
K>
Table 4-57. Selected results from epidemiologic studies of TCE exposure and liver cancer (continued)
Study
population
Exposure group
Liver and intrahepatic bile
ducts
Primary liver
Gallbladder and extrahepatic bile ducts
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Reference
Geographic studies
Residents in two study areas in Endicott, NY
0.71 (0.09, 2.56)
<6
ATSDR (2006b)
Residents in 13 census tracts inRedlands, CA
1.29 (0.74, 2.05)k
28
Morgan and
Cassidy (2002)
Finnish residents
Residents of Hausjarvi
0.7 6 (0.3, 1.4)
7
Vartiainen et al.
(1993)
Residents of Huttula
0.6(0.2, 1.3)
6
S?
>!
rs
s;
>1
£
J
00 ^3
^ §
>!
O
a
ca
a
§•
>!
o
o
p
H
o
o
2;. _
a
Cs
o
a
>j
S"
(3s
H
1—1 s
H 2.
Wf
V
o
c
o
H
W
a ICD-7, 155 and 156; Primary liver (155.0), gallbladder, and biliary passages (155.1), and liver secondary and unspecified (156).
b ICD-7, 155; Primary liver, gallbladder, and biliary passages.
0 Internal referents, workers without TCE exposure.
d Proportional mortality ratio (PMR).
e Logistic regression analysis with a 0-yr lag for TCE exposure.
f Logistic regression analysis with a 15-yr lag for TCE exposure.
8 Risk ratio from Cox Proportional Hazard Analysis, stratified by age, sex, and decade Environmental Health Strategies (1997).
hMorganetal. (1998) do not identify if SIR is for liver and biliary passage orprimary liver cancer; identified as primary liver inNRC (2006).
:Radican et al. (2008) provide results for TCE exposure for follow-up through 1990, comparing the Poisson model rate ratios as reported by Blair et al. (1998)
with Cox model hazard ratios. Relative risk from Cox model adjusted for age and gender for liver and intrahepatic bile duct cancer was 1.2 (95% CI: 0.5, 3.4)
and for primary liver cancer was 1.3 (95% CI: 0.1, 12.0).
J Odds ratio.
k 99% CIs.
exp = exposures, GE = General Electric, IBM = International Business Machines Corporation.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
degree and duration of exposure and time since first exposure with a statistically significant but
imprecise (wide confidence intervals) liver cancer risk for those with the highest exposure and
longest time since first exposure. This observation is consistent with association with TCE, but
with uncertainty given one TCE exposed case in the highest exposure group and correlation
between TCE, cutting fluids, and radiation exposures.
Observations in these studies provide some evidence of susceptibility of liver, gallbladder
and biliary tract; observations consistent with pharmacokinetic processing of TCE and the
extensive intra- and extrahepatic recirculation of metabolites. Magnitude of risk of gallbladder
and biliary tract cancer is slightly higher than that for primary liver cancer in Raaschou-Nielsen
et al. (2003) the study with the most cases. Observations in Blair et al. (1998), Hansen et al.
(2001), and Radican et al. (2008), three smaller studies, suggest slightly larger risk ratios for
primary liver cancer compared to gallbladder and biliary tract cancer. Overall, these studies are
not highly informative for cross-organ comparison of relative magnitude of susceptibility.
The largest geographic studies (Lee et al., 2003; Morgan and Cassady, 2002) are also
suggestive of association with the risk ratio (mortality odds ratio) in Lee et al. (2003) as
statistically significantly elevated. The geographic studies do not include a characterization of
TCE exposure to individual subjects other than residency in a community with groundwater
contamination by TCE with potential for exposure misclassification bias dampening
observations; these studies lack characterization of TCE concentrations in drinking water and
exposure characteristics such as individual consumption patterns. For this reason, observations
in Morgan and Cassidy (2002) and Lee et al. (2003) are noteworthy, particularly if positive bias
leading to false positive finding is considered minimal, and the lack of association with liver
cancer in the two other community studies (ATSDR, 2006b; Vartiainen et al., 1993) does not
detract from Morgan and Cassidy (2002) or Lee et al. (2003). Lee et al. (2003), however, do not
address possible confounding related to hepatitis viral infection status, a risk factor for liver
cancer, or potential misclassification due to the inclusion of secondary liver cancer among the
case series, factors which may amplify observed association.
Meta-analysis is adopted as a tool for examining the body of epidemiologic evidence on
liver cancer and TCE exposure, to identify possible sources of heterogeneity and as an additional
means to identify cancer hazard. The meta-analyses of the overall effect of TCE exposure on
liver (and gall bladder/biliary passages) cancer suggest a small, statistically significant increase
in risk. The summary estimate from the primary random effects meta-analysis of the 9 (all
cohort) studies is 1.29 (95% CI: 1.07, 1.56) (see Figure 4-3). The study of Raaschou-Nielsen
et al. (2003) contributes about 57% of the weight; its removal from the analysis decreases
somewhat the RRm estimate and is no longer statistically significant (RRm = 1.22; 95% CI:
This document is a draft for review purposes only and does not constitute Agency policy.
283 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Study
Anttila (1965)
feelson <1964:1
Boice (1689)
Boice (2006)
Greenland (1964)
Hansen (2001)
Morgan (1098)
R a ase h o u- N i e Ise n (2003)
Radiean <2008)
OVERALL
0.1
Figure 4-3. Relative risk estimates of liver and biliary tract cancer and
overall TCE exposure. Random effects model; fixed effect model same. The
summary estimate is in the bottom row, represented by the diamond. Symbol
sizes reflect relative weights of the studies.
0.93, 1.61). The summary estimate was not overly influenced by any other single study, nor was
it overly sensitive to individual RR estimate selections. There is no evidence of publication bias
in this data set, and no observable heterogeneity (/ = 0%) across the study results.
Examination of sites individually (i.e., primary liver and intrahepatic bile ducts separate
from the combined liver and gallbladder/biliary passage grouping) resulted in the RRm estimate
for liver cancer alone (for the three studies for which the data are available; for the other studies,
results for the combined grouping were used) slightly lower than the one based entirely on
results from the combined cancer categories and was just short of statistical significance (1.25;
95% CI: 0.99, 1.57). This result is driven by the fact that the risk ratio estimate from the large
Raaschou-Nielsen et al. (2003) study decreased from 1.35 for liver and gall bladder/biliary
passage cancers combined to 1.28 for liver cancer alone.
The RRm estimate from the random effects meta-analysis of liver cancer in the highest
exposure groups in the six studies which provide risk estimates associated with highest exposure.
TCE Exposure ami Livei Cancel
Rtlative Risk and 95* CI
RR
LCL
UCL
1.89
0.86
3.59
1.41
0.38
3.60
0.81
0.46
1.33
1.28
0.35
3.27
0.54
0.11
2.33
2.10
0.70
5.00
1.48
0.58
3.01
1.35
1.03
1.77
1.12
0.57
2.18
1.20
1.07
1.56
10
This document is a draft for review purposes only and does not constitute Agency policy.
284 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
TCE Exposure annull
1
1.00
0.32
3.10
Zhao i2005)null
I
1.00
0.08
11.00
OVERALL
1.28
0.93
1.77
0.1 1 10
Figure 4-4. Meta-analysis of liver cancer and TCE exposure—highest
exposure groups, with assumed null RR estimates for Hansen and Zhao (see
Appendix C text). Random effects model; fixed effect model same. The
summary estimate is in the bottom row, represented by the diamond. Symbol
sizes reflect relative weights of the studies.
primary liver cancer is 1.32 (95% CI: 0.93, 1.86), slightly lower than the RRm estimate for liver
and gallbladder/biliary cancer and any TCE exposure of 1.33 (95% CI: 1.09, 1.64), and not
statistically significant (see Figure 4-4). Again, the RRm estimate of the highest-exposure
groups is dominated by one study (Raaschou-Nielsen et al., 2003). Two studies lack reporting of
liver cancer risk associated with highest exposure, so consideration of reporting bias (considered
the primary analysis) lead to a result of 1.28 (95% CI: 0.93, 1.77), similar to that estimated in the
more restricted set of studies presenting risk ratios association with highest exposure groups in
published papers.
Different exposure metrics are used in the various studies, and the purpose of combining
results across the different highest exposure groups is not to estimate an RRm associated with
some level of exposure, but rather to examine impacts of combining RR estimates that should be
less affected by exposure misclassification. In other words, the highest exposure category is
This document is a draft for review purposes only and does not constitute Agency policy.
285 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
more likely to represent a greater differential TCE exposure compared to people in the referent
group than the exposure differential for the overall (typically any vs. none) exposure comparison.
Thus, if TCE exposure increases the risk of liver and gallbladder/biliary cancer, the effects
should be more apparent in the highest exposure groups. The findings of a lower RRm
associated with highest exposure group reflects observations in Radican et al. (2008) and
Raaschou-Nielsen et al. (2003), the study contributing greatest weight to the meta-analysis, that
RR estimates for the highest-exposure groups, although greater than 1.0, are less than the RR
estimates with any TCE exposure.
Thus, while the finding of an elevated and statistically significant RRm for liver and
gallbladder/biliary cancer and any TCE exposure provides evidence of association, the statistical
significance of the summary estimates is dependent on one study, which provides the majority of
the weight in the meta-analyses. Furthermore, combining results from the highest-exposure
groups yields lower RRm estimates than for an overall effect. These results do not rule out an
effect of TCE on liver cancer, because the liver cancer results are relatively underpowered with
respect to numbers of studies and number of cases; overall, the meta-analysis provides only
minimal support for association between TCE exposure and liver and gallbladder/biliary cancer.
NRC (2006) deliberations on trichloroethylene commented on two prominent evaluations
of the then-current TCE epidemiologic literature using meta-analysis techniques, Wartenberg
et al. (2000) and Kelsh et al. (2005), submitted by Exponent-Health Sciences to NRC during
their deliberations and published afterwards in the open literature as Alexander et al. (2007a)
adding the than-published study of Boice et al. (2006b). NRC (2006) found weaknesses in the
techniques used in Wartenberg et al. (2000) and the Exponent analyses. EPA staff conducted
their analysis according to NRC (2006) suggestions for transparency, systematic review criteria,
and examination of both cohort and case-control studies. The EPA analysis of liver cancer
considered a similar set of studies as Alexander et al. (2007a) although treatment of these studies
differs between analyses. Alexander et al. (2007a) in their Table 2, for example, present
summary relative risk estimates, grouping of studies with differing exposure potentials, for
example, including liver and biliary cancer risk estimates for all subjects, those exposed and
unexposed to TCE, in Boice et al.(1999), Blair et al. (1998), Morgan et al. (1998) and Boice
et al. (2006b), with biomarker studies (Anttila et al., 1995; Axelson et al., 1994; Hansen et al.,
2001). The inclusion of risk estimates for subjects who have little to no TCE exposure over
background levels has the potential to introduce misclassification bias and dampen observed risk
ratios. Potential bias from exposure misclassification may substantial in Alexander et al. (2007a)
since the percentage of TCE exposed subjects to all cohort subjects in the four studies was 3, 23,
51 and 68% in Boice et al. (1999), Morgan et al. (1998), Blair et al. (1998) and Boice et al.
This document is a draft for review purposes only and does not constitute Agency policy.
286 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
(2006b), respectively, and is a likely alternative explanation for observed inconsistency across
occupational groups reported by the authors. Another difference between the EPA and previous
meta-analyses is their treatment of Ritz (1999a), included in Wartenberg et al. (2000), Kelsh
et al. (2005), and Alexander et al. (2007a), but not in this analysis. For a grouping of studies
with subcohorts most similar to those in EPA's analysis, summary liver and gall bladder/biliary
tract cancer risk estimates for overall TCE exposure for TCE subcohorts is of a similar
magnitude as that observed in EPA's updated and expanded analysis, Wartenberg et al. (2000),
1.1 (95% CI: 0.3, 4.8) for incidence and 1.1 (95% CI: 0.7, 1.7) for mortality, Kelsh et al. (2005),
1.32 (95% CI: 1.05, 1.66) and Alexander et al. (2007a), 1.30 (95% CI: 1.09-1.55).
4.5.3. Experimental Studies of Trichloroethylene (TCE) in Rodents—Introduction
The previous sections have described available human data for TCE-induced noncancer
effects (e.g., disturbances in bile production) and whether an increased risk of liver cancer in
humans has been established from analysis of the epidemiological literature. A primary concern
for effects on the liver comes from a large database in rodents indicating that, not only TCE, but
a number of its metabolites are capable of inducing hepatocellular adenomas and carcinomas in
rodent species. Thus, many of rodent bioassays have focused on the study of liver cancer for
TCE and its metabolites and possible early effects specifically that may be related to tumor
induction.
This section describes the hazard data for TCE effects in the rodent liver and inferences
from studies of its metabolites. For more detailed descriptions of the issues providing context for
these data in terms the state of the science of liver physiology (see Section E. 1), cancer (see
Section E.3), liver cancer (see Section E.3), and the MOA of liver cancer and other TCE-induced
effects (see Section E.3.4), please see Appendix E. A more comprehensive review of individual
studies of TCE-induced liver effects in laboratory animals is also provided in Section E.2 that
includes detailed analyses of the strengths and the limitations of these studies. Issues have been
raised regarding the relevance of mouse liver tumor data to human liver cancer risk that are
addressed in Sections E.3.2 and E.3.3. Given that activation of the PPARa receptor has received
great attention as a potential MOA for TCE induced liver tumors, the current status of that
hypothesis is reviewed in Section E.3.4.1. Finally, comparative studies of TCE metabolites and
the similarities and differences of such study results are described in summary sections of
Appendix E (i.e., Section E.2.4) as well as discussions of proposed MOAs for TCE-induced liver
cancer (i.e., Sections E.2.4 and E.3.4.2).
This document is a draft for review purposes only and does not constitute Agency policy.
287 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
A number of acute and subchronic studies have been undertaken to describe the early
changes in the rodent liver after TCE administration with the majority using the oral gavage
route of administration. Several key issues affect the interpretation of these data. The few
drinking water studies available for TCE have recorded significant loss of TCE through
volatilization in drinking water solutions and thus, this route of administration is generally not
used. Some short-term studies of TCE have included detailed examinations while others have
reported primarily liver weight changes as a marker of TCE response. The matching and
recording of age, but especially initial and final body weight, for control and treatment groups is
of particular importance for studies using liver weight gain as a measure of TCE response as
differences in these parameters affect TCE-induced liver weight gain. Most data are for TCE
exposures of at least 10 days to 42 days. For many of the subchronic inhalation studies
(Kjellstrand et al., 1981, 1983a, b), issues associated with whole-body exposures make
determination of dose levels more difficult. The focus of the long-term studies of TCE is
primarily detection and characterization of liver tumor formation.
For gavage experiments, death due to gavage errors and specifically from use of this
route of administration, especially at higher TCE exposure concentrations, has been a recurring
problem, especially in rats. Unlike inhalation exposures, the effects of vehicle can also be an
issue for background liver effects in gavage studies. Concerns regarding effects of oil vehicles,
especially corn oil, have been raised (Charbonneau et al., 1991; Kim et al., 1990a). Several oral
studies in particular document that use of corn oil as the vehicle for TCE gavage dosing induces
a different pattern of toxicity, especially in male rodents (see Merrick et al., 1989, Section
E.2.2.1). Several studies also report the effects of corn oil on hepatocellular DNA synthesis and
indices of lipid peroxidation (Channel et al., 1998; Rusyn et al., 1999). For example, Rusyn
et al. (1999) report that a single dose of dietary corn oil increases hepatocyte DNA synthesis 24
hours after treatment by ~3.5-fold of control, activates of NF-kB to a similar extent ~2 hours
after treatment almost exclusively in Kupffer cells, and induces an ~3-fourfold increase of
control NF-kB in hepatocytes after 8 hours and an increase in tumor necrosis factor (TNF)-a
mRNA between 8 and 24 hours after a single dose in female rats.
In regard to studies that have used the i.p. route of administration, as noted by
Kawamoto et al. (1988b), injection of TCE may result in paralytic ileus and peritonitis and that
subcutaneous treatment paradigm will result in TCE not immediately being metabolized but
retained in the fatty tissue. Wang and Stacey (1990) state that "intraperitoneal injection is not
particularly relevant to humans" and suggest that intestinal interactions require consideration in
responses such as increase serum bile acid.
This document is a draft for review purposes only and does not constitute Agency policy.
288 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
While studies of TCE metabolites have been almost exclusively conducted via drinking
water, and thus, have avoided vehicle effects and gavage error, they have issues of palatability at
high doses and decreased drinking water consumption as a result that not only raises issues of the
resulting internal dose of the agent but also of effects of drinking water reduction.
Although there are data for both mice and rats for TCE exposure and studies of its
metabolites, the majority of the available information has been conducted in mice. This is
especially the case for long-term studies of DCA and TCA in rats. There is currently one study
each available for TCA and DCA in rats and both were conducted with such few numbers of
animals that the ability to detect and discern whether there was a treatment-related effect are very
limited (DeAngelo et al., 1996, 1997; Richmond et al., 1995).
With regard to the sensitivity of studies used to detect a response, there are issues
regarding not only the number of animals used but also the strain and weight of the animals. For
some studies of TCE strains were used that have less background rate of liver tumor
development and carcinogenic response. As for the B6C3F1 mouse, the strain most used in the
bioassays of TCE metabolites, the susceptibility of the B6C3F1 to hepatocarcinogenicity has
made the strain a sensitive biomarker for a variety of hepatocarcinogens. Moreover, Leakey
et al. (2003b) demonstrated that increased body weight at 45 weeks of life is an accurate
predictor of large background tumor rates. Unfortunately a 2-year study of chloral hydrate
(George et al., 2000) and the only available 2-year study of TCA (DeAngelo et al., 2008), which
used the same control animals, were both conducted in B6C3F1 mice that grew very large
(-50 g) and prone to liver cancer (64% background incidence of hepatocellular adenomas and
carcinomas) and premature mortality. Thus, these bioassays are of limited value for
determination of the dose-response for carcinogenicity.
Finally, as discussed below, the administration of TCE to laboratory animals as well as
environmental exposure of TCE in humans are effectively coexposure studies. TCE is
metabolized to a number of hepatoactive as well as hepatocarcinogenic agents. A greater
variability of response is expected than from exposure to a single agent making it particularly
important to look at the TCE database in a holistic fashion rather than the results of a single
study, especially for quantitative inferences. This approach is particularly useful given that the
number of animals in treatment groups in a variety of TCE and TCE metabolite studies have
been variable and small for control and treatment groups. Thus, their statistical power was not
only limited for detection of statistically significant changes but also in many cases to be able to
determine whether there is not a treatment related effect (i.e., Type II error for power
calculation). Section E.2.4.2 provides detailed analyses of the database for liver weight
induction by TCE and its metabolites in mice and the results of those analyses are described
This document is a draft for review purposes only and does not constitute Agency policy.
289 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
below. Specifically, the relationship of liver weight induction, but also other endpoints such as
peroxisomal enzyme activation and increases in DNA synthesis to liver tumor responses are also
addressed as well.
4.5.4. Trichloroethylene (TCE)-Induced Liver Noncancer Effects
A number of effects have been studied as indicators of TCE effects on the liver but also
as proposed events whose sequellae could be associated with resultant liver tumors after chronic
TCE exposure in rodents. Similar effects have been studied in rodents exposed to TCE
metabolites which may be useful for not only determining whether such effects are associated
with liver tumors induced by these metabolites but also if they are similar to what has been
observed for TCE. Summaries of the laboratory animal studies of TCE noncancer effects in the
liver are provided in Table 4-58 (oral studies) and Table 4-59 (inhalation studies), along with the
types of effects discussed in the subsections below for each study.
4.5.4.1.1. Liver Weight
Increases in liver weight in mice, rats, and gerbils have been reported as a result of acute
and short-term, and subchronic TCE treatment by inhalation and oral routes of exposure (Berman
et al., 1995; Buben and O'Flaherty, 1985; Dees and Travis, 1993; Elcombe et al., 1985; Goel et
al., 1992; Goldsworthy and Popp, 1987; Kjellstrand et al., 1983a; Kjellstrand et al., 1983b;
Kjellstrand et al., 1981b; Laughter et al., 2004; Melnick et al., 1987; Merrick et al., 1989;
Nakajima et al., 2000; Nunes et al., 2001; Tao et al., 2000a; Tucker et al., 1982). The extent of
TCE-induced liver weight gain is dependent on species, strain, gender, nutrition status, duration
of exposure, route of administration, vehicle used in oral studies, and the concentration of TCE
administered. Of great importance to the determination of the magnitude of response is whether
the dose of TCE administered also affects whole-body weight, and thus, liver weight and the
percentage liver/body weight ratio. Therefore, studies which employed high enough doses to
induce whole-body weight loss generally showed a corresponding decrease in percentage
liver/body weight at such doses and "flattening" of the dose-response curve, while studies which
did not show systemic toxicity reported liver/body weight ratios generally proportional to dose.
Chronic studies, carried out for longer durations, that examine liver weight are few and often
confounded by the presence of preneoplastic foci or tumors that also affect liver weight after an
extended period of TCE exposure. The number of studies that examine liver weight changes in
the rat are much fewer than for mouse. Overall, the database for mice provides data for
This document is a draft for review purposes only and does not constitute Agency policy.
290 DRAFT—DO NOT CITE OR QUOTE
-------
1 examination of the differences in TCE-induced effects from differing exposure levels, durations
2 of exposure, vehicle, strain, and gender. One study provided a limited examination of
3 TCE-induced liver weight changes in gerbils.
This document is a draft for review purposes only and does not constitute Agency policy.
291 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-58. Oral studies of TCE-induced liver effects in mice and rats
Reference
Animals (sex)
Exposure route
Dose/exposure concentration
Exposed
Section(s) where noncancer liver effects are
discussed
Berman et al.
(1995)
F344 rats (F)
Corn oil gavage
0, 150, 500, 1,500 or 5,000 mg/kg for
once d
0, 50, 150, 500 or 1,500 mg/kg-day
for 14 d]
8/group]
4.5.4.1 Liver weight
4.5.4.2 Cytotoxicity and histopathology
E.2.1.11. Berman etal. (1995)
Buben and
O'Flaherty
(1985)
Swiss-Cox mice
(M)
Corn oil gavage
0,100, 200, 400,800,1,600, 2,400,
or 3.200 mg/kg-day, 5 d/wk for
6 wk
12-15/
group
4.5.4.1 Liver weight
4.5.4.2 Cytotoxicity and histopathology
E.2.2.7. Buben and O'Flaherty (Buben and
O'Flaherty, 1985)
Channel et al.
(1998)
B6C3Fl/CrlBR
mice (M)
Corn oil gavage
0 (water), 0 (corn oil), 400, 800, or
1,200 mg/kg-day, 5 d/wk for up to
8 wk
77/group
4.5.4.2 Cytotoxicity and histopathology
4.5.4.3 Measures of DNA Synthesis, Cellular
Proliferation, and Apoptosis
4.5.4.4 Peroxisome proliferation and related
effects
4.5.4.5 Oxidative stress
E.2.2.8. Channel et al. (1998)
Dees and Travis
(1993)
B6C3F1 mice (M
andF)
Corn oil gavage
0, 100, 250, 500, or 1,000 mg/kg-day
for 10 d
5/group
4.5.4.1 Liver weight
4.5.4.2 Cytotoxicity and histopathology
4.5.4.3 Measures of DNA Synthesis
E.2.1.9. Dees and Travis (1993)
Elcombe et al.
(1985)
B6C3F1and
Alderley Park
(Swiss) mice (M)
Osborne-Mendel
and Alderley
Park (Wistar)
rats (M)
Corn oil gavage
0, 500, 1,000, or 1,500 mg/kg-day for
10 d
6-10/
group
4.5.4.1 Liver weight
4.5.4.2 Cytotoxicity and histopathology
4.5.4.3 Measures of DNA Synthesis, Cellular
Proliferation, and Apoptosis
4.5.4.4 Peroxisome proliferation and related
effects
E.2.1.8. Elcombe et al. (1985)
Goel et al.
(1992)
Swiss albino
mice (M)
Groundnut oil
gavage
0, 500, 1,000, or 2,000
mg/kg-day, 5 d/wk for 28 d.
6/group
4.5.4.1 Liver weight
4.5.4.2 Cytotoxicity and histopathology
4.5.4.4 Peroxisome proliferation and
related effects
E.2.2.2. Goel etal. (1992)
-------
L/l
K>
Table 4-58. Oral studies of TCE-induced liver effects in mice and rats (continued)
Reference
Animals (sex)
Exposure
route
Dose/exposure concentration
Exposed
Section(s) where noncancer liver
effects are discussed
Goldsworthy
and Popp
(1987)
F344 rats (M)
B6C3F1 mice
(M)
Corn oil or
methyl
cellulose
gavage
1,000 mg/kg-day for 10 d
5-7/group
4.5.4.1 Liver weight
4.5.4.4 Peroxisome proliferation and
related effects
E.2.1.7. Goldsworthy and Popp (1987)
Laughter et al.
(2004)
Sv/129 and
PPARa-null
mice (M)
Methyl-
cellulose
gavage
0-1,500 mg/kg-day for 3 d;
and 5 d/wk for 3 wk
4-5/group
4.5.4.1 Liver weight
4.5.4.2 Cytotoxicity and histopathology
4.5.4.3 Measures of DNA Synthesis,
Cellular Proliferation, and Apoptosis
4.5.4.4 Peroxisome proliferation and
related effects
E.2.1.13. Laughter et al. (2004)
Melnick et al.
(1987)
F344 rats (M)
Micro-
encapsulated
in feed
Corn oil
gavage
0, 0.055, 1.10, 2.21, or 4.41%
in feed for 14 d, equivalent to
0, 600, 1.300, 2.200, or 4.800
mg/kg-day
10/group
4.5.4.1 Liver weight
4.5.4.2 Cytotoxicity and histopathology
4.5.4.4 Peroxisome proliferation and
related effects
E.2.1.12. Melnick etal. (1987)
Merrick et al.
(1989)
B6C3F1 mice
(M and F)
Corn oil and
20% emulphor
in water
gavage
Males: 0, 600, 1,200, or
2,400 mg/kg-day
Females: 0, 450, 900, or
1,800 mg/kg-day
12/group
4.5.4.1 Liver weight
4.5.4.2 Cytotoxicity and histopathology
E.2.2.1. Merrick et al. (1989)
Mirsalis et al.
(1989)
B6C3F1 mice
(M and F)
F344 rats (M)
Corn oil
gavage
0, 50, 200, or 1,000 mg/kg
(single dose)
3/group
4.5.4.3 Measures of DNA Synthesis,
Cellular Proliferation, and Apoptosis
E.2.4.1. Summary of Results for Short-
term Effects of Trichloroethylene (TCE)
S?
>!
rs
s;
>1
£
J
VO ^3
00 O
>!
>1
O
a
a
§•
>!
o
0
H
1 5
81
w (V
2; ^
M s
H 2.
W £
V
o
c
o
H
W
-------
L/l
K>
Table 4-58. Oral studies of TCE-induced liver effects in mice and rats (continued)
Reference
Animals (sex)
Exposure
route
Dose/exposure concentration
Exposed
Section(s) where noncancer liver
effects are discussed
Nakajima et al.
(2000)
Sv/129 and
PPARa-null
mice (M and
F)
Corn oil
gavage
0 or 750 mg/kg-day for 14 d
6/sex/
group
4.5.4.1 Liver weight
4.5.4.4 Peroxisome proliferation and
related effects
E.2.1.10. Nakajima et al. (2000)
NTP (1990)
B6C3F1 mice
(M and F)
F334/N rats
(M and F)
Corn oil
gavage
Mice: 0, 375-6,000
mg/kg-day, 5 d/wk, 13 wk
Rats: 0, 62.5-1,000
mg/kg-day, 5 d/wk, 13 wk
10/group
4.5.4.2 Cytotoxicity and histopathology
E.2.2.12.1 13-wk studies
NTP (1990)
B6C3F1 mice
(M and F)
F334/N rats
(M and F)
Corn oil
gavage
Mice: 0, or 1,000 mg/kg-day,
5 d/wk, 103 wk
Rats: 0, 500, or 1,000
mg/kg-day, 5 d/wk, 103 wk
50/group
4.5.4.2 Cytotoxicity and histopathology
E.2.2.12.2 2-yr studies
Nunes et al.
(2001)
Sprague-
Dawley rats
(M)
Corn oil
gavage
2,000 mg/kg-day on d 10-16
(with and without lead
carbonate pretreatment for 9 d)
10/group
4.5.4.1 Liver weight
4.5.4.2 Cytotoxicity and histopathology
E.2.1.4. Nunes etal. (2001)
Tao et al.
(2000a)
B6C3F1 mice
(F)
Corn oil
gavage
1,000 mg/kg-day for 5 d
4-6/group
4.5.4.1 Liver weight
E.2.1.5. Tao et al., (2000a)
Tucker et al.
(1982)
CD-I mice (M
and F)
Drinking
water with 1%
emulphor
0 (untreated), 0 (vehicle), 0.1,
1.0, 2.5, or 5 mg/ml for 4 or 6
mo.
M: 0, 0, 18.4, 216.7, 393.0,or
660.2 mg/kg-day
F: 0, 0, 17.9, 193.0, 437.1,or
793.3 mg/kg-day
140/group
untreated
and TCE-
treated
260/group
vehicle-
treated
4.5.4.1 Liver weight
E.2.1.6. Tucker etal. (1982)
S?
>!
rs
s;
>1
£
J
\Q ^3
K O
^ >!
>1
o
a
a
§•
>!
o
0
H
1 5
81
w (V
2; ^
M s
H 2.
«£
V
o
c
o
H
W
Bolded study(ies) carried forward for consideration in dose-response assessment (see Section 5).
-------
Table 4-59. Inhalation and i.p. studies of TCE-induced liver effects in mice and rats
Reference
Animals (sex)
Exposure route
Dose/exposure concentration
Exposed
Section(s) where noncancer liver effects are
discussed
Hamdan and
Stacey (1993)
Sprague-Dawley
rats (M)
i.p. in corn oil
0 or 131 mg/kg
6/group
4.5.4.6 Bile production
E.2.6. Serum Bile Acid Assays
Kaneko et al.
(2000)
MRL-lpr/lpr
mice (M)
Inhalation
0, 500, 1,000, or 2,000 ppm, 4 h/d,
6 d/wk, for 8 wk
5/group
4.5.4.2 Cytotoxicity and histopathology
Kjellstrand et al.
(1981b)
NMRI mice.
Sprague-Dawley
rats
Mongolian
gerbils
Inhalation
150 ppm continuous for 2-30 d
4-12/group
4.5.4.1 Liver weight
E.2.2.3. Kjellstrand et al., (1981b)
Kjellstrand et al.
(1983b)
wild, C57B1,
DBA, B6CBA,
A/sn, NZB, and
NMRI mice (M
andF)
Inhalation
150 ppm continuous for 30 d
6/group
4.5.4.1 Liver weight
E.2.2.5. Kjellstrand et al., (1983b)
Kjellstrand et al.
(1983a)
NMRI mice (M
and F)
Inhalation
0-3.600 ppm, variable time periods
of 1-24 h/d, for 30 or 120 d.
10-20/
group
4.5.4.1 Liver weight
4.5.4.2 Cytotoxicity and histopathology
E.2.2.6. Kjellstrand et al., (1983a)
Kumar et al.
(2001a)
Wistar rats (M)
Inhalation
376 ppm, 4 h/d, 5 d/wk, 8-24 wk
6/group
4.5.4.2 Cytotoxicity and histopathology
E.2.2.10. Kumar et al., 2001
Okino et al.
(1991)
Wistar rats (M)
Inhalation
0, 500 (8 h), 2,000 (2 h or 8 h), or
8,000 ppm (2 h) (single exposure)
5/group
4.5.4.2 Cytotoxicity and histopathology
E.2.1.3. Okino etal. (1991)
Ramdhan et al.
(2008)
SV/129 mice (M)
CYP2El-null
mice (M)
Inhalation
0, 1,000, or 2,000 ppm, 8 h/d, 7 d
6/group
4.5.4.2 Cytotoxicity and histopathology
4.5.6.2.1. Hepatomegally- qualitative and
quantitative comparisons
4.5.6.2.2. Cytotoxicity
E.2.1.14. Ramdhan et al. (2008)
-------
Table 4-59. Inhalation and i.p. studies of TCE-induced liver effects in mice and rats (continued)
Reference
Animals (sex)
Exposure route
Dose/exposure concentration
Exposed
Section(s) where noncancer liver effects are
discussed
Ramdhan et al.
(2010)
Sv/129,
PPARa-null, and
hPPARa mice
(M)
Inhalation
0, 1,000, or 2,000 ppm. 8 h/d, 7 d
6/group]
4.5.4.1 Liver weight
4.5.4.2 Cytotoxicity and histopathology
4.5.6.2.1. Hepatomegally- qualitative and
quantitative comparisons
4.5.6.2.2. Cytotoxicity
4.5.7.2. Peroxisome Proliterator Activated
Receptor Alpha (PPARa) Receptor Activation
E.2.1.15. Ramdhan etal. (2010)
Toraason et al.
(1999)
Fischer rats (M)
i.p. in Alkamuls/
water
0, 100, 500, or 1,000 mg/kg
6/group
4.5.4.5 Oxidative stress
E.2.4.3. Summary of Trichloroethylene (TCE)
Subchronic and Chronic Studies
E.3.4.2.3. Oxidative Stress
Wang and Stacey
(1990)
Sprague-Dawley
rats (M)
i.p. in corn oil
Inhalation
i.p.: 0, 1.3-1,314 mg/kg-day for 3 d
Inhalation: 0, 200,or 1,000 ppm, 6
h/d for 28 d
4-6/group
4.5.4.6 Bile production
E.2.2. Subchronic and Chronic Studies of
Trichloroethylene (TCE)
Watanabe and
Fukui (2000)
ddY mice (M)
i.p. in corn oil
0, 158 mg/kg (single dose)
4/group
4.5.4.4 Peroxisome proliferation and related
effects
Woolhiser et al.
(2006)
Sprague-Dawley
rats(F)
Inhalation
0,100,300, or 1,000 ppm, 6 h/d,
5 d/wk, for 4 wk
16/group
4.5.4.1 Liver weight
E.2.2.4. Woolhiser et al. (2006)
Bolded study(ies) carried forward for consideration in dose-response assessment (see Section 5).
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
TCE-induced increases in liver weight have been reported to occur quickly.
Kjellstrand et al. (1981b) reported liver weight increases after 2 days inhalation exposure in
NMRI mice, Laughter et al. (2004) reported increased liver weight in SV129 mice in their 3-days
study (see below), and Tao et al. (2000a) reported a increased in percentage liver/body weight
ratio in female B6C3Flmice for after 5 days. Elcombe et al. (1985) and Dees and Travis (1993)
reported gavage results in mice and rats after 10 days exposure to TCE which showed
TCE-induced increases in liver weight. Tucker et al. (1982) reported that 14 days of exposure to
24 mg/kg and 240 mg/kg TCE via gavage to induce a dose-related increase in liver weight in
male CD-I mice but did not show the data.
For mice, the inhalation studies of Kjellstrand et al provided the most information on the
affect of duration of exposure, dose of exposure, strain tested, gender, initial weight, and
variability in response between experiments on TCE-induced liver weight increases. These
experiments also provided results that were independent of vehicle effect. Although the
determination of the exact magnitude of response is limited by experimental design,
Kjellstrand et al. (1981b) reported that in NMRI mice, continuous TCE inhalation exposure
induced increased percentage liver/body weight by 2 days and that by 30 days (the last recorded
data point) the highest percentage liver/body weight ratio was reported (~1.75-fold over controls)
in both male and female mice. Kjellstrand et al. (1983b) exposed seven different strains of mice
(wild, C57BL, DBA, B6CBA, A/sn, NZB, NMRI) to 150-ppm TCE for 30 days and
demonstrated that strain, gender, and toxicity, as reflected by changes in whole-body weight,
affected the percentage liver/body weight ratios induced by 30 days of continuous TCE
exposure. In general for the seven strains of mice examined, female mice had the less variable
increases in TCE-induced liver weight gain across duplicate experiments than male mice. For
instance, in strains that did not exhibit changes in body weight (reflecting systemic toxicity) in
either gender (wild-type and DBA), 150-ppm TCE exposure for 30 days induced 1.74- to
1.87-fold of control percentage liver/body weight ratios in female mice and 1.45- to 2.00-fold of
control percentage liver/body weight ratios in male mice. The strain with the largest
TCE-induced increase in percentage liver/body weight increase was the NZB strain (~2.08-fold
of control for females and 2.34- to 3.57-fold of control for males). Kjellstrand et al. (1983a)
provided dose-response information for the NMRI strain of mice (A Swiss-derived strain) that
indicated dose-related increases in percentage liver/body weight ratios between 37- and 300-ppm
TCE exposure for 30 days. The 150-ppm dose was reported to induce a 1.66- and 1.69-fold
increases in percentage liver/body weight ratios in male and female mice, respectively.
Interestingly, they also reported similar liver weight increases among groups with the same
This document is a draft for review purposes only and does not constitute Agency policy.
297 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
cumulative exposure, but with different daily exposure durations (1 hour/day at 3,600 ppm to
24 hours/day at 150 ppm for 30 days).
Not only have most gavage experiments have been carried out in male mice, which
Kjellstrand et al. (1983b) had demonstrated to have more variability in response than females,
but also vehicle effects were noted to occur in experiments that examined them. Merrick et al.
(1989) reported that corn oil induced a similar increase in percentage liver/body weight ratios in
female mice fed TCE in emulphor and corn oil for 4 weeks, male mice TCE administered in the
corn oil vehicle induced a greater increase in liver weight than emulphor but less mortality at a
high does.
Buben and O'Flaherty (1985) treated male, outbred Swiss-Cox mice for 6 weeks at doses
ranging from 100-3,200 mg/kg-day, and reported increased liver/body-weight ratios at all tested
doses (1.12- to 1.75-fold of controls). Given the large strain differences observed by Kjellstrand
et al. (1983b), the use of predominantly male mice, and the effects of vehicle in gavage studies,
interstudy variability in dose-response relationships is not surprising.
Dependence of PPARa activation for TCE-liver weight gain has been investigated in
PPARa null mice by Nakajima et al. (2000), Laughter et al. (2004), and Ramdhan et al. (2010),
the latter of which also investigated PPARa null mice with human PPARa inserted. Nakajima
et al. (2000) reported that after 2 weeks of 750 mg/kg TCE exposure to carefully matched SV129
wild-type or PPARa-null male and female mice (n = 6 group), there was a reported 1.50-fold
increase in wild-type and 1.26-fold of control percentage liver/body weight ratio in PPARa-null
male mice. For female mice, there was ~1.25-fold of control percentage liver/body weight ratios
for both wild-type and PPARa-null mice. Thus, TCE-induced liver weight gain was not
dependent on a functional PPARa receptor in female mice and some portion of it may have been
in male mice. Both wild-type male and female mice were reported to have similar increases in
the number of peroxisome in the pericentral area of the liver and TCE exposure and, although
increased twofold, were still only -4% of cytoplasmic volume. Female wild-type mice were
reported to have less TCE-induced elevation of very long chain acyl-CoA synthetase, D-type
peroxisomal bifunctional protein, mitochondrial trifunctional protein a subunits a and P, and
cytochrome P450 4A1 than males mice, even though peroxisomal volume was similarly elevated
in male and female mice. The induction of PPARa protein by TCE treatment was also reported
to be slightly less in female than male wild-type mice (2.17- vs. 1.44-fold of control induction,
respectively). Thus, differences between genders in this study were for increased liver weight
were not associated with differences in peroxisomal volume in the hepatocytes but there was a
gender-related difference in induction of enzymes and proteins associated with PPARa.
This document is a draft for review purposes only and does not constitute Agency policy.
298 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
The study of Laughter et al. (2004) used SV129 wild-type and PPARa-null male mice
treated with three daily doses of TCE in 0.1% methyl cellulose for either 3 days (1,500 mg/kg
TCE) or 3 weeks (0, 10, 50, 125, 500, 1,000, or 1,500 mg/kg TCE 5 days a week). However, the
paradigm is not strictly comparable to other gavage paradigms due to the different dose vehicle
and the documented impacts of vehicles such as corn oil on TCE-induced effects. In addition, no
initial or final body weights of the mice were reported and thus, the influence of differences in
initial body weight on percentage liver/body weight determinations could not be ascertained.
While control wild-type and PPARa-null mice were reported to have similar percentage
liver/body weight ratios (i.e., -4.5%) at the end of the 3-day study, at the end of the 3-week
experiment the percentage liver/body weight ratios were reported to be larger in the control
PPARa-null male mice (5.1%). TCE treatment for 3 days was reported for percentage liver/body
weight ratio to be 1.4-fold of control in the wild-type mice and 1.07-fold of control in the null
mice. After 3 weeks of TCE exposure at varying concentrations, wild-type mice were reported
to have percentage liver/body weight ratios that were within -2% of control values with the
exception of the 1,000 mg/kg and 1,500 mg/kg treatment groups (-1.18- and 1.30-fold of
control, respectively). For the PPARa-null mice the variability in percentage liver/body weight
ratios were reported to be greater than that of the wild-type mice in most of the TCE groups and
the baseline levels of percentage liver/body weight ratio for control mice 1.16-fold of that of
wild-type mice. TCE exposure was apparently more toxic in the PPARa-null mice. Decreased
survival at the 1,500 mg/kg TCE exposure level resulted in the prevention of recording of
percentage liver/body weight ratios for this group. At 1,000 mg/kg TCE exposure level, there
was a reported 1.10-fold of control percentage liver/body weight ratio in the PPARa-null mice.
None of the increases in percentage liver/body weight in the null mice were reported to be
statistically significant by Laughter et al. (2004). However, the power of the study was limited
due to low numbers of animals and increased variability in the null mice groups. The percentage
liver/body weight ratio after TCE treatment reported in this study was actually greater in the
PPARa-null mice than the wild-type male mice at the 1,000 mg/kg TCE exposure level
(5.6 ± 0.4% vs. 5.2 ± 0.5%), for PPARa-null and wild-type mice, respectively) resulting in a
1.18-fold of wild-type and 1.10-fold of PPARa-null mice. Although the results reported in
Laughter et al. (2004) for DCA and TCA were not conducted in experiments that used the same
paradigm, the TCE-induced increase in percentage liver/body weight more closely resembled the
dose-response pattern for DCA than for DCA wild-type SV129 and PPARa-null mice.
Ramdhan et al. (2010) examined TCE-induced hepatice steatosis and toxicity using male
wild type, PPARa-null, and human PPARa inserted ("humanized") mice exposed to high
inhalation concentrations of TCE for 7 days. Significant differences were observed among
This document is a draft for review purposes only and does not constitute Agency policy.
299 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
control mice for each genotype with reduced body weight in untreated humanized mice.
Liver/body weight ratios were 11% higher in untreated PPARa-null mice than wild type mice.
Higher levels of liver triglycerides and hepatic steatosis were reported in the untreated
humanized mice and PPARa null mice than wild type mice. Background expression of a number
of genes and protein expression levels were significantly different between the untreated strains.
In particular, human PPARa protein levels were > 10-fold greater in the humanized mice than
mouse PPARa in untreated wild type mice. Insertion of human PPARa in the null mice did not
return the mice to a normal state. Both PPARa null and humanized mice were more susceptible
to TCE toxicity. Hepatomegally was induced in all strains to a similar extent after TCE
exposure. However, urinary TCA concentrations were reported to be significantly lower and
trichloroethanol levels significantly higher in TCE-treated PPARa-null mice in comparison to
treated wild type mice. This difference was not related to changes in expression of metabolic
enzymes.
No study examined strain differences among rats, and cross-study comparisons are
confounded by heterogeneity in the age of animals, dosing regimen, and other design
characteristics that may affect the degree of response. For rats, TCE-induced percentage
liver/body weight ratios were reported to range from 1.16- to 1.46-fold of control values
depending on the study paradigm. The studies which employed the largest range of exposure
concentrations (Berman et al., 1995; Melnick et al., 1987) examined four doses in the rat. In
general, there was a dose-related increase in percentage liver/body weight in the rat, especially at
doses that did not cause concurrent decreased survival or significant body weight loss. For
gerbils, Kjellstrand et al. (1981b) reported a similar value of ~1.25-fold of control percentage
liver/body weight as for S-D rats exposed to 150 ppm TCE continuously for 30 days. Woolhiser
et al. (2006) also reported inhalation TCE exposure to increase the percentage liver/body weight
ratios in female Sprague-Dawley rats although this strain appeared to be less responsive that
others tested for induction of hepatomegaly from TCA exposure and to also be less prone to
spontaneous liver cancer.
The size of the liver is under tight control and after cessation of a mitogenic stimulus or
one inducing hepatomegaly, the liver will return to its preprogrammed size (see Appendix E).
The increase in liver weight from TCE-exposure also appears to be reversible. Kjellstrand et al.
(1981b) reported a reduction in liver weight gain increases after cessation of TCE exposure for 5
or 30 days in male and female mice. However, experimental design limitations precluded
discernment of the magnitude of decrease. Kjellstrand et al. (1983a) reported that mice exposed
to 150 ppm TCE for 30 days and then examined 120 days after the cessation of exposure, had
liver weights were 1.09-fold of control for TCE-exposed female mice and the same as controls
This document is a draft for review purposes only and does not constitute Agency policy.
300 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
for TCE-exposed male mice. However, the livers were not the same as untreated liver in terms
of histopathology. The authors reported that "after exposure to 150 ppm for 30 days, followed
by 120 days of rehabilitation, the morphological picture was similar to that of the air-exposure
controls except for changes in cellular and nuclear sizes." Qualitatively, the reduction in liver
weight after treatment cessation is consistent with the report of Elcombe et al. (1985) in Alderly
Park mice. The authors report that the reversibility of liver effects after the administration of
TCE to Alderly Park mice for 10 consecutive days. Effects upon liver weight, DNA
concentration, and tritiated thymidine incorporation 24 and 48 hours after the last dose of TCE
were reported to still be apparent. However, 6 days following the last dose of TCE, all of these
parameters were reported to return to control values with the authors not showing the data to
support this assertion. Thus, cessation of TCE exposure would have resulted in a 75% reduction
in liver weight by 4 days in mice exposed to the highest TCE concentration. Quantitative
comparisons are not possible because Elcombe et al. (1985) did not report data for these results
(e.g., how many animals, what treatment doses, and differences in baseline body weights) and
such a large decrease in such a short period of time needs to be verified.
4.5.4.1.2. Cytotoxicity and Histopathology
Acute exposure to TCE appears to induce low cytotoxicity below subchronically lethal
doses. Relatively high doses of TCE appear necessary to induce cytotoxicity after a single
exposure with two available studies reported in rats. Okino et al. (1991) reported small increases
in the incidence of hepatocellular necrosis in male Wistar rats exposed to 2,000 ppm (8 hours)
and 8,000 ppm (2 hours), but not at lower exposures. In addition, "swollen" hepatocytes were
noted at the higher exposure when rats were pretreated with ethanol or Phenobarbital. Serum
transaminases increased only marginally at the 8,000-ppm exposure, with greater increases with
pretreatments. Berman et al. (1995) reported hepatocellular necrosis, but not changes in serum
markers of necrosis, after single gavage doses of 1,500 and 5,000 mg/kg TCE in female F344
rats. However, they did not report any indications of necrosis after 14 days of treatment at
50-1,500 mg/kg-day nor the extent of necrosis.
At acute and subchronic exposure periods to multiple doses, the induction of cytotoxicity,
though usually mild, appears to differ depending on rodent species, strain, dosing vehicle and
duration of exposure, and the extent of reporting to vary between studies. For instance,
Elcombe et al. (1985) and Dees and Travis (1993), which used the B6C3F1 mouse strain and
corn oil vehicle, reported only slight or mild necrosis after 10 days of treatment with TCE at
doses up to 1,500 mg/kg-day. Elcombe et al. (1985) also reported cell hypertrophy in the
This document is a draft for review purposes only and does not constitute Agency policy.
3 01 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
centrilobular region. Dees and Travis (1993) reported some loss of vacuolization in hepatocytes
of mice treated at 1,000 mg/kg-day. Laughter et al. (2004) reported that "wild-type" SV129
mice exposed to 1,500 mg/kg TCE exposure for 3 weeks exhibited mild granuloma formation
with calcification or mild hepatocyte degeneration but gave no other details or quantitative
information as to the extent of the lesions or what parts of the liver lobule were affected. The
authors noted that "wild-type mice administered 1,000 and 1,500 mg/kg exhibited centrilobular
hypertrophy" and that "the mice in the other groups did not exhibit any gross pathological
changes" after TCE exposure. Channel et al. (1998) reported no necrosis in B6C3F1 mice
treated by 400-1,200 mg/kg-day TCE by corn oil gavage for 2 days to 8 weeks.
However, as stated above, Merrick et al. (1989) reported that corn oil resulted in more
hepatocellular necrosis, as described by small focal areas of 3-5 hepatocytes, in male B6C3F1
mice than use of emulphor as a vehicle for 4-week TCE gavage exposures. Necrotic hepatocytes
were described as surrounded by macrophages and polymorphonuclear cells. The authors
reported that visible necrosis was observed in 30-40% of male mice administered TCE in corn
oil but not that there did not appear to be a dose-response. For female mice, the extent of
necrosis was reported to be 0 for all control and TCE treatment groups using either vehicle.
Serum enzyme activities for alanine aminotransferase (ALT), AST, and LDH (markers of liver
toxicity) showed that there was no difference between vehicle groups at comparable TCE
exposure levels for male or female mice. Except for LDH levels in male mice exposed to TCE
in corn oil there was not a correlation with the extent of necrosis and the patterns of increases in
ALT and AST enzyme levels.
Ramdhan et al. (2008) assessed TCE-induced hepatotoxicity by measuring plasma ALT
and AST activities and histopathology in Sv/129 mice treated by inhalation exposure, which are
not confounded by vehicle effects. Despite high variability and only six animals per dose group,
all three measures showed statistically significant increases at the high dose of 2,000 ppm
(8 hours/day for 7 days), although a nonstatistically significant elevation is evident at the low
dose of 1,000 ppm. Even at the highest dose, cytotoxicity was not severe, with ALT and AST
measures increased twofold or less and an average histological score less than two (range 0-4).
Using the same paradigm, Ramdhan et al. (2010) also reported increased in AST and
ALT liver injury biomarkers to be significantly increased in all exposed mice (Sv/129 wild type,
PPARa-null and humanized PPARa mice) relative to controls (41-74% and 36-79% higher,
respectively) with mean levels within each group higher, though not statistically significantly
different, with exposure to 2,000 versus 1,000 ppm TCE. . Steatosis scores were reported to be
significantly higher in the 2,000 versus 1,000 ppm TCE exposures to PPARa-null mice. The
authors reported steatosis scored to be significantly correlated with liver triglyceride levels of all
This document is a draft for review purposes only and does not constitute Agency policy.
3 02 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
mice examined in the study (r = 0.75). Macrovesicular steatosis was reported to occur more
frequently in hPPARa than PPARa-null mice. Necrosis scores were reported to be significantly
higher in TCE exposed mice relative to controls in all three genotype mice and to be significantly
higher with 2,000 versus 1,000 ppm TCE exposure in wild type mice and hPPARa mice.
Inflammation scores were reported to be significantly higher with exposed group than control
with 2,000 ppm TCE exposure than controls for each genotype group with a difference between
the 2,000 ppm and 1,000 ppm exposure groups in wild type mice.
Kjellstrand et al. (1983a) exposed male and female NRMI mice to 150 ppm for
30-120 days. Kjellstrand et al. (1983a) reported more detailed light microscopic findings from
their study and stated that
After 150 ppm exposure for 30 days, the normal trabecular arrangement of the liver cells
remained. However, the liver cells were generally larger and often displayed a fine vacuolization
of the cytoplasm. The nucleoli varied slightly to moderately in size and shape and had a finer,
granular chromatin with a varying basophilic staining intensity. The Kupffer cells of the sinusoid
were increased in cellular and nuclear size. The intralobular connective tissue was infiltrated by
inflammatory cells. There was not sign of bile stasis. Exposure to TCE in higher or lower
concentrations during the 30 days produced a similar morphologic picture. After intermittent
exposure for 30 days to a time-weighted-average concentration of 150 ppm or continuous
exposure for 120 days, the trabecular cellular arrangement was less well preserved. The cells had
increased in size and the variations in size and shape of the cells were much greater. The nuclei
also displayed a greater variation in basophilic staining intensity, and often had one or two
enlarged nucleoli. Mitosis was also more frequent in the groups exposed for longer intervals. The
vacuolization of the cytoplasm was also much more pronounced. Inflammatory cell infiltration in
the interlobular connective tissue was more prominent. After exposure to 150 ppm for 30 days,
followed by 120 days of rehabilitation, the morphological picture was similar to that of the
air-exposure controls except for changes in cellular and nuclear sizes.
Although not reporting comparisons between male and female mice in the results section
of the paper for TCE-induced histopathological changes, the authors stated in the discussion
section that "However, liver mass increase and the changes in liver cell morphology were similar
in TCE-exposed male and female mice." Kjellstrand et al. (1983a) did not present any
quantitative data on the lesions they describe, especially in terms of dose-response. Most of the
qualitative description presented was for the 150-ppm exposure level and the authors suggest that
lower concentrations of TCE give a similar pathology as those at the 150-ppm level, but do not
present data to support that conclusion. Although stating that Kupffer cells were reported to be
increased in cellular and nuclear size, no differential staining was applied light microscopy
sections to distinguish Kupffer from endothelial cells lining the hepatic sinusoid in this study.
Without differential staining such a determination is difficult at the light microscopic level.
This document is a draft for review purposes only and does not constitute Agency policy.
303 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Indeed, Goel et al. (1992) describe proliferation of "sinusoidal endothelial cells" after
1,000 and 2,000 mg/kg-day TCE exposure for 28 days in male Swiss mice. They reported that
histologically, "the liver exhibits swelling, vacuolization, widespread degeneration/necrosis of
hepatocytes as well as marked proliferation of endothelial cells of hepatic sinusoids at 1,000 and
2,000 mg/kg TCE doses." Only one figure is given, at the light microscopic level, in which it is
impossible to distinguish endothelial cells from Kupffer cells and no quantitative measures or
proliferation were examined or reported to support the conclusion that endothelial cells are
proliferating in response to TCE treatment. Similarly, no quantitative analysis regarding the
extent or location of hepatocellular necrosis was given. The presence or absence of
inflammatory cells were not noted by the authors as well. In terms of white blood cell count, the
authors note that it is slightly increased at 500 mg/kg-day but decreased at 1,000 and
2,000 mg/kg-day TCE, perhaps indicating macrophage recruitment from blood to liver and
kidney, which was also noted to have pathology at these concentrations of TCE.
The inflammatory cell infiltrates described in the Kjellstrand et al. (1983a) study are
consistent with invasion of macrophages and well as polymophorphonuclear cells into the liver,
which could activate resident Kupffer cells. Although not specifically describing the changes as
consistent with increased polyploidization of hepatocytes, the changes in cell size and especially
the continued change in cell size and nuclear staining characteristics after 120 days of cessation
of exposure are consistent with changes in polyploidization induced by TCE. Of note is that in
the histological description provided by the authors, although vacuolization is reported and
consistent with hepatotoxicity or lipid accumulation, which is lost during routine histological
slide preparation, there is no mention of focal necrosis or apoptosis resulting from these
exposures to TCE.
Buben and O'Flaherty (1985) reported liver degeneration "as swollen hepatocytes" and to
be common with treatment of TCE to Male Swiss-Cox mice after 6 weeks. They reported that
"Cells had indistinct borders; their cytoplasm was clumped and a vesicular pattern was apparent.
The swelling was not simply due to edema, as wet weight/dry weight ratios did not increase."
Karyorrhexis (the disintegration of the nucleus) was reported to be present in nearly all
specimens and suggestive of impending cell death. No Karyorrhexis, necrosis, or polyploidy
was reported in controls, but a low score Karyorrhexis was given for 400 mg/kg TCE and a
slightly higher one given for 1,600 mg/kg TCE. Central lobular necrosis reported to be present
only at the 1,600 mg/kg TCE exposure level and assigned a low score. Polyploidy was described
as characteristic in the central lobular region but with low score for both 400 mg/kg and
1,600 mg/kg TCE exposures. The authors reported that "hepatic cells had two or more nuclei or
had enlarged nuclei containing increased amounts of chromatin, suggesting that a regenerative
This document is a draft for review purposes only and does not constitute Agency policy.
3 04 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
process was ongoing" and that there were no fine lipid droplets in TCE exposed animals. The
finding of "no polyploidy" in control mouse liver in the study of Buben and O'Flaherty (1985) is
unexpected given that binucleate and polyploid hepatocytes are a common finding in the mature
mouse liver. It is possible that the authors were referring to unusually high instances of
"polyploidy" in comparison to what would be expected for the mature mouse. The score given
by the authors for polyploidy did not indicate a difference between the two TCE exposure
treatments and that it was of the lowest level of severity or occurrence. No score was given for
centrolobular hypertrophy although the DNA content and liver weight changes suggested a
dose-response. The "Karyrrhexis" described in this study could have been a sign of cell death
associated with increased liver cell number or dying of maturing hepatocytes associated with the
increased ploidy, and suggests that TCE treatment was inducing polyploidization. Consistent
with enzyme analyses, centrilobular necrosis was only seen at the highest dose and with the
lowest qualitative score, indicating that even at the highest dose there was little toxicity.
At high doses, Kaneko et al. (2000) reported sporadic necrosis in male Mrl-lpr/lpr mice,
which are "genetically liable to autoimmune disease," exposed to 500-2,000 ppm, 4hours/day,
6 days/week, for 8 weeks (n = 5). Dose-dependent mild inflammation and associated changes
were reported to be found in the liver. The effects on hepatocytes were reported to be minimal
by the authors with 500-ppm TCE inducing sporadic necrosis in the hepatic lobule. Slight
mobilization and activation of sinusoid lining cells were also noted. These pathological features
were reported to increase with dose.
NTP (1990), which used the B6C3F1 mouse strain, reported centrilobular necrosis in
6/10 male and 1/10 female B6C3F1 mice treated at a dose of 6,000 mg/kg-day for up to
13 weeks (all the male mice and 8 of the 10 female mice died in the first week of treatment). At
3,000 mg/kg-day exposure level, although centrilobular necrosis was not observed, 2/10 males
had multifocal areas of calcification in their livers, which the authors suggest is indicative of
earlier hepatocellular necrosis. However, only 3/10 male mice at this dose survived to the end of
the 13-week study.
For the NTP (1990) 2-year study, B6C3F1 mice were reported to have no
treatment-related increase in necrosis in the liver. A slight increase in the incidence of focal
necrosis was noted TCE-exposed male mice (8 vs. 2%) with a slight reduction in fatty
metamorphosis in treated male mice (zero treated vs. two control animals) and in female mice a
slight increase in focal inflammation (29 vs. 19% of animals) and no other changes. Therefore,
this study did not show concurrent evidence of liver toxicity with TCE-induced neoplasia after
2 years of TCE exposure in mice.
This document is a draft for review purposes only and does not constitute Agency policy.
305 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
For the more limited database in rats, there appears to be variability in reported TCE
induced cytotoxicity and pathology. Nunes et al. (2001) reported no gross pathological changes
in rats gavaged with corn oil or with corn oil plus 2,000 mg/kg TCE for 7 days. Goldsworthy
and Popp (1987) gave no descriptions of liver histology given in this report for TCE-exposed
animals or corn-oil controls. Kjellstrand et al. (1981b) gave also did not give histological
descriptions for livers of rats in their inhalation study.
Elcombe et al. (1985) provided a description of the histopathology at the light
microscopy level in Osborne-Mendel rats, and Alderly Park rats exposed to TCE via gavage for
10 days. However, they did not provide a quantitative analysis or specific information regarding
the variability of response between animals within group and there was no indication by the
authors regarding how many rats were examined by light microscopy. Hematoxylin and eosin
sections from Osborne-Mendel rats were reported to show that
Livers from control rats contained large quantities of glycogen and isolated inflammatory foci, but
were otherwise normal. The majority of rats receiving 1,500 mg/kg body weight TCE showed
slight changes in centrilobular hepatocytes. The hepatocytes were more eosinophilic and
contained little glycogen. At lower doses these effects were less marked and were restricted to
fewer animals. No evidence of treatment-related hepatotoxicity (as exemplified by single cell or
focal necrosis) was seen in any rat receiving TCE. H&E [hematoxylin and eosin] sections from
Alderly Park Rats showed no signs of treatment-related hepatotoxicity after administration of
TCE. However, some signs of dose-related increase in centrilobular eosinophilia were noted.
Thus, both mice and rats were reported to exhibit pericentral hypertrophy and
eosinophilia as noted from the histopathological examination in Elcombe et al. (1985).
Berman et al. (1995) reported that for female rats exposed to TCE for 14 days
hepatocellular necrosis was noted to occur in the 1,500 and 5,000 mg/kg groups in 6/7 and
6/8 female rats, respectively but not to occur in lower doses. The extent of necrosis was not
noted by the authors for the two groups exhibiting a response after 1 day of exposure. Serum
enzyme levels, indicative of liver necrosis, were not presented and because only positive results
were presented in the paper, presumed to be negative. Therefore, the extent of necrosis was not
of a magnitude to affect serum enzyme markers of cellular leakage.
Melnick et al. (1987) reported that the only treatment-related lesion observed
microscopically in rats from either dosed-feed or gavage groups was individual cell necrosis of
the liver with the frequency and severity of this lesion similar at each dosage levels of TCE
microencapsulated in the feed or administered in corn oil. The severity for necrosis was only
mild at the 2.2 and 4.8 g/kg feed groups and for the six animals in the 2.8 g/kg group corn oil
group. The individual cell necrosis was reported to be randomly distributed throughout the liver
This document is a draft for review purposes only and does not constitute Agency policy.
306 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
lobule with the change to not be accompanied by an inflammatory response. The authors also
reported that there was no histologic evidence of cellular hypertrophy or edema in hepatic
parenchymal cells. Thus, although there appeared to be TCE-treatment related increases in focal
necrosis after 14 days of exposure, the extent was mild even at the highest doses and involved
few hepatocytes.
For the 13-week NTP study (1990), only control and high dose F344/N rats were
examined histologically. Pathological results were reported to reveal that 6/10 males and
6/10 female rats had pulmonary vasculitis at the highest concentration of TCE. This change was
also reported to have occurred in 1/10 control male and female rats. Most of those animals were
also reported to have had mild interstitial pneumonitis. The authors report that viral titers were
positive during this study for Sendai virus.
Kumar et al. (2001) reported that male Wistar rats exposed to 376 ppm, 4 hours/day,
5 days/week for 8-24 weeks showed evidence of hepatic toxicity. The authors stated that, "after
8 weeks of exposure enlarged hepatocytes, with uniform presence of fat vacuoles were found in
all of the hepatocytes affecting the periportal, midzonal, and centriolobular areas, and fat
vacuoles pushing the pyknosed nuclei to one side of hepatocytes. Moreover, congestion was not
significant. After exposure of 12 and 24 weeks, the fatty changes became more progressive with
marked necrosis, uniformly distributed in the entire organ." No other description of pathology
was provided in this report. In regard to the description of fatty change, the authors only did
conventional H&E staining of sections with no precautions to preserve or stain lipids in their
sections. However, as noted below, the NCI study also reports long-term TCE exposure in rats
to result in hepatocellular fatty metamorphosis. The authors provided a table with histological
scoring of simply + or - for minimal, mild or moderate effects and do not define the criteria for
that scoring. There is also no quantitative information given as to the extent, nature, or location
of hepatocellular necrosis. The authors report "no change was observed in glutamic oxoacetate
transaminase and glutamic pyruvated transaminase levels of liver in all the three groups. The
GSH level was significantly decreased while "total sulphydryl" level was significantly increased
during 8, 12, and 24 weeks of TCE exposure. The acid and alkaline phosphatases were
significantly increased during 8, 12, and 24 weeks of TCE exposure." The authors present a
series of figures that are poor in quality to demonstrate histopathological TCE-induced changes.
No mortality was observed from TCE exposure in any group despite the presence of liver
necrosis.
Thus, in this limited database that spans durations of exposure from days to 24 weeks and
uses differing routes of administration, generally high doses for long durations of exposure are
required to induce hepatotoxicity from TCE exposure in the rat. The focus of 2-year bioassays in
This document is a draft for review purposes only and does not constitute Agency policy.
307 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
rats has been the detection of a cancer response with little or no reporting of noncancer pathology
in most studies. Henschler et al. (1984) and Fukuda et al. (1983) do not report noncancer
histopathology, but do both report rare biliary cell derived tumors in rats in relatively insensitive
assays. For male rats, noncancer pathology in the NCI (1976) study was reported to include
increased fatty metamorphosis after TCE exposure and angiectasis or abnormally enlarged blood
vessels. Angiectasis can be manifested by hyperproliferation of endothelial cells and dilatation
of sinusoidal spaces. For the NTP (1990) study there was little reporting of non-neoplastic
pathology or toxicity and no report of liver weight at termination of the study. In the NTP
(1988) study, the 2 year study of TCE exposure reported no evidence of TCE-induced liver
toxicity described as non-neoplastic changes in ACI, August, Marshal, and Osborne-Mendel rats.
Interestingly, for the control animals of these four strains there was, in general, a low background
level of focal necrosis in the liver of both genders. Obviously, the negative results in this
bioassay for cancer are confounded by the killing of a large portion of the animals accidently by
experimental error but TCE-induced overt liver toxicity was not reported.
In sum, the cytotoxic effects in the liver of TCE treatment appear include little or no
necrosis in the rodent liver, but rather, a number of histological changes such as mild focal
hepatocyte degeneration at high doses, cellular "swelling" or hypertrophy, and enlarged nuclei.
Histological changes consistent with increased polyploidization and specific descriptions of
TCE-induced polyploidization have been noted in several experiments. Several studies note
proliferation of nonparenchymal cells after TCE exposure as well. These results are more
consistently reported in mice, but also have been reported in some studies at high doses in rats,
for which fewer studies are available. In addition, the increase in cellular and nuclear sizes
appeared to persist after cessation of TCE treatment. In neither rats nor mice is there evidence
that TCE treatment results in marked necrosis leading to regenerative hyperplasia.
4.5.4.1.3. Measures of DNA Synthesis, Cellular Proliferation, and Apoptosis
The increased liver weight observed in rodents after TCE exposure may result from either
increased numbers of cells in the liver, increased size of cells in the liver, or a combination of
both. Studies of TCE in rodents have studied whole liver DNA content of TCE-treated animals
to determine whether the concentration of DNA per gram of liver decreases as an indication of
hepatocellular hypertrophy (Buben and O'Flaherty, 1985; Dees and Travis, 1993; Elcombe et al.,
1985). While the slight decreases observed in some studies are consistent with hypertrophy, the
large variability in controls and lack of dose-response limits the conclusions that can be drawn
from these data. In addition, multiple factors beyond hypertrophy affect DNA concentration in
This document is a draft for review purposes only and does not constitute Agency policy.
308 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
whole-liver homogenates, including changes in ploidy and the number of hepatocytes and
nonparenchymal cells.
The incorporation of tritiated thymidine or BrdU has also been analyzed in whole liver
DNA and in individual hepatocytes as a measure of DNA synthesis. Such DNA synthesis can
occur from either increased numbers of hepatocytes in the liver or by increased polyploidization.
Section E. 1.1 describes polyploidization in human and rodent liver and its impacts on liver
function, while Sections E.3.1.2 and E.3.3.1 discuss issues of target cell identification for liver
cancer and changes in ploidy as a key even in liver cancer using animals models, respectively.
Along with changes in cell size (hypertrophy), cell number (cellular proliferation), and the DNA
content per cell (cell ploidy), the rate of apoptosis has also been noted or specifically examined
in some studies of TCE and its metabolites. All of these phenomena have been identified in
proposed hypotheses as key events possibly related to carcinogenicity. In particular, changes in
cell proliferation and apoptosis have been postulated to be part of the MOA for PPARa-agonists
by Klaunig et al. (2003) (see Section E.3.4).
In regard to early changes in DNA synthesis, the data for TCE are very limited
Mirsalis et al. (1989) reported measurements of in vivo-in vitro hepatocyte DNA repair and
S-phase DNA synthesis in primary hepatocytes from male Fischer 344 rats and male and female
B6C3F1 mice administered single doses of TCE by gavage in corn oil. They reported negative
results 2-12 hours after treatment from 50-1,000 mg/kg TCE in rats and mice (male and female)
for unscheduled DNA synthesis and repair using three animals per group. After 24 and 48 hours
of 200 or 1,000 mg/kg TCE in male mice (n = 3) and after 48 hours of 200 (n = 3) or
1,000 (n = 4) mg/kg TCE in female mice, similar values of 0.30-0.69% of hepatocytes were
reported as undergoing DNA synthesis in primary culture. Only the 1,000 mg/kg TCE dose in
male mice at 48 hours was reported to give a result considered to be positive (-2.2% of
hepatocytes) but no statistical analyses were performed on these measurements. These results
are limited by both the number of animals examined and the relevance of the paradigm.
As noted above, TCE treatment in rodents has been reported to result in hepatocellular
hypertrophy and increased centrilobular eosinophilia. Elcombe et al. (1985) reported a small
decrease in DNA content with TCE treatment (consistent with hepatocellular hypertrophy) that
was not dose-related, increased tritiated thymidine incorporation in whole mouse liver DNA that
was that was treatment but not dose-related (i.e., a two-, two-, and fivefold of control in mice
treated with 500, 1,000, and 1,500 mg/kg TCE), and slightly increased numbers of mitotic
figures that were treatment but not dose-related and not correlated with DNA synthesis as
measured by thymidine incorporation. Elcombe et al., reported no difference in response
between 500 and 1,000 mg/kg TCE treatments for tritiated thymidine incorporation. Dees and
This document is a draft for review purposes only and does not constitute Agency policy.
309 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Travis (1993) also reported that incorporation of tritiated thymidine in DNA from mouse liver
was elevated after TCE treatment with the mean peak level of tritiated thymidine incorporation
occurred at 250 mg/kg TCE treatment level and remaining constant for the 500 and 1,000 mg/kg
treated groups. Dees and Travis (1993) specifically report that mitotic figures, although very
rare, were more frequently observed after TCE treatment, found most often in the intermediate
zone, and found in cells resembling mature hepatocytes. They reported that there was little
tritiated thymidine incorporation in areas near the bile duct epithelia or close to the portal triad in
liver sections from both male and female mice. Channel et al. (1998) reported proliferating cell
nuclear antigen (PCNA) positive cells, a measure of cells that have undergone DNA synthesis,
was elevated only on Day 10 (out of the 21 studied) and only in the 1,200 mg/kg-day TCE
exposed group with a mean of -60 positive nuclei per 1,000 nuclei for six mice (-6%). Given
that there was little difference in PCNA positive cells at the other TCE doses or time points
studied, the small number of affected cells in the liver could not account for the increase in liver
size reported in other experimental paradigms at these doses. The PCNA positive cells as well as
"mitotic figures" were reported to be present in centrilobular, midzonal, and periportal regions
with no observed predilection for a particular lobular distribution. No data were shown
regarding any quantitative estimates of mitotic figures and whether they correlated with PCNA
results. Thus, whether the DNA synthesis phases of the cell cycle indicated by PCNA staining
were indentifying polyploidization or increased cell number cannot be determined.
For both rats and mice, the data from El combe et al. (1985) showed that tritiated
thymidine incorporation in total liver DNA observed after TCE exposure did not correlate with
mitotic index activity in hepatocytes. Both Elcombe et al. (1985) and Dees and Travis (1993)
reported a small mitotic indexes and evidence of periportal hepatocellular hypertrophy from TCE
exposure. Neither mitotic index or tritiated thymidine incorporation data support a correlation
with TCE-induced liver weight increase in the mouse, but rather the increase to be most likely
due to hepatocellular hypertrophy. If higher levels of hepatocyte replication had occurred
earlier, such levels were not sustained by 10 days of TCE exposure. These data suggest that
increased tritiated thymidine levels were targeted to mature hepatocytes and in areas of the liver
where greater levels of polyploidization occur (see Section E. 1.1). Both Elcombe et al. (1985)
and Dees and Travis (1993) show that tritiated thymidine incorporation in the liver was -twofold
greater than controls between 250-1,000 mg/kg TCE, a result consistent with a doubling of
DNA. Thus, given the normally quiescent state of the liver, the magnitude of this increase over
control levels, even if a result of proliferation rather than polyploidization, would be confined to
a very small population of cells in the liver after 10 days of TCE exposure.
This document is a draft for review purposes only and does not constitute Agency policy.
310 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Laughter et al. (2004) reported that there was an increase in DNA synthesis after aqueous
gavage exposure to 500 and 1,000 mg/kg TCE given as three boluses a day for 3 weeks with
BrdU given for the last week of treatment. An examination of DNA synthesis in individual
hepatocytes was reported to show that 1 and 4.5% of hepatocytes had undergone DNA synthesis
in the last week of treatment for the 500 and 1,000 mg/kg doses, respectively. Again, this level
of DNA synthesis is reported for a small percentage of the total hepatocytes in the liver and not
reported to be a result of regenerative hyperplasia.
Finally, Dees and Travis (1993) and Channel et al. (1998) reported evaluating changes in
apoptosis with TCE treatment. Dees and Travis (1993) enumerated identified by either
hematoxylin and eosin or feulgen staining in male and female mice after 10 days of TCE
treatment by. Only zero or one apoptosis was observed per 100 high power (400x) fields in
controls and all dose groups except for those given 1,000 mg/kg-day, in which eight or nine
apoptoses per 100 fields were reported. None of the apoptoses were in the intermediate zones
where mitotic figures were observed, and all were located near the central veins. This is the
same region where one would expect endogenous apoptoses as hepatocytes "stream" from the
portal triad toward the central vein (Schwartz-Arad et al., 1989). In addition, this is the same
region where Buben and O'Flaherty (1985) noted necrosis and polyploidy. By contrast Channel
et al. (1998) reported no significant differences in apoptosis at any treatment dose (400-1,200
mg/kg-day) examined after any time from 2 days to 4 weeks.
4.5.4.1.4. Peroxisomal Proliferation and Related Effects
Numerous studies have reported that TCE administered to mice and rats by gavage leads
to proliferation of peroxisomes in hepatocytes. Some studies have measured changes in the
volume and number of peroxisomes as measures of peroxisome proliferation while others have
measured peroxisomal enzyme activity such catalase and cyanide-insensitive PCO. Like liver
weight, the determination of a baseline level of peroxisomal volume, number, or enzyme activity
can be variable and have great effect on the ability to determine the magnitude of a
treatment-related effect.
Elcombe et al. (1985) reported increases in the percentage of the cytoplasm occupied by
peroxisomes in B6C3F1 and Alderley Park mice treated for 10 days at 500-1,500 mg/kg-day.
Although the increase over controls appeared larger in the B6C3F1 strain, this is largely due to
the twofold smaller control levels in that strain, as the absolute percentage of peroxisomal
volume was similar between strains after treatment. All these results showed high variability, as
evidenced from the reported standard deviations. Channel et al. (1998) found a similar absolute
This document is a draft for review purposes only and does not constitute Agency policy.
311 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
percentage of peroxisomal volume after 10 days treatment in the B6C3F1 mouse at
1,200 mg/kg-day TCE but with the percentage in vehicle controls similar to the Alderley-Park
mice in the Elcombe et al. (1985) study. Interestingly, Channel et al. (1998) found that the
increase in peroxisomes peaked at 10 days, with lower values after 6 and 14 days of treatment.
Furthermore, the vehicle control levels also varied almost twofold depending on the number of
days of treatment. Nakajima et al. (2000), who treated male wild-type SV129 mice at
750 mg/kg-day for 14 days, found even higher baseline values for the percentage of peroxisomal
volume, but with an absolute level after treatment similar to that reported by Channel et al.
(1998) in B6C3F1 mice treated at 1,200 mg/kg-day TCE for 14 days. Nakajima et al. (2000)
also noted that the treatment-related increases were smaller for female wild-type mice, and that
there were no increases in peroxisomal volume in male or female PPARa-null mice, although
vehicle control levels were slightly elevated (not statistically significant). Only Elcombe et al.
(1985) examined peroxisomal volume in rats, and reported smaller treatment-related increases in
two strains (OM and AP), but higher baseline levels. In particular, at 1,000 mg/kg-day, after
10 days treatment, the percentage peroxisomal volume was similar in OM and AP rats, with
similar control levels as well. While the differences from treatment were not statistically
significant, only five animals were used in each group, and variability, as can be seen by the
standard deviations, was high, particularly in the treated animals.
The activities of a number of different hepatic enzymes have also been as markers for
peroxisome proliferation and/or activation of PPARa. The most common of these are catalase
and cyanide-insensitive PCO. In various strains of mice (B6C3F1, Swiss albino, SV129
wild-type) treated at doses of 500-2,000 mg/kg-day for 10-28 days, increases in catalase activity
have tended to be more modest (1.3- to 1.6-fold of control) as compared to increases in PCO
(1.4- to 7.9-fold of control) (Elcombe et al., 1985; Goel et al., 1992; Goldsworthy and Popp,
1987; Laughter et al., 2004; Nakajima et al., 2000; Watanabe and Fukui, 2000). In rats, Elcombe
et al. (1985) reported no increases in catalase or PCO activity in Alderley-Park rats treated at
1,000 mg/kg-day TCE for 10 days. In F344 rats, Goldsworthy and Popp (1987) and Melnick
et al. (1987) reported increases of up to twofold in catalase and 4.1-fold in PCO relative to
controls treated at 600-4,800 mg/kg-day for 10-14 days. The changes in catalase were similar
to those in mice at similar treatment levels, with 1.1- to 1.5-fold of control enzyme activities at
doses of 1,000-1,300 mg/kg-day (Elcombe et al., 1985; Melnick et al., 1987). However, the
changes in PCO were smaller, with 1.1- to 1.8-fold of control activity at these doses, as
compared to 6.3- to 7.9-fold of control in mice (Goldsworthy and Popp, 1987; Melnick et al.,
1987).
This document is a draft for review purposes only and does not constitute Agency policy.
312 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
In SV129 mice, Nakajima et al. (2000) and Laughter et al. (2004) investigated the
dependence of these changes on PPARa by using a null mouse. Nakajima et al. reported that
neither male nor female wild-type or PPARa null mice had significant increases in catalase after
14 days of treatment at 750 mg/kg-day. However, given the small number of animals (4 per
group) and the relatively small changes in catalase observed in other (wild-type) strains of mice,
this study had limited power to detect such changes. Several other markers of peroxisome
proliferation, including acyl-CoA oxidase and CYP4A1 (PCO was not investigated), were
induced by TCE in male wild-type mice, but not in male null mice or female mice of either type.
Unfortunately, none of these markers have been investigated using TCE in female mice of any
other strain, so it is unclear whether the lack of response is characteristic of female mice in
general, or just in this strain. Interestingly, as noted above, liver/body weight ratio increases
were observed in both sexes of the null mice in this study. Laughter et al. (2004) only quantified
activity of the peroxisome proliferation marker PCO in their study, and found in null mice a
slight decrease (0.8-fold of control) at 500 mg/kg-day TCE and an increase (1.5-fold of control)
at 1,500 mg/kg-day TCE after 3 weeks of treatment, with neither statistically significant
(4-5 mice per group). However, baseline levels of PCO were almost twofold higher in the null
mice, and the treated wild-type and null mice differed in PCO activity by only about 1.5-fold.
In sum, oral administration of TCE for up to 28 days causes proliferation of peroxisomes
in hepatocytes along with associated increases in peroxisomal enzyme activities in both mice and
rats. Male mice tend to be more sensitive in that at comparable doses, rats and female mice tend
to exhibit smaller responses. For example, for peroxisomal volume and PCO, the fold-increase
in rats appears to be lower by three- to sixfold than that in mice, but, for catalase, the changes
were similar between mice in F344 rats. No inhalation or longer-term studies were located, and
only one study examined these changes at more than one time-point. Therefore, little is known
about the route-dependence, time course, and persistence of these changes. Finally, two studies
in PPARa-null mice (Laughter et al., 2004; Nakajima et al., 2000) found diminished responses in
terms of increased peroxisomal volume and peroxisomal enzyme activities as compared to
wild-type mice, although there was some confounding due to baseline differences between null
and wild-type control mice in several measures.
4.5.4.1.5. Oxidative Stress
Several studies have attempted to study the possible effects of "oxidative stress" and
DNA damage resulting from TCE exposures. The effects of induction of metabolism by TCE, as
well as through coexposure to ethanol, have been hypothesized to in itself increase levels of
This document is a draft for review purposes only and does not constitute Agency policy.
313 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
"oxidative stress" as a common effect for both exposures (see Sections E.3.4.2.3 and E.4.2.4).
Oxidative stress has been hypothesized to be a key event or MOA for peroxisome proliferators as
well, but has been found to neither be correlated with cell proliferation nor carcinogenic potency
of peroxisome proliferators (see Section E.3.4.1.1). As a MOA, it is not defined or specific as
the term "oxidative stress" is implicated as part of the pathophysiologic events in a multitude of
disease processes and is part of the normal physiologic function of the cell and cell signaling.
In regard to measures of oxidative stress, Rusyn et al. (2006) noted that although an
overwhelming number of studies draw a conclusion between chemical exposure, DNA damage,
and cancer based on detection of 8-hydroxy-2' deoxyguanosine (8-OHdG), a highly mutagenic
lesion, in DNA isolated from organs of in vivo treated animals, a concern exists as to whether
increases in 8-OHdG represent damage to genomic DNA, a confounding contamination with
mitochondrial DNA, or an experimental artifact. As noted in Sections E.2.1.1 and E.2.2.11,
studies of TCE which employ the i.p. route of administration can be affected by inflammatory
reactions resulting from that routes of administration and subsequent toxicity that can involve
oxygen radical formation from inflammatory cells. Finally, as described in Section E.2.2.8, the
study by Channel et al. (1998) demonstrated that corn oil as vehicle had significant effects on
measures of "oxidative stress" such as thiobarbiturate acid-reactive substances (TBARS).
The TBARS results presented by Channel et al. (1998) indicate suppression of TBARS
with increasing time of exposure to corn oil alone with data presented in such a way for 8-OHdG
and total free radical changes that the pattern of corn oil administration was obscured. It was not
apparent from that study that TCE exposure induced oxidative damage in the liver.
Toraason et al. (1999) measured 8-OHdG and a "free radical-catalyzed isomer of
arachidonic acid and marker of oxidative damage to cell membranes, 8-Epi-prostaglandin F2a
(8-epiPGF)", excretion in the urine and TBARS (as an assessment of malondialdehyde and
marker of lipid peroxidation) in the liver and kidney of male Fischer rats exposed to single i.p.
injections in of TCE in Alkamuls vehicle. Using this paradigm, 500-mg/kg TCE was reported to
induce Stage II anesthesia and a 1,000 mg/kg TCE to induce Level III or IV (absence of reflex
response) anesthesia and burgundy colored urine with 2/6 rats at 24 hours comatose and
hypothermic. The animals were sacrificed before they could die and the authors suggested that
they would not have survived another 24 hours. Thus, using this paradigm there was significant
toxicity and additional issues related to route of exposure. Urine volume declined significantly
during the first 12 hours of treatment and while water consumption was not measured, it was
suggested by the authors to be decreased due to the moribundity of the rats. Given that this study
examined urinary markers of "oxidative stress" the effects on urine volume and water
consumption, as well as the profound toxicity induced by this exposure paradigm, limit the
This document is a draft for review purposes only and does not constitute Agency policy.
314 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
interpretation of the study. The issues of bias in selection of the data for this analysis, as well as
the issues stated above for this paradigm limit interpretation of these data while the authors
suggest that evidence of oxidative damage was equivocal.
4.5.4.1.6. Bile Production
Effects of TCE exposure in humans and in experimental animals is presented in
Section E.2.6. Serum bile acids (SBA) have been suggested as a sensitive indicator of
hepatotoxicity to a variety of halogenated solvents with an advantage of increased sensitivity and
specificity over conventional liver enzyme tests that primarily reflect the acute perturbation of
hepatocyte membrane integrity and "cell leakage" rather than liver functional capacity (i.e.,
uptake, metabolism, storage, and excretion functions of the liver) (Neghab et al., 1997) Bai et al.,
1992b. While some studies have reported negative results, a number of studies have reported
elevated SBA in organic solvent-exposed workers in the absence of any alterations in normal
liver function tests. These variations in results have been suggested to arise from failure of some
methods to detect some of the more significantly elevated SBA and the short-lived and reversible
nature of the effect (Neghab et al., 1997). Neghab et al. (1997) have reported that occupational
exposure to l,l,2-trichloro-l,2,2-trifluoroethane and trichloroethylene has resulted in elevated
SBA and that several studies have reported elevated SBA in experimental animals to chlorinated
solvents such as carbon tetrachloride, chloroform, hexachlorobutadiene, tetrachloroethylene,
1,1,1-trichloroethane, and trichloroethylene at levels that do not induce hepatotoxicity (Bai et al.,
1992a; Bai et al., 1992b; Hamdan and Stacey, 1993; Wang and Stacey, 1990). Toluene, a
nonhalogenated solvent, has also been reported to increase SBA in the absence of changes in
other hepatobiliary functions (Neghab and Stacey, 1997). Thus, disturbance in SBA appears to
be a generalized effect of exposure to chlorinated solvents and nonchlorinated solvents and not
specific to TCE exposure.
Wang and Stacey (1990) administered TCE in corn oil via i.p. injection to male
Sprague-Dawley rats with liver enzymes and SBA examined 4 hours after the last TCE
treatment. The limitations of i.p injection experiments have already been discussed. While
reporting no overt liver toxicity there was, generally, a reported dose-related increase in cholic
acid, chenodeoxycholic acid, deoxycholic acid, taurocholic acid, tauroursodeoxycholic acid with
cholic acid and taurochlolic acid increased at the lowest dose. The authors report that
"examination of liver sections under light microscopy yielded no consistent effects that could be
ascribed to trichloroethylene." In the same study a rats were also exposed to TCE via inhalation
and using this paradigm, cholic acid and taurocholic acid were also significantly elevated but the
This document is a draft for review purposes only and does not constitute Agency policy.
315 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
large variability in responses between rats and the low number of rats tested in this paradigm
limit its ability to determine quantitative differences between groups. Nevertheless, without the
complications associated with i.p. exposure, inhalation exposure of TCE at relatively low
exposure levels that were not associated with other measures of toxicity were associated with
increased SB A level.
Hamdan and Stacey (1993) administered TCE in corn oil (1 mmol/kg) in male Sprague-
Dawley rats and followed the time-course of SBA elevation, TCE concentration, and
trichloroethanol in the blood up to 16 hours. Liver and blood concentration of TCE were
reported to peak at 4 hours while those of trichloroethanol peaked at 8 hours after dosing. TCE
levels were not detectable by 16 hours in either blood or liver while those of trichloroethanol
were still elevated. Elevations of SBA were reported to parallel those of TCE with cholic acid
and taurochloate acid reported to show the highest levels of bile acids. The authors state that
liver injury parameters were checked and found unaffected by TCE exposure but did not show
the data. Thus, it was TCE concentration and not that of its metabolite that was most closely
related to changes in SBA and after a single exposure and the effect appeared to be reversible. In
an in vitro study by Bai and Stacey (1993), TCE was studied in isolated rat hepatocytes with
TCE reported to cause a dose-related suppression of initial rates of cholic acid and taurocholic
acid but with no significant effects on enzyme leakage and intracellular calcium contents, further
supporting a role for the parent compound in this effect.
4.5.4.1.7. Summary: Trichloroethylene (TCE)-Induced Noncancer Effects in Laboratory
Animals
In laboratory animals, TCE leads to a number of structural changes in the liver, including
increased liver weight, small transient increases in DNA synthesis, cytomegaly in the form of
"swollen" or enlarged hepatocytes, increased nuclear size probably reflecting polyploidization,
and proliferation of peroxisomes. Liver weight increases proportional to TCE dose are
consistently reported across numerous studies, and appear to be accompanied by periportal
hepatocellular hypertrophy. There is also evidence of increased DNA synthesis in a small
portion of hepatocytes at around 10 days in vivo exposure. The lack of correlation of
hepatocellular mitotic figures with whole liver DNA synthesis or DNA synthesis observed in
individual hepatocytes supports the conclusion that cellular proliferation is not the predominant
cause of increased DNA synthesis. The lack of correlation of whole liver DNA synthesis and
those reported for individual hepatocytes suggests that nonparenchymal cells also contribute to
such synthesis. Indeed, nonparenchymal cell activation or proliferation has been noted in several
This document is a draft for review purposes only and does not constitute Agency policy.
316 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
studies. Moreover, the histological descriptions of TCE exposed liver are consistent with and in
some cases specifically note increased polyploidy after TCE exposure. Interestingly, changes in
TCE-induced hepatocellular ploidy, as indicated by histological changes in nuclei, have been
noted to remain after the cessation of exposure. In regard to apoptosis, TCE has been reported to
either not change apoptosis or to cause a slight increase at high doses. Some studies have also
noted effects from dosing vehicle alone (such as corn oil in particular) not only on liver
pathology, but also on DNA synthesis.
Available data also suggest that TCE does not induce substantial cytotoxicity, necrosis, or
regenerative hyperplasia, as only isolated, focal necroses and mild to moderate changes in serum
and liver enzyme toxicity markers having been reported. Data on peroxisome proliferation,
along with increases in a number of associated biochemical markers, show effects in both mice
and rats. These effects are consistently observed across rodent species and strains, although the
degree of response at a given mg/kg-day dose appears to be highly variability across strains, with
mice on average appearing to be more sensitive.
In addition, like humans, laboratory animals exposed to TCE have been observed to have
increased serum bile acids, though the toxicologic importance of these effects is unclear.
4.5.5. Trichloroethylene (TCE)-Induced Liver Cancer in Laboratory Animals
For 2-year or lifetime studies of TCE exposure a consistent hepatocarcinogenic response
has been observed using mice of differing strains and genders and from differing routes of
exposure. However, some rat studies have been confounded by mortality from gavage error or
the toxicity of the dose of TCE administered. In some studies, a relative insensitive strain of rat
has been used. However, in general it appears that the mouse is more sensitive than the rat to
TCE-induced liver cancer. Three studies give results the authors consider to be negative for
TCE-induced liver cancer in mice, but have either design and/or reporting limitations, or are in
strains and paradigms with apparent low ability for liver cancer induction or detection. Findings
from these studies are shown in Tables 4-60 through 4-65, and discussed below.
4.5.5.1.1. Negative or Inconclusive Studies of Mice and Rats
Fukuda et al. (1983) reported a 104-week inhalation bioassay in female Crj:CD-l (ICR)
mice and female Crj:CD (S-D) rats exposed to 0-, 50-, 150-, and 450-ppm TCE (n = 50). There
were no reported incidences of mice or rats with liver tumors for controls indicative of relatively
insensitive strains and gender used in the study for liver effects. While TCE was reported to
This document is a draft for review purposes only and does not constitute Agency policy.
317 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
induce a number of other tumors in mice and rats in this study, the incidence of liver tumors was
less than 2% after TCE exposure. Of note is the report of cystic cholangioma reported in one
group of rats.
Henschler et al. (1980) exposed NMRI mice and WIST random bred rats to 0-, 100-, and
500-ppm TCE for 18 months (n = 30). Control male mice were reported to have
one hepatocellular carcinoma and one hepatocellular adenoma with the incidence rate unknown.
In the 100-ppm TCE exposed group, two hepatocellular adenomas and one mesenchymal liver
tumor were reported. No liver tumors were reported at any dose of TCE in female mice or
controls. For male rats, only one hepatocellular adenomas at 100 ppm was reported. For female
rats no liver tumors were reported in controls, but one adenoma and one cholangiocarcinoma was
reported at 100-ppm TCE and at 500-ppm TCE, two cholangioadenomas, a relatively rare biliary
tumor, was reported. The difference in survival in mice, did not affect the power to detect a
response, as was the case for rats. However, the low number of animals studied, abbreviated
This document is a draft for review purposes only and does not constitute Agency policy.
318 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
Table 4-60. Summary of liver tumor findings in gavage studies of
trichloroethylene by NTP (1990)a
Sex
Dose (mg/kg)b
Adenoma
(overall; terminal0)
Adenocarcinoma
(overall; terminal0)
1/d, 5 d/wk, 103-wk study, F344/N rats
Male
0
NAd
0/49
500
NA
0/49
1,000
NA
1/49
Female
0
NA
0/50
500
NA
1/48
1,000
NA
1/48
1/d, 5 d/wk, 103-wk study, B6C3F1 mice
Male
0
7/48; 6/33
8/48; 6/33
1,000
14/50; 6/16
31/50; 14/16f
Female
0
4/48; 4/32
2/48; 2/32
1,000
16/49; 11/23°
13/49; 8/23g
aLiver tumors not examined in 13-wk study, so data shown only for 103-wk study.
b Corn oil vehicle.
0 Terminal values not available for rats.
d Data not available.
sp< 0.003.
V< 0.001.
gp< 0.002.
This document is a draft for review purposes only and does not constitute Agency policy.
319 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-61. Summary of liver tumor findings in gavage studies of
2 trichloroethylene by NCI (1976)
3
Sex
Dose (mg/kg)a
Hepatocarcinoma
1/d, 5 d/wk, 2-yr study, Osborn-Mendel rats
Males
0
0/20
549
0/50
1,097
0/50
Females
0
0/20
549
1/48
1,097
0/50
1/d, 5 d/wk, 2-yr study, B6C3F1 mice
Males
0
1/20
1,169
26/50b
2,339
31/48b
Females
0
0/20
869
4/50
1,739
1 l/47b
4
5 a Treatment period was 48 wk for rats, 66 wk for mice. Doses were changed several times during the study
6 based on monitoring of body weight changes and survival. Dose listed here is the time-weighted average
7 dose over the days on which animals received a dose.
8 b/?<0.01.
9
This document is a draft for review purposes only and does not constitute Agency policy.
320 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-62. Summary of liver tumor incidence in gavage studies of
2 trichloroethylene by NTP (1988)
3
Sex
Dose (mg/kg)a
Adenoma
Adenocarcinoma
1/d, 5 d/wk, 2-yr study, ACI rats
Male
0
0/50
1/50
500
0/49
1/49
1,000
0/49
1/49
Female
0
0/49
2/49
500
0/46
0/46
1,000
0/39
0/39
1/d, 5 d/wk, 2-yr study, August rats
Male
0
0/50
0/50
500
0/50
1/50
1,000
0/48
1/48
Female
0
0/48
2/48
500
0/48
0/48
1,000
0/50
0/50
1/d, 5 d/wk, 2-yr study, Marshall rats
Male
0
1/49
1/49
500
0/50
0/50
1,000
0/47
1/47
Female
0
0/49
0/49
500
0/48
0/48
1,000
0/46
0/46
1/d, 5 d/wk, 2-yr study, Osborne-Mendel rats
Male
0
1/50
1/50
500
1/50
0/50
1,000
1/49
2/49
Female
0
0/50
0/50
500
0/48
2/48
1,000
0/49
2/49
4
This document is a draft for review purposes only and does not constitute Agency policy.
321 DRAFT—DO NOT CITE OR QUOTE
-------
aCorn oil vehicle.
This document is a draft for review purposes only and does not constitute Agency policy.
322 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-63. Summary of liver tumor findings in inhalation studies of
trichloroethylene by Maltoni et al. (1988)a
Sex
Concentration (ppm)
Hepatoma
7 h/d, 5 d/wk, 8-wk exposure, observed for lifespan, Swiss mice
Male
0
1/100
100
3/60
600
4/72
Female
0
1/100
100
1/60
600
0/72
7 h/d, 5 d/wk, 78-wk exposure, observed for lifespan, Swiss mice
Male
0
4/90
100
2/90
300
8/90
600
13/90
Female
0
0/90
100
0/90
300
0/90
600
1/90
7 h/d, 5 d/wk, 78-wk exposure, observed for lifespan, B6C3F1 miceb
Male
0
1/90
100
1/90
300
3/90
600
6/90
Female
0
3/90
100
4/90
300
4/90
600
9/90
Three inhalation experiments in this study found no hepatomas: BT302 (8-wk exposure to 0, 100, 600 ppm
in Sprague-Dawley rats); BT303 (8-wk exposure to 0, 100, or 600 ppm in Swiss mice); and BT304 (78-wk
exposure to 0, 100, 300, or 600 ppm in Sprague-Dawley rats).
Female incidences are from experiment BT306, while male incidences are from experiment BT306bis,
which was added to the study because of high, early mortality due to aggressiveness and fighting in males
in experiment BT306.
This document is a draft for review purposes only and does not constitute Agency policy.
323 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-64. Summary of liver tumor findings in inhalation studies of
2 trichloroethylene by Henschler et al. (1980)a and Fukuda et al. (1983)
3
Sex
Concentration (ppm)
Adenomas
Adenocarcinomas
6 h/d, 5 d/wk, 18-mo exposure, 30-mo observation, Han:NMRI mice (Henschler et al., 1980)
Males
0
1/30 b
1/30
100
2/29 b
0/30
500
0/29
0/30
Females
0
0/29
0/29
100
0/30
0/30
500
0/28
0/28
6 h/d, 5 d/wk, 18-mo exposure, 36-mo observation, Han:WIST rats (Henschler et al., 1980)
Males
0
1/29
0/29
100
1/30
0/30
500
0/30
0/30
Females
0
0/28
0/28
100
1/30
1/30
500
2/30
0/30
7 h/d, 5 d/wk, 2-yr study, Crj:CD (S-D) rats (Fukuda et al., 1983)
Females
0
0/50
0/50
50
1/50
0/50
150
0/47
0/47
450
0/51
1/50
7 h/d, 5 d/wk, 2-yr study, Crj:CD (ICR) mice (Fukuda et al., 1983)
Females
0
0/49
0/49
50
0/50
0/50
150
0/50
0/50
450
1/46
0/46
4
5 a Henschler et al. (1980) observed no liver tumors in control or exposed Syrian hamsters.
6 b One additional hepatic tumor of undetermined class not included.
7
This document is a draft for review purposes only and does not constitute Agency policy.
324 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-65. Summary of liver tumor findings in gavage studies of
2 trichloroethylene by Henschler et al. (1984)a
3
Sex
(TCE conc.)
TCE (Stabilizers if present)
Benignb
Malignant0
5 d/wk, 18-mo exposure, 24-mo observation, Swiss mice (Henschler et al., 1984)
Males
(2.4g/kg BW)
Control (none)
5/50
0/50
TCE (triethanolamine)
7/50
0/50
TCE (industrial)
9/50
0/50
TCE (epichlorohydrin (0.8%))
3/50
1/50
TCE (1,2-epoxybutane [0.8%])
4/50
0/50
TCE (both epichlorohydrin [0.25%]
and 1,2-epoxybutane [0.25%])
5/50
0/50
Females
(1.8 g/kg BW)
Control (none)
1/50
0/50
TCE (triethanolamine)
7/50
0/50
TCE (industrial)
9/50
0/50
TCE (epichlorohydrin (0.8%))
3/50
0/50
TCE (1,2-epoxybutane (0.8%))
2/50
0/50
TCE (both epichlorohydrin (0.25%)
and 1,2-epoxybutane (0.25%))
4/50
1/50
4
5 a Henschler et al. (1984) Due to poor condition of the animals resulting from the nonspecific toxicity of high doses
6 of TCE and/or the additives, gavage was stopped for all groups during wk 35-40, 65 and 69-78, and all doses
7 were reduced by a factor of 2 from the 40th wk on.
8 b Includes hepatocellular adenomas, hemangioendothelioma, cholangiocellular adenoma.
9 0 Includes hepatocellular carcinoma, malignant hemangiosarcoma, cholangiocellular carcinoma.
10
11 Cone. = concentration.
12
13
14 exposure duration, low survival in rats, and absent background response (suggesting low intrinsic
15 sensitivity to this endpoint) suggest a study of limited ability to detect a TCE carcinogenic liver
16 response. Of note is that despite their limitations, both Fukuda et al. (1983) and Henschler et al.
17 (1980) report rare biliary cell derived tumors in TCE-exposed rats.
18 Van Duuren et al. (1979), exposed mice to 0.5 mg/mouse to TCE via gavage once a week
19 in 0.1 mL trioctanion (// = 30). Inadequate design and reporting of this study limit that ability to
20 use the results as an indicator of TCE carcinogenicity.
This document is a draft for review purposes only and does not constitute Agency policy.
325 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
The NCI (1976) study of TCE was initiated in 1972 and involved the exposure of
Osborn-Mendel rats to varying concentrations of TCE. A low incidence of liver tumors was
reported for controls and carbon tetrachloride positive controls in rats from this study. The
authors concluded that due to mortality, "the test is inconclusive in rats." They note the
insensitivity of the rat strain used to the positive control of carbon tetrachloride exposure.
The NTP (1990) study of TCE exposure in male and female F344/N rats, and B6C3F1
mice (500 and 1,000 mg/kg for rats) is limited in the ability to demonstrate a dose-response for
hepatocarcinogenicity. For rats, the NTP (1990) study reported no treatment-related
non-neoplastic liver lesions in males and a decrease in basophilic cytological change reported
from TCE-exposure in female rats. The results for detecting a carcinogenic response in rats were
considered to be equivocal because both groups receiving TCE showed significantly reduced
survival compared to vehicle controls and because of a high rate (e.g., 20% of the animals in the
high-dose group) of death by gavage error.
The NTP (1988) study of TCE exposure in four strains of rats to
"diisopropylamine-stabilized TCE" was also considered inadequate for either comparing or
assessing TCE-induced liver carcinogenesis in these strains of rats because of chemically
induced toxicity, reduced survival, and incomplete documentation of experimental data. TCE
gavage exposures of 0, 500, or 1,000 mg/kg-day (5 days/week, for 103 weeks) male and female
rats was also marked by a large number of accidental deaths (e.g., for high-dose male Marshal
rats 25 animals were accidentally killed).
Maltoni et al. (1986) reported the results of several studies of TCE via inhalation and
gavage in mice and rats. A large number of animals were used in the treatment groups but the
focus of the study was detection of a neoplastic response with only a generalized description of
tumor pathology phenotype given and limited reporting of non-neoplastic changes in the liver.
Accidental death by gavage error was reported not to occur in this study. In regards to effects of
TCE exposure on rat survival, "a nonsignificant excess in mortality correlated to TCE treatment
was observed only in female rats (treated by ingestion with the compound)".
For rats, Maltoni et al. (1986) reported four liver angiosarcomas (one in a control male
rat, one both in a TCE-exposed male and female at 600 ppm TCE for 8 weeks, and one in a
female rat exposed to 600-ppm TCE for 104 weeks), but the specific results for incidences of
hepatocellular "hepatomas" in treated and control rats were not given. Although the Maltoni
et al. (1986) concluded that the small number was not treatment related, the findings were
brought forward because of the extreme rarity of this tumor in control Sprague-Dawley rats,
untreated or treated with vehicle materials. In rats treated for 104 weeks, there was no report of a
TCE treatment-related increase in liver cancer in rats. This study only presented data for positive
This document is a draft for review purposes only and does not constitute Agency policy.
326 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
findings so it did not give the background or treatment-related findings in rats for liver tumors in
this study. Thus, the extent of background tumors and sensitivity for this endpoint cannot be
determined. Of note is that the Sprague-Dawley strain used in this study was also noted in the
Fukuda et al. (1983) study to be relatively insensitive for spontaneous liver cancer and to also be
negative for TCE-induced hepatocellular liver cancer induction in rats. However, like Fukuda
et al. (1983) and Henschler et al. (1980), that reported rare biliary tumors in insensitive strains of
rat for hepatocellular tumors, Maltoni et al. (1986) reported a relatively rare tumor type,
angiosarcoma, after TCE exposure in a relatively insensitive strain for "hepatomas." As noted
above, many of the rat studies were limited by premature mortality due to gavage error or
premature mortality (Henschler et al., 1980; NCI, 1976; NTP, 1988, 1990), which was reported
not occur in Maltoni et al. (1986).
4.5.5.1.2. Positive Trichloroethylene (TCE) Studies of Mice
In the NCI (1976) study of TCE exposure in B6C3F1 mice, TCE was reported to increase
incidence of hepatocellular carcinomas in both doses and both genders of mice (-1,170 and
2,340 mg/kg for males and 870 and 1,740 mg/kg for female mice). Hepatocellular carcinoma
diagnosis was based on histologic appearance and metastasis to the lung. The tumors were
described in detail and to be heterogeneous "as described in the literature" and similar in
appearance to tumors generated by carbon tetrachloride. The description of liver tumors in this
study and tendency to metastasize to the lung are similar to descriptions provided by
Maltoni et al. (1986) for TCE-induced liver tumors in mice via inhalation exposure.
The NTP (1990) study of TCE exposure in male and female B6C3F1 mice (1,000 mg/kg
for mice) reported decreased latency of liver tumors, with animals first showing carcinomas at
57 weeks for TCE-exposed animals and 75 weeks for control male mice. The administration of
TCE was also associated with increased incidence of hepatocellular carcinoma (tumors with
markedly abnormal cytology and architecture) in male and female mice. Hepatocellular
adenomas were described as circumscribed areas of distinctive hepatic parenchymal cells with a
perimeter of normal appearing parenchyma in which there were areas that appeared to be
undergoing compression from expansion of the tumor. Mitotic figures were sparse or absent but
the tumors lacked typical lobular organization. Hepatocellular carcinomas had markedly
abnormal cytology and architecture with abnormalities in cytology cited as including increased
cell size, decreased cell size, cytoplasmic eosinophilia, cytoplasmic basophilia, cytoplasmic
vacuolization, cytoplasmic hyaline bodies, and variations in nuclear appearance. Furthermore, in
many instances several or all of the abnormalities were present in different areas of the tumor
This document is a draft for review purposes only and does not constitute Agency policy.
327 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
and variations in architecture with some of the hepatocellular carcinomas having areas of
trabecular organization. Mitosis was variable in amount and location. Therefore, the phenotype
of tumors reported from TCE exposure was heterogeneous in appearance between and within
tumors. However, because it consisted of a single-dose group in addition to controls, this study
is limited of limited utility for analyzing the dose-response for hepatocarcinogenicity. There was
also little reporting of non-neoplastic pathology or toxicity and no report of liver weight at
termination of the study.
Maltoni et al. (1986) reported the results of several studies of TCE in mice. A large
number of animals were used in the treatment groups but the focus of the study was detection of
a neoplastic response with only a generalized description of tumor pathology phenotype given
and limited reporting of non-neoplastic changes in the liver. There was no accidental death by
gavage error reported to occur in mice but, a "nonsignificant" excess in mortality correlated to
TCE treatment was observed in male B6C3F1 mice. TCE-induced effects on body weight were
reported to be absent in mice except for one experiment (BT 306 bis) in which a slight nondose
correlated decrease was found in exposed animals. "Hepatoma" was the term used to describe
all malignant tumors of hepatic cells, of different subhistotypes, and of various degrees of
malignancy and were reported to be unique or multiple, and have different sizes (usually
detected grossly at necropsy) from TCE exposure. In regard to phenotype tumors were described
as usual type observed in Swiss and B6C3F1 mice, as well as in other mouse strains, either
untreated or treated with hepatocarcinogens and to frequently have medullary (solid), trabecular,
and pleomorphic (usually anaplastic) patterns. Swiss mice from this laboratory were reported to
have a low incidence of hepatomas without treatment (1%). The relatively larger number of
animals used in this bioassay (n = 90-100), in comparison to NTP standard assays, allows for a
greater power to detect a response.
TCE exposure for 8 weeks via inhalation at 100 or 600 ppm may have been associated
with a small increase in liver tumors in male mice in comparison to concurrent controls during
the life span of the animals. In Swiss mice exposed to TCE via inhalation for 78 weeks, there a
reported increase in hepatomas associated with TCE treatment that was dose-related in male but
not female Swiss mice. In B6C3F1 mice exposed via inhalation to TCE for 78 weeks, increases
in hepatomas were reported in both males and females. However, the experiment in males was
repeated with B6C3F1 mice from a different source, since in the first experiment more than half
of the mice died prematurely due to excessive fighting. Although the mice in the two
experiments in males were of the same strain, the background level of liver cancer was
significantly different between mice from the different sources (1/90 vs. 19/90), though the early
mortality may have led to some censoring. The finding of differences in response in animals of
This document is a draft for review purposes only and does not constitute Agency policy.
328 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
the same strain but from differing sources has also been reported in other studies for other
endpoints. However, for both groups of male B6C3F1 mice the background rate of liver tumors
over the lifetime of the mice was no greater than about 20%.
There were other reports of TCE carcinogenicity in mice from chronic exposures that
were focused primarily on detection of liver tumors with limited reporting of tumor phenotype or
non-neoplastic pathology. Herren-Freund et al. (1987) reported that male B6C3F1 mice given
40 mg/L TCE in drinking water had increased tumor response after 61 weeks of exposure.
However, concentrations of TCE fell by about V2 at this dose of TCE during the twice a week
change in drinking water solution so the actual dose of TCE the animals received was less than
40 mg/L. The percentage liver/body weight was reported to be similar for control and
TCE-exposed mice at the end of treatment. However, despite difficulties in establishing
accurately the dose received, an increase in adenomas per animal and an increase in the number
of animals with hepatocellular carcinomas were reported to be associated with TCE exposure
after 61 weeks of exposure and without apparent hepatomegaly. Anna et al. (1994) reported
tumor incidences for male B6C3F1 mice receiving 800 mg/kg-day TCE via gavage (5 days/week
for 76 weeks). All TCE-treated mice were reported to be alive after 76 weeks of treatment.
Although the control group contained a mixture of exposure durations (76-134 weeks) and
concurrent controls had a very small number of animals, TCE-treatment appeared to increase the
number of animals with adenomas, the mean number of adenomas and carcinomas, but with no
concurrent TCE-induced cytotoxicity.
4.5.5.1.3. Summary: Trichloroethylene (TCE)-Induced Cancer in Laboratory Animals
Chronic TCE bioassays have consistently reported increased liver tumor incidences in
both sexes of B6C3F1 mice treated by inhalation and gavage exposure in a number of bioassays.
The only inhalation study of TCE in Swiss mice also showed an effect in males. Data in the rat,
while not reporting statistically significantly increased risks, are not entirely adequate due to low
numbers of animals, inadequate reporting, use of insensitive bioassays, increased systemic
toxicity, and/or increased mortality. Notably, several studies in rats noted a few very rare types
of liver or biliary tumors (cystic cholangioma, cholangiocarcinoma, or angiosarcomas) in treated
animals.
4.5.6. Role of Metabolism in Liver Toxicity and Cancer
It is generally thought that TCE oxidation by CYPs is necessary for induction of
hepatotoxicity and hepatocarcinogenicity (Bull, 2000). Direct evidence for this hypothesis is
This document is a draft for review purposes only and does not constitute Agency policy.
329 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
limited, e.g., the potentiation of hepatotoxicity by pretreatment with CYP inducers such as
ethanol and phenobarbital (Nakajima et al., 1988; Okino et al., 1991). Rather the presumption
that CYP-mediated oxidation is necessary for TCE hepatotoxicity and hepatocarcinogenicity is
largely based on similar effects (e.g., increases in liver weight, peroxisome proliferation, and
hepatocarcinogenicity) having been observed with TCE's oxidative metabolites. The discussion
below focuses the similarities and differences between the major effects in the liver of TCE and
of the oxidative metabolites CH, TCA, and DCA. In addition, CH is largely converted to TCOH,
TCA, and possibly DCA. DCA has been used in human clinical practice for a variety of severe
illnesses and no data on liver effects in humans have been reported (U.S. EPA, 2003b).
However, as noted in EPA (2003b), data on DCA in humans are scarce and complicated by the
fact that available studies have predominantly focused on individuals who have a pre-existing
(usually severe) disease.
4.5.6.1.1. Pharmacokinetics of Chloral Hydrate (CH), Trichloroacetic Acid (TCA), and
Dichloroacetic Acid (DCA) From Trichloroethylene (TCE) Exposure
As discussed in Section 3, in vivo data confirm that CH and TCA, are oxidative
metabolites of TCE, with available data on TCA incorporated into the PBPK modeling. In
addition, there are indirect data suggesting the formation of DCA. However, direct in vivo
evidence of the formation of DCA is confounded by its rapid clearance at low concentrations,
and analytical artifacts in its detection in vivo that have yet to be entirely resolved. PBPK
modeling (see Section 3.5) predicts that the proportions of TCE metabolized to CH and TCA
varies considerably in mice (ranging from 15-97 and 4-38%, respectively) and rats (ranging
7-75 and 0.5-22%, respectively). Therefore, a range of smaller concentrations of TCA or CH
may be relevant for comparisons with TCE-induced liver effects. For example, for
1,000 mg/kg-day oral doses of TCE, the relevant comparisons would be approximately
0.25-1.5 g/L in drinking water for TCA and CH. For DCA a corresponding range is harder to
determine and has been suggested to be an upper limit of about 12% following oral exposures
(Barton et al., 1999). This is consistent with the range estimated from PBPK modeling
attributing all of the "untracked" oxidation (i.e., not producing TCOH or TCA) to DCA (95% CI:
0.2-16%), see Figure 3-22).
Two studies have used analytic methods for DCA that are considered more reliable and
less confounded by artifactual formation. Kim et al. (2009), which was published too late to be
incorporated into the PBPK model, used an empirical pharmacokinetic model to analyze data on
male B6C3F1 mice exposed to a single dose of 2,100 mg/kg TCE by gavage. Peak levels of
This document is a draft for review purposes only and does not constitute Agency policy.
330 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
TCA and DC A were found to be 64 |ig/ml and 18 ng/ml, respectively, a difference of more than
3.000-fold. The kinetic rate constant they estimated for TCE -> DCA were more than
five orders of magnitude smaller than the kinetic rate constant estimated for TCE -> TCA.
These data all suggest that DCA is a minor metabolite of TCE as compared to TCA at high doses
of around 2,000 mg/kg. Delinsky et al. (2005) reported that in male Sprague-Dawley rats, after a
single 2,000 mg/kg dose by oral gavage, peak levels of DCA were 39.5 ng/ml. Delinsky et al.
(2005) did not report TCA levels for comparison. The only data available in rats in this range of
oral gavage doses (coincidentally also in male Sprague-Dawley rats) reported peak levels of
TCA of 24 and 60 mg/ml at oral gavage doses of 600 and 3,000 mg/kg, respectively (Larson and
Bull, 1992b). This suggests a difference between DCA and TCA levels in rats exposed to TCE
of about 1,000-fold, albeit with more uncertainty as compared to Kim et al. (2009), in which both
were measured simultaneously in the same animals. However, liver toxicity in both rats and
mice is evident at much lower doses, so additional data are needed to inform whether the relative
amount of TCA and DCA changes at lower exposures.
4.5.6.1.2. Comparisons Between Trichloroethylene (TCE) and Trichloroacetic Acid (TCA),
Dichloroacetic Acid (DCA), and Chloral Hydrate (CH) Noncancer Effects
4.5.6.1.3. Hepatomegaly—qualitative and quantitative comparisons
As discussed above, TCE causes hepatomegaly in rats, mice, and gerbils under both acute
and chronic dosing. Data from a few available studies suggest that oxidative metabolism is
important for mediating these effects. Buben and O'Flaherty (1985) collected limited
pharmacokinetic data in a sample of the same animals for which liver weight changes were being
assessed. While liver weight increases had similarly strong correlations with applied dose and
urinary metabolites for doses up to 1,600 mg/kg-day (R of 0.97 for both), above that dose, the
linear relationship was maintained with urinary metabolites but not with applied dose. Ramdhan
et al. (2008) conducted parallel experiments at TCE 1,000 and 2,000 ppm (8 hours/day, 7 days)
in wild-type and CYP2El-null mice, which did not exhibit increased liver/body weight ratios
with TCE treatment and excreted twofold lower amounts of oxidative metabolites TCA and
TCOH in urine as compared to wild-type mice. However, among control mice, those with the
null genotype had 1.32-fold higher absolute liver weights and 1.18-fold higher liver/body weight
ratios than wild-type mice, reducing the sensitivity of the experiment, particularly with only
six mice per dose group.
This document is a draft for review purposes only and does not constitute Agency policy.
3 31 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Ramdhan et al. (2008) reported that stated that urinary TCA levels in wild type mice were
incorrectly reported by Ramdhan et al. (2008) but were corrected in this study. The authors
reported no differences in urinary volume by genotype or exposure but did not show the data.
TCA and trichloroethanol were detected in all exposed mice with no significant differences
between the 1,000 and 2,000 ppm TCE levels. TCA concentrations were reported to be
significantly lower and trichloroethanol levels significantly higher in PPARa-null mice relative
to wild type mice with no differences in genotype between the sum of total TCA and
trichloroethanol concentrations between genotypes. The authors reported that they measured
hepatic protein expression of CYP2E1 and ALDH2 enzymes and did not observe a significant
difference among controls (data not shown) and that TCE exposure did not alter hepatic CYP2E1
expression but did decrease ALDH2 expression to a comparable extent in all mouse lines (data
not shown). Thus, changes in urinary TCA levels in the differing strains were not related to
changes in expression of these metabolic enzymes.
As stated above, hepatomegally was increased by TCE exposure in all three strains. TCE
at both 1,000 and 2,000 ppm significantly increased liver weight in the three mouse lines to a
similar extent (i.e., 38 and 49% in wild type mice, 20 and 37% in PPAR-null mice, and 28 and
32% in hPPARa mice). The increases were not statistically significant between doses within
each strain. Liver/body weight ratios were also significantly increased with TCE exposure at
1,000 and 2,000 ppm relative to controls (i.e., 38 and 43% in wild type mice, 24 and 36% in
PPARa-null mice, and 27 and 39% in hPPARa mice, respectively). The difference between
2,000 and 1,000 ppm TCE exposure was statistically significant in PPARa-null mice. As to the
nature of the hepatomegally induced under these conditions, hepatic triglyceride levels were
reported to be significantly correlated with liver/body weight ratios of all mice used in the study
(r = 0.54).
With respect to oxidative metabolites themselves, data from CH studies are not
informative—either because data were not shown (Sanders et al., 1982a) or, because at the time
points measured, liver weight increases are substantially confounded by foci and carcinogenic
lesions (Leakey et al., 2003a). TCA and DCA have both been found to cause hepatomegaly in
mice and rats, with mice being more sensitive to this effect. DCA also increases liver/body
weight ratios in dogs, but TCE and TCA have not been tested in this species (Cicmanec et al.,
1991).
As noted above, TCE-induced changes in liver weight appear to be proportional to the
exposure concentration across route of administration, gender and rodent species. As an
indication of the potential contribution of TCE metabolites to this effect, a quantitative
This document is a draft for review purposes only and does not constitute Agency policy.
332 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
comparison of the shape of the dose-response curves for liver weight induction for TCE and its
metabolites is informative. The analysis below was reported in Evans et al. (2009).
A number of short-term (<4 weeks) studies of TC A and DCA in drinking water have
attempted to measure changes in liver weight induction, with the majority of these studies being
performed in male B6C3F1 mice. Studies conducted from 14-30 days show a consistent
increase in percentage liver/body weight induction by TCA or DCA. However, as stated in
many of the discussions of individual studies (see Appendix E), there is a limited ability to detect
a statistically significant change in liver weight change in experiments that use a relatively small
number of animals or do not match control and treatment groups for age and weight. The
experiments of Buben and O'Flaherty used 12-14 mice per group giving it a greater ability to
detect a TCE-induced dose-response. However, many experiments have been conducted with
4-6 mice per dose group. For example, the data from DeAngelo et al. (2008) for TCA-induced
percentage liver/body weight ratio increases in male B6C3F1 mice were only derived from
five animals per treatment group after 4 weeks of exposure. The 0.05 and 0.5 g/L exposure
concentrations were reported to give a 1.09- and 1.16-fold of control percentage liver/body
weight ratios which were consistent with the increases noted in the cross-study database above.
However, a power calculation shows that the Type II error (which should be >50% and thus,
greater than the chances of "flipping a coin") was only a 6 and 7% and therefore, the designed
experiment could accept a false null hypothesis. In addition, some experiments took greater care
to age and weight match the control and treatment groups before the start of treatment.
Therefore, given these limitations and the fact that many studies used a limited range of
doses, an examination of the combined data from multiple studies (Carter et al., 1995; DeAngelo
et al., 1989; 2008; Kato-Weinstein et al., 2001; Parrish et al., 1996; Sanchez and Bull, 1990) can
best inform/discern differences in DCA and TCA dose-response relationships for liver weight
induction (described in more detail in Section E.2.4.2). The dose-response curves for similar
concentrations of DCA and TCA are presented in Figure 4-5 for durations of exposure from
14-28 days in the male B6C3F1 mouse, which was the most common sex and strain used. As
noted in Appendix E, there appears to be a linear correlation between dose in drinking water and
liver weight induction up to 2 g/L of DCA. However, the shape of the dose-response curve for
TCA appears to be quite different. Lower concentrations of TCA induce larger increase that
does DCA, but the TCE response reaches an apparent plateau while that of DCA continues to
increase the response. TCA studies did not show significant duration-dependent difference in
liver weight induction in this duration range. Short duration studies (10-42 days) were selected
because (1) in chronic studies, liver weight increases are confounded by tumor burden,
This document is a draft for review purposes only and does not constitute Agency policy.
333 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
(2) multiple studies are available, and (3) TCA studies do not show significant
duration-dependent differences in this duration range.
Of interest is the issue of how the dose-response curves for TCA and DC A compare to
that of TCE in a similar model and dose range. Since TCA and DCA have strikingly different
dose-response curves, which one if either best fits that of TCE and thus, can give insight as to
which is causative agent for TCE's effects in the liver? The carcinogenicity of chronic TCE
exposure has been predominantly studies in two mouse strains, Swiss and B6C3F1, both of
which reportedly developed liver tumors. Rather than administered in drinking water, oral TCE
studies have been conducted via oral gavage and generally in corn oil for 5 days of exposure per
2.0
1.8
1.6
%
T3 1.4
£
1.2
1.0
0.0 0.5 1.0 1.5 2.0 2.5
Concentration of DCA or TCA (g/l)
Figure 4-5. Comparison of average fold-changes in relative liver weight to
control and exposure concentrations of 2 g/L or less in drinking water for
TCA and DCA in male B6C3F1 mice for 14-30 days (Carter et al., 1995;
DeAngelo et al., 1989; 2008; Kato-Weinstein et al., 2001; Parrish et al., 1996;
Sanchez and Bull, 1990).
week. Factors adding to the increased difficulty in establishing the dose-response relationship
for TCE across studies and for comparisons to the DCA and TCA database include vehicle
effects, the difference between daily and weekly exposures, the dependence of TCE effects in the
liver on its metabolism to a variety of agents capable inducing effects in the liver, differences in
response between strains, and the inherent increased variability in use of the male mouse model.
In particular, these factors would add variability to any effort at a combined analysis, and make a
This document is a draft for review purposes only and does not constitute Agency policy.
334 DRAFT—DO NOT CITE OR QUOTE
DCA
TCA
-------
1 consistent dose-response pattern more difficult to discern. Nonetheless, despite such differences
2 in exposure route, vehicle, etc., a consistent pattern of dose-response emerges from combining
3 the available TCE data. The effects of oral exposure to TCE from 10-42 days on liver weight
4 induction is shown below in Figure 4-6 using the data of Elcombe et al. (1985), Dees and Travis
5 (1993), Goel et al. (1992), Merrick et al. (1989), Goldsworthy and Popp (1987), and Buben and
6 O'Flaherty (1985). Oral TCE administration in male B6C3F1 and Swiss mice appeared to
7 induce a dose-related increase in percentage liver/body weight that was generally proportional to
8 the increase in magnitude of dose, though as expected, with more variability than observed for a
9 similar exercise for DCA or TCA in drinking water. Some of the variability is due to the
2.0
• B6C3F1
Regression
O)
0
500
1000
1500
2000
2500
3000
20 Concentration of TCE (mg/kg/day)
2.0
• B6C3F1 and Swiss
Plot 2 Regr
O)
0
500
1000
1500
2000
2500
3000
^ i Concentration of TCE (mg/kg/day)
This document is a draft for review purposes only and does not constitute Agency policy.
335 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
Figure 4-6. Comparisons of fold-changes in average relative liver weight and
gavage dose of (top panel) male B6C3F1 mice for 10-28 days of exposure
(Dees and Travis, 1993; Elcombe et al., 1985; Goldsworthy and Popp, 1987;
Merrick et al., 1989) and (bottom panel) in male B6C3F1 and Swiss mice.
inclusion of the 10 day studies, since as discussed in Section E.2.4.2, there was a greater increase
in TCE-induced liver weight at 28-42 days of exposure Swiss mice than the 10-day data in
B6C3F1 mice, and Kjellstrand et al. (1981b) noted that TCE-induced liver weight increases are
still increasing at 10 days inhalation exposure. A strain difference is not evident between the
Swiss and B6C3F1 males, as both the combined TCE data and that for only B6C3F1 mice show
similar correlation with the magnitude of dose and magnitude of percentage liver/body weight
increase. The correlation coefficients for the linear regressions presented for the B6C3F1 data
2 2
are R = 0.861 and for the combined data sets is R = 0.712. Comparisons of the slopes of the
dose-response curves suggest a greater consistency between TCE and DCA than between TCE
and TCA. There did not appear to be evidence of a plateau with higher TCE doses, and the
degree of fold-increase rises to higher levels with TCE than with TCA in the same strain of
mouse.
A more direct comparison would be on the basis of dose rather than drinking water
concentration. The estimations of internal dose of DCA or TCA from drinking water studies,
while varying considerably (DeAngelo et al., 1989; 2008), nonetheless suggest that the doses of
TCE used in the gavage experiments were much higher than those of DCA or TCA. However,
only a fraction of ingested TCE is metabolized to DCA or TCA, as, in addition to oxidative
metabolism, TCE is also cleared by GSH conjugation and by exhalation. While DCA dosimetry
is highly uncertain (see Sections 3.3 and 3.5), the mouse PBPK model, described in Section 3.5
was calibrated using extensive in vivo data on TCA blood, plasma, liver, and urinary excretion
data from inhalation and gavage TCE exposures, and makes robust predictions of the rate of
TCA production. If TCA were predominantly responsible for TCE-induced liver weight
increases, then replacing administered TCE dose (e.g., mg TCE/kg/day) by the rate of TCA
produced from TCE (mg TCA/kg/day) should lead to dose-response curves for increased liver
weight consistent with those from directly administered TCA. Figure 4-7 shows this comparison
using the PBPK model-based estimates of TCA production for four TCE studies from
28-42 days in the male NMRI, Swiss, and B6C3F1 mice (Kjellstrand et al., 1983a (Buben and
O'Flaherty, 1985; Goel et al., 1992; Merrick et al., 1989) and four oral TCA studies in B6C3F1
male mice at 2 g/L or lower drinking water exposure (DeAngelo et al., 1989; 2008; Kato-
Weinstein et al., 2001; Parrish et al., 1996) from 14-28 days of exposure. The selection of the
This document is a draft for review purposes only and does not constitute Agency policy.
336 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
2.5
O)
'a>
>>
T3
O
_Q
-C
O)
'a>
a>
>
0)
w
CO
0)
2
o
~ TCE Studies [28-42 d]
o TCA Studies [14-28 d]
Linear (TCA Studies [14-28 d])
Linear (TCE Studies [28-42 d])
1.5 -
100
400
200 300
mg TCA/kg-d
(produced [TCE studies] or administered [TCA studies])
500
28-42 day data for TCE was intended to address the decreased opportunity for full expression of
response at 10 days. PBPK modeling predictions of daily internal doses of TCA in terms of
mg/kg-day via produced via TCE metabolism would be are indeed lower than the TCE
concentrations in terms of mg/kg-day given orally by gavage. The predicted internal dose of
TCA from TCE exposure studies are of a comparable range to those predicted from TCA
drinking water studies at exposure concentrations in which palpability has not been an issue for
estimation of internal dose. Thus, although the TCE data are for higher exposure concentrations,
they are predicted to produce comparable levels of TCA internal dose estimated from direct TCA
administration in drinking water.
This document is a draft for review purposes only and does not constitute Agency policy.
337 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
Figure 4-7. Comparison of fold-changes in relative liver weight for data sets
in male B6C3F1, Swiss, and NRMI mice between TCE studies Kjellstrand
et al., 1983a (Buben and O'Flaherty, 1985; Goel et al., 1992; Merrick et al.,
1989) [duration 28-42 days]) and studies of direct oral TCA administration
to B6C3F1 mice (DeAngelo et al., 1989; DeAngelo et al., 2008; Kato-
Weinstein et al., 2001; Parrish et al., 1996)[duration 14-28 days]). Abscissa
for TCE studies consists of the median estimates of the internal dose of TCA
predicted from metabolism of TCE using the PBPK model described in
Section 3.5 of the TCE risk assessment. Lines show linear regression with
intercept fixed at unity. All data were reported fold-change in mean liver
weight/body weight ratios, except for Kjellstrand et al. (1983b), with were the
fold-change in the ratio of mean liver weight to mean body weight. In addition, in
Kjellstrand et al. (1983b), some systemic toxicity as evidence by decreased total
body weight was reported in the highest-dose group.
Figure 4-7 clearly shows that for a given amount of TCA produced from TCE, but going
through intermediate metabolic pathways, the liver weight increases are substantially greater
than, and highly inconsistent with, that expected based on direct TCA administration. In
particular, the response from direct TCA administration appears to "saturate" with increasing
TCA dose at a level of about 1.4-fold, while the response from TCE administration continues to
increase with dose to 1.75-fold at the highest dose administered orally in Buben and O'Flaherty
(1985) and over twofold in the inhalation study of Kjellstrand et al. (1983b). Because TCA liver
concentrations are proportional to the dose TCA, and do not depend on whether it is
administered in drinking water or internally produced in the liver, the results of the comparison
using the TCA liver dose-metric are identical.
Furthermore, while as noted previously, oral studies appear to report a linear relationship
between TCE exposure concentration and liver weight induction, the inclusion of inhalation
studies on the basis of internal dose led to a highly consistent dose-response curve for among
TCE study. Therefore, it is unlikely that differing routes of exposure can explain the
inconsistencies in dose-response.
The bioavailability of TCA, which in the above analysis is assumed to be 100%, is
another factor that may impact the dose-response. Sweeney et al. (2009), in an analysis of the
potential role of TCA in the liver carcinogenesis of tetrachloroethylene, identified a number of
previously unpublished TCA kinetic data in mice exposed to TCA via drinking water for
3-14 days. They concluded that fractional absorption of TCA via drinking water exposures is
much less than 100%—about 29% at low exposures and decreasing with increasing dose.
However, the conclusions of the Sweeney et al. (2009) were based on the Hack et al. (2006) TCE
PBPK model, which had a number of deficiencies, as noted in Section 3.5 and Appendix A.
This document is a draft for review purposes only and does not constitute Agency policy.
338 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Therefore, as discussed in Appendix A, Chiu (Chiu, In Press)(in press) reanalyzed those data
using the updated TCE PBPK model of Evans et al. (2009) and Chiu et al. (2009) and concluded
that while there was evidence of reduced absorption (80-90% at low exposures, and decreasing
with increasing dose), it was not as low as that estimated by Sweeney et al. (2009). As discussed
in Appendix A, it may be more accurate to characterize the fractional absorption as an empirical
parameter reflecting unaccounted-for biological processes as well as experimental variation.
Chiu (Chiu, In Press) also reanalyzed the data on TCE- and TCA-induced hepatomegaly
using the central estimates of the fractional absorption of TCA inferred from the analysis
described above. Figure 4-8 shows the results, comparing a fixed fractional absorption of 95%
with the fitted fractional absorption from Chiu (Chiu, In Press), here plotted using area-under-
the-curve (AUC) of TCA in the liver as the dose-metric. For reference, the dose-response for
administered TCA with an assumption of fixed, nearly complete absorption (analogous to the
results from Evans et al., 2009, Figure 4-7) is also included. While the reduced fractional
absorption inferred from drinking water data reported by Sweeney et al. (2009) accounts for part
of the difference in dose-responses between TCE- and TCA-induced hepatomegaly reported by
Evans et al. (2009), it does not appear to be able to account for the entire difference. In
particular, the fraction of hepatomegaly contributed by TCA is about 0.20 assuming nearly
complete absorption, as compared to about 0.33 assuming the best-fitting fractional absorption
inferred from the PBPK
This document is a draft for review purposes only and does not constitute Agency policy.
339 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
2.5
• TCE Studies
~ TCA Studies,
Fabs=0.95
O TCA Studies, Fabs
0 500 1000 1500 2000 2500 3000 3500
Daily AUC of TCA in liver (mg-hr/L-d)
Figure 4-8. Comparison of hepatomegaly as a function of AUC of TCA in
liver, using values for the TCA drinking water fractional absorption (Fabs).
Fold-changes in relative liver weight for data sets in male B6C3F1, Swiss, and
NRMI mice between TCE studies Kjellstrand et al., 1983a (Buben and
OTlaherty, 1985; Goel et al., 1992; Merrick et al., 1989) [duration 28-42 days]
and studies of direct oral TCA administration to B6C3F1 mice (DeAngelo et al.,
1989; DeAngelo et al., 2008; Kato-Weinstein et al., 2001; Parrish et al., 1996)
Green, 2003b [duration 14-28 days]. Linear regressions were compared using
ANOVA to assess whether the TCE studies were consistent with the TCA studies,
using TCA as the dose-metric. For each analysis of drinking water fraction
absorption, ANOVA /^-values were <10 4 when comparing the assumption that all
the data had a common slope with the assumption that TCE and TCA data had
different slopes.
model-based analysis. The inability of TCA to account for TCE-induced hepatomegaly is
confirmed statistically by ANOVA, with /^-values of <10 4, Therefore, assuming a reduced TCA
bioavailability does not change the conclusion that the available data are inconsistent with the
toxicological hypothesis that TCA can fully account for TCE-induced hepatomegaly.
This document is a draft for review purposes only and does not constitute Agency policy.
340 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Additional analyses do, however, support a role for oxidative metabolism in
TCE-induced liver weight increases, and that the parent compound TCE is not the likely active
moiety (suggested previously by Buben and O'Flaherty (1985)). In particular, the same studies
are shown in Figure 4-9 using PBPK-model based predictions of the AUC of TCE in blood and
total oxidative metabolism, which produces chloral, TCOH, DCA, and other metabolites in
addition to TCA. The dose-response relationship between TCE blood levels and liver weight
increase, while still having a significant trend, shows substantial scatter and a low R of 0.43.
On the other hand, using total oxidative metabolism as the dose-metric leads to substantially
more consistency dose-response across studies, and a much tighter linear trend with an R of
0.90 (see Figure 4-9). A similar consistency is observed using liver-only oxidative metabolism
as the dose-metric, with R of 0.86 (not shown). Thus, while the slope is similar between liver
weight increase and TCE concentration in the blood and liver weight increase and rate of total
oxidative metabolism, the data are a much better fit for total oxidative metabolism.
Figure 4-9. Fold-changes in relative liver weight for data sets in male
B6C3F1, Swiss, and NRMI mice reported by TCE studies of duration
28-42 days (Kjellstrand et al., 1983a(Buben and O'Flaherty, 1985; Goel et
al., 1992; Merrick et al., 1989) using internal dose-metrics predicted by the
PBPK model described in Section 3.5: (A) dose-metric is the median estimate
of the daily AUC of TCE in blood, (B) dose-metric is the median estimate of
the total daily rate of TCE oxidation. Lines show linear regression. Use of
liver oxidative metabolism as a dose-metric gives results qualitatively similar to
(B), with R2 = 0.86.
This document is a draft for review purposes only and does not constitute Agency policy.
341 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Although the qualitative similarity to the linear dose-response relationship between DCA
and liver weight increases is suggestive of DCA being the predominant metabolite responsible
for TCE liver weight increases, due to the highly uncertain dosimetry of DCA derived from
TCE, this hypothesis cannot be tested on the basis of internal dose. Similarly, another TCE
metabolite, chloral hydrate, has also been reported to induce liver tumors in mice, however, there
are no adequate comparative data to assess the nature of liver weight increases induced by this
TCE metabolite (see Section E.2.5 and Section 4.5.1.2.4 below). Whether its formation in the
liver after TCE exposure correlates with TCE-induced liver weight changes cannot be
determined.
4.5.6.1.4. Cytotoxicity
As discussed above, TCE has sometimes been reported to cause minimal/mild focal
hepatocellular necrosis or other signs of hepatic injury, albeit of low frequency and mostly at
doses >1,000 mg/kg-day (Dees and Travis, 1993; Elcombe et al., 1985) or at exposures
>1,000 ppm in air (Ramdhan et al., 2010; Ramdhan et al., 2008) from 7-10 days of exposure.
Data from available studies are supportive of a role for oxidative metabolism in TCE-induced
cytotoxicity in the liver, though they are not informative as to the actual active moiety(ies).
Buben and O'Flaherty (1985) noted a strong correlation (R-squared of between
glucose-6-phosphatase inhibition and total urinary oxidative metabolites). Ramdhan et al. (2008)
conducted parallel experiments at TCE 1,000 and 2,000 ppm (8 hours/day, 7 days) in wild-type
and CYP2El-null mice, the latter of which did not exhibit hepatotoxicity (assessed by serum
ALT, AST, and histopathology) and excreted twofold lower amounts of oxidative metabolites
TCA and TCOH in urine as compared to wild-type mice. In addition, urinary TCA and TCOH
excretion was correlated with serum ALT and AST measures, though the R-squared values
(square of the reported correlation coefficients) were relatively low (0.54 and 0.67 for TCOH and
TCA, respectively). Ramdhan et al. (2010) reported that TCA and TCOH were detected in the
urine of wild type and PPARa-null and humanized mice after TCE exposure with no significant
differences between the 1,000 and 2,000 ppm TCE treatments. TCA concentrations were
significantly lower and TCOH concentrations higher in exposed PPARa-null mice relative to
wild type mice. They stated that urinary TCA levels in wild type mice were incorrectly reported
by Ramdhan et al. (2008) but have been corrected in this study. AST and ALT levels were
significantly increased in all exposed mice relative to control with mean levels between 1,000
and 2,000 ppm TCE exposures higher but not significantly different (41-74% and 36-79%
higher, respectively). Although increased, such increases were small. Necrosis scores were
This document is a draft for review purposes only and does not constitute Agency policy.
342 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
reported to be significantly higher in TCE exposed mice relative to controls in all three genotype
mice and to be significantly higher with 2,000 versus 1,000 ppm TCE exposure in wild type mice
and hPPARa mice. Inflammation scores were reported to be significantly higher with exposed
group than control with 2,000 ppm TCE exposure than controls for each genotype group with a
difference between the 2,000 ppm and 1,000 ppm exposure groups in wild type mice. However,
necrosis and inflammation score means at the highest TCE exposure levels in any mouse strain
were minimal (only occasional necrotic cells in any lobule) for necrosis and mild for
inflammation (<2 foci/field ).
With respect to CH (166 mg/kg-day) and DCA (-90 mg/kg-day), Daniel et al. (1992)
reported that after drinking water treatment, hepatocellular necrosis and chronic active
inflammation were reported to be mildly increased in both prevalence and severity in all treated
groups after 104 weeks of exposure. The histological findings, from interim sacrifices (n = 5),
were considered by the authors to be unremarkable and were not reported. TCA has not been
reported to induce necrosis in the liver under the conditions tested. Relatively high doses of
DCA (>1 g/L in drinking water) appear to result in mild focal necrosis with attendant reparative
proliferation at lesion sites, but no such effects were reported at lower doses (<0.5 g/L in
drinking water) more relevant for comparison with TCE (DeAngelo et al., 1999; Sanchez and
Bull, 1990; Stauber et al., 1998). Enlarged nuclei and changes consistent with increased ploidy,
are further discussed below in the context of DNA synthesis.
4.5.6.1.5. DNA synthesis and polyploidization
The effects on DNA synthesis and polyploidization observed with TCE treatment have
similarly been observed with TCA and DCA. With respect to CH, George et al. (2000) reported
that CH exposure did not alter DNA synthesis in rats and mice at any of the time periods
monitored (all well past 2 weeks), with the exception of 0.58 g/L chloral hydrate at 26 weeks
slightly increasing hepatocyte labeling (~two-threefold of controls) in rats and mice but the
percentage labeling still representing 3% or less of hepatocytes.
In terms of whole liver or hepatocyte label incorporation, the most comparable exposure
duration between TCE, TCA, and DCA studies is the 10- and 14-day period. Several studies
have reported that in this time period, peak label incorporation into individual hepatocytes and
whole liver for TCA and DCA have already passed (Carter et al., 1995; Pereira, 1996; Sanchez
and Bull, 1990; Styles et al., 1991). A direct time-course comparison is difficult, since data at
earlier times for TCE are more limited.
This document is a draft for review purposes only and does not constitute Agency policy.
343 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
There are conflicting reports of DNA synthesis induction in individual hepatocytes for up
to 14 days of DCA or TCA exposure. In particular, Sanchez and Bull (1990) reported tritiated
thymidine incorporation in individual hepatocytes up to 2 g/L exposure to DCA or TCA induced
little increase in DNA synthesis except in instances and in close proximity to areas of
proliferation/necrosis for DCA treatment after 14 days of exposure in male mice. The largest
percentage of hepatocytes undergoing DNA synthesis for any treatment group was less than
1% of hepatocytes. However, they reported treatment- and exposure duration-changes in hepatic
DNA incorporation of tritiated thymidine for DCA and TCA. For TCA treatment, the largest
increases over control levels for hepatic DNA incorporation (at the highest dose) was a threefold
increase after 5 days of treatment and a twofold increase over controls after 14 days of treatment.
For DCA whole-liver tritiated thymidine incorporation was only slightly elevated at necrogenic
concentrations and decreased at the 0.3 g/L non-necrogenic level after 14 days of treatment. In
contrast to Sanchez and Bull (1990), Stauber and Bull (1997) reported increased tritiated
thymidine incorporation for individual hepatocytes after 14 days of treatment with 2 g/L DCA or
TCA in male mice. They used a more extended period of tritiated thymidine exposure of
3-5 days and so these results represent aggregate DNA synthesis occurring over a more extended
period of time. A "1-day labeling index" was reported as less than 1% for the highest level of
increased incorporation. However, after 14 days, the labeling index was reported to be increased
by ~3.5-fold for TCA and ~5.5-fold for DCA over control values. After 28 days, the labeling
index was reported to be decreased ~2.3-fold by DCA and increased ~2.5-fold after treatment
with TCA. Pereira (1996) reported that for female B6C3F1 mice, 5-day incorporation of BrDU,
as a measure of DNA synthesis, was increased at 0.86 g/L and 2.58 g/L DCA treatment for
5 days (-twofold at the highest dose) but that by Day 12 and 33 levels had fallen to those of
controls. For TCA exposures, 0.33 g/L, 1.10 g/L and 3.27 g/L TCA all gave a similar -threefold
increase in BrdU incorporation by 5 days, but that by 12 and 33 days were not changed from
controls. Nonetheless, what is consistent is that these data report that, similar to TCE-exposed
mice at 10 days of exposure, cells undergoing DNA synthesis in DCA- or TCA-exposed mice for
up to 14 days of exposure to be confined to a very small population of cells in the liver. Thus,
these data are consistent with hypertrophy being primarily responsible for liver weight gains as
opposed to increases in cell number in mice.
Interestingly, a lack of correlation between whole liver label incorporation and that in
individual hepatocytes has been reported by several studies of DCA (Carter et al., 1995; Sanchez
and Bull, 1990). For example, Carter et al. (1995) reported no increase in labeling of
hepatocytes in comparison to controls for any DCA treatment group from 5-30 days of DCA
exposure. Rather than increase hepatocyte labeling, DCA induced no change from Days 5
This document is a draft for review purposes only and does not constitute Agency policy.
344 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
though 15 but significantly decreased levels between days 20 and 30 for 0.5 g/L that were similar
to those observed for the 5 g/L exposures. However, for whole liver DNA tritiated thymidine
incorporation, Carter et al. (1995) reported 0.5g/L DCA treatments to show trends of initial
inhibition of DNA tritiated thymidine incorporation followed by enhancement of labeling that
was not statistically significant from 5-30 days of exposure. Examination of individual
hepatocytes does not include the contribution of nonparenchymal cell DNA synthesis that would
be detected in whole liver DNA. As noted above, proliferation of the nonparenchymal cell
compartment of the liver has been noted in several studies of TCE in rodents, and thus, this is
one possible reason for the reported discrepancy.
Another possible reason for this inconsistency with DCA treatment is polyploidization, as
was suggested above for TCE. Although this was not examined for DCA or TCA exposure by
Sanchez and Bull (1990), Carter et al. (1995) reported that hepatocytes from both 0.5 and 5 g/L
DCA treatment groups had enlarged, presumably polyploidy nuclei, with some hepatocyte nuclei
labeled in the mid-zonal area. There were statistically significant changes in cellularity, nuclear
size, and multinucleated cells during 30 days exposure to DCA. The percentage of
mononucleated cells hepatocytes was reported to be similar between control and DCA treatment
groups at 5- and 10-day exposure. However, at 15 days and beyond DCA treatments were
reported to induce increases in mononucleated hepatocytes with later time periods to also
showing DCA-induced increases nuclear area, consistent with increased polyploidization without
mitosis. The consistent reporting of an increasing number of mononucleated cells between 15
and 30 days could be associated with clearance of mature hepatocytes as suggested by the report
of DCA-induced loss of cell nuclei. The reported decrease in the numbers of binucleate cells in
favor of mononucleate cells is not typical of any stage of normal liver growth (Brodsky and
Uryvaeva, 1977). The pattern of consistent increase in percentage liver/body weight induced by
0.5 g/L DCA treatment from days 5 though 30 was not consistent with the increased numbers of
mononucleate cells and increase nuclear area reported from Day 20 onward. Specifically, the
large differences in liver weight induction between the 0.5 g/L treatment group and the 5 g/L
treatment groups at all times studied also did not correlate with changes in nuclear size and
percentage of mononucleate cells. Thus, increased liver weight was not a function of cellular
proliferation, but probably included both aspects of hypertrophy associated with polyploidization
and increased glycogen deposition (see below) induced by DCA. Carter et al. (1995) suggested
that although there is evidence of DCA-induced cytotoxicity (e.g., loss of cell membranes and
apparent apoptosis), the 0.5 g/L exposure concentration has been shown to increase
hepatocellular lesions after 100 weeks of treatment without concurrent peroxisome proliferation
or cytotoxicity (DeAngelo et al., 1999).
This document is a draft for review purposes only and does not constitute Agency policy.
345 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
In sum, the observation of TCE-treatment related changes in DNA content, label
incorporation, and mitotic figures are generally consistent with patterns observed for both TCA
and DCA. In all cases, hepatocellular proliferation is confined to a very small fraction of
hepatocytes, and hepatomegaly observed with all three treatments probably largely reflects
cytomegaly rather than cell proliferation. Moreover, label incorporation likely largely reflects
polyploidization rather than hepatocellular proliferation, with a possible contribution from
nonparenchymal cell proliferation. As with TCE, histological changes in nuclear sizes and
number also suggest a significant degree of treatment-related polyploidization, particularly for
DCA.
4.5.6.1.6. Apoptosis
As for apoptosis, Both El combe et al. (1985) and Dees and Travis (1993) reported no
changes in apoptosis other than increased apoptosis only at a treatment level of 1,000-mg/kg
TCE. Dees and Travis (1993) reported that increased apoptoses from TCE exposure "did not
appear to be in proportion to the applied TCE dose given to male or female mice." Channel et al.
(1998) reported that there was no significant difference in apoptosis between TCE treatment and
control groups with data not shown. However, the extent of apoptosis in any of the treatment
groups, or which groups and timepoints were studied for this effect cannot be determined. While
these data are quite limited, it is notable that peroxisome proliferators have been suggested
inhibit, rather than increase, apoptosis as part of their carcinogenic MOA (Klaunig et al., 2003).
However, for TCE metabolites, DCA has been most studied, though it is clear that age
and species affect background rates of apoptosis. Snyder et al. (1995), in their study of DCA,
report that control mice were reported to exhibit apoptotic frequencies ranging from
-0.04-0.085%, that over the 30-day period of their study the frequency rate of apoptosis
declined, and suggest that this pattern is consistent with reports of the livers of young animals
undergoing rapid changes in cell death and proliferation. They reported rat liver to have a
greater the estimated frequency of spontaneous apoptosis (~0.1%) and therefore, greater than that
of the mouse. Carter et al. (1995) reported that after 25 days of 0.5 g/L DCA treatment apoptotic
bodies were reported as well as fewer nuclei in the pericentral zone and larger nuclei in central
and midzonal areas. This would indicate an increase in the apoptosis associated with potential
increases in polyploidization and cell maturation. However, Snyder et al. (1995) report that mice
treated with 0.5 g/L DCA over a 30-day period had a similar trend as control mice of decreasing
apoptosis with age. The percentage of apoptotic hepatocytes decreased in DCA-treated mice at
the earliest time point studied and remained statistically significantly decreased from controls
This document is a draft for review purposes only and does not constitute Agency policy.
346 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
from 5-30 days of exposure. Although the rate of apoptosis was very low in controls, treatment
with 0.5g/L DCA reduced it further (-30-40% reduction) during the 30-day study period. The
results of this study not only provide a baseline of apoptosis in the mouse liver, which is very
low, but also to show the importance of taking into account the effects of age on such
determinations. The significance of the DCA-induced reduction in apoptosis reported in this
study, from a level that is already inherently low in the mouse, for the MOA for induction of
DCA-induce liver cancer is difficult to discern.
4.5.6.1.7. Glycogen accumulation
As discussed in Sections E.3.2 and E.3.4.2.1, glycogen accumulation has been described
to be present in foci in both humans and animals as a result from exposure to a wide variety of
carcinogenic agents and predisposing conditions in animals and humans. The data from
Elcombe et al. (1985) included reports of TCE-induced pericentral hypertrophy and eosinophilia
for both rats and mice but with "fewer animals affected at lower doses." In terms of glycogen
deposition, Elcombe report "somewhat" less glycogen pericentrally in the livers of rats treated
with TCE at 1,500 mg/kg than controls with less marked changes at lower doses restricted to
fewer animals. They do not comment on changes in glycogen in mice. Dees and Travis (1993)
reported TCE-induced changes to "include an increase in eosinophilic cytoplasmic staining of
hepatocytes located near central veins, accompanied by loss of cytoplasmic vacuolization."
Since glycogen is removed using conventional tissue processing and staining techniques, an
increase in glycogen deposition would be expected to increase vacuolization and thus, the report
from Dees and Travis is consistent with less not more glycogen deposition. Neither study
produced a quantitative analysis of glycogen deposition changes from TCE exposure. Although
not explicitly discussing liver glycogen content or examining it quantitatively in mice, these
studies suggest that TCE-induced liver weight increases did not appear to be due to glycogen
deposition after 10 days of exposure and any decreases in glycogen were not necessarily
correlated with the magnitude of liver weight gain either.
For TCE and TCA 500 mg/kg treatments in mice for 10 days, changes in glycogen were
not reported in the general descriptions of histopathological changes (Dees and Travis, 1993;
Elcombe et al., 1985; Styles et al., 1991) or were specifically described by the authors as being
similar to controls (Nelson et al., 1989). However, for DCA, glycogen deposition was
specifically noted to be increased with treatment, although no quantitative analyses was
presented that could give information as to the nature of the dose-response (Nelson et al., 1989).
This document is a draft for review purposes only and does not constitute Agency policy.
347 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
In regard to cell size, although increased glycogen deposition with DCA exposure was
noted by Sanchez and Bull (1990) to occur to a similar extent in B6C3F1 and Swiss Webster
male mice despite differences in DCA-induced liver weight gain. Lack of quantitative analyses
of that accumulation in this study precludes comparison with DCA-induced liver weight gain.
Carter et al. (1995) reported that in control mice there was a large variation in apparent glycogen
content and also did not perform a quantitative analysis of glycogen deposition. The variability
of this parameter in untreated animals and the extraction of glycogen during normal tissue
processing for light microscopy make quantitative analyses for dose-response difficult unless
specific methodologies are employed to quantitatively assess liver glycogen levels as was done
by Kato-Weinstein et al. (2001) and Pereira et al. (2004b).
Bull et al. (1990) reported that glycogen deposition was uniformly increased from 2 g/L
DCA exposure with photographs of TCA exposure showing slightly less glycogen staining than
controls. However, the abstract and statements in the paper suggest that there was increased
PAS positive material from TCA treatment that has caused confusion in the literature in this
regard. Kato-Weinstein et al. (2001) reported that in male B6C3F1 mice exposed to DCA and
TCA, the DCA treatment increased glycogen and TCA decreased glycogen content of the liver
by using both chemical measurement of glycogen in liver homogenates and by using
ethanol-fixed sections stained with PAS, a procedure designed to minimize glycogen loss.
Kato-Weinstein et al. (2001) reported that glycogen rich and poor cells were scattered
without zonal distribution in male B6C3F1 mice exposed to 2 g/L DCA for 8 weeks. For TCA
treatments, they reported centrilobular decreases in glycogen and -25% decreases in whole liver
by 3 g/L TCA. Kato-Weinstein et al. (2001) reported whole liver glycogen to be increased
~1.50-fold of control (90 vs. 60 mg glycogen/g liver) by 2 g/L DCA after 8 weeks exposure male
B6C3F1 mice with a maximal level of glycogen accumulation occurring after 4 weeks of DCA
exposure. Pereira et al. (2004b) reported that after 8 weeks of exposure to 3.2 g/L DCA liver
glycogen content was 2.20-fold of control levels (155.7 vs. 52.4 mg glycogen/g liver) in female
B6C3F1 mice. Thus, the baseline level of glycogen content reported by (-60 mg/g) and the
increase in glycogen after DCA exposure was consistent between Kato-Weinstein et al. (2001)
and Pereira et al. (2004b). However, the increase in liver weight reported by Kato-Weinstein
et al. (2001) of 1.60-fold of control percentage liver/body weight cannot be accounted for by the
1.50-fold of control glycogen content. Glycogen content only accounts for 5% of liver mass so
that 50% increase in glycogen cannot account for the 60% increase liver mass induced by 2 g/L
DCA exposure for 8 weeks reported by Kato-Weinstein (2001). Thus, DCA-induced increases
in liver weight are occurring from other processes as well. Carter et al. (2003) and DeAngelo
et al. (1999) reported increased glycogen after DCA treatment at much lower doses after longer
This document is a draft for review purposes only and does not constitute Agency policy.
348 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
periods of exposure (100 weeks). Carter reported increased glycogen at 0.5 g/L DCA and
DeAngelo et al. (1999) at 0.03 g/L DCA in mice. However, there is no quantitation of that
increase.
4.5.6.1.8. Peroxisome proliferation and related effects
TCA and DCA have both been reported to induce peroxisome proliferation or increase in
related enzyme markers in rodent hepatocytes (DeAngelo et al., 1989; 1997; Mather et al., 1990;
Parrish et al., 1996). Between TCA and DCA, both induce peroxisome proliferation in various
strains of mice, but it clear that TCA and DCA are weak PPARa agonists and that DCA is
weaker than TCA in this regard (Nelson et al., 1989) using a similar paradigm.
George et al. (2000) reported that CH exposure did not hepatic PCO activity in rats and
mice at any of the time periods monitored. It is notable that the only time at which DNA
synthesis index was (slightly) increased, at 26 weeks, there remained a lack of induction of PCO.
A number of measures that may be related to peroxisome proliferation were investigated in
Leakey et al. (2003a). Of the enzymes associated with PPARa agonism (total CYP, CYP2B
isoform, CYP4A, or lauric acid P-hydroxylase activity), only CYP4A and lauric acid
P-hydroxylase activity were significantly increased at 15 months of exposure in the
dietary-restricted group administered the highest dose (100 mg/kg CH) with no other groups
reported showing a statistically significant increased response (n = 12/group). There is an issue
of interpretation of peroxisomal enzyme activities and other enzymes associated with PPARa
receptor activation to be a relevant event in liver cancer induction at a time period in which
tumors or foci are already present. Although not statistically significant, the 100 mg/kg CH
exposure group of ad libitum-fed mice also had an increase in CH-induced increases of CYP4A
and lauric acid P-hydroxylase activity. Seng et al. (2003) described CH toxicokinetics and
peroxisome proliferation-associated enzymes in mice at doses up to 1,000 mg/kg-day for
2 weeks with dietary control or caloric restriction. Lauric acid P-hydroxylase and PCO activities
were reported to be induced only at doses >100 mg/kg in all groups, with dietary-restricted mice
showing the greatest induction. Differences in serum levels of TCA, the major metabolite
remaining 24 hours after dosing, were reported not to correlate with hepatic lauric acid
P-hydroxylase activities across groups.
Direct quantitative inferences regarding the magnitude of response in these studies in
comparison to TCE, however, are limited by possible variability and confounding. In particular,
many studies used cyanide-insensitive PCO as a surrogate for peroxisome proliferation, but the
utility of this marker may be limited for a number of reasons. First, several studies have shown
This document is a draft for review purposes only and does not constitute Agency policy.
349 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
that this activity is not well correlated with the volume or number of peroxisomes that are
increased as a result of exposure to TCE or it metabolites (Elcombe et al., 1985; Nakajima et al.,
2000; Nelson et al., 1989). In addition, this activity appears to be highly variable both as a
baseline measure and in response to chemical exposures. Laughter et al. (2004) presented data
showing WY-14,643 induced increases in PCO activity that varied up to sixfold between
different experiments in wild-type mice. They also showed that, in some instances, PCO activity
in untreated PPARa-null mice was up to sixfold greater than that in wild-type mice. Parrish
et al. (1996) noted that control values between experiments varied as much as a factor of twofold
for PCO activity and thus, their data were presented as percentage of concurrent controls.
Furthermore, Melnick et al. (1987) reported that corn oil administration alone can elevate PCO
(as well as catalase) activity, and corn oil has also been reported to potentiate the induction of
PCO activity of TCA in male mice (DeAngelo et al., 1989). Thus, quantitative inferences
regarding the magnitude of response in these studies are limited by a number of factors. For
example, in the studies reported in DeAngelo et al. (2008) a small number of animals was
studied for PCO activity at interim sacrifices (n = 5). PCO activity varied 2.7-fold as baseline
controls. Although there was a 10-fold difference in TCA exposure concentration, the increase
in PCO activity at 4 weeks was 1.3-, 2.4-, and 5.3-fold of control. More information on the
relationship of PCO enzyme activity and its relationship to carcinogenicity is discussed in
Section E.3.4 and below.
4.5.6.1.9. Oxidative stress
Very limited data are available as to oxidative stress and related markers induced by the
oxidative metabolites of TCE. As discussed in Appendix E, above, there are limited data that do
not indicate significant oxidative stress and associated DNA damage associated with acute and
subacute TCE treatment. In regard to DCA and TCA, Larson and Bull (1992b) exposed male
B6C3F1 mice or Fischer 344 rats to single doses TCA or DCA in distilled water by oral gavage
(n = 4). In the first experiment, TBARS was measured from liver homogenates and assumed to
be malondialdehyde. The authors stated that a preliminary experiment had shown that maximal
TBARS was increased 6 hours after a dose of DCA and 9 hours after a dose of TCA in mice and
that by 24 hours TBARS concentrations had declined to control values. Time-course
information in rats was not presented. A dose of 100 mg/kg DCA (rats or mice) or TCA (mice)
did not elevate TBARS concentrations over that of control liver with this concentration of TCA
not examined in rats. For TCA, there was a slight dose-related increase in TBARS over control
values starting at 300 mg/kg in mice with the increase in TBARS increasing at a rate that was
This document is a draft for review purposes only and does not constitute Agency policy.
350 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
lower than the magnitude of increase in dose. Of note, is the report that the induction of TBARS
in mice is transient and has subsided within 24 hours of a single dose of DCA or TCA, that the
response in mice appeared to be slightly greater with DCA than TCA at similar doses, and that
for DCA, there was similar TBARS induction between rats and mice at similar dose levels.
Austin et al. (1996) appears to a follow-up publication of the preliminary experiment
cited in Larson and Bull (1992b). Male B6C3F1 mice were treated with single doses of DCA or
TCA via gavage with liver examined for 8-OHdG. The authors stated that in order to conserve
animals, controls were not employed at each time point. There was a statistically significant
increase over controls in 8-OHdG for the 4- and 6-hour time points for DCA (-1.4- and 1.5-fold
of control, respectively) but not at 8 hours in mice. For TCA, there was a statistically significant
increase in 8-OHdG at 8 and 10 hours for TCA (-1.4- and 1.3-fold of control, respectively).
Consistent results as to low, transient increases in markers of "oxidative stress" were also
reported by Parrish et al. (1996), who in addition to examining oxidative stress alone, attempted
to examine its possible relationship to PCO and liver weight in male B6C3F1 mice exposed to
TCA or DCA for 3 or 10 weeks (n = 6). The dose-related increase in PCO activity at 21 days for
TCA was reported to not be increased similarly for DCA. Only the 2.0 g/L dose of DCA was
reported to induce a statistically significant increase at 21-days of exposure of PCO activity over
control (-1.8-fold of control). After 71 days of treatment, TCA induced dose-related increases in
PCO activities that were approximately twice the magnitude as that reported at 21 days.
Treatments with DCA at the 0.1 and 0.5 g/L exposure levels produced statistically significant
increase in PCO activity of-1.5- and 2.5-fold of control, respectively. The administration of
1.25 g/L clofibric acid in drinking water, used as a positive control, gave -six-sevenfold of
control PCO activity at 21 and 71 days exposure. Parrish et al. (1996) reported that laurate
hydroxylase activity was reported to be elevated significantly only by TCA at 21 days and to
approximately the same extent (-1.4- to 1.6-fold of control) increased at all doses tested and at
71 days both the 0.5 and 2.0 g/L TCA exposures to a statistically significant increase in laurate
hydroxylase activity (i.e., 1.6- and 2.5-fold of control, respectively). No change was reported
after DCA exposure. Laurate hydroxylase activity within the control values varying 1.7-fold
between 21 and 71 days experiments. Levels of 8-OHdG in isolated liver nuclei were reported to
not be altered from 0.1, 0.5, or 2.0 g/L TCA or DCA after 21 days of exposure and this negative
result was reported to remain even when treatments were extended to 71 days of treatment. The
authors noted that the level of 8-OHdG increased in control mice with age (i.e., -twofold
increase between 71-day and 21-day control mice). Thus, the increases in PCO activity noted for
DCA and TCA were not associated with 8-OHdG levels (which were unchanged) and also not
with changes laurate hydrolase activity observed after either DCA or TCA exposure. Of note, is
This document is a draft for review purposes only and does not constitute Agency policy.
3 51 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
that the authors report taking steps to minimize artifactual responses for their 8-OHdG
determinations. The authors concluded that their data suggest that peroxisome proliferative
properties of TCA were not linked to oxidative stress or carcinogenic response.
4.5.6.1.10. Comparisons of Trichloroethylene (TCE)-Induced Carcinogenic Responses With
Trichloroacetic Acid (TCA), Dichloroacetic Acid (DCA), and Chloral Hydrate (CH)
Studies
4.5.6.1.11. Studies in rats
As discussed above, data on TCE carcinogenicity in rats, while not reporting statistically
significantly increased risks, are not entirely adequate due to low numbers of animals, increased
systemic toxicity, and/or increased treatment-related or accidental mortality. Notably, several
studies in rats noted a few very rare types of liver or biliary tumors (cystic cholangioma,
cholangiocarcinoma, or angiosarcomas) in treated animals. For TCA, DCA and CH, there are
even fewer studies in rats, so there is a very limited ability to assess the consistency or lack
thereof in rat carcinogenicity among these compounds.
For TCA, the only available study in rats (DeAngelo et al., 1997) has been frequently
cited in the literature to indicate a lack of response in this species for TCA-induced liver tumors.
However, this study does report an apparent dose-related increase in multiplicity of adenomas
and an increase in carcinomas over control at the highest dose. The use by DeAngelo et al.
(1997) of a relatively low number of animals per treatment group (n = 20-24) limits this study's
ability to determine a statistically significant increase in tumor response. Its ability to determine
an absence of treatment-related effect is similarly limited. In particular, a power calculation of
the study shows that for most endpoints (incidence and multiplicity of all tumors at all exposure
DCA concentrations), the Type II error, which should be >50%, was less than 8%. The only
exception was for the incidence of adenomas and adenomas and carcinomas for the 0.5 g/L
treatment group (58%), at which, notably, there was a reported increase in reported adenomas or
adenomas and carcinomas combined over control (15 vs. 4%). Therefore, the likelihood of a
false null hypothesis was not negligible. Thus, while suggesting a lower response than for mice
for liver tumor induction, this study is inconclusive for determining of whether TCA induces a
carcinogenic response in the liver of rats.
For DCA, there are two reported long-term studies in rats (DeAngelo et al., 1996;
Richmond et al., 1995) that appear to have reported the majority of their results from the same
data set and which consequently were subject to similar design limitations and DCA-induced
This document is a draft for review purposes only and does not constitute Agency policy.
352 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
neurotoxicity in this species. DeAngelo et al. (1996) reported increased hepatocellular adenomas
and carcinomas in male F344 rats exposed to DCA for 2 years. However, the data from
exposure concentrations at a 5 g/L dose had to be discarded and the 2.5 g/L DCA dose had to be
continuously lowered during the study due to neurotoxicity. There was a DCA-induced
increased in adenomas and carcinomas combined reported for the 0.5 g/L DCA (24.1 vs.
4.4% adenomas and carcinomas combined in treated vs. controls) and an increase at a variable
dose started at 2.5 g/L DCA and continuously lowered (28.6 vs. 3.0% adenomas and carcinomas
combined in treated vs. controls). Only combined incidences of adenomas and carcinomas for
the 0.5 g/L DCA exposure group was reported to be statistically significant by the authors
although the incidence of adenomas was 17.2 versus 4% in treated versus control rats.
Hepatocellular tumor multiplicity was reported to be increased in the 0.5 g/L DCA group
(0.31 adenomas and carcinomas/animal in treated vs. 0.04 in control rats) but was reported by the
authors to not be statistically significant. At the starting dose of 2.5 g/L that was continuously
lowered due to neurotoxicity, the increased multiplicity of hepatocellular carcinomas was
reported by the authors to be to be statistically significant (0.25 carcinomas/animals vs. 0.03 in
control) as well as the multiplicity of combined adenomas and carcinomas (0.36 adenomas and
carcinomas/animals vs. 0.03 in control rats). Issues that affect the ability to determine the nature
of the dose-response for this study include (1) the use of a small number of animals (n = 23,
n = 21, and n = 23 at final sacrifice for the 2.0 g/L NaCl control, 0.05 g/L and 0.5 g/L treatment
groups) that limit the power of the study to both determine statistically significant responses and
to determine that there are not treatment-related effects (i.e., power) (2) apparent addition of
animals for tumor analysis not present at final sacrifice (i.e., 0.05 and 0.5 g/L treatment groups),
and (3) most of all, the lack of a consistent dose for the 2.5 g/L DCA exposed animals.
Similar issues are present for the study of Richmond et al. (1995) which was conducted
by the same authors as DeAngelo et al. (1996) and appeared to be the same data set. There was a
small difference in reports of the results between the two studies for the same data for the 0.5 g/L
DCA group in which Richmond et al. (1995) reported a 21% incidence of adenomas and
DeAngelo et al. (1996) reported a 17.2% incidence. The authors did not report any of the results
of DCA-induced increases of adenomas and carcinomas to be statistically significant. The same
issues discussed above for DeAngelo et al. (1996) apply to this study. Similar to the DeAngelo
et al. (1997) study of TCA in rats, the use in these DCA studies (DeAngelo et al., 1996;
Richmond et al., 1995) of relatively small numbers of rats limits the detection of
treatment-related effects and the ability to determine whether there was no treatment related
effects (Type II error), especially at the low concentrations of DCA exposure.
This document is a draft for review purposes only and does not constitute Agency policy.
353 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
For CH, George et al. (2000) exposed male F344/N rats to CH in drinking water for
2 years. Groups of animals were sacrificed at 13, 26, 52, and 78 weeks following the initiation
of dosing, with terminal sacrifices at Week 104. Only a few animals received a complete
pathological examination. The number of animals surviving >78 weeks and the number
examined for hepatocellular proliferative appeared to differ (42-44 animals examined but
32-35 surviving till the end of the experiment). Only the lowest treatment group had increased
liver tumors which were marginally significantly increased.
Leuschner and Beuscher (1998) examined the carcinogenic effects of CH in male and
female Sprague-Dawley rats (69-79 g, 25-29 days old at initiation of the experiment)
administered 0, 15, 45, and 135 mg/kg CH in unbuffered drinking water 7 days/week
(n = 50/group) for 124 weeks in males and 128 weeks in females. Two control groups were
noted in the methods section without explanation as to why they were conducted as two groups.
The authors report no substance-related influence on organ weights and no macroscopic evidence
of tumors or lesions in male or female rats treated with CH for 124 or 128 weeks. However, no
data are presented on the incidence of tumors in either treatment or control groups. The authors
did report a statistically significant increase in the incidence of hepatocellular hypertrophy in
male rats at the 135 mg/kg dose (14/50 animals vs. 4/50 and 7/50 in Controls I and II). For
female rats, the incidence of hepatocellular hypertrophy was reported to be 10/50 rats (Control I)
and 16/50 (Control II) rats with 18/50, 13/50 and 12/50 female rats having hepatocellular
hypertrophy after 15, 45, and 135 mg/kg CH, respectively. The lack of reporting in regard to
final body weights, histology, and especially background and treatment group data for tumor
incidences, limit the interpretation of this study. Whether this paradigm was sensitive for
induction of liver cancer cannot be determined.
Therefore, given the limitations in the available studies, a comparison of rat liver
carcinogenicity induced by TCE, TCA, DCA, and CH reveals no strong inconsistencies, but nor
does it provide much insight into the relative importance of different TCE metabolites in liver
tumor induction.
4.5.6.1.12. Studies in mice
Similar to TCE, the bioassay data in mice for DCA, TCA, and CH are much more
extensive and have shown that all three compounds induce liver tumors in mice. Several 2-year
bioassays have been reported for CH (Daniel et al., 1992; George et al., 2000; Leakey et al.,
2003a). For many of the DCA and TCA studies, the focus was not carcinogenic dose-response
but rather investigation of the nature of the tumors and potential MO As in relation to TCE. As a
This document is a draft for review purposes only and does not constitute Agency policy.
354 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
result, studies often employed relatively high concentrations of DC A or TCA and/or were
conducted for a year or less. As shown previously in Section 4.5.4.2.1, the dose-response curves
for increased liver weight for TCE administration in male mice are more similar to those for
DCA administration and TCE oxidative metabolism than for direct TCA administration
(inadequate data were available for CH). An analogous comparison for DCA-, TCA-, and
CH-induced tumors would be informative, ideally using data from 2-year studies.
4.5.6.1.12.1. Trichloroethylene (TCE) carcinogenicity dose-response data
Unfortunately, the database for TCE, while consistently showing an induction of liver
tumors in mice, is very limited for making inferences regarding the shape of the dose-response
curve. For many of these experiments multiplicity was not given only liver tumor incidence.
NTP (1990), Bull et al. (2002), Anna et al. (1994) conducted gavage experiments in which they
only tested one dose of-1,000 mg/kg-day TCE. NCI (1976) tested two doses that were adjusted
during exposure to an average of 1,169 and 2,339 mg/kg-day in male mice with only twofold
dose spacing in only two doses tested. Maltoni et al. (1986) conducted inhalation experiments in
two sets of B6C3F1 mice and one set of Swiss mice at three exposure concentrations that were
threefold apart in magnitude between the low and mid-dose and twofold apart in magnitude
between the mid- and high-dose. However, for one experiment in male B6C3F1 mice (BT306),
the mice fought and suffered premature mortality and for two the experiments in B6C3F1 mice,
although using the same strain, the mice were obtained from differing sources with very different
background liver tumor levels. For the Maltoni et al. (1988) study a general descriptor of
"hepatoma" was used for liver neoplasia rather than describing hepatocellular adenomas and
carcinomas so that comparison of that data with those from other experiments is difficult. More
importantly, while the number of adenomas and carcinomas may be the same between treatments
or durations of exposure, the number of adenomas may decrease as the number of carcinomas
increase during the course of tumor progression. Such information is lost by using only a
hepatoma descriptor.
Given the limited database, it would be useful if different studies could be combined to
yield a more comprehensive dose-response curve, as was done for liver weight, above. However,
this is probably not appropriate for several reasons. First, only NTP (1990) was performed with
dosing duration and time of sacrifice both being the "standard" 104 weeks. NCI (1976), Maltoni
et al. (1986), Anna et al. (1994), and Bull et al. (2002) all had shorter dosing periods and either
longer (Maltoni et al., 1986) or shorter (the other three studies) observation times. Therefore,
because of potential dose-rate effects and differences in the degree of expression of TCE-induced
This document is a draft for review purposes only and does not constitute Agency policy.
355 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
tumors, it is difficult to even come up with a comparable administered dose-metric across studies.
Moreover, the background tumor incidences are substantially different across experiments, even
controlling for mouse strain and sex. For example, across gavage studies in male B6C3F1 mice,
the incidence of hepatocellular carcinomas ranged from 1.2-16.7% (Anna et al., 1994; NCI,
1976; NTP, 1990) and the incidence of adenomas ranged from 1.2-14.6% (Anna et al., 1994;
NTP, 1990) in control B6C3F1 mice. After -1,000 mg/kg-day TCE treatment, the incidence of
carcinomas ranged from 19.4-62%) (Anna et al., 1994; Bull et al., 2002; NCI, 1976; NTP, 1990),
with three of the studies (Anna et al., 1994; NCI, 1976; NTP, 1990) reporting a range of
incidences between 42.8-62.0%). The incidence of adenomas ranged from 28-66.7% (Anna et
al., 1994; Bull et al., 2002; NTP, 1990). In the Maltoni et al. (1986) inhalation study as well,
male B6C3F1 mice from two different sources had very different control incidences of hepatomas
(-2% vs. about -20%).
Therefore, only data from the same experiment in which more than a single exposed dose
group was used provide reliable data on the dose-response relationship for TCE
hepatocarcinogenicity, and incidences from these experiments are shown in Figures 4-10 and
4-11. Except for one of the two Maltoni et al. (1986) inhalation experiments in male B6C3F1
mice, all of these data sets show relatively proportional increases with dose, albeit with
somewhat different slopes as may be expected across strains and sexes. Direct comparison is
difficult, since the "hepatomas" reported by Maltoni et al. (1986) are much more heterogeneous,
including neoplastic nodules, adenomas, and carcinomas, than the carcinomas reported by NCI
(1976). Nonetheless, although the data limitations preclude a conclusive statement, these data
are generally consistent with the linear relationship observed with TCE-induced liver weight
changes.
4.5.6.1.12.2. Dichloroacetic acid (DCA) carcinogenicity dose-response data
With respect to DCA, Pereira (1996) reported that for 82 week exposure to DCA in
female B6C3F1 mice, DCA exposure concentrations of 0, 2, 6.67, and 20 mmol/L (0, 0.26, 0.86,
and 2.6 g/L) led to close proportionally increasing adenoma prevalences of 2.2, 6, 25, and 84.2%,
though adenoma multiplicity increased more than linearly between the highest two doses.
Unfortunately, too few carcinomas were observed at these doses and duration to meaningfully
inform the shape of the dose-response relationship. More useful is DeAngelo et al. (1999),
which reported on a study of DCA hepatocarcinogenicity in male B6C3F1 mice over a lifetime
exposure. DeAngelo et al. (1999) used 0.05 g/L, 0.5 g/L, 1.0 g/L, 2.0 g/L and 3.5 g/L exposure
concentrations of DCA in their 100-week dirking water study. The number of animals at final
This document is a draft for review purposes only and does not constitute Agency policy.
356 DRAFT—DO NOT CITE OR QUOTE
-------
1 sacrifice was generally low in the DCA treatment groups and variable. The multiplicity or
2 number of hepatocellular carcinomas/animals was reported to be significantly increased over
3 controls in a dose-related manner at all DCA treatments including 0.05 g/L DCA, and a
4 no-observed-effect level (NOEL) reported not to be observed by the authors. Between the 0.5
5 g/L and 3.5 g/L exposure concentrations of DCA the magnitude of increase in multiplicity was
6 similar to the increases in magnitude in dose. The incidence of hepatocellular carcinomas were
7 reported to be increased at all doses as well but not reported to be statistically significant at the
8 0.05 g/L exposure concentration. However, given that the number of mice examined for this
9 response (n = 33), the power of the experiment at this dose was only 16.9% to be able to
10
11
12
13
14
15
16
17
18
Figure 4-10. Dose-response relationship, expressed as (A) percentage
incidence and (B) fold-increase over controls, for TCE hepatocarcinogenicity
in NCI (1976). For comparison, incidences of carcinomas for NTP (1990), Anna
et al. (1994), and Bull et al. (2002) are included, but without connecting lines
since they are not appropriate for assessing the shape of the dose-response
relationship.
19
This document is a draft for review purposes only and does not constitute Agency policy.
357 DRAFT—DO NOT CITE OR QUOTE
-------
Figure 4-11. Dose-response relationship, expressed as (A) incidence and
(B) fold-increase over controls, for TCE hepatocarcinogenicity in Maltoni
et al. (1986). Note that the BT306 experiment reported excessive mortality due to
fighting, and so the paradigm was repeated in experiment BT306bis using mice
from a different source.
This document is a draft for review purposes only and does not constitute Agency policy.
358 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
determine that there was not a treatment related effect. Indeed, Figure 4-12 replots the data from
DeAngelo et al. (1999) with an abscissa drawn to scale (unlike the figure in the original paper,
which was not to scale), suggests even a slightly greater than linear effect at the lowest dose
(0.05 g/L, or 8 mg/kg-day) as compared to the next lowest dose (0.5 g/L, or 84 mg/kg-day),
though of course the power of such a determination is limited. The authors did not report the
incidence or multiplicity of adenomas for the 0.05 g/L exposure group in the study or the
incidence or multiplicity of adenomas and carcinomas in combination. For the animals surviving
from 79-100 weeks of exposure, the incidence and multiplicity of adenomas peaked at 1 g/L
while hepatocellular carcinomas continued to increase at the higher doses. This would be
expected where some portion of the adenomas would either regress or progress to carcinomas at
the higher doses.
Figure 4-12. Dose-response data for hepatocellular carcinomas (HCC)
(A) incidence and (B) multiplicity, induced by DCA from DeAngelo et al.
(1999). Drinking water concentrations were 0, 0.05, 0.5, 1, 2, and 3.5 g/L, from
which daily average doses were calculated using observed water consumption in the
study.
Associations of DCA carcinogenicity with various noncancer, possibly precursor, effects
was also investigated. Importantly, the doses that induced tumors in DeAngelo et al. (1999)
were reported to not induce widespread cytotoxicity. An attempt was also made to relate
differing exposure levels to subchronic changes and peroxisomal enzyme induction.
Interestingly, DeAngelo et al. (1999) reported that peroxisome proliferation was significantly
increased at 3.5 g/L DCA only at 26 weeks, not correlated with tumor response, and to not be
This document is a draft for review purposes only and does not constitute Agency policy.
359 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
increased at either 0.05 g/L or 0.5 g/L treatments. The authors concluded that DCA-induced
carcinogenesis was not dependent on peroxisome proliferation or chemically sustained
proliferation, as measured by DNA synthesis. Slight hepatomegaly was present by 26 weeks in
the 0.5 g/L group and decreased with time. By contrast, increases in both percentage liver/body
weight and the multiplicity of hepatocellular carcinomas increased proportionally with DCA
exposure concentration after 79-100 weeks of exposure. DeAngelo et al. (1999) presented a
figure comparing the number of hepatocellular carcinomas/animal at 100 weeks compared with
the percentage liver/body weight at 26 weeks that showed a linear correlation (r = 0.9977) while
peroxisome proliferation and DNA synthesis did not correlate with tumor induction profiles.
The proportional increase in liver weight with DCA exposure was also reported for shorter
durations of exposure as noted previously. Therefore, for DCA, both tumor incidence and liver
weight appear to increase proportionally with dose.
4.5.6.1.12.3. Trichloroacetic acid (TCA) carcinogenicity dose-response data
With respect to TCA, Pereira (1996) reported that for 82 week exposure to TCA in
female B6C3F1 mice, TCA exposure concentrations of 0, 2, 6.67, and 20 mmol/L (0, 0.33, 1.1,
and 3.3 g/L) led to increasing incidences and multiplicity of adenomas and of carcinomas (see
Figure 4-13). DeAngelo et al. (2008) reported the results of three experiments exposing male
B6C3F1 mice to neutralized TCA in drinking water (incidences also in Figure 4-13). Rather
than using five exposure levels that were generally twofold apart, as was done in DeAngelo et al.
(1999) for DCA, DeAngelo et al. (2008) studied only three doses of TCA that were an order of
magnitude apart which limits the elucidation of the shape of the dose-response curve. In
addition, the 104-week data, DeAngelo et al. (2008) contained two studies, each conducted in a
separate laboratories—the two lower doses were studied in one study and the highest dose in
another. The first 104-week study was conducted using 2 g/L NaCl, or 0.05, 0.5, or 5 g/L TCA
in drinking water for 60 weeks (Study #1) while the other two were conducted for a period of
104 weeks (Study #2 with 2.5 g/L neutralized acetic acid or 4.5 g/L TCA exposure groups and
Study #3 with deionized water, 0.05 g/L TCA and 0.5 g/L TCA exposure groups). In addition, a
relatively small number of animals were used for the determination of a tumor response (n ~ 30
at final necropsy).
This document is a draft for review purposes only and does not constitute Agency policy.
360 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Figure 4-13. Reported incidences of hepatocellular carcinomas (HCC) and
hepatocellular adenomas plus carcinomas (HCA + HCC) in various studies
in B6C3F1 mice (DeAngelo et al., 2008; Pereira, 1996). Combined
HCA + HCC were not reported in Pereira (1996).
In Study #1, the incidence data for adenomas observed at 60 weeks at 0.05 g/L, 0.5 g/L
and 5.0 g/L TCA were 2.1-, 3.0- and 5.4-fold of control values, with similar fold increases in
multiplicity. As shown by Pereira (1996), 60 weeks does not allow for full tumor expression, so
whether the dose-response relationship is the same at 104 weeks is not certain. For instance,
Pereira (1996) examined the tumor induction in female B6C3F1 mice and demonstrated that foci,
adenoma, and carcinoma development in mice are dependent on duration of exposure (period of
observation in controls). In control female mice a 360- versus 576-day observation period
showed that at 360 days no foci or carcinomas and only 2.5% of animals had adenomas whereas
by 576 days of observation, 11% had foci, 2% adenomas, and 2% had carcinomas. For DCA and
TCA treatments, foci, adenomas, and carcinoma incidence and multiplicity did not reach full
expression until 82 weeks at the three doses employed. Although the numbers of animals were
relatively low and variable at the two highest doses (18-28 mice) there were 50-53 mice studied
at the lowest dose level and 90 animals studied in the control group.
Therefore, the 104-week DeAngelo et al. (2008) data from Studies #2 and #3 would
generally be preferred for elucidating the TCA dose-response relationship. However, Study #2
was only conducted at one dose, and although Study #3 used lower doses, it exhibited
extraordinarily high control incidences of liver tumors. In particular, while the incidence of
adenomas and carcinomas was 12% in Study #2, it was reported to be 64% in Study #3. The
This document is a draft for review purposes only and does not constitute Agency policy.
3 61 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
mice in Study #3 were of very large size (weighing -50 g at 45 weeks) as compared to Study #1,
Study #2, or most other bioassays in general, and the large background rate of tumors reported is
consistent with the body-weight-dependence observed by Leakey et al. (2003b).
To put into context the 64% incidence data for carcinomas and adenomas reported in
DeAngelo et al. (2008) for the control group of Study #3, other studies cited in this review for
male B6C3F1 mice show a much lower incidence in liver tumors with (1) NCI (1976) study of
TCE reporting a colony control level of 6.5% for vehicle and 7.1%> incidence of hepatocellular
carcinomas for untreated male B6C3F1 mice (n = 70-77) at 78 weeks, (2) Herren-Freund et al.
(1987) reporting a 9% incidence of adenomas in control male B6C3F1 mice with a multiplicity
of 0.09 ± 0.06 and no carcinomas (n = 22) at 61 weeks, (3) NTP (1990) reporting an incidence of
14.6% adenomas and 16.6%> carcinomas in male B6C3F1 mice after 103 weeks (n = 48), and
(4) Maltoni et al. (1986) reporting that B6C3F1 male mice from the "NCI source" had a
1.1% incidence of "hepatoma" (carcinomas and adenomas) and those from "Charles River Co."
had a 18.9% incidence of "hepatoma" during the entire lifetime of the mice (n = 90 per group).
The importance of examining an adequate number of control or treated animals before
confidence can be placed in those results in illustrated by Anna et al. (1994) in which at
76 weeks 3/10 control male B6C3F1 mice that were untreated and 2/10 control animals given
corn oil were reported to have adenomas but from 76-134 weeks, 4/32 mice were reported to
have adenomas (multiplicity of 0.13 ± 0.06) and 4/32 mice were reported to have carcinomas
(multiplicity of 0.12 ± 0.06). Thus, the reported combined incidence of carcinomas and
adenomas of 64% reported by DeAngelo et al. (2008) for the control mice of Study # 3, not only
is inconsistent and much higher than those reported in Studies #1 and #2, but also much higher
than reported in a number of other studies of TCE.
Therefore, this large background rate and the increased mortality for these mice limit
their use for determining the nature of the dose-response for TCA liver carcinogenicity. At the
two lowest doses of 0.05 g/L and 0.5 g/L TCA from Study #3, the differences in the incidences
and multiplicities for all tumors were twofold at 104 weeks. However, there was no difference in
any of the tumor results (i.e., adenoma, carcinoma, and combinations of adenoma and carcinoma
incidence and multiplicity) between the 4.5 g/L dose group in Study #2 and the 0.5 g/L dose
group in Study #3 at 104 weeks. By contrast, at 60 weeks of exposure, but within the same study
(Study #1), there was a twofold increase in multiplicity for adenomas, and for adenomas and
carcinomas combined between the 0.5 and 5.0 g/L TCA exposure groups. These results are
consistent with the two highest exposure levels reaching a plateau of response after a long
enough duration of exposure for full expression of the tumors (i.e., -90% of animals having liver
tumors at the 0.5 g/L and 5 g/L exposures). However, whether such a plateau would have been
This document is a draft for review purposes only and does not constitute Agency policy.
3 62 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
observed in mice with a more "normal" body weight, and hence a lower background tumor
burden cannot be determined.
Because of the limitations of different studies, it is difficult to discern whether the liver
tumor dose-response curves of TCA and DCA are different in a way analogous to that for liver
weight (see Figure 4-14). Certainly, it is clear that at the same concentration in drinking water or
estimated applied dose, DCA is more potent than TCA, as DCA induces nearly 100% incidence
of carcinomas at a lower dose than TCA. Therefore, like with liver weight gains, DCA has a
steeper dose-response function than TCA. However, the evidence for a "plateau" in tumor
response at high doses with TCA, as was observed for liver weight, is equivocal, as it is
confounded by the highly varying background tumor rates and the limitations of the available
study paradigms.
100%
90%
80%
g 70%
9 30%
200 400
mg/kg-d
600
DeAngelo et al. (2008) (TCA Study #2)
DeAngelo et al. (2008) (TCA Study #3)|
DeAngelo et al. (1999) (DCA)
Figure 4-14. Reported incidence of hepatocellular carcinomas induced by
DCA and TCA in 104-week studies (DeAngelo et al., 2008; DeAngelo et al.,
1999). Only carcinomas were reported in DeAngelo et al. (1999), so combined
adenomas and carcinomas could not be compared.
DeAngelo et al. (2008) attempt to identify a NOEL for turnorigenicity using tumor
multiplicity data and estimated TCA dose. However, it is not an appropriate descriptor for these
data, especially given that "statistical significance" of the tumor response is the determinant used
by the authors to support the conclusions regarding a dose in which there is no TCA-induced
effect. Due to issues related to the appropriateness of use of the concurrent control in Study #3,
This document is a draft for review purposes only and does not constitute Agency policy.
363 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
only the 60-week experiment (i.e., Study # 1) is useful for the determination of tumor
dose-response. Not only is there not allowance for full expression of a tumor response at the
60-week time point but a power calculation of the 60-week study shows that the Type II error,
which should be >50% and thus, greater than the chances of "flipping a coin," was 41 and 71%
for incidence and 7 and 15% for multiplicity of adenomas for the 0.05 and 0.5 g/L TCA exposure
groups. For the combination of adenomas and carcinomas, the power calculation was 8 and 92%
for incidence and 6 and 56% for multiplicity at 0.05 and 0.5 g/L TCA exposure. Therefore, the
designed experiment could accept a false null hypothesis, especially in terms of tumor
multiplicity, at the lower exposure doses and erroneously conclude that there is no response due
to TCA treatment.
In terms of correlations with other noncancer, possibly precursor effects, DeAngelo et al.
(2008) also reported that PCO activity, which varied 2.7-fold as baseline controls, was 1.3-, 2.4-,
and 5.3-fold of control for the 0.05, 0.5, and 5 g/L TCA exposure groups in Study #1 at 4 weeks
was for adenomas incidence 2.1-, 3.0-, and 5.4-fold of control and not similar at the lowest dose
level at 60 weeks. However, it is not clear whether the similarly between PCO and
carcinogenicity at 60 weeks would persist for tumor incidence at 104 weeks. DeAngelo et al.
(2008) report a regression analyses that compare "percent of hepatocellular neoplasia," indicated
by tumor multiplicity, with TCA dose, represented by estimations of the TCA dose in
mg/kg-day, and with PCO activity for the 60-week and 104-week data. Whether adenomas and
carcinomas combined or individual tumor type were used in these analysis was not reported by
the authors. However, it would be preferable to compare "precursor" levels of PCO at earlier
time points, rather than at a time when there was already a significant tumor response. In
addition, linear regression analyses of these data are difficult to interpret because of the wide
dose spacing of these experiments. In such a situation, for a linear regression, control and 5 g/L
exposure levels will basically determine the shape of the dose-response curve since the 0.05 g/L
and 0.5 g/L exposure levels are so close to the control (zero) value. Thus, dose-response appears
to be linear between control and the 5.0 g/L value with the two lowest doses not affectively
changing the slope of the line (i.e., "leveraging" the regression). Moreover, at the 5 g/L dose
level, there is potential for effects due to palatability, as reported in one study in which drinking
water consumption declined at this concentration (DeAngelo et al., 2008). Thus, the value of
these analyses is limited by (1) use of data from Study # 3 in a tumor prone mouse that is not
comparable to those used in Studies #1 and #2, (2) the appropriateness of using PCO values from
later time points and the variability in PCO control values, (3) the uncertainty of the effects of
palatability on the 5 g/L TCA results which were reported in one study to reduce drinking water
consumption, and (4) the dose-spacing of the experiment.
This document is a draft for review purposes only and does not constitute Agency policy.
3 64 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
4.5.6.1.12.4. Chloral hydrate (CH) carcinogenic dose-response
Although a much more limited database in rodents than for TCA or DCA, there is
evidence that chloral hydrate is also a rodent liver hepatocarcinogen (see also Section E.2.5 and
Caldwell and Keshava [2006]).
Daniel et al. (1992) exposed adult male B6C3F1 28-day-old mice to 1 g/L CH in drinking
water for 30 and 60 weeks (n = 5 for interim sacrifice) and for 104 weeks (n = 40). The
concentration of CH was 1 g/L and estimated to provide a 166-mg/kg-day dose. It is not clear
from the report what control group better matched the CH group, as the mean initial body
weights of the groups as well as the number of animals varied considerably in each group (i.e.,
-40% difference in mean body weights at the beginning of the study). Liver tumors were
increased by CH treatment. The percentage incidence of liver carcinomas and adenomas in the
surviving animals was 15% in control and 71% in CH-treated mice and the incidence of
hepatocellular carcinoma reported to be 46% in the CH-treated group. The number of
tumors/animals was also significantly increased with CH treatment. However, because this was
a single dose study, a comparison with the dose-response relationship with TCE, TCA, or DCA
is not feasible.
George et al. (2000) exposed male B6C3F1 mice to CH in drinking water for 2 years.
Groups of animals were sacrificed at 26, 52, and 78 weeks following the initiation of dosing,
with terminal sacrifices at Week 104. Only a few animals received a complete pathological
examination. Preneoplastic foci and adenomas were reported to be increased in the livers of all
CH treatment groups at 104 weeks. The percentage incidence of hepatocellular adenomas was
reported to be 21.4, 43.5, 51.3, and 50% in control, 13.5, 65.0 and 146.6 mg/kg-day CH
treatment groups, respectively. The percentage incidence of hepatocellular carcinomas was
reported to be 54.8, 54.3, 59.0 and 84.4% in these same groups. The resulting percentage
incidence of hepatocellular adenomas and carcinomas was reported to be 64.3, 78.3, 79.5 and
90.6%. Of concern is the reporting of a 64% incidence of hepatocellular carcinomas and
adenomas in the control group of mice for this experiment, which is the same as that for another
study published by this same laboratory (DeAngelo et al., 2008). DeAngelo et al. (2008) did not
identify them as being contemporaneous studies or sharing controls, but a comparison of the
control data published by DeAngelo et al. (2008) for TCA and that published by George et al.
(2000) for the CH studies shows them to be the same data set. Therefore, as discussed above,
this data set was derived from B6C3F1 mice that were large (-50 g) and resultantly tumor prone,
making determinations of the dose-response of CH from this experiment difficult. Therefore, for
the purposes of comparison of dose-response relationships, this study has the same limitations as
the DeAngelo et al. (2008) study, discussed above.
This document is a draft for review purposes only and does not constitute Agency policy.
365 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Leakey et al. (2003a) studied the effects of CH exposure (0, 25, 50, and 100 mg/kg-day,
5 days/week, 104-105 weeks via gavage) in male B6C3F1 mice with dietary control used to
manipulate body growth (n = 48 for 2-year study and n = 12 for the 15-month interim study).
Dietary control was reported to decrease background liver tumor rates (decreased by 15-20%)
and was reported to be associated with decreased variation in liver-to-body weight ratios, thereby
potentially increasing assay sensitivity. In dietary-controlled groups and groups fed ad libitum,
liver adenomas and carcinomas (combined) were reported to be increased with CH treatment.
With dietary restriction there was a more discernable CH tumor-response with overall tumor
incidence reduced, and time-to-tumor increased by dietary control in comparison to ad libitum
fed mice. Incidences of hepatocellular adenoma and carcinoma overall rates were reported to be
33, 52, 49, and 46% for control, 25, 50, and 100 mg/kg adlibitum-fed mice, respectively. For
dietary controlled mice the incidence rates were reported to be 22.9, 22.9, 29.2, and 37.5% for
controls, 25, 50, and 100 mg/kg CH, respectively. Body weights were matched and carefully
controlled in this study. These data are shown in Figure 4-15, relative to control incidences. It is
evident from these data that dietary control significantly changes the apparent shape of the
dose-response curve, presumably by reducing variability between animals. While the ad libitum
dose groups had an apparent "saturation" of response, this was not evident with the dietary
controlled group. Of note all the other bioassays for TCE, TCA, DCA, and CH were in ad
libitum fed mice. Therefore, it is difficult to compare the dose-response curves for CH-treated
mice on dietary restriction to those fed ad libitum. However, the rationale for dietary restriction
in the B6C3F1 mouse is to prevent the types of weight gain and corresponding high background
tumor levels observed in DeAngelo et al. (2008) and George et al. (2000). As stated previously,
most other studies of TCA, DCA, and TCE had background levels that, while varied, were lower
than the ad libitum fed mice studied in Leakey et al. (2003a).
Of note is that incidences of adenomas and carcinomas combined do not show
differences in tumor progression as carcinomas may increase and adenomas may regress. Liver
weight increases at 15-months did not correlate with 2-year tumor incidences in the ad libitum
group, but a consistent dose-response shape between these two measures is evident in the dietary
controlled group. However, of note is the reporting of liver weight at 15 months is for a time
period in which foci and liver tumors have been reported to have already occurred in other
studies, so hepatomegaly in the absence of these changes is hard to detect.
In terms of other noncancer effects that may be associated with tumor induction, it is
notable that while dietary restriction reduced the overall level of CH-mediated tumor induction,
it led to greater CH-mediated induction of peroxisome proliferation-associated enzymes.
Moreover, between control groups, dietary restricted mice appeared to have higher levels of
This document is a draft for review purposes only and does not constitute Agency policy.
366 DRAFT—DO NOT CITE OR QUOTE
-------
60%
50%
0)
g 40%
;o
c 30%
<
? 20%
o
X
10%
0%
0
20
40
60
80
100
CH mg/kg-d
ad libitum -H- dietary control
This document is a draft for review purposes only and does not constitute Agency policy.
367 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Figure 4-15. Effects of dietary control on the dose-response curves for
changes in liver tumor incidences induced by CH in diet (Leakey et al.,
2003a).
lauric acid co-hydrolase activity than ad libilum-fed mice. Seng et al. (2003) report that lauric
acid P-hydroxylase and PCO were induced only at exposure levels >100 mg/kg CH, again with
dietary restricted groups showing the greatest induction. Such data argue against the role of
peroxisome proliferation in CH-liver tumor induction in mice.
Leakey et al. (2003a) gave no descriptions of liver pathology were given other than
incidence of mice with fatty liver changes. Hepatic malondialdehyde concentration in ad libitum
fed and dietary controlled mice did not change with CH exposure at 15 months but the dietary
controlled groups were all approximately half that of the ad libitum-fed mice. Thus, while
overall increased tumors observed in the ad libitum diet correlated with increased
malondialdehyde concentration, there was no association between CH dose and malondialdehyde
induction for either diet.
Overall, from the CH studies in mice, there is an apparent increase in liver adenomas and
carcinomas induced by CH treatment by either drinking water or gavage with all available
studies performed in male B6C3F1 mice. However, the background levels of hepatocellular
adenomas and carcinomas in these mice in George et al. (2000) and body-weight data from this
study are high, consistent with the association between large body weight and background tumor
susceptibility shown with dietary control (Leakey et al., 2003a). With dietary control, Leakey
et al. (2003a) report a dose-response relationship between exposure and tumor incidence that is
proportional to dose.
4.5.6.1.12.5. Degree of concordance among trichloroethylene (TCE), trichloroacetic acid
(TCA), dichloroacetic acid (DCA), and chloral hydrate (CH) dose-response relationships
A quantitative comparison of the carcinogenicity dose-response relationships among
TCE, TCA, DCA, and CH—analogous to the quantitative comparison between TCE and TCA
hepatomegaly—was considered. This first step in such a comparison would an examination of
the dose-response data for TCE alone to see if they are consistent with a single dose-response
relationship. As shown in Figure 4-10 and 4-11, there is substantial variability among the
available liver tumor dose-response data that was not observed for hepatomegaly. The strain of
mice used in the bioassays not only had a difference in TCE liver tumor response but also a
difference in background liver tumor incidence. Differences in exposure paradigms in the
This document is a draft for review purposes only and does not constitute Agency policy.
368 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
bioassays also leads to difference in tumor incidence and reporting. In addition, unlike the case
with TCE hepatomegaly data in mice, the TCE dose-response data for liver tumors in mice
exposed via inhalation and gavage are not consistent with a common dose-response curve even
on an internal dose basis (e.g., Rhomberg, 2000; Section 5.2). This heterogeneity is also evident
for the TCA dose-response data, as shown in Figure 4-13, which may in part be due to the
differences in study duration. Furthermore, among all the available cancer bioassay data for
TCE, TCA, DCA, and CH, the control incidences for background liver tumors vary from about
1% to over 50%, and difference of more than 50-fold that adds substantial uncertainty to any
joint analysis. Therefore, differences within and across the databases of these compounds, such
as the comparability of study durations, control tumor incidences, and carcinogenic potency,
preclude either a quantitative analysis or a definitive conclusion. This question may be better
addressed experimentally where similar animals are exposed to different compounds in the same
experimental setting.
4.5.6.1.13. Inferences from liver tumor phenotype and genotype
A number of studies have investigation tumor phenotypes, such as c-Jun staining,
tincture, and dysplacity, or genotypes, such as H-ras mutations, to inform both the identification
of the active agents of TCE liver tumor induction as well as what MOA(s) may be involved.
4.5.6.1.13 .1. Tumor phenotype—staining and appearance
The descriptions of tumors in mice reported by the NCI, NTP, and Maltoni et al studies
are also consistent with phenotypic heterogeneity as well as spontaneous tumor morphology (see
Section E.3.4.1.5). As noted in Section E.3.1, hepatocellular carcinomas observed in humans are
also heterogeneous. For mice, Maltoni et al. (1986) described malignant tumors of hepatic cells
to be of different subhistotypes, and of various degrees of malignancy and were reported to be
unique or multiple, and have different sizes (usually detected grossly at necropsy) from TCE
exposure. In regard to phenotype, tumors were described as usual type observed in Swiss and
B6C3F1 mice, as well as in other mouse strains, either untreated or treated with
hepatocarcinogens and to frequently have medullary (solid), trabecular, and pleomorphic
(usually anaplastic) patterns. For the NCI (1976) study, the mouse liver tumors were described
in detail and to be heterogeneous "as described in the literature" and similar in appearance to
tumors generated by carbon tetrachloride. The description of liver tumors in this study and
tendency to metastasize to the lung are similar to descriptions provided by Maltoni et al. (1986)
This document is a draft for review purposes only and does not constitute Agency policy.
369 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
for TCE-induced liver tumors in mice via inhalation exposure. The NTP (1990) study reported
TCE exposure to be associated with increased incidence of hepatocellular carcinoma (tumors
with markedly abnormal cytology and architecture) in male and female mice. Hepatocellular
adenomas were described as circumscribed areas of distinctive hepatic parenchymal cells with a
perimeter of normal appearing parenchyma in which there were areas that appeared to be
undergoing compression from expansion of the tumor. Mitotic figures were sparse or absent but
the tumors lacked typical lobular organization. Hepatocellular carcinomas were reported to have
markedly abnormal cytology and architecture with abnormalities in cytology cited as including
increased cell size, decreased cell size, cytoplasmic eosinophilia, cytoplasmic basophilia,
cytoplasmic vacuolization, cytoplasmic hyaline bodies and variations in nuclear appearance.
Furthermore, in many instance several or all of the abnormalities were reported to be present in
different areas of the tumor and variations in architecture with some of the hepatocellular
carcinomas having areas of trabecular organization. Mitosis was variable in amount and
location. Therefore, the phenotype of tumors reported from TCE exposure was heterogeneous in
appearance between and within tumors from all three of these studies.
Caldwell and Keshava (2006) report "that Bannasch (2001) and Bannasch et al. (2001)
describe the early phenotypes of preneoplastic foci induced by many oncogenic agents
(DNA-reactive chemicals, radiation, viruses, transgenic oncogenes and local hyperinsulinism) as
insulinomimetic. These foci and tumors have been described by tincture (after hematoxylin and
eosin staining of structural contents) as primarily eosinophilic (pink, reflecting eosin staining,
e.g., staining of intracellular and extracellular protein), basophilic (blue, reflecting hematoxylin
staining, e.g., staining of ribosomes and arginine rich basic nucleorprotein such as histones), and
to be heterogeneous. Primary eosin staining is associated with a less malignant state of the
hepatocyte with increased ribosomal content, decreased glycogen content, and increased
basophilia of the cytoplasm by hematoxylin staining to be indicative of a more malignant state or
tumor progression (Bannasch, 2001; Carter et al., 2003). Several studies do identify foci and
tumors as primarily eosinophilic or basophilic, but do not give specific criteria for how a foci or
tumor (which can be and usually is made up of a mixture of phenotypically heterogenous cells)
are assigned to be one category or another. Caldwell and Keshava (2006) noted that the tumors
observed after TCE exposure are consistent with the description for the main tumor lines of
development described by Bannasch et al. (2001) (see Section 3.4.1.5). Thus, the response of
liver to DCA (glycogenesis with emergence of glycogen poor tumors) is similar to the
progression of preneoplastic foci to tumors induced from a variety of agents and conditions
associated with increased cancer risk." Furthermore Caldwell and Keshava (2006) note that Bull
et al. (2002) report expression of insulin receptor to be elevated in tumors of control mice or
This document is a draft for review purposes only and does not constitute Agency policy.
370 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
mice treated with TCE, TCA and DCA but not in nontumor areas suggesting that this effect is
not specific to DCA.
There is a body of literature that has focused on the effects of TCE and its metabolites
after rats or mice have been exposed to "mutagenic" agents to "initiate" hepatocarcinogenesis
and this is discussed in Section E.4.2. TCE and its metabolites were reported to affect tumor
incidence, multiplicity, and phenotype when given to mice as a coexposure with a variety of
"initiating" agents and with other carcinogens. Pereira and Phelps (1996) reported that methyl
nitrosourea (MNU) alone induced basophilic foci and adenomas. MNU and low concentrations
of DCA or TCA in female mice were reported to induce heterogeneous for foci and tumor with a
higher concentration of DCA inducing more eosinophilic and a higher concentration of TCA
inducing more tumors that were basophilic. Pereira et al. (2001) reported that not only dose, but
gender also affected phenotype in mice that had already been exposed to MNU and were then
exposed to DCA. As for other phenotypic markers, Lantendresse and Pereira (1997) reported
that exposure to MNU and TCA or DCA induced tumors that had some commonalities, were
heterogeneous, but for female mice were overall different between DCA and TCA as
coexposures with MNU.
With regard to the phenotype of TCA and DCA-induced tumors, Stauber and Bull (1997)
reported the for male B6C3F1 mice, DCA-induced "lesions" contained a number of smaller
lesions that were heterogeneous and more eosinophilic with larger "lesions" tending to less
numerous and more basophilic. For TCA results using this paradigm, the "lesions" were
reported to be less numerous, more basophilic, and larger than those induced by DCA. Carter
et al. (2003) used tissues from the DeAngelo et al. (1999) and examined the heterogeneity of the
DCA-induced lesions and the type and phenotype of preneoplastic and neoplastic lesions pooled
across all time points. Carter et al. (2003) examined the phenotype of liver tumors induced by
DCA in male B6C3F1 mice and the shape of the dose-response curve for insight into its MOA.
They reported a dose-response of histopathologic changes (all classes of premalignant lesions
and carcinomas) occurring in the livers of mice from 0.05-3.5 g/L DCA for 26-100 weeks and
suggest foci and adenomas demonstrated neoplastic progression with time at lower doses than
observed DCA genotoxicity. Preneoplastic lesions were identified as eosinophilic, basophilic
and/or clear cell (grouped with clear cell and mixed cell) and dysplastic. Altered foci were
50% eosinophilic with about 30% basophilic. As foci became larger and evolved into
carcinomas they became increasingly basophilic. The pattern held true throughout the exposure
range. There was also a dose and length of exposure related increase in atypical nuclei in
"noninvolved" liver. Glycogen deposition was also reported to be dose-dependent with
periportal accumulation at the 0.5 g/L exposure level. Carter et al. (2003) suggested that size and
This document is a draft for review purposes only and does not constitute Agency policy.
3 71 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
evolution into a more malignant state are associated with increasing basophilia, a conclusion
consistent with those of Bannasch (1996) and that there a greater periportal location of lesions
suggestive as the location from which they arose. Consistent with the results of DeAngelo et al.
(1999), Carter et al. (2003) reported that DCA (0.05-3.5 g/L) increased the number of lesions
per animal relative to animals receiving distilled water, shortened the time to development of all
classes of hepatic lesions, and that the phenotype of the lesions were similar to those
spontaneously arising in controls. Along with basophilic and eosinophilic lesions or foci,
Carter et al. (2003) concluded that DCA-induced tumors also arose from isolated, highly
dysplastic hepatocytes in male B6C3F1 mice chronically exposed to DCA suggesting another
direct neoplastic conversion pathway other than through eosinophilic or basophilic foci.
Rather than male B6C3F1 mice, Pereira (1996) studied the dose-response relationship for
the carcinogenic activity of DCA and TCA and characterized their lesions (foci, adenomas and
carcinomas) by tincture in females (the generally less sensitive gender). Like the studies of TCE
by Maltoni et al. (1986), female mice were also reported to have increased liver tumors after
TCA and DCA exposures. Pereira (1996) pool lesions were pooled for phenotype analysis so the
affect of duration of exposure could not be determined nor adenomas separated from carcinomas
for "tumors." However, as the concentration of DCA was decreased the number of foci was
reported by Pereira (1996) to be decreased but the phenotype of the foci to go from primarily
eosinophilic foci (i.e., -95% eosinophilic at 2.58 g/L DCA) to basophilic foci (-57%
eosinophilic at 0.26 g/L). For TCA the number of foci was reported to -40 basophilic and
-60 eosinophilic regardless of dose. Spontaneously occurring foci were more basophilic by a
ratio of 7/3. Pereira (1996) described the foci of altered hepatocytes and tumors induced by
DCA in female B6C3F1 mice to be eosinophilic at higher exposure levels but at lower or
intermittent exposures to be half eosinophilic and half basophilic. Regardless of exposure level,
half of the TCA-induced foci were reported to be half eosinophilic and half basophilic with
tumors 75%) basophilic. In control female mice, the limited numbers of lesions were mostly
basophilic, with most of the rest being eosinophilic with the exception of a few mixed tumors.
The limitations of descriptions tincture and especially for inferences regarding peroxisome
proliferator from the description of "basophilia" is discussed in Section E.3.4.1.5.
Thus, the results appear to differ between male and female B6C3F1 mice in regard to
tincture for DCA and TCA at differing doses. What is apparent is that the tincture of the lesions
is dependent on the stage of tumor progression, agent (DCA or TCA), gender, and dose. Also
what is apparent from these studies is the both DCA and TCA are heterogeneous in their tinctoral
characteristics.
This document is a draft for review purposes only and does not constitute Agency policy.
3 72 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Overall, tumors induced by TCA, DCA, CH, and TCE are all heterogeneous in their
physical and tinctural characteristics in a manner this not markedly distinguishable from
spontaneous lesions or those induced by a wide variety of chemical carcinogens. For instance,
Daniel et al. (1992), which studies DCA and CH carcinogenicity (discussed above) noted that
morphologically, there did not appear to be any discernable differences in the visual appearance
of the DCA- and CH-induced tumors. Therefore, these data do not provide strong insights into
elucidating the active agent(s) for TCE hepatocarcinogenicity or their MOA(s).
4.5.6.1.13.2. C-Jun staining
Stauber and Bull (1997) reported that in male B6C3F1 mice, the oncoproteins c-Jun and
c-Fos were expressed in liver tumors induced by DCA but not those induced by TCA. Although
Bull et al. (2004) have suggested that the negative expression of c-Jun in TCA-induced tumors
may be consistent with a characteristic phenotype shown in general by peroxisome proliferators
as a class, as pointed out by Caldwell and Keshava (2006), there is no supporting evidence of
this. Nonetheless, the observation that TCA and DCA have different levels of oncogene
expression led to a number of follow-up studies by this group. No data on oncoprotein
immunostaining are available for CH.
Stauber et al. (1998) studied induction of "transformed" hepatocytes by DCA and TCE
treatment in vitro, including an examination of c-Jun staining. Stauber et al. (1998) isolated
primary hepatocytes from 5-8 week old male B6C3F1 mice (n = 3) and subsequently cultured
them in the presence of DCA or TCA. In a separate experiment 0.5 g/L DCA was given to mice
as pretreatment for 2 weeks prior to isolation. The authors assumed that the
anchorage-independent growth of these hepatocytes was an indication of an "initiated cell."
After 10 days in culture with DCA or TCA (0, 0.2, 0.5 and 2.0 mM), concentrations of 0.5 mM
or more DCA and TCA both induced an increase in the number of colonies that was statistically
significant, with DCA showing dose-dependence as well as slightly greater overall increases than
TCA. In a time course experiment the number of colonies from DCA treatment in vitro peaked
by 10 days and did not change through Days 15-25 at the highest dose and, at lower
concentrations of DCA, increased time in culture induced similar peak levels of colony
formation by Days 20-25 as that reached by 10 days at the higher dose. Therefore, the number
of colonies formed was independent of dose if the cells were treated long enough in vitro.
However, not only did treatment with DCA or TCA induce anchorage independent growth but
untreated hepatocytes also formed larger numbers of colonies with time, although at a lower rate
than those treated with DCA. The level reached by untreated cells in tissue culture at 20 days
This document is a draft for review purposes only and does not constitute Agency policy.
373 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
was similar to the level induced by 10 days of exposure to 0.5 mM DCA. The time course of
TCA exposure was not tested to see if it had a similar effect with time as did DCA. The colonies
observed at 10 days were tested for c-Jun expression with the authors noting that "colonies
promoted by DCA were primarily c-Jun positive in contrast to TCA promoted colonies that were
predominantly c-Jun negative." Of the colonies that arose spontaneously from tissue culture
conditions, 10/13 (76.9%) were reported to be c-Jun +, those treated with DCA 28/34 (82.3%)
were c-Jun +, and those treated with TCA 5/22 {22.1%) were c-Jun +. Thus, these data show
heterogeneity in cell in colonies but with more that were c-Jun + colonies occurring by tissue
culture conditions alone than in the presence of DCA, rather than in the presence of TCA.
Bull et al. (2002) administered TCE, TCA, DCA, and combinations of TCA and DCA to
male B6C3F1 mice by daily gavage (TCE) or drinking water (TCA, DCA, and TCA + DCA) for
52-79 weeks, in order to compare a number of tumor characteristics, including c-Jun expression,
across these different exposures. Bull et al. (2002) reported lesion reactivity to c-Jun antibody to
be dependent on the proportion of the DCA and TCA administered after 52 weeks of exposure.
Given alone, DCA was reported to produce lesions in mouse liver for which approximately half
displayed a diffuse immunoreactivity to a c-Jun antibody, half did not, and none exhibited a
mixture of the two. After TCA exposure alone, no lesions were reported to be stained with this
antibody. When given in various combinations, DCA and TCA coexposure induced a few
lesions that were only c-Jun+, many that were only c-Jun-, and a number with a mixed
phenotype whose frequency increased with the dose of DCA. For TCE exposure of 79 weeks,
TCE-induced lesions were reported to also have a mixture of phenotypes (42% c-Jun+, 34%
c-Jun-, and 24% mixed) and to be most consistent with those resulting from DCA and TCA
coexposure but not either metabolite alone.
A number of the limitations of the experiment are discussed in Caldwell et al. (2008).
Specifically, for the DCA and TCA exposed animals, the experiment was limited by low
statistical power, a relatively short duration of exposure, and uncertainty in reports of lesion
prevalence and multiplicity due to inappropriate lesions grouping (i.e., grouping of hyperplastic
nodules, adenomas, and carcinomas together as "tumors"), and incomplete histopathology
determinations (i.e., random selection of gross lesions for histopathology examination). For
determinations of immunoreactivity to c-Jun, Bull et al. (2002) combined hyperplastic nodules,
adenomas, and carcinomas in most of their treatment groups, so differences in c-Jun expression
across differing types of lesions were not discernable.
Nonetheless, these data collectively strongly suggest that TCA is not the sole agent of
TCE-induced mouse liver tumors. In particular, TCE-induced tumors that were, in order of
frequency, c-Jun+, c-Jun-, and of mixed phenotype, while c-Jun+ tumors have never been
This document is a draft for review purposes only and does not constitute Agency policy.
3 74 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
observed with TCA treatment. Nor do these data support DCA as the sole contributor, since
mixed phenotypes were not observed with DCA treatment.
4.5.6.1.13.3. Tumor genotype: H-ras mutation frequency and spectrum
An approach to determine the potential MOAs of DCA and TCA through examination of
the types of tumors each "induced" or "selected" was to examine H-ras activation (Anna et al.,
1994; Bull et al., 2002; Ferreira-Gonzalez et al., 1995; Nelson et al., 1989). No data of this type
were available for CH. This approach has also been used to try to establish an H-ras activation
pattern for "genotoxic" and "nongenotoxic" liver carcinogens compounds and to make
inferences concerning peroxisome proliferator-induced liver tumors. However, as noted by
Stanley et al. (1994), the genetic background of the mice used and the dose of carcinogen may
affect the number of activated H-ras containing tumors which develop. In addition, the stage of
progression of "lesions" (i.e., foci vs. adenomas vs. carcinomas) also has been linked the
observance of H-ras mutations. Fox et al. (1990) note that tumors induced by phenobarbital
(0.05% drinking water [H2O], 1 year), chloroform (200 mg/kg corn oil gavage, two times weekly
for 1 year) or ciprofibrate (0.0125% diet, 2 years) had a much lower frequency of H-ras gene
activation than those that arose spontaneously (2-year bioassays of control animals) or induced
with the "genotoxic" carcinogen benzidine-2 hydrochloric acid (HC1) (120 ppm, drinking H20,
1 year) in mice. In that study, the term "tumor" was not specifically defined but a correlation
between the incidence of H-ras gene activation and development of either a hepatocellular
adenoma or hepatocellular carcinoma was reported to be made with no statistically significant
difference between the frequency of H-ras gene activation in the hepatocellular adenomas and
carcinomas. Histopathological examination of the spontaneous tumors, tumors induced with
benzidine-2 HC1, Phenobarbital, and chloroform was not reported to reveal any significant
changes in morphology or staining characteristics. Spontaneous tumors were reported to have
64% point mutation in codon 61 (n = 50 tumors examined) with a similar response for Benzidine
of 59%) (n = 22 tumors examined), whereas for Phenobarbital the mutation rate was
7%> (n= 15 tumors examined), chloroform 21%> (n = 24 tumors examined) and ciprofibrate
21%) (n = 39 tumors examined). The ciprofibrate-induced tumors were reported to be more
eosinophilic as were the surrounding normal hepatocytes.
Hegi et al. (1993) tested ciprofibrate-induced tumors in the NIH3T3 cotransfection-nude
mouse tumorigenicity assay, which the authors state is capable of detecting a variety of activated
protooncogenes. The tumors examined (ciprofibrate-induced or spontaneously arising) were
taken from the Fox et al. study (1990), screened previously, and found to be negative for H-ras
This document is a draft for review purposes only and does not constitute Agency policy.
375 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
activation. With the limited number of samples examined, Hegi et al concluded that ras
protooncogene activation or activation of other protooncogenes using the nude mouse assay were
not frequent events in ciprofibrate-induced tumors and that spontaneous tumors were not
promoted with it. Using the more sensitive methods, the H-ras activation rate was reported to be
raised from 21-31% for ciprofibrate-induced tumors and from 64-66% for spontaneous tumors.
Stanley et al. (1994) studied the effect of methylclofenapate (MCP) (25 mg/kg for up to 2 years),
a peroxisome proliferator, in B6C3F1 (relatively sensitive) and C57BL/10J (relatively resistant)
mice for H-ras codon 61 point mutations in MCP-induced liver tumors (hepatocellular adenomas
and carcinomas). In the B6C3F1 mice the number of tumors with codon 61 mutations was 11/46
and for C57BL/10J mice 4/31. Unlike the findings of Fox et al. (1990), Stanley et al. (1994)
reported an increase in the frequency of mutation in carcinomas, which was reported to be twice
that of adenomas in both strains of mice, indicating that stage of progression was related to the
number of mutations in those tumors, although most tumors induced by MCP did not have this
mutation.
Anna et al. (1994) reported that the H-ras codon 61 mutation frequency was not
statistically different in liver tumors from DCA and TCE-treated mice from a highly variable
number of tumors examined. From their concurrent controls, they reported that H-ras codon
61 mutations in 17% (n = 6) of adenomas and 100% (n = 5) of carcinomas. For historical
controls (published and unpublished), they reported mutations in 73% (// = 33) of adenomas and
mutations in 70% (n = 30) of carcinomas. For tumors from TCE-treated animals, they reported
mutations in 35% (// = 40) of adenomas and 69% (n = 36) of carcinomas, while for DCA-treated
animals, they reported mutations in 54% (// = 24) of adenomas and in 68% (n = 40) of
carcinomas. Anna et al. (1994) reported more mutations in TCE-induced carcinomas than
adenomas. In regard to mutation spectra in H-ras oncogenes in control or spontaneous tumors,
the patterns were slightly different but those from TCE treatment were mostly similar to that of
DCA-induced tumors (0.5% in drinking water).
The study of Ferreira -Gonzalez (1995) in male B6C3F1 mice has the advantage of
comparison of tumor phenotype at the same stage of progression (hepatocellular carcinoma), for
allowance of the full expression of a tumor response (i.e., 104 weeks), and an adequate number
of spontaneous control lesions for comparison with DCA or TCA treatments. However, tumor
phenotype at an end stage of tumor progression may not be indicative of earlier stages of the
disease process. In spontaneous liver carcinomas, 58% were reported to show mutations in H-61
as compared with 50% of tumor from 3.5 g/L DCA-treated mice and 45% of tumors from
4.5 g/L TCA-treated mice. A number of peroxisome proliferators have been reported to have a
much smaller mutation frequency that spontaneous tumors (e.g., 13-24%) H-ras codon 61
This document is a draft for review purposes only and does not constitute Agency policy.
376 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
mutations after methylclofenopate depending on mouser strain, Stanely et al. [1994]: 21-31% for
ciprofibrate-induced tumors and from 64-66% for spontaneous tumors, Fox et al. [1990] and
Hegi et al [1993]). Thus, there was a heterogeneous response for this phenotypic marker for the
spontaneous, DCA-, and TCA- treatment induced hepatocellular carcinomas had similar patterns
H-ras mutations that differed from the reduced H-ras mutation frequencies reported for a number
of peroxisome proliferators.
In his review, Bull (2000) suggested "the report by Anna et al. (1994) indicated that
TCE-induced tumors possessed a different mutation spectra in codon 61 of the H-ras oncogene
than those observed in spontaneous tumors of control mice." Bull (2000) stated that "results of
this type have been interpreted as suggesting that a chemical is acting by a mutagenic
mechanism" but went on to suggest that it is not possible to a priori rule out a role for selection
in this process and that differences in mutation frequency and spectra in this gene provide some
insight into the relative contribution of different metabolites to TCE-induced liver tumors. Bull
(2000) noted that data from Anna et al. (1994), Ferreira-Gonzalez et al. (1995), and Maronpot
et al. (1995b) indicated that mutation frequency in DCA-induced tumors did not differ
significantly from that observed in spontaneous tumors. Bull (2000) also noted that the mutation
spectra found in DCA-induced tumors has a striking similarity to that observed in TCE-induced
tumors, and DCA-induced tumors were significantly different than that of TCA-induced liver
tumors.
Bull et al. (2002) reported that mutation frequency spectra for the H-ras codon 61 in
mouse liver "tumors" induced by TCE (n = 37 tumors examined) were reported to be
significantly different than that for TCA (n = 41 tumors examined), with DCA-treated mice
tumors giving an intermediate result (n = 64 tumors examined). In this experiment,
TCA-induced "tumors" were reported to have more mutations in codon 61 (44%) than those
from TCE (21%) and DCA (33%). This frequency of mutation in the H-ras codon 61 for TCA is
the opposite pattern as that observed for a number of peroxisome proliferators in which the
number of mutations at H-ras codon 61 in tumors has been reported to be much lower than
spontaneously arising tumors (see above). Bull et al. (2002) noted that the mutation frequency
for all TCE, TCA or DCA tumors was lower in this experiment than for spontaneous tumors
reported in other studies (they had too few spontaneous tumors to analyze in this study), but that
this study utilized lower doses and was of shorter duration than that of Ferreira-Gonzalez (1995).
Furthermore, the disparities from previous studies may also be impacted by lesion grouping,
mentioned above, in which lower stages of progression are grouped with more advanced stages.
Overall, in terms of H-ras mutation, TCE-induced tumors appears to be more like
DCA-induced tumors (which are consistent with spontaneous tumors), or those resulting from a
This document is a draft for review purposes only and does not constitute Agency policy.
377 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
coexposure to both DCA and TCA (Bull et al., 2002), than from those induced by TCA. As
noted above, Bull et al. (2002) reported the mutation frequency spectra for the H-ras codon 61 in
mouse liver tumors induced by TCE to be significantly different than that for TCA, with
DCA-treated mice tumors giving an intermediate result and for TCA-induced tumors to have a
H-ras profile that is the opposite than those of a number of other peroxisome proliferators. More
importantly, however, these data, along with the measures discussed above, show that mouse
liver tumors induced by TCE are heterogeneous in phenotype and genotype in a manner similar
to that observed in spontaneous tumors.
4.5.6.1.14. "Stop" experiments
Several stop experiments, in which treatment is terminated early in some dose groups,
have attempted to ascertain the whether progression differences exist between TCA and DCA.
After 37 weeks of treatment and then a cessation of exposure for 15 weeks, Bull et al. (1990)
reported that after combined 52 week period, liver weight and percentage liver/body weight were
reported to still be statistically significantly elevated after DCA or TCA treatment. The authors
partially attribute the remaining increases in liver weight to the continued presence of
hyperplastic nodules in the liver. In terms of liver tumor induction, the authors stated that
"statistical analysis of tumor incidence employed a general linear model ANOVA with contrasts
for linearity and deviations from linearity to determine if results from groups in which treatments
were discontinued after 37 weeks were lower than would have been predicted by the total dose
consumed." The multiplicity of tumors (incidence was not used) observed in male mice exposed
to DCA or TCA at 37 weeks and then sacrificed at 52 weeks were compared with those exposed
for a full 52 weeks. The response in animals that received the shorter duration of DCA exposure
was very close to that which would be predicted from the total dose consumed by these animals.
By contrast, the response to TCA exposure for the shorter duration was reported by the authors
to deviate significantly (p = 0.022) from the linear model predicted by the total dose consumed.
However, in the prediction of "dose-response," foci, adenomas, and carcinomas were combined
into one measure. Therefore, foci, a certain percentage of which have been commonly shown to
spontaneously regress with time, were included in the calculation of total "lesions." Moreover,
only a sample of lesions were selected for histological examination, and as is evident in the
sample, some lesions appeared "normal" upon microscopic examination (see below). Therefore,
while suggesting that cessation of exposure diminished the number of "lesions," methodological
limitations temper any conclusions regarding the identity and progression of lesion with
continuous versus noncontinuous DCA and TCA treatment.
This document is a draft for review purposes only and does not constitute Agency policy.
378 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Additionally, Bull et al. (1990) noted that after stopping treatment, DC A lesions appeared
to arrest their progression in contrast to TCA lesions, which appeared to progress. In particular,
among those in the stop treatment group (at 2 g/L) with 0/19 lesions examined histologically
were carcinomas, while in the continuous treatment groups, a significant fraction of lesions
examined were carcinomas at the higher exposure (6/23 at 2 g/L). By contrast, at terminal
sacrifice, TCA lesions a larger fraction of the lesions examined were carcinomas in the stop
treatment group (3/5 at 2 g/L) than in the continuous treatment group (2/7 and 4/16 at 1 g/L and
2 g/L, respectively).
However, as mentioned above, these inferences are based on examination of only a
subset of lesions. Specifically, for TCA treatment the number of animals examined for
determination of which "lesions" were foci, adenomas, and carcinomas was 11 out of the
19 mice with "lesions" at 52 weeks while all four mice with lesions after 37 weeks of exposure
and 15 weeks of cessation were examined. For DCA treatment the number of animals examined
was only 10 out of 23 mice with "lesions" at 52 weeks while all 7 mice with lesions after 37
weeks of exposure and 15 weeks of cessation were examined. Most importantly, when lesions
were examined microscopically, some did not all turn out to be preneoplastic or neoplastic—for
example, two lesions appeared "to be histologically normal" and one necrotic.
While limited, the conclusions of Bull et al. (1990) are consistent with later experiments
performed by Pereira and Phelps (1996). They noted that in MNU-treated mice that were then
treated with DCA, the yield of altered hepatocytes decreases as the tumor yields increase
between 31 and 51 weeks of exposure suggesting progression of foci to adenomas, but that
adenomas did not appear to progress to carcinomas. For TCA, Pereira and Phelps (1996)
reported that "MNU-initiated" adenomas promoted with TCA continued to progress. However,
the use of MNU initiation complicates direct comparisons with treatment with TCA or DCA
alone.
No similar data comparing stop and continued treatment of TCE are available to assess
the consistency or lack-thereof with TCA or DCA. Moreover, the informative of such a
comparison would be limited by designs of the available TCA and DCA studies, which have
used higher concentrations in conjunction with the much lower durations of exposure. While
higher doses allow for responses to be more easily detected, it introduces uncertainty as to the
effects of the higher doses alone. In addition, because the overall duration of the experiments is
also generally much less than 104 weeks, it is not possible to discern whether the differences in
results between those animals in which treatment was suspended in comparison to those in which
had not had been conducted would persist with longer durations.
This document is a draft for review purposes only and does not constitute Agency policy.
379 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
4.5.6.1.15. Conclusions Regarding the Role of Trichloroacetic Acid (TCA), Dichloroacetic
Acid (DCA), and Chloral Hydrate (CH) in Trichloroethylene (TCE)-Induced Effects in the
Liver
In summary, it is likely that oxidative metabolism is necessary for TCE-induced effects in
the liver. However, the specific metabolite or metabolites responsible for both noncancer and
cancer effects is less clear. TCE, TCA, and DCA exposures have all been associated with
induction of peroxisomal enzymes but are all weak PPARa agonists. The available data strongly
support TCA not being the sole or predominant active moiety for TCE-induced liver effects.
With respect to hepatomegaly, TCE and TCA dose-response relationships are quantitatively
inconsistent, for TCE leads to greater increases in liver/body weight ratios that expected from
predicted rates of TCA production. In fact, above a certain dose of TCE, liver/body weight
ratios are greater than that observed under any conditions studied so far for TCA. Histological
changes and effects on DNA synthesis are generally consistent with contributions from either
TCA or DCA, with a degree of polyploidization, rather than cell proliferation, likely to be
significant for TCE, TCA, and DCA. With respect to liver tumor induction, TCE leads to a
heterogeneous population of tumors, not unlike those that occur spontaneously or that are
observed following TCA-, DCA-, or CH-treatment. Moreover, some liver phenotype
experiments, particularly those utilizing immunostaining for c-Jun, support a role for both DCA
and TCA in TCE-induced tumors, with strong evidence that TCA cannot solely account for the
characteristics of TCE-induced tumors. In addition, H-ras mutation frequency and spectrum of
TCE-induced tumors more closely resembles that of spontaneous tumors or of those induced by
DCA, and were less similar in comparison to that of TCA-induced tumors. The heterogeneity of
TCE-induced tumors is similar to that observed to be induced by a broad category of
carcinogens, and to that observed in human liver cancer. Overall, then, it is likely that multiple
TCE metabolites, and therefore, multiple pathways, contribute to TCE-induced liver tumors.
4.5.7. Mode of Action (MOA) for Trichloroethylene (TCE) Liver Carcinogenicity
This section will discuss the evidentiary support for several hypothesized modes of action
for liver carcinogenicity (including mutagenicity and peroxisome proliferation, as well as several
additional proposed hypotheses and key events with limited evidence or inadequate experimental
support), following the framework outlined in the Cancer Guidelines (U.S. EPA, 2005c, d).9
9 As recently reviewed (Guvton et al.. 20081 the approach to evaluating mode of action information described in
EPA. 2005c. d) considers the issue of human relevance of a hypothesized mode of
action in the context of hazard evaluation. This excludes, for example, consideration of toxicokinetic differences
across species; specifically, the Cancer Guidelines state, "the toxicokinetic processes that lead to formation or
distribution of the active agent to the target tissue are considered in estimating dose but are not part of the mode of
action." In addition, information suggesting quantitative differences in the occurrence of a key event between test
This document is a draft for review purposes only and does not constitute Agency policy.
380 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
4.5.7.1.1. Mutagenicity
The hypothesis is that TCE acts by a mutagenic mode of action in TCE-induced
hepatocarcinogenesis. According to this hypothesis, the key events leading to TCE-induced liver
tumor formation constitute the following: TCE oxidative metabolite CH, after being produced in
the liver, cause direct alterations to DNA (e.g., mutation, DNA damage, and/or micronuclei
induction). Mutagenicity is a well established cause of carcinogenicity.
4.5.7.1.2. Experimental support for the hypothesized mode of action
The genotoxicity, as described by the ability of TCE, CH, TCA, and DCA to induce
mutations, was discussed previously in Section 4.2. The strongest data for mutagenic potential
are for CH, thought to be a relatively short-lived intermediate in the metabolism of TCE that is
rapidly converted to TCA and TCOH in the liver (see Section 3.3). CH causes a variety of
genotoxic effects in available in vitro and in vivo assays, with particularly strong data as to its
ability to induce aneuploidy. It has been argued that CH mutagenicity is unlikely to be the cause
of TCE carcinogenicity because the concentrations required to elicit these responses are
generally quite high, several orders of magnitude higher that achieved in vivo (Moore and
Harrington-Brock, 2000). For example, peak concentrations of CH in the liver of around
2-3 mg/kg have been reported after TCE administration at doses that are hepatocarcinogenic in
chronic bioassays (Abbas and Fisher, 1997; Greenberg et al., 1999). Assuming a liver density of
about 1 kg/L, these concentrations are orders of magnitude less than the minimum concentrations
reported to elicit genotoxic responses in the Ames test and various in vitro measures of
micronucleus, aneuploidy, and chromosome aberrations, which are in the 100-1,000 mg/L range.
However, it is not clear how much of a correspondence is to be expected from concentrations in
genotoxicity assays in vitro and concentrations in vivo, as reported in vivo CH concentrations are
in whole-liver homogenate while in vitro concentrations are in culture media. In addition, a few
in vitro studies have reported positive results at concentrations as low as 1 or 10 mg/L, including
Furnus et al. (1990) for aneuploidy in Chinese hamster CHED cells (10 mg/L), Eichenlaub-Ritter
et al. (1996) for bivalent chromosomes in meiosis I in MF1 mouse oocytes (10 mg/L), and
Gibson et al. (1995) for cell transformation in Syrian hamster embryo cells after 7 day treatment.
Moreover, some in vivo genotoxicity assays of CH reported positive results at doses similar to
species and humans are noted for consideration in the dose-response assessment, but is not considered in human
This document is a draft for review purposes only and does not constitute Agency policy.
3 81 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
those eliciting a carcinogenic response in chronic bioassays. For example, Nelson and Bull
(1988) reported increased DNA single strand breaks at 100 CH mg/kg (oral) in male B6C3F1
mice, although the result was not replicated by Chang et al. (1992). In another example, four of
six in vivo mouse genotoxicity studies reported that CH induced micronuclei in mouse
bone-marrow erythrocytes, with the lowest effective doses in positive studies ranging from
83-500 mg/kg (positive: Russo and Levis (1992), Russo et al. (1992), Marrazzini et al. (1994),
Beland et al. (1999); negative: Leuschner and Leuschner (1991), Leopardi et al. (1993)).
However, the use of i.p. administration in these and many other in vivo genotoxicity assays
complicates the comparison with carcinogenicity data. Also, it is difficult with the available data
to assess the contributions from the genotoxic effects of CH along with those from the genotoxic
and nongenotoxic effects of other oxidative metabolites (discussed below in Sections 4.5.5.2 and
4.5.5.3).
Furthermore, altered DNA methylation, another heritable mechanism by which gene
expression may be altered, is discussed below in the in Section 4.5.1.3.2.6. As discussed
previously, the differential patterns of H-ras mutations observed in liver tumors induced by TCE,
TCA, and DCA may be more indicative of tumor selection and tumor progression resulting from
exposure to these agents rather than a particular mechanism of tumor induction. The state of the
science of cancer and the role of epigenetic changes, in addition to genetic changes, in the
initiation and progression of cancer and specifically liver cancer, are discussed in Section E.3.1.
Therefore, while data are insufficient to conclude that a mutagenic MOA mediated by CH
is operant, a mutagenic MOA, mediated either by CH or by some other oxidative metabolite of
TCE, cannot be ruled out.
4.5.7.1.3. Peroxisome Proliferator Activated Receptor Alpha (PPARa) Receptor Activation
The hypothesis is that TCE acts by a PPARa agonism MOA in TCE-induced
hepatocarcinogenesis. According to this hypothesis, the key events leading to TCE-induced liver
tumor formation constitute the following: the TCE oxidative metabolite TCA, after being
produced in the liver, activates the PPARa receptor, which then causes alterations in cell
proliferation and apoptosis and clonal expansion of initiated cells. This MOA is assumed to
apply only to the liver.
relevance determination.
This document is a draft for review purposes only and does not constitute Agency policy.
3 82 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
4.5.7.1.4. Experimental support for the hypothesized mode of action
Proliferation of peroxisomes and increased activity of a number of related marker
enzymes has been observed in rodents treated with TCE, TCA, and DCA. The
peroxisome-related effects of TCE are most likely mediated primarily through TCA based on
TCE metabolism producing more TCA than DCA and the lower doses of TCA required to elicit
a response relative to DCA. However, Bull (2004a) and Bull et al. (2004) have recently
suggested that peroxisome proliferation occurs at higher exposure levels than those that induce
liver tumors for TCE and its metabolites. They report that a direct comparison in the no-effect
level or low-effect level for induction of liver tumors in the mouse and several other endpoints
shows that, for TCA, liver tumors occur at lower concentrations than peroxisome proliferation in
vivo but that PPARa activation occurs at a lower dose than either tumor formation or peroxisome
proliferation. A similar comparison for DCA shows that liver tumor formation occurs at a much
lower exposure level than peroxisome proliferation or PPARa activation. In vitro transactivation
studies have shown that human and murine versions of PPARa are activated by TCA and DCA,
while TCE itself is relatively inactive in the in vitro system, at least with mouse PPARa
(Maloney and Waxman, 1999; Zhou and Waxman, 1998). In addition, Laughter et al. (2004)
reported that the responses of acyl CoA oxidase (ACO), PCO, and CYP4A induction by TCE,
TCA, and DCA were substantially diminished in PPARa-null mice. Therefore, evidence
suggests that TCE, through its metabolites TCA and DCA, activate PPARa, and that at doses
relevant to TCE-induced hepatocarcinogenesis, the role of TCA in PPARa agonism is likely to
predominate.
It has been suggested that PPARa receptor activation is both the MOA for TCA liver
tumor induction as well as the MOA for TCE liver tumor induction, as a result of the metabolism
of TCE to TCA (Corton, 2008; NRC, 2006). Section E.3.4 addressed the status of the PPARa
MOA hypothesis for liver tumor induction and provides a more detailed discussion. However, as
discussed previously and in Section E.2.1.10, TCE-induced increases in liver weight have been
reported in male and female mice that do not have a functional PPARa receptor (Nakajima et al.,
2000). The dose-response for TCE-induced liver weight increases differs from that of TCA (see
Section E.2.4.2). The phenotype of the tumors induced by TCE have been described to differ
from those by TCA and to be more like those occurring spontaneously in mice, those induced by
DCA, or those resulting from a combination of exposures to both DCA and TCA (see
Section E.2.4.4). As to whether TCA induces tumors through activation of the PPARa receptor,
the tumor phenotype of TCA-induced mouse liver tumors has been reported to have a different
pattern of H-ras mutation frequency from other peroxisome proliferators (see Section E.2.4.4,
Bull et al., 2002; Fox et al., 1990; Hegi et al., 1993; Stanley et al., 1994). While TCE, DCA, and
This document is a draft for review purposes only and does not constitute Agency policy.
383 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
TCA are weak peroxisome proliferators, liver weight induction from exposure to these agents
has not correlated with increases in peroxisomal enzyme activity (e.g., PCO activity) or changes
in peroxisomal number or volume. By contrast, as discussed above, liver weight induction from
subchronic exposures appears to be a more accurate predictor of carcinogenic response for DC A,
TCA and TCE in mice (see also Section E.2.4.4). The database for cancer induction in rats is
much more limited than that of mice for determination of a carcinogenic response to these
chemicals in the liver and the nature of such a response.
While many compounds known to cause rodent liver tumors with long-term treatment
also activate the nuclear receptor PPARa, the mechanisms by which PPARa activation
contributes to tumorigenesis are not completely known (Klaunig et al., 2003; NRC, 2006; Yang
et al., 2007). As reviewed by Keshava and Caldwell (2006), PPARa activation leads to a highly
pleiotropic response and may play a role in toxicity in multiple organs as well as in multiple
chronic conditions besides cancer (obesity, atherosclerosis, diabetes, inflammation). Klaunig
et al. (2003) and NRC (2006) proposed that the key causal events for PPARa agonist-induced
liver carcinogenesis, after PPARa activation, are perturbation of cell proliferation and/or
apoptosis, mediated by gene expression changes, and selective clonal expansion. It has also been
proposed that sufficient evidence for this MOA consists of evidence of PPARa agonism (i.e., in
a receptor assay) in combination with either light- or electron-microscopic evidence for
peroxisome proliferation or both increased liver weight and one more of the in vivo markers of
peroxisome proliferation (Klaunig et al., 2003). However, it should be noted that peroxisome
proliferation and in vivo markers such as PCO are not considered causal events (Klaunig et al.,
2003; NRC, 2006), and that their correlation with carcinogenic potency is poor (Marsman et al.,
1988). Therefore, for the purposes of this discussion, peroxisome proliferation and its markers
are considered indicators of PPARa activation, as it is well established that these highly specific
effects are mediated through PPARa (Klaunig et al., 2003; Peters et al., 1997).
As recently reviewed by Guyton et al. (2009), recent data suggest that PPARa activation
along with these hypothesized causal events may not be sufficient for carcinogenesis. In
particular, Yang et al. (2007) reported comparisons between mice treated with Wy-14643 and
transgenic mice in which PPARa was constitutively activated in hepatocytes without the
presence of ligand. Yang et al. (2007) reported that, in contrast to Wy-14643-treatment, the
transgene did not induce liver tumors at 11 months, despite inducing PPARa-mediated effects of
a similar type and magnitude seen in response to tumorigenic doses of Wy-14643 in wild-type
mice (decreased serum fatty acids, induction of PPARa target genes, altered expression of
cell-cycle control genes, and a sustained increase in cellular proliferation). Nonetheless, it is
important to discuss the extent to which PPARa activation mediates the effects proposed by
This document is a draft for review purposes only and does not constitute Agency policy.
3 84 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Klaunig et al. (2003) and NRC (2006), even if the hypothesized sequence of key events may not
be sufficient for carcinogenesis. Investigation continues into additional events that may also
contribute, such as nonparenchymal cell activation and micro-RNA-based regulation of
protooncogenes (Shah et al., 2007; Yang et al., 2007). Specifically addressed below are gene
expression changes, proliferation, clonal expansion, and mutation frequency or spectrum.
With respect to gene expression changes due to TCE, Laughter et al. (2004) evaluated
transcript profiles induced by TCE in wild-type and PPARa-null mice. As noted in
Sections E.3.4.1.3 and E.3.1.2, there are limitations to the interpretation of such studies, some of
which are discussed below. Also noted in Appendix E are discussions of how studies of
peroxisome proliferators, indicate of the need for phenotypic anchoring, especially since gene
expression is highly variable between studies and within studies using the same experimental
paradigm. Section E.3.4 in also provides detailed discussions of the status of the PPARa
hypothesis. Of note, all null mice at the highest TCE dose (1,500 mg/kg-day) were moribund
prior to the end of the planned 3-week experiment (Laughter et al., 2004), and it was proposed
that this may reflect a greater sensitivity in PPARa-null mice to hepatotoxins due to defects in
tissue repair abilities. Laughter et al. (2004) also noted that four genes known to be regulated by
other peroxisome proliferators also had altered expression with TCE treatment in wild-type, but
not null mice. Ramdhan et al. (2010) report that not only do PPARa-null mice, but also
humanized mice (PPARa-null mice with inserted human PPARa) have underlying dysregulation
of lipid metabolism and gene expression. However, in a comparative analysis, Bartosiewicz
et al. (2001) concluded that TCE induced a different pattern of transcription than two other
peroxisome proliferators, di(2-ethylhexyl) phthalate (DEHP) and clofibrate. In addition,
Keshava and Caldwell (2006) compared gene expression data from Wy-14643, dibutyl phthalate
(DBP), gemifibrozil (GEM), and DEHP, and noted a lack of consistent results across PPARa
agonists. Thus, available data are insufficient to conclude that TCE gene expression changes are
similar to other PPAR agonists, or even that there are consistent changes (beyond the in vivo
markers of peroxisome proliferation, such as ACO, PCO, CYP4A, etc.) among different
agonists. It should also be noted that Laughter et al. (2004) did not compare baseline (i.e.,
control levels of) gene expression between null and wild-type control mice, hindering
interpretation of these results (Keshava and Caldwell, 2006). The possible relationship between
PPARa activation and hypomethylation are discussed below in Section 4.5.7.1.9.
In terms of proliferation, mitosis itself has not been examined in PPARa-null mice, but
BrdU incorporation, a measure of DNA synthesis that may reflect cell division, polyploidization,
or DNA repair, was observed to be diminished in null mice as compared to wild-type mice at 500
and 1,000 mg/kg-day TCE (Laughter et al., 2004). However, BrdU incorporation in null mice
This document is a draft for review purposes only and does not constitute Agency policy.
385 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
was still about threefold higher than controls, although it was not statistically significantly
different due to the small number of animals, high variability, and the two- to threefold higher
baseline levels of BrdU incorporation in control null mice as compared to control wild-type
mice. Therefore, while PPARa appears to contribute to the short-term increase in DNA
synthesis observed with TCE treatment, these results cannot rule out other contributing
mechanisms. However, since it is likely that both cellular proliferation and increased ploidy
contribute to the observed TCE-induced increases in DNA synthesis, it is not clear to whether the
observed decrease in BrdU incorporation is due to reduced proliferation, reduced
polyploidization, or both.
With respect to clonal expansion, it has been suggested that tumor characteristics such as
tincture (i.e., the staining characteristics light microscopy sections of tumor using H&E stains)
and oncogene mutation status can be used to associate chemical carcinogens with a particular
MOA such as PPARa agonism (Klaunig et al., 2003; NRC, 2006). This approach is problematic
primarily because of the lack of specificity of these measures. For example, with respect to
tincture, it has been suggested that TCA-induced foci and tumors resemble those of other
peroxisome proliferators in basophilia and lack of expression of GGT and GST-pi. However, as
discussed in Caldwell and Keshava (2006), the term "basophilic" in describing foci and tumors
can be misleading, because, for example, multiple lineages of foci and tumors exhibit basophilia,
including those not associated with peroxisome proliferators (Bannasch, 1996; Bannasch et al.,
2001; Carter et al., 2003). Moreover, a number of studies indicate that foci and tumors induced
by other "classic" peroxisome proliferators may have different phenotypic characteristics from
that attributed to the class through studies of WY-14643, including DEHP (Voss et al., 2005) and
clofibric acid (Michel et al., 2007). Furthermore, even the combination of GGT and GST-pi
negative, basophilic foci are nonspecific to peroxisome proliferators, as they have been observed
in rats treated with AfBl and AfBl plus PB, none of which are peroxisome proliferators (Grasl-
Kraupp et al., 1993)(Kraupp-Grasl et al., 1990). Finally, while Bull et al. (2004) suggested that
negative expression of c-jun in TCA-induced tumors may be consistent with a characteristic
phenotype of peroxisome proliferators, no data could be located to support this statement.
Therefore, of phenotypic information does not appear to be reliable for associating a chemical
with a PPARa agonism MOA.
Mutation frequency or spectrum in oncogenes has also been suggested to be an indicator
of a PPARa agonism MOA being active (NRC, 2006), with the idea being that specific
genotypes are being promoted by PPARa agonists. Although not a highly specific marker, H-ras
codon 61 mutation frequency and spectra data do not support a similarity between mutations in
TCE-induced, TCA-, or DCA- tumors and those due to other peroxisome proliferators. For
This document is a draft for review purposes only and does not constitute Agency policy.
386 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
example, while ciprofibrate and methylclofenopate had lower mutation frequencies than
historical controls (Hegi et al., 1993; Stanley et al., 1994), TCA-induced tumors had mutation
frequencies similar to or higher than historical controls (Bull et al., 2002; Ferreira-Gonzalez et
al., 1995). Anna et al. (1994) and Ferreira-Gonzalez et al. (1995) also reported TCE and
DCA-induced tumors to have mutation frequencies similar to historical controls, although Bull
et al. (2002) reported lower frequencies for these chemicals. However, the data reported by Bull
et al. (2002) consist of mixed lesions at different stages of progression, and such differing stages,
in addition to differences in genetic background and dose, can influence the frequency of H-ras
mutations (Stanley et al., 1994). In addition, a greater frequency of mutations was reported in
carcinomas than adenomas, and Bull et al. (2002) stated that this suggested that H-ras mutations
were a late event. Moreover, Fox et al. (1990) noted that tumors induced by phenobarbital,
chloroform, and ciprofibrate all had a much lower frequency of H-ras gene activation than those
that arose spontaneously, so this marker does not have good specificity. Mutation spectrum is
similarly of low utility for supporting a PPARa agonism MOA. First, because many peroxisome
proliferators been reported to have low frequency of mutations, the comparison of mutation
spectrum would be limited to a small fraction tumors. In addition to the low power due to small
numbers, the mutation spectrum is relatively nonspecific, as Fox et al. (1990) reported that of the
tumors with mutations, the spectra of the peroxisome proliferator ciprofibrate, historical controls,
and the genotoxic carcinogen benzidine-2 HC1 were similar.
In summary, TCE clearly activates PPARa, and some of the effects contributing to
tumorigenesis that Klaunig et al. (2003) and NRC (2006) propose to be the result of PPARa
agonism are observed with TCE, TCA, or DCA treatment. While this consistency is supportive a
role for PPARa, all of the proposed key causal effects with the exception of PPARa agonism
itself are nonspecific, and may be caused by multiple mechanisms. There is more direct
evidence that several of these effects, including alterations in gene expression and changes in
DNA synthesis, are mediated by multiple mechanisms in the case of TCE, and a causal linkage
to PPARa specifically is lacking. Therefore, because, as discussed further in the MOA
discussion below, there are multiple lines of evidence supporting the role of multiple pathways
of TCE-induced tumorigenesis, the hypothesis that PPARa agonism and the key causal events
proposed by Klaunig et al. (2003) and NRC (2006) constitute the sole or predominant MOA for
TCE-induced carcinogenesis is considered unlikely.
Furthermore, as reviewed by Guyton et al. (2009), recent data strongly suggest that
PPARa and key events hypothesized by Klaunig et al. (2003) are not sufficient for
carcinogenesis induced by the purported prototypical agonist Wy-14643. Therefore, the
This document is a draft for review purposes only and does not constitute Agency policy.
387 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
proposed PPARa MOA is likely "incomplete" in the sense that the sequence of key eventsio
necessary for cancer induction has not been identified. A recent 2-year bioassay of the
peroxisome proliferator DEHP showed that it can induce a liver tumor response in mice lacking
PPARa similar to that in wild-type mice (Ito et al., 2007). Klaunig et al. (2003) previously
concluded that PPARa agonism was the sole MOA for DEHP-induced liver tumorigenesis based
on the lack of tumors in PPARa-null mice after 11 months treatment with Wy-14643 (Peters et
al., 1997). They also assumed that due to the lack of markers of PPARa agonism in PPARa-null
mice after short-term treatment with DEHP (Ward et al., 1998), a long-term study of DEHP in
PPARa-null mice would yield the same results as for Wy-14643. However, due the finding by
Ito et al. (2007) that PPARa-null mice exposed to DEHP do develop liver tumors, they
concluded that DEHP can induce liver tumors by multiple mechanisms (Ito et al., 2007;
Takashima et al., 2008). Hence, since there is no 2-year bioassay in PPARa-null mice exposed
to TCE or its metabolites, it is not justifiable to use a similar argument based on Peters et al.
(1997) and short-term experiments to suggest that the PPARa MOA is operative. Therefore, the
conclusion is supported that the hypothesized PPARa MOA is inadequately specified because
the data do not adequately show the proposed key events individually being required for
hepatocarcinogenesis, nor do they show the sequence of key events collectively to be sufficient
for hepatocarcinogenesis.
4.5.7.1.5. Quantitative relationships between key events and tumor induction
The issues of whether there is a quantitative relationship between hypothesized key
events and tumor induction were recently examined in Guyton et al. (2009) and are discussed
below. Furthermore, IARC has recently concluded that additional mechanistic information has
become available, including studies with DEHP in PPARa-null mice, studies with several
transgenic mouse strains, carrying human PPARalpha or with hepatocyte-specific constitutively
activated PPARa and a study in humans exposed to DEHP from the environment that has
changed its conclusions regarding the relevance of rodent tumor data to human risk (Grosse et
al., 2011). Data from these new studies suggest that many molecular signals and pathways in
several cell types in the liver, rather than a single molecular event, contribute to cancer
10 As defined by the a "key event" is "an empirically observable precursor step
that is itself a necessary element of the mode of action or is a biologically based marker for such an element," and
the term "mode of action" (MOA) is defined as "a sequence of key events and processes, starting with interaction of
an agent with a cell, proceeding through operational and anatomical changes, and resulting in cancer formation."
Therefore, a single key event alone is necessary, but not necessarily sufficient for carcinogenesis; however, the
sequence of key events constituting a MOA needs to be sufficient for carcinogenesis.
This document is a draft for review purposes only and does not constitute Agency policy.
388 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
development in rodents with IARC concluding that the human relevance of the molecular events
leading to DEHPinduced cancer in several target tissues (e.g., liver and testis) in rats or mice
could not be ruled out, resulting in the evaluation of DEHP as a Group-2B agent, rather than
Group 3.
This following discussion is from Guyton et al. (2009):
Are key or associative events in the PPAR-a activation MOA quantitatively predictive of
hepatocarcinogenicity?
Another question to consider is whether potency for PPAR-a activation or its attendant
sequelae is quantitatively associated with carcinogenic activity or potency. If so, differences in
sensitivity for carcinogenesis (such as may occur across species) could be predicted using
quantitative information about the key events alone. If robust correlations were established, then
they could potentially be used either to quantitatively account for pharmacodynamic differences
that impact carcinogenic potency or as precursor events in nonlinear dose response assessment.
However, there are limitations in the dose-response data available for analyses of quantitative
relationships between potencies for precursor events in the proposed PPAR-a activation MOA and
for liver tumor induction. Most tumor data, including for the best characterized PPAR-a agonists,
are for exposure concentrations inducing well above 50% tumor incidence with less-than-lifetime
administration. Precursor events have typically been studied at a single dose, often eliciting a near
maximal response, thus precluding benchmark-based comparisons across studies. This is
especially true for Wy-14,643, which has been administered most often at only one exposure
concentration (1,000 ppm) that elicits a 100% tumor incidence after 1 year or less (Peters et al.,
1997) and that also appears to be necrogenic (Woods et al., 2007a). On the other hand,
hypothesized precursor events such as hepatomegaly, peroxisome proliferation, and increased
DNA synthesis appear to have reached their maximal responses at 50 ppm Wy-14,643, with some
statistically significant responses as low as 5 ppm (Marsman et al., 1992; Wada et al., 1992).
Potencies across compounds have rarely been compared in a single study using the same
experimental paradigm. These deficits in the database notwithstanding, provided below is an
assessment of the quantitative predictive power of the potency for four proposed data elements for
establishing the hypothesized MOA for hepatocarcinogenesis: PPAR-a activation in mice; and
hepatomegaly, DNA synthesis, and increased peroxisome proliferation in rats.
PPAR- a activation in mice
Table 2 [reproduced as Table 4-66] presents data for four peroxisome proliferators in order of
decreasing potency for inducing mouse liver tumors. These compounds were selected because of
their importance to environmental human health risk assessments and because data to derive
receptor activation potency indicators were available from a single study (Maloney and Waxman,
1999). The transactivation potencies of MEHP, Wy-14,643, dichloroacetic acid (DCA), and TCA
for the mouse PPAR-a were monitored using a luciferase reporter gene containing multiple PPAR
response elements derived from the rat hydratase/dehyrogenase promoter in transiently transfected
COS-1 monkey kidney cells. The derived potency indicators were compared to the TD50 (i.e., the
daily dose inducing tumors in half of the mice that would otherwise have remained tumor-free)
from the Carcinogenic Potency Database (CPDB) of Gold et al. (2005). Note that for Wy-14,643,
the dose listed yielded a maximal response and thus represents an upper limit to the TD50
(indicated by "<")• Two estimates of PPAR-a transactivation potency are given, the first based on
50% of the maximal response (i.e., EC50) and the second based on the effective concentration
required for a 2-fold increase in activity (i.e., EC2-fold) (Maloney and Waxman, 1999). Orally
administered DEHP undergoes presystemic hydrolysis catalyzed by lipase to MEHP in the gut,
with mice exhibiting higher lipase activities in the small intestine compared to rats and marmosets
(Kessler et al., 2004; Pollack et al., 1985) Ito et al. 2005). Therefore, because the mouse liver is
likely exposed predominantly to MEHP rather than DEHP and unmetabolized be
This document is a draft for review purposes only and does not constitute Agency policy.
389 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
explained by pharmacokinetics, i.e., hepatic conversion of DEHP to its mono-ester MEHP, since
studies in rats demonstrate that orally administered DEHP undergoes presystemic hydrolysis to
MEHP in the gut (Kessler et al., 2004; Pollack et al., 1985). Possible explanations for these
results include one or more of the following: (1) the transactivation assay is not an accurate
quantitative indicator of in vivo receptor activation, (2) the rate and nature of effects downstream
of PPAR-a activation depends on the ligand or, (3) there are rate-limiting events independent of
PPAR-a activation that contribute to mouse hepatocarcinogenesis by the agonists examined.
Hepatomegaly, DNA synthesis, and peroxisome proliferation in rats
Table 1 [reproduced as 4-67] compares potency indicators for various precursor effects at
the TD50 for four PPAR-a agonists and rat hepatocarcinogens. Our analysis of whether there are
consistent levels of in vivo precursor effect induction across peroxisome proliferators at the TD50
does not include all of the data from a similar, prior analysis by Ashby et al. (1994) for several
reasons. First, unlike the CPDB, Ashby et al. (1994) did not adjust carcinogenicity data for
less-than-lifetime dosing, which is relevant for most compounds. Second, for those mouse
carcinogens reported in the CPDB, only acute data are available regarding DNA synthesis effects
from Ashby et al.. Therefore, our analysis was restricted to rat precursor and potency data for the
four compounds Wy-14,643, nafenopin, clofibrate, and DEHP and included both 1-week and
13-week data to separately address transient and sustained changes in DNA synthesis. Even for
this small set of compounds, several limitations in the rat database were apparent. Because no
single study provided comparative data for the precursor endpoints of interest, four separate
reports were used. In the Wada et al. (1992) and Tanaka et al. (1992) studies of Wy-14,643 and
clofibrate, respectively, administered doses were within 10% of the TD50. However, nafenopin
data were only available at a single dose of 500 ppm (Lake et al., 1993), which was linearly
interpolated to the TD50. The highest administered dose of DEHP was 12,500 ppm (David et al.,
1999), a dose notably below the TD50, and thus a lower limit based on the assumption of
monotonicity with dose is shown. A further data limitation is that in the CPDB, only the TD50 for
one of the four compounds, DEHP, incorporates data from studies administering more than one
dose for two years.
The results shown in Table 1 [reproduced as Table 4-67] indicate that potency for the
occurrence of short-term in vivo markers of PPAR-a activation varies widely in magnitude and
lacks any apparent correlation with carcinogenic potency. Such differences have been noted
previously. Similar to the results presented in Table 1 [reproduced as Table 4-67], Marsman et al.
(1988) noted that although DEHP (12,000 ppm) and Wy-14,643 (1,000 ppm) induced a similar
extent of hepatomegaly and peroxisome proliferation (measured either morphologically or
biochemically) after 1 year, the frequency of hepatocelluar lesions was over 100-fold higher in
Wy-14,643- relative to DEHP-exposed rats. In addition, a higher labeling index was reported for
12,500 ppm DEHP than the maximal level attained after 50-1,000 ppm Wy-14,643 (David et al.,
1999; Tanaka et al., 1992; Wada et al., 1992). We did not examine such differences in maximal
responses in our analysis. We also do not present differences in response with dose and time seen
among PPAR-a agonists, which are prominent enough to prevent displaying dose-response data
on a common scale. For instance, labeling index is increased in a dose-dependent manner at 1
week by clofibrate (1.500, 4.500 and 9.000 ppm) but is decreased compared with controls at 13
weeks at the two higher doses (Tanaka et al., 1992). Together, these findings underscore the
significant chemical-specific quantitative differences in these markers that limit their utility for
predicting carcinogenic dose-response relationships.
Table 4-66. Potency indicators for mouse hepatocarcinogenicity and in vitro
transactivation of mouse PPARa for four PPARa agonists
Carcinogenic potency
Transactivation potency
indicators
indicators (jiM)
This document is a draft for review purposes only and does not constitute Agency policy.
390 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
(mg/kg-day)
Chemical
td50
EC50
ECtwofold
Hepatocarcinogens
Wy-14,643
<10.8
0.63
-0.4
DCA
119
-300
-300
TCA
584
-300
-300
DEHP/MEHP
700
-0.7
-0.7
Source: reproduced from Table 2 of Guyton et al. (2009).
Note: TD50 = the daily dose inducing tumors in half of the mice that would otherwise have remained
tumor-free, estimated from the Carcinogenic Potency Database (Gold et al. 2005). EC50 = the
effective concentration yielding 50% of the maximal response; ECY,o[o|d the effective
concentration required for a twofold increase in activity. Transactivation potencies were estimated
from Maloney and Waxman (1999). The "<" symbol denotes an upper limit due to maximal
response. A symbol indicates that the transactivation potency was approximated from figures
in Maloney and Waxman (1999).
MEHP = monoethylhexyl phthalate.
Table 4-67. Potency indicators for rat hepatocarcinogenicity and common
short-term markers of PPARa activation for four PPARa agonists
Chemical
Tumor TD50
(ppm in diet)
Fold-increase over control at tumor TD50
1 wk
13 wk
RLW
LI
PCO
RLW
LI
PCO
Wy-14,643
109
1.8
12
13
2.6
6.8
39
Nafenopin
275
1.4
3.6
7.6
1.5
1.12
6.7
Clofibrate
4.225
1.4
4.4
4.2
1.4
0.95
3.7
DEHP
17.900
>1.4
>19
>3.6
>1.9
>1.25
>4.9
Source: reproduced from Table 1 of Guyton et al. (2009).
Note: For ease of comparison with precursor effect studies, administered doses for the tumor TD50s in the
Carcinogenic Potency Database were back-converted to equivalent ppm in diet using the formula of Gold et al.
(2005), i.e., TD50 (mg/kg-day) = TD50(ppm in diet) x 0.04 (for male rats). Administered doses for precursor data
on Wy-14,643 (Wada et al., 1992) and clofibrate (Tanaka et al., 1992) were within 10% of the TD50. Because
nafenopin precursor data were only available at 0 and 500 ppm (Lake et al., 1993), these doses were linearly
interpolated to the TD50. Because the highest administered dose of DEHP in precursor effect studies was
12,500 ppm (David et al., 1999), a lower limit is shown, based on the assumption of monotonicity with dose.
RLW = relative liver weight, LI = labeling index, PCO = cyanide insensitive palmitoyl CoA oxidation.
This document is a draft for review purposes only and does not constitute Agency policy.
3 91 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
4.5.7.1.6. Additional Proposed Hypotheses and Key Events with Limited Evidence or
Inadequate Experimental Support
Several effects that been hypothesized to be associated with liver cancer induction are
discussed in more detail below, including increased liver weight, DNA hypomethylation, and
pathways involved in glycogen accumulation such as insulin signaling proteins. As discussed
above, TCE and its metabolites reportedly increase nuclear size and ploidy in hepatocytes, and
these effects likely account for much of the increases in labeling index and DNA synthesis
caused by TCE. Importantly, these changes appear to persist with cessation of treatment, with
liver weights, but not nuclear sizes, returning to control levels (Kjellstrand et al., 1983a). In
addition, glycogen deposition, DNA synthesis, increases in mitosis, or peroxisomal enzyme
activity do not appear correlated with TCE-induced liver weight changes.
4.5.7.1.7. Increased liver weight
Increased liver weight or liver/body weight ratios (hepatomegaly) is associated with
increased risk of liver tumors in rodents, but it is relatively nonspecific (Allen et al., 2004). The
evidence presented above for TCE and its metabolites suggest a similarity in dose-response
between liver weight increases at short-term durations of exposure and liver tumor induction
observed from chronic exposure. Liver weight increases may results from several concurrent
processes that have been associated with increase cancer risk (e.g., hyperplasia, increased ploidy,
and glycogen accumulation) and when observed after chronic exposure may result from the
increased presence of foci and tumors themselves. Therefore, there are inadequate data to
adequately define a MOA hypothesis for hepatocarcinogenesis based on liver weight increases.
4.5.7.1.8. "Negative selection"
As discussed above, TCE, TCA, and DCA all cause transient increases in DNA synthesis.
This DNA synthesis has been assumed to result from proliferation of hepatocytes. However, the
dose-related TCA- and DCA-induced increases in liver weight not correlate with patterns of
DNA synthesis; moreover, there have been reports that DNA synthesis in individual hepatocytes
does not correlate with whole liver DNA synthesis measures (Carter et al., 1995; Sanchez and
Bull, 1990). With continued treatment, decreases in DNA synthesis have been reported for DCA
(Carter et al., 1995). More importantly, several studies show that transient DNA synthesis is
confined to a very small population of cells in the liver in mice exposed to TCE for 10 days or to
This document is a draft for review purposes only and does not constitute Agency policy.
3 92 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
DC A or TCA for up to 14 days of exposure. Therefore, generalized mitogenic stimulation is not
likely to play a role in TCE-induced liver carcinogenesis.
Bull has proposed that the TCE metabolites TCA and DCA may contribute to liver tumor
induction through so-called "negative selection" by way of several possible processes (Bull,
2000). First, it is hypothesized that the mitogenic stimulation by continued TCA and DCA
exposure is down-regulated in normal hepatocytes, conferring a growth advantage to initiated
cells that either do not exhibit the down-regulation of response or are resistant to the
down-regulating signals. This is implausible as both the normal rates of cell division in the liver
and the TCE-stimulated increases are very low. Polyploidization has been reported to decrease
the normal rates of cell division even further. That the transient and relatively low level of DNA
synthesis reported for TCE, DCA, and TCA is reflective of proliferation rather than
polyploidization is not supported by data on mitosis. A mechanism for such "down-regulation"
has not been identified experimentally.
A second proposed contributor to "negative-selection" is direct enhancement by TCA and
DCA in the growth of certain populations of initiated cells. While differences in phenotype of
end stage tumors have been reported between DCA and TCA, the role of selection and
emergence of potentially different foci has not been elucidated. Neither have pathway
perturbations been identified that are common to liver cancer in human and rodent for TCE,
DCA, and TCA. The selective growth of clones of hepatocytes that may progress fully to cancer
is a general feature of cancer and not specific to at TCE, TCA, or DCA MOA.
A third proposed mechanism by which TCE may enhance liver carcinogenesis within this
"negative selection" paradigm is through changing apoptosis. However, as stated above, TCE
has been reported to either not change apoptosis or to cause a slight increase at high doses.
Rather than increases in apoptosis, peroxisome proliferators have been suggested to inhibit
apoptosis as part of their carcinogenic MOA. However, the age and species studied appear to
greatly affect background rates of apoptosis (Snyder et al., 1995) with the rat having a greater
rate of apoptosis than the mouse. DCA has been reported to induce decreases in apoptosis in the
mouse (Carter et al., 1995; Snyder et al., 1995). However, the significance of the DCA-induced
reduction in apoptosis, from a level that is already inherently low in the mouse, for the MOA for
induction of DCA-induce liver cancer is difficult to discern.
Therefore, for a MOA for hepatocarcinogenesis based on "negative selection," there are
inadequate data to adequately define the MOA hypothesis, or the available data do not support
such a MOA being operative.
This document is a draft for review purposes only and does not constitute Agency policy.
393 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
4.5.7.1.9. Polyploidization
Polyploidization may be an important key event in tumor induction. For example, in
addition to TCE, partial hepatectomy, nafenopin, methylclofenopate, DEHP, diethylnitrosamine,
A'-nitrosomorpholine, and various other exposures that contribute to liver tumor induction also
shift the hepatocyte ploidy distribution to be increasingly diploid or polypoid (Hasmall and
Roberts, 2000; Melchiorri et al., 1993; Miller et al., 1996; Styles et al., 1988; Vickers and Lucier,
1996). As discussed by Gupta (2000), "[wjorking models indicate that extensive polyploidy
could lead to organ failure, as well as to oncogenesis with activation of precancerous cell
clones." However, the mechanism(s) by which increased polypoidy enhances carcinogenesis is
not currently understood. Due to increased DNA content, polypoid cells will generally have
increased gene expression. However, polyploid cells are considered more highly differentiated
and generally divide more slowly and are more likely to undergo apoptosis, perhaps thereby
indirectly conferring a growth advantage to initiated cells (see Section E. 1). Of note is that
changes in ploidy have been observed in transgenic mouse models that are also prone to develop
liver cancer (see Section E.3.3.1). It is likely that polyploidization occurs with TCE exposure
and it is biologically plausible that polyploidization can contribute to liver carcinogenesis,
although the mechanism(s) is (are) not known. However, whether polyploidization is necessary
for TCE-induced carcinogenesis is not known, as no experiment in which polyploidization
specifically is blocked or diminished has been performed and the extent of polyploidization has
not been quantified. Therefore, there are inadequate data to adequately define a MOA
hypothesis for hepatocarcinogenesis based on polyploidization.
4.5.7.1.10. Glycogen storage
As discussed above, several studies have reported that DCA causes accumulation of
glycogen in mouse hepatocytes. Such glycogen accumulation has been suggested to be
pathogenic, as it is resistant to mobilization by fasting (Kato-Weinstein et al., 1998). In humans,
glycogenesis due to glycogen storage disease or poorly controlled diabetes has been associated
with increased risk of liver cancer (Adami et al., 1996; La Vecchia et al., 1994; Rake et al., 2002;
Wideroff et al., 1997). Glycogen accumulation has also been reported to occur in rats exposed to
DCA.
For TCE exposure in mice or rats, glycogen content of hepatocytes has been reported to
be somewhat less than or the same as controls, or not remarked upon in the studies. TCA
exposure has been reported to decrease glycogen content in rodent hepatocytes while DCA has
been reported to increase it (Kato-Weinstein et al., 2001). There is also evidence that
This document is a draft for review purposes only and does not constitute Agency policy.
3 94 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
DCA-induced increases in glycogen accumulation are not proportional to liver weight increases
and only account for a relatively small portion of increases in liver mass. DCA-induced
increases in liver weight are not a function of cellular proliferation but probably include
hypertrophy associated with polyploidization, increased glycogen deposition and other factors.
While not accounting for increases in liver weight, excess glycogen can still be not only
be pathogenic but a predisposing condition for hepatocarcinogenesis. Some hypotheses
regarding the possible relationship between glycogenesis and carcinogenesis have been posed
that lend them biological plausibility. Evert et al. (2003), using an animal model of hepatocyte
exposure to a local hyperinsulinemia from transplanted islets of Langerhans with remaining
tissue is hypoinsulinemic, reported that insulin induces alterations resembling preneoplastic foci
of altered hepatocytes that develop into hepatocellular tumors in later stages of carcinogenesis.
Lingohr et al. (2001) suggest that normal hepatocytes down-regulate insulin-signaling proteins in
response to the accumulation of liver glycogen caused by DCA and that the initiated cell
population, which does not accumulate glycogen and is promoted by DCA treatment, responds
differently from normal hepatocytes to the insulin-like effects of DCA. Bull et al. (2002)
reported increased insulin receptor protein expression in tumor tissues regardless of whether they
were induced by TCE, TCA, or DCA. Given the greater activity of DCA relative to TCA on
carbohydrate metabolism, it is unclear whether changes in these pathways are causes or simply
reflect the effects of tumor progression. Therefore, it is biologically plausible that changes in
glycogen status may occur from the opposing actions of TCE metabolites, but changes in
glycogen content due to TCE exposure has not been quantitatively studied. The possible
contribution of these effects to TCE-induced hepatocarcinogenesis is unclear. Therefore, there
are inadequate data to adequately define a MOA hypothesis for TCE-induced
hepatocarcinogenesis based on changes in glycogen storage or even data to support increased
glycogen storage to result from TCE exposure.
4.5.7.1.11. Inactivation of GST-zeta
DCA has been shown to inhibit its own metabolism in that pretreatment in rodents prior
to a subsequent challenge dose leads to a longer biological half-life (Schultz et al., 2002). This
self-inhibition is hypothesized to occur through inactivation of GST-zeta (Schultz et al., 2002).
In addition, TCE has been shown to cause the same prolongation of DCA half-life in rodents,
suggesting that TCE inhibits GST-zeta, probably through the formation of DCA (Schultz et al.,
2002). DCA-induced inhibition of GST-zeta has also been reported in humans, with GST-zeta
polymorphisms reported to influence the degree of inactivation (Blackburn et al., 2001;
This document is a draft for review purposes only and does not constitute Agency policy.
395 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Blackburn et al., 2000; Tzeng et al., 2000). Board et al. (2001) report one variant to have
significantly higher activity with DC A as a substrate than other GST-zeta isoforms, which could
affect DC A susceptibility.
GST-zeta, which is identical to maleylacetoacetate isomerase, is part of the tyrosine
catabolism pathway which is disrupted in Type 1 hereditary tyrosinemia, a disease associated
with the development of hepatocellular carcinoma at a young age (Tanguay et al., 1996). In
particular, GST-zeta metabolizes maleylacetoacetate (MAA) to fumarylacetoacetate (FAA) and
maleylacetone (MA) to fumarylacetone (Cornett et al., 1999; Tanguay et al., 1996). It has been
suggested that the increased cancer risk with this disease, as well as through DC A exposure,
results from accumulation of MAA and MA, both alkylating agents, or FAA, which displays
apoptogenic, mutagenic, aneugenic, and mitogenic activities (Bergeron et al., 2003; Cornett et
al., 1999; Jorquera and Tanguay, 2001; Kim et al., 2000; Tanguay et al., 1996). However, the
possible effects of DCA through this pathway will depend on whether MAA, MA, or FAA is the
greater risk factor, since inhibition of GST-zeta will lead to greater concentrations of MAA and
MA and lower concentrations of FAA. Therefore, if MAA is the more active agent, DCA may
increase carcinogenic risk, while if FAA is the more active, DCA may decrease carcinogenic
risk. Tzeng et al. (2000) propose the later based on the greater genotoxicity of FAA, and in fact
suggest that DCA may "merit consideration for trial in the clinical management of hereditary
tyrosinemia type 1."
Therefore, TCE-induced inactivation GST-zeta, probably through formation of DCA,
may play a role in TCE-induced hepatocarcinogenesis. However, this mode of action is not
sufficiently delineated at this point for further evaluation, as even the question of whether its
actions through this pathway may increase or decrease cancer risk has yet to be experimentally
tested.
4.5.7.1.12. Oxidative stress
Several studies have attempted to study the possible effects of "oxidative stress" and
DNA damage resulting from TCE exposures. The effects of induction of metabolism by TCE, as
well as through coexposure to ethanol, have been hypothesized to in itself increase levels of
"oxidative stress" as a common effect for both exposures (see Section E.4.2.4). In terms of
contributing to a carcinogenic MO A, the term "oxidative stress" is a somewhat nonspecific term,
as it is implicated as part of the pathophysiologic events in a multitude of disease processes and
is part of the normal physiologic function of the cell and cell signaling. Commonly, it appears to
refer to the formation of reactive oxygen species leading to cellular or DNA damage. As
This document is a draft for review purposes only and does not constitute Agency policy.
396 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
discussed above, however, measures of oxidative stress induced by TCE, TCA, and DCA appear
to be either not apparent, or at the very most transient and nonpersistent with continued treatment
(Channel et al., 1998; Larson and Bull, 1992b; Parrish et al., 1996; Toraason et al., 1999).
Therefore, while the available data are limited, there is insufficient evidence to support a role for
such effects in TCE-induced liver carcinogenesis.
Oxidative stress has been hypothesized to be part of the MOA for peroxisome
proliferators, but has been found to neither be correlated with cell proliferation nor carcinogenic
potency of peroxisome proliferators (see Section E.3.4.1.1). For instance, Parrish et al. (1996)
reported that increases in PCO activity noted for DCA and TCA were not associated with
8-OHdG levels (which were unchanged) and also not with changes laurate hydrolase activity
observed after either DCA or TCA exposure. The authors concluded that their data do not
support an increase in steady state oxidative damage to be associated with TCA initiation of
cancer and that extension of treatment to time periods sufficient to insure peroxisome
proliferation failed to elevate 8-OHdG in hepatic DNA. The authors thus, suggested that
peroxisome proliferative properties of TCA were not linked to oxidative stress or carcinogenic
response.
4.5.7.1.13. Changes in gene expression (e.g., hypomethylation)
Studies of gene expression as well as considerations for interpretation of studies of using
the emerging technologies of DNA, siRNA, and miRNA microarrays for MOA analyses are
included in Sections E.3.1.2 and E.3.4.2.2. Caldwell and Keshava (2006) and Keshava and
Caldwell (2006) report on both genetic expression studies and studies of changes in methylation
status induced by TCE and its metabolites as well as differences and difficulties in the patterns of
gene expression between differing PPARa agonists. In particular are concerns for the
interpretation of studies which employ pooling of data as well as interpretation of "snapshots in
time of multiple gene changes." For instance, in the Laughter et al. (2004) study, it is not clear
whether transcription arrays were performed on pooled data as well as the issue of phenotypic
anchoring as data on percentage liver/body weight indicates significant variability within TCE
treatment groups, especially in PPARa-null mice. For studies of gene expression using
microarrays Bartosiewicz et al. (2001) used a screening analysis of 148 genes for
xenobiotic-metabolizing enzymes, DNA repair enzymes, heat shock proteins, cytokines, and
housekeeping gene expression patterns in the liver in response TCE. The TCE-induced gene
induction was reported to be highly selective; only Hsp 25 and 86 and Cyp2a were up-regulated
at the highest dose tested. Collier et al. (2003) reported differentially expressed mRNA
This document is a draft for review purposes only and does not constitute Agency policy.
397 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
transcripts in embryonic hearts from Sprague-Dawley rats exposed to TCE with sequences
down-regulated with TCE exposure appearing to be those associated with cellular housekeeping,
cell adhesion, and developmental processes. TCE was reported to induce up-regulated
expression of numerous stress-response and homeostatic genes.
For the Laughter et al. (2004) study, transcription profiles using macroarrays containing
approximately 1,200 genes were reported in response to TCE exposure with 43 genes reported to
be significantly altered in the TCE-treated wild-type mice and 67 genes significantly altered in
the TCE-treated PPARa knockout mice. However, the interpretation of this information is
difficult because in general, PPARa knockout mice have been reported to be more sensitive to a
number of hepatotoxins partly because of defects in the ability to effectively repair tissue damage
in the liver (Mehendale, 2000; Shankar et al., 2003) and because a comparison of gene
expression profiles between controls (wild-type and PPARa knockout) were not reported. As
reported by Voss et al. (2006), dose-, time course-, species-, and strain-related differences should
be considered in interpreting gene array data. The comparison of differing PPARa agonists
presented in Keshava and Caldwell (2006) illustrate the pleiotropic and varying liver responses
of the PPARa receptor to various agonists, but did not imply that these responses were
responsible for carcinogenesis.
As discussed above in Section E.3.3.5, Aberrant DNA methylation is a common hallmark
of all types of cancers, with hypermethylation of the promoter region of specific tumor
suppressor genes and DNA repair genes leading to their silencing (an effect similar to their
mutation) and genome-wide hypomethylation (Ballestar and Esteller, 2002; Berger and
Daxenbichler, 2002; Herman et al., 1998; Pereira et al., 2004a; Rhee et al., 2002). Whether
DNA methylation is a consequence or cause of cancer is a long-standing issue (Ballestar and
Esteller, 2002). Fraga et al. (2005; 2004) reported global loss of monoacetylation and
trimethylation of histone H4 as a common hallmark of human tumor cells; they suggested,
however, that genomewide loss of 5-methylcytosine (associated with the acquisition of a
transformed phenotype) exists not as a static predefined value throughout the process of
carcinogenesis but rather as a dynamic parameter (i.e., decreases are seen early and become more
marked in later stages).
DNA methylation is a naturally occurring epigenetic mechanism for modulating gene
expression, and disruption of this mechanism is known to be relevant to human carcinogenesis.
As reviewed by Calvisi et al. (2007),
[a]berrant DNA methylation occurs commonly in human cancers in the forms of genome-wide
hypomethylation and regional hypermethylation. Global DNA hypomethylation (also known as
demethylation) is associated with activation of protooncogenes, such as c-Jun, c-Myc, and
This document is a draft for review purposes only and does not constitute Agency policy.
398 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
c-HA-Ras, and generation of genomic instability. Hypermethylation on CpG islands located in the
promoter regions of tumor suppressor genes results in transcriptional silencing and genomic
instability.
While clearly associated with cancer, it has not been conclusively established whether these
epigenetic changes play a causative role or are merely a consequence of transformation
(Tryndyak et al., 2006). However, as Calvisi et al. (2007) note, "Current evidence suggests that
hypomethylation might promote malignant transformation via multiple mechanisms, including
chromosome instability, activation of protooncogenes, reactivation of transposable elements, and
loss of imprinting."
Although little is known about how it occurs, a hypothesis has also been proposed that
that the toxicity of TCE and its metabolites may arise from its effects on DNA methylation
status. In regard to methylation studies, many are coexposure studies as they have been
conducted in initiated animals with some studies being very limited in their reporting and
conduct. Caldwell and Keshava (2006) review the body of work regarding TCE, DC A, and
TCA. Methionine status has been noted to affect the emergence of liver tumors (Counts et al.,
1996). Tao et al. (2000a) and Pereira et al. (2004b) have studied the effects of excess methionine
in the diet to see if it has the opposite effects as a deficiency (i.e., and reduction in a carcinogenic
response rather than enhancement). However, Tao et al. (2000a) report that the administration of
excess methionine in the diet is not without effect and can result in percentage liver/body weight
ratios. Pereira et al. (2004b) report that methionine treatment alone at the 8 g/kg level was
reported to increase liver weight, decrease lauryl-CoA activity and to increase DNA methylation.
Pereira et al. (2004b) reported that very high level of methionine supplementation to an
AIN-760A diet, affected the number of foci and adenomas after 44 weeks of coexposure to
3.2 g/L DCA. However, while the highest concentration of methionine (8.0 g/kg) was reported
to decrease both the number of DCA-induce foci and adenomas, the lower level of methionine
coexposure (4.0 g/kg) increased the incidence of foci. Coexposure of methionine (4.0 or
8.0 g/kg) with 3.2 g/L DCA was reported to decrease by -25% DCA-induced glycogen
accumulation, increase mortality, but not to have much of an effect on peroxisome enzyme
activity (which was not elevated by more than 33% over control for DCA exposure alone). The
authors suggested that their data indicate that methioninine treatment slowed the progression of
foci to tumors. Given that increasing hypomethylation is associated with tumor progression,
decreased hypomethylation from large doses of methionine are consistent with a slowing of
progression. Whether, these results would be similar for lower concentrations of DCA and lower
concentrations of methionine that were administered to mice for longer durations of exposure,
cannot be ascertained from these data. It is possible that in a longer-term study, the number of
This document is a draft for review purposes only and does not constitute Agency policy.
399 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
tumors would be similar. Finally, a decrease in tumor progression by methionine
supplementation is not shown to be a specific event for the MOA for DCA-induced liver
carcinogenicity.
Tao et al. (2000a) reported that 7 days of gavage dosing of TCE (1,000 mg/kg in corn
oil), TCA (500 mg/kg, neutralized aqueous solution), and DCA (500 mg/kg, neutralized aqueous
solution) in 8-week old female B6C3F1 mice resulted in not only increased liver weight but also
increased hypomethylation of the promoter regions of c-jun and c-myc genes in whole liver
DNA. However, data were shown for 1-2 mice per treatment. Treatment with methionine was
reported to abrogate this response only at a 300 mg/kg i.p dose with 0-100 mg/kg doses of
methionine having no effect. Ge et al. (2001a) reported DCA- and TCA-induced DNA
hypomethylation and cell proliferation in the liver of female mice at 500 mg/kg and decreased
methylation of the c-myc promoter region in liver, kidney and urinary bladder. However,
increased cell proliferation preceded hypomethylation. Ge et al. (2002) also reported
hypomethylation of the c-myc gene in the liver after exposure to the peroxisome proliferators
2,4-dichlorophenoxyacetic acid (1,680 ppm), DBP (20,000 ppm), gemfibrozil (8,000 ppm), and
Wy-14,643 (50-500 ppm, with no effect at 5 or 10 ppm) after 6 days in the diet. Caldwell and
Keshava (2006) concluded that hypomethylation did not appear to be a chemical-specific effect
at these concentrations. As noted Section E.3.3.5, chemical exposure to a number of differing
carcinogens have been reported to lead to progressive loss of DNA methylation..
After initiation by A'-methyl-A-nitrosourea (25 mg/kg) and exposure to 20 mmol/L DCA
or TCA (46 weeks), Tao et al. (2004a) report similar hypomethylation of total mouse liver DNA
by DCA and TCA with tumor DNA showing greater hypomethylation. A similar effect was
noted for the differentially methylated region-2 of the insulin-like growth factor-II (IGF-II) gene.
The authors suggest that hypomethylation of total liver DNA and the IGF-II gene found in
nontumorous liver tissue would appear to be the result of a more prolonged activity and not cell
proliferation, while hypomethylation of tumors could be an intrinsic property of the tumors. As
pointed out by Caldwell and Keshava (2006) over expression of IGF-II gene in liver tumors and
preneoplastic foci has been shown in both animal models of hepatocarcinogenesis and humans,
and may enhance tumor growth, acting via the over-expressed IGF-I receptor (Scharf et al.,
2001; Werner and Le Roith, 2000).
Diminished hypomethylation was observed in Wy-14643-treated PPARa-null mice as
compared to wild-type mice, suggestive of involvement of PPARa in mediating hypomethylation
(Pogribny et al., 2007), but it is unclear how relevant these results are to TCE and its metabolites.
First, the doses of Wy-14643 administered are associated with substantial liver necrosis and
mortality with long-term treatment (Woods et al., 2007a), adding confounding factors the
This document is a draft for review purposes only and does not constitute Agency policy.
400 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
interpretation of their results. Hypomethylation by Wy-14643 progressively increased with time
up to 5 months (Pogribny et al., 2007), consistent with the sustained DNA synthesis caused by
Wy-14643 and a role for proliferation in causing hypomethylation. Regardless, as discussed
above, it is unlikely that PPARa is the mediator of the observed transient increase in DNA
synthesis by DCA, so even if it is important for hypomethylation by TCA, there may be more
than one pathway for this effect.
To summarize, aberrant DNA methylation status, including hypomethylation, is clearly
associated with both human and rodent carcinogenesis. Hypomethylation itself appears to be
sufficient for carcinogenesis, as diets deficient in choline and methionine that induce
hypomethylation have been shown to cause liver tumors in both rats and mice (Ghoshal and
Farber, 1984; Henning and Swendseid, 1996; Mikol et al., 1983; Wainfan and Poirier, 1992).
However, it is not known to what extent hypomethylation is necessary for TCE-induced
carcinogenesis. However, as noted by Bull (2004a) and Bull et al. (2004), the doses of TCA and
DCA that have been tested for induction of hypomethylation are quite high compared to doses at
which tumor induction occurs—at least 500 mg/kg-day. Whether these effects are still manifest
at lower doses relevant to TCE carcinogenicity, particularly with respect to DCA, has not been
investigated. Finally, the role of PPARa in modulating hypomethylation, possibly through
increased DNA synthesis as suggested by experiments with Wy-14643, are unknown for TCE
and its metabolites.
4.5.7.1.14. Cytotoxicity
Cytotoxicity and subsequent induction of reparative hyperplasia have been proposed as
key events for a number of chlorinated solvents, such as chloroform and carbon tetrachloride..
However, as discussed above and discussed by Bull (2004a) and Bull et al. (2004), TCE
treatment at doses relevant to liver carcinogenicity results in relatively low cytotoxicity. While a
number of histological changes with TCE exposure are observed, in most cases necrosis is
minimal or mild, associated with vehicle effects, and with relatively low prevalence. This is
consistent with the low prevalence of necrosis observed with TCA and DCA treatment at doses
relevant to TCE exposure. Therefore, it is unlikely that cytotoxicity and reparative hyperplasia
play a significant role in TCE carcinogenicity
This document is a draft for review purposes only and does not constitute Agency policy.
401 DRAFT—DO NOT CITE OR QUOTE
-------
4.5.7.1.15. Mode of Action (MOA) Conclusions
1 The conclusions regarding the MOA for TCE-induced liver carcinogenesis described in
2 the preceding sections are summarized in Table 4-68. Overall, although a role for many of the
3 proposed key events discussed above cannot be ruled out, there are inadequate data to support
4 the conclusion that any of the particular MOA hypotheses reviewed above are operant. The
5 available data do suggest that the MOA of liver tumors induced by TCE is complex, as it is
6 likely that key events from several pathways may operate. Nonetheless, because a collection of
7 key events sufficient to induce liver tumors has not been identified, the answer to the first key
This document is a draft for review purposes only and does not constitute Agency policy.
402 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-68. Summary of mode of action conclusions for TCE-induced liver carcinogenesis
Hypothesized MOA and key
events
Evidence that TCE or TCE
metabolites induces key events
Necessity of MOAs key
events for carcinogenesis
Sufficiency of MOA for
carcinogenesis
Summary and section
reference
Mutagenicity induced by one
or more metabolites advances
acquisition of multiple critical
traits contributing to
carcinogenesis
In rodents, TCE binds to and/or
induces damage in DNA and
chromosome structure. TCE has a
limited ability to induce mutation in
bacterial systems. Oxidative
metabolites, particularly CH, can
cause a variety of genotoxic effects
(including aneuploidy) in available in
vitro and in vivo assays.
No TCE-specific studies
No TCE-specific studies;
mutagenicity is assumed to
cause cancer, as a
sufficient cause
Data are inadequate to support
a conclusion that a mutagenic
MOA mediated by CH is
operant; however, a mutagenic
MOA, mediated either by CH
or other oxidative metabolites
of TCE, cannot be ruled out.
Section 4.5.7.1 ; see also
4.2.1.5
PPARa activation:
• TCE oxidative metabolites
(e.g., TCA), after being
produced in the liver,
activate PPARa
• Alterations in cell
proliferation and apoptosis
• Clonal expansion of
initiated cells
TCE, TCA and DCA activate
PPARa, induce peroxisome
proliferation and hepatocyte
proliferation in mice and rats (e.g.,
DeAngelo et al., 2008; Dees and
Travis, 1994; Elcombe et al., 1985;
Goel et al., 1992; Goldsworthy and
Popp, 1987; Laughter et al., 2004;
Nakajima et al., 2000; Pereira, 1996;
Sanchez and Bull, 1990; Stauber and
Bull, 1997; Watanabe and Fukui,
2000).
No studies of TCE or its
metabolites (e.g., cancer
bioassays in PPARa-null
mice). TCE induces
increases in liver weight in
male and female mice lacking
a functional PPARa receptor
(Nakajima et al., 2000;
Ramdhan et al., 2010) and in
humanized null mice
(Ramdhan et al., 2010).
Liver tumor response from
WY dramatically diminished
in PPARa-null mice (Peters
et al., 1997); however, liver
tumor response from DEHP
unchanged in PPARa-null
mice (Ito et al., 2007). Thus,
inferences regarding TCE are
not possible.
No TCE-specific studies;
PPARa activation in a
transgenic mouse model
caused all the key events in
the MOA, but not
carcinogenesis, suggesting
that the MOA is not
sufficient for
carcinogenesis (Yang et al.,
2007). Consistent with
hypothesis that TCE liver
carcinogenesis involves
multiple mechanisms.
It is unlikely that PPARa
agonism and its sequellae
constitute the sole or
predominant MOA for TCE-
induced carcinogenesis.
Section 4.5.7.2
-------
Table 4-68. Summary of mode of action conclusions for TCE-induced liver carcinogenesis (continued)
Hypothesized MOA and key
events
Evidence that TCE or TCE
metabolites induces key events
Necessity of MOAs key
events for carcinogenesis
Sufficiency of MOA for
carcinogenesis
Summary and section
reference
Liver weight increases
TCE increases liver weight in mice,
rats and gerbils following acute,
short-term and subchronic exposures
(Berman et al., 1995; Buben and
O'Flaherty, 1985; Dees and Travis,
1993, 1994; Elcombe et al., 1985;
Goel et al., 1992; Goldsworthy and
Popp, 1987; Kjellstrand et al., 1983a;
Kjellstrand et al., 1983b; Kjellstrand
et al., 1981b; Laughter et al., 2004;
Melnick et al., 1987; Merrick et al.,
1989; Nakajima et al., 2000; Nunes et
al., 2001; Ramdhan et al., 2010; Tao
et al., 2000a; Tucker et al., 1982).
Similarity in dose-response between
liver weight increases at short-term
durations of exposure and liver tumor
induction observed from chronic
exposure to TCE and metabolites.
No TCE-specific studies
No TCE-specific studies
Data are inadequate to define
a MOA hypothesis for
hepatocarcinogenesis based on
liver weight increases.
Section 4.5.7.3.1
"Negative selection" confers a
growth advantage to initiated
cells
Transient DNA synthesis is confined
to a very small population of cells in
mouse liver (e.g., Dees and Travis,
1993; Elcombe et al., 1985; Laughter
et al., 2004).
No TCE-specific studies
No TCE-specific studies
Data are inadequate to define
a MOA hypothesis for
hepatocarcinogenesis based on
"negative selection".
Section 4.5.7.3.3
Polyploidization
Polyploidization likely occurs with
TCE exposure, although the evidence
is limited (Buben and O'Flaherty,
1985).
No TCE-specific studies
No TCE-specific studies
Although it is biologically
plausible that polyploidization
can contribute to liver
carcinogenesis, inadequate
data are available to support
this hypothesized MOA for
TCE.
Section 4.5.7.3.3
-------
Table 4-68. Summary of mode of action conclusions for TCE-induced liver carcinogenesis (continued)
Hypothesized MOA and key
events
Evidence that TCE or TCE
metabolites induces key events
Necessity of MOAs key
events for carcinogenesis
Sufficiency of MOA for
carcinogenesis
Summary and section
reference
Glycogen storage
DCA increases glycogen deposition
(Nelson et al., 1989); For TCE and
TCA, effects on glycogen were either
not reported (Dees and Travis, 1993;
Elcombe et al., 1985; Styles et al.,
1991) or were described as similar to
controls (Nelson et al., 1989).
No TCE-specific studies
No TCE-specific studies
Data are inadequate to define
a MOA hypothesis for TCE-
induced hepatocarcinogenesis
based on changes in glycogen
storage.
Section 4.5.7.3.4
Inactivation of GST-zeta
TCE prolongs DCA half-life in
rodents, suggesting that TCE may
inhibit GST-zeta, likely through the
formation of DCA (Schultz et al.,
2002).
No TCE-specific studies
No TCE-specific studies
Data are inadequate to define
a MOA hypothesis for TCE-
induced hepatocarcinogenesis
based on inactivation of GST-
zeta.
Section 4.5.7.3.5
Oxidative stress
Measures of oxidative stress induced
by TCE, TCA, and DCA either do
not occur, or are transient and do not
persistent with continued treatment
(Channel et al., 1998; Larson and
Bull, 1992b; Parrishetal., 1996).
No TCE-specific studies
No TCE-specific studies
Available data are limited to
support a role for oxidative
stress in TCE-induced liver
carcinogenesis.
Section 4.5.7.3.6
Epigenetic changes,
particularly DNA methylation,
induced by one or more
metabolites (TCA, DCA, and
other reactive species) advance
acquisition of multiple critical
traits contributing to
carcinogenesis
TCE, TCA and DCA decrease global
DNA methylation and promoter
hypomethylation (e.g., of c-myc) in
mouse liver (Ge et al., 2001b; Tao et
al., 1998; Tao et al., 2004a)
No TCE-specific studies
No TCE-specific studies
Although it is biologically
plausible that epigenetic
changes contribute to liver
carcinogenesis, inadequate
data are available to support
this hypothesized MOA for
TCE.
Section 4.5.7.3.7
-------
Table 4-68. Summary of mode of action conclusions for TCE-induced liver carcinogenesis (continued)
Hypothesized MOA and key
events
Evidence that TCE or TCE
metabolites induces key events
Necessity of MOAs key
events for carcinogenesis
Sufficiency of MOA for
carcinogenesis
Summary and section
reference
Cytotoxicity and reparative
hyperplasia:
• One or more reactive
intermediates induces
hepatotoxicity
• Reparative hyperplasia
ensues
TCE treatment at doses relevant to
liver carcinogenicity results in
relatively low cytotoxicity (Bull,
2004a; Bull et al., 2004).
No TCE-specific studies
No TCE-specific studies
It is unlikely that cytotoxicity
and reparative hyperplasia
play a significant role in TCE
carcinogenicity.
Section 4.5.7.3.8
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
question "2. Is the hypothesized mode of action sufficiently supported in the test animals?" is
"no" at this time. Consequently, the other key questions of "2. Is the hypothesized mode of
action relevant to humans ?" and "3. Which populations or lifestages can be particularly
susceptible to the hypothesized mode of action?" will not be discussed in a MOA-specific
manner. Rather, they are discussed below in more general terms, first qualitatively and then
quantitatively, using available relevant data.
4.5.7.1.16. Qualitative human relevance and susceptibility
No data exist that suggests that TCE-induced liver tumorigenesis is caused by processes
that irrelevant in humans. In addition, as discussed above, several of the other effects such as
polyploidization, changes in glycogen storage, and inhibition of GST-zeta—are either clearly
related to human carcinogenesis or areas of active research as to their potential roles. For
example, the effects of DCA on glycogen storage parallel the observation that individuals with
conditions that lead to glycogenesis appear to be at an increased risk of liver cancer (Adami et
al., 1996; La Vecchia et al., 1994; Rake et al., 2002; Wideroff et al., 1997). In addition, there
may be some relationship between the effects of DCA and the mechanism of increased liver
tumor risk in childhood in those with Type 1 hereditary tyrosinemia, though the hypotheses
needs to be tested experimentally. Similarly, with respect to PPARa activation and downstream
events hypothesized to be causally related to liver carcinogenesis, it is generally acknowledged
that "a point in the rat/mouse key events cascade where the pathway is biologically precluded in
humans cannot be identified, in principle" (Klaunig et al., 2003; NRC, 2006).
In terms of human relevance and susceptibility, it is also useful to briefly review what is
known about human HCC. A number of risk factors have been identified for human
hepatocellular carcinoma, including ethanol consumption, hepatitis B and C virus infection,
aflatoxin B1 exposure, and, more recently, diabetes and perhaps obesity (El-Serag and Rudolph,
2007). However, it is also estimated that a substantial minority of HCC patients, perhaps
15-50%, have no established risk factors (El-Serag and Rudolph, 2007). In addition, cirrhosis is
present in a large proportion of HCC patients, but the prevalence of HCC without underlying
cirrhosis, while not precisely known, is still significant, with estimates based on relatively small
samples ranging from 7-54% (Fattovich et al., 2004).
However, despite the identification of numerous factors that appear to play a role in the
human risk of HCC, the mechanisms are still largely unclear (Yeh et al., 2007). Interestingly,
the observation by Leakey et al. (2003a; 2003b) that body weight significantly and strongly
impacts background liver tumor rates in B6C3F1 mice parallels the observed epidemiologic
This document is a draft for review purposes only and does not constitute Agency policy.
407 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
associations between liver cancer and obesity (review in 2007). This concordance suggests that
similar pathways may be involved in spontaneous liver tumor induction between mice and
humans. The extent to which TCE exposure may interact with known risk factors for HCC
cannot be determined at this point, but several hypotheses can be posed based on existing data.
If TCE affects some of the same pathways involved in human HCC, as suggested in the
discussion of several TCE-induced effects above, then TCE exposure may lead a risk that is
additive to background.
As discussed above, there are several parallels between the possible key events in
TCE-induced liver tumors in mice and what is known about mechanisms of human HCC, though
none have been experimentally tested. Altered ploidy distribution and DNA hypomethylation
are commonly observed in human HCC (Calvisi et al., 2007; Lin et al., 2003; Zeppa et al., 1998).
Interestingly, El-Serag and Rudolph (2007) have been suggested that the risk of HCC increases
with cirrhosis in part because the liver parenchymal cells have decreased proliferative capacity,
resulting in an altered milieu that promotes tumor cell proliferation. This description suggests a
similarity in mode of action, though via different mechanisms, with the "negative selection"
hypothesis proposed by Bull (2000) for TCE and its metabolites although for TCE changes in
apoptosis and cell proliferation have not been noted or examined to such an extent to provide
evidence of a similar environment. Increased ploidy decreases proliferative capacity, so that
may be another mechanism through which the effects of TCE mimic the conditions thought to
facilitate the induction of human HCC.
In sum, from the perspective of hazard characterization, the available data support the
conclusion that the mode of action for TCE-induced mouse liver tumors is relevant to humans.
No data suggest that any of the key events are biologically precluded in humans, and a number of
qualitative parallels exist between hypotheses for the mode of action in mice and what is known
about the etiology and induction of human HCC. A number of risk factors have been identified
that appear to modulate the risk of human HCC, and these may also modulate the susceptibility
to the effects from TCE exposure. As noted in Section E.4, TCE exposure in the human
population is accompanied not only by external exposures to its metabolites, but brominated
analogues of those metabolites that are also rodent carcinogens, a number of chlorinate solvents
that are hepatocarcinogenic and alcohol consumption. The types of tumors and the heterogeneity
of tumors induced by TCE in rodents parallel those observed in humans (see Section E.3.1.8).
The pathways identified for induction of cancer in humans for cancer are similar to those for the
induction of liver cancer (see Section E.3.2.1). However, while risk factors have been identified
for human liver cancer that have similarities to TCE-induced effects and those of its metabolites,
This document is a draft for review purposes only and does not constitute Agency policy.
408 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
both the mechanism for human liver cancer induction and that for TCE-induced liver
carcinogenesis in rodents are not known.
4.5.7.1.17. Quantitative species differences
As a precursor to the discussion of quantitative differences between humans and rodents
and among humans, it should be noted that an adequate explanation for the difference in
response for TCE-liver cancer induction between rats and mice has yet to be established or for
that difference to be adequately described given the limitations in the rat database. For TCA,
there is only one available long-term study in rats that, while suggestive that TCA is less potent
in rats than mice, is insufficient to determine if there was a TCA-induced effect or what its
magnitude may be. While some have proposed that the lower rate of TCA formation in rats
relative to mice would explain the species difference, PBPK modeling suggests that the
differences (three-fivefold) may be inadequate to fully explain the differences in carcinogenic
potency. Moreover, inferences from comparing the effects of TCE and TCA on liver weight,
using PBPK model-based estimates of TCA internal dose-metrics as a result of TCE or TCA
administration, indicate that TCA is not likely to play a predominant role in hepatomegaly.
Combined with the qualitative correlation between rodent hepatomegaly and
hepatocarcinogenesis observed across many chemicals, this suggests that TCA similarly is not a
predominant factor in TCE-induced hepatocarcinogenesis. Indeed, there are multiple lines of
evidence that TCA is insufficient to account for TCE-induced tumors, including data on tumor
phenotype (e.g., c-Jun immunostaining) and genotype (e.g., H-ras mutation frequency and
spectrum). For DC A, only a single experiment in rats is available (reported in two publications),
and although it suggests lower hepatocarcinogenic potency in rats relative to mice, its relatively
low power limits the inferences that can be made as to species differences.
As TCA induces peroxisome proliferation in the mouse and the rat, some have suggested
that difference in peroxisomal enzyme induction is responsible for the difference in susceptibility
to TCA liver carcinogenesis. The study of DeAngelo et al. (1989) has been cited in the literature
as providing evidence of differences between rats and mice for peroxisomal response to TCA.
However, data from the most resistant strain of rat (Sprague-Dawley) have been cited in
comparisons of peroxisomal enzyme effects but the Osborne-Mendel and F344 rat were not
refractory and showed increased PCO activity so it is not correct to state that the rat is refractory
to TCA-induction of peroxisome activity (see Section E.2.3.1.5). In addition, as discussed
above, inferences based on PCO activity are limited by its high variability, even in control
animals, as well as its not necessarily being predictive of the peroxisome number or cytoplasmic
volume.
This document is a draft for review purposes only and does not constitute Agency policy.
409 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
The same assumption of lower species sensitivity by measuring peroxisome proliferation
has been applied to humans, as peroxisome proliferation caused by therapeutic PPARa agonists
such as fibrates in humans is generally lower (
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
the different pathways that may be responsible for this heterogeneity (Chen et al., 2002;
Feitelson et al., 2002; Yeh et al., 2007).
Appropriate quantitative data are generally lacking on interspecies differences in the
occurrence of most other proposed key events, although many have argued that there are
significant quantitative differences between rodents and humans related to PPARa activation
(Klaunig et al., 2003; NRC, 2006). For instance, it has been suggested that lower levels of
PPARa receptor in human hepatocytes relative to rodent hepatocytes contributes to lower human
sensitivity (Klaunig et al., 2003; Palmer et al., 1998; Tugwood et al., 1996). However, out of a
small sample of human livers (n = 6) show similar protein levels to mice (Walgren et al., 2000b).
Another proposed species difference has been ligand affinity, but while transactivation assays
showed greater affinity of Wy-14643 and perfluorooctanoic acid for rodent relative to human
PPARa, they showed TCA and DCA had a similar affinities between species (Maloney and
Waxman, 1999). Furthermore, it is not clear that receptor-ligand kinetics (capacity and affinity)
are rate-limiting for eliciting hepatocarcinogenic effects, as it is known that maximal receptor
occupation is not necessary for a maximal receptor mediated response (Stephenson, 1956) (see
also review by Danhof et al., 2007).
There is also limited in vivo and in vitro data suggesting that increases in cell
proliferation mediated by PPARa agonists are diminished in humans and other primates relative
to rodents (Hoivik et al., 2004; Klaunig et al., 2003; NRC, 2006). However, Walgren et al.
(2000a) reported that TCA and DCA were not mitogenic in either human or rodent hepatocytes
in vitro. Furthermore, TCE, TCA, and DCA all induce only transient increases in cell
proliferation, so the relevance to TCE of interspecies differences from PPARa agonists that to
produce sustained proliferation, such as Wy-14643, is not clear. In addition, comparisons
between primate and rodent models should take into account the differences in the ability to
respond to any mitogenic stimulation (see Section E.3.2). Primate and human liver respond
differently (and much more slowly) to a stimulus such as partial hepatectomy.
Recent studies in "humanized" mice (PPARa-null mice in which a human PPARa gene
was subsequently inserted and expressed in the liver) reported that treatment with a PPARa
agonist lead to greatly lower incidence of liver tumors as compared to wild-type mice (Morimura
et al., 2006). However, these experiments were performed with WY-14643 at a dose causing
systemic toxicity (reduced growth and survival), had a duration of less than 1 year, and involved
a limited number of animals. In addition, because liver tumors in mice at less than 1 year are
extremely rare, the finding a one adenoma in WY-14643-treated humanized mice suggests
carcinogenic potential that could be further realized with continued treatment (Keshava and
Caldwell, 2006). In addition, Yang et al. (2007) recently noted that let-7C, a microRNA
This document is a draft for review purposes only and does not constitute Agency policy.
411 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
involved in cell growth and thought to be a regulatory target of PPARa (Shah, 2008), was
inhibited by Wy-14643 in wild-type mice, but not in "humanized mice" in which had human
PPARa was expressed throughout the body on a PPARa-null background. However, these
humanized mice had about a 20-fold higher baseline expression of let-7C, as reported in control
mice, potentially masking any treatment effects. More generally, it is not known to what extent
PPARa-related events are rate-limiting in TCE-induced liver tumorigenesis, for which multiple
pathways appear to be operative. So even if quantitative differences mediated by PPARa were
well estimated, they would not be directly usable for dose-response assessment in the absence of
way to integrate the contributions from the different pathways.
In sum, the only quantitative data and inter- and intraspecies susceptibility suitable for
consideration in dose-response assessment are differences background liver tumor risk. These
may modulate the effects of TCE if relative risk, rather than additional risk, is the appropriate
common inter- and intraspecies metric. However, the extent to which relative risk would provide
a more accurate estimate of human risk is unknown.
4.6. IMMUNOTOXICITY AND CANCERS OF THE IMMUNE SYSTEM
Chemical exposures may result in a variety of adverse immune-related effects, including
immunosuppression (decreased host resistance), autoimmunity, and allergy-hypersensitivity, and
may result in specific diseases such as infections, systemic or organ-specific autoimmune
diseases, or asthma. Cell-mediated immune response, such as activation of macrophages, natural
killer cells, and cytokine production, can also influence a broader range of diseases, such as
cancer. Measures of immune function (e.g., T-cell counts, immunoglobulin [Ig] E levels,
specific autoantibodies, cytokine levels) may provide evidence of an altered immune response
that precedes the development of clinically expressed diseases. The first section of this section
discusses effects relating to immunotoxicity, including risk of autoimmune diseases, allergy and
hypersensitivity, measures of altered immune response, and lymphoid cancers. Studies
pertaining to effects in humans are presented first, followed by a section discussing relevant
studies in animals. The second section of this section discusses evidence pertaining to
trichloroethylene in relation to lymphoid tissue cancers, including childhood leukemia.
This document is a draft for review purposes only and does not constitute Agency policy.
412 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
4.6.1. Human Studies
4.6.1.1.1. Noncancer Immune-Related Effects
4.6.1.1.2. Immunosuppression, asthma, and allergies
In 1982, Lagakos et al. conducted a telephone survey of residents of Woburn,
Massachusetts, collecting information on residential history and history of 14 types of medically
diagnosed conditions (Lagakos et al., 1986). The survey included 4,978 children born since
1960 who lived in Woburn before age 19. Completed surveys were obtained from
approximately 57% of the town residences with listed phone numbers. Two of the wells
providing the town's water supply from 1964-1979 had been found to be contaminated with a
number of solvents, including tetrachloroethylene (21 ppb) and trichloroethylene (267 ppb) (as
cited in Lagakos et al., 1986). Lagakos et al. used information from a study by the
Massachusetts Department of Environmental Quality and Engineering to estimate the
contribution of water from the two contaminated wells to the residence of each participant, based
on zones within the town receiving different mixtures of water from various wells, for the period
in which the contaminated wells were operating. This exposure information was used to
estimate a cumulative exposure based on each child's length of residence in Woburn. A higher
cumulative exposure measure was associated with conditions indicative of immunosuppression
(e.g., bacterial or viral infections) or hypersensitivity (e.g., asthma). In contrast, a recent study
using the National Health and Nutrition Examination Survey data collected from 1999-2000 in a
representative sample of the U.S. population (n = 550) did not find an association between TCE
exposure and self-report of a history of physician-diagnosed asthma (OR: 0.94, 95% CI: 0.77,
1.14) (Arif and Shah, 2007). TCE exposure, as well as exposure to 9 other volatile organic
compounds, was determined through a passive monitor covering a period of 48-72 hours. No
clear trend was seen with self-reported wheeze episodes (OR: 1.29, 95% CI: 0.98, 1.68 for one to
two episodes; OR: 0.21, 95% CI: 0.04, 10.05 for three or more episodes in the past 12 months).
Allergy and hypersensitivity, as assessed with measures of immune system parameters or
immune function tests (e.g., atopy) in humans, have not been extensively studied with respect to
the effects of trichloroethylene (see Table 4-69). Lehmann et al. reported data pertaining to
immunoglobulin E (IgE) levels and response to specific antigens in relation to indoor levels of
volatile organic compounds among children (age 36 months) selected from a birth cohort study
in Leipzig, Germany (Lehmann et al., 2001). Enrollment into the birth cohort occurred between
1995 and 1996. The children in this allergy study represent a higher-risk group for development
of allergic disease, with eligibility criteria that were based on low birth weight (between 1,500
This document is a draft for review purposes only and does not constitute Agency policy.
399 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
and 2,500 g), or cord blood IgE greater than 0.9 kU/L with double positive family history of
atopy. These eligibility criteria were met by 429 children; 200 of these children participated in
the allergy study described below, but complete data (IgE and volatile organic compound
measurements) were available for only 121 of the study participants. Lehmann et al. measured
26 volatile organic compounds via passive indoor sampling in the child's bedroom for a period
of 4 weeks around the age of 36 months. The median exposure of trichloroethylene was 0.42
[j,g/m3 (0.17 (J,g/m3 and 0.87 (J,g/m3 for the 25th and 75th percentiles, respectively). Blood samples
were taken at the 36-month-study examination and were used to measure the total IgE and
specific IgE antibodies directed to egg white, milk, indoor allergens (house dust mites, cats, and
molds), and outdoor allergens (timothy-perennial grass and birch trees). There was no
association between trichloroethylene exposure and any of the allergens tested in this study,
although some of the other volatile organic compounds (e.g., toluene, 4-ethyltoluene) were
associated with elevated total IgE levels and with sensitization to milk or eggs.
4.6.1.1.3. Generalized hypersensitivity skin diseases, with or without hepatitis
Occupational exposure to trichloroethylene has been associated with a severe,
generalized skin disorder that is distinct from contact dermatitis in the clinical presentation of the
skin disease (which often involves mucosal lesions), and in the accompanying systemic effects
that can include lymphadenopathy, hepatitis, and other organ involvement. Kamijima et al.
recently reviewed case reports describing 260 patients with trichloroethylene-related generalized
This document is a draft for review purposes only and does not constitute Agency policy.
400 DRAFT—DO NOT CITE OR QUOTE
-------
to
Table 4-69. Studies of immune parameters (IgE antibodies and cytokines) and trichloroethylene in humans
Parameter,
source of data
Results
Reference, location, diagnosis
period, sample size, age
IgE antibodies
blood sample, indoor air sampling
of 28 volatile organic chemicals in
child's bedroom
Trichloroethylene exposure not associated with sensitization to
indoor or outdoor allergens
Lehmann et al. (2001)
Germany. 1997-1999. n = 121
36-mo old children
Cytokine secreting CD3+ T-cell
populations
cord blood, indoor air sampling of
28 volatile organic chemicals in
child's bedroom 4 wk after birth
In CD3+ cord blood cells, some evidence of association
between increasing trichloroethylene levels and
decreased IL-4 >75th percentile OR: 0.6 (95% CI: 0.2, 2.1),
<25th percentile OR 4.4 (95% CI: 1.1, 17.8)
increased IFN-y >75* percentile OR: 3.6 (95% CI: 0.9, 14.9)
<25th percentile OR: 0.7 (95% CI: 0.2, 2.2)
Similar trends not seen with tumor necrosis factor-a or IL-2
Lehmann et al. (2002)
Germany. 1995-1996. n = 85
newborns
Cytokine secreting CD3+ and
CD8+ T-cell populations
blood sample, indoor air sampling
of 28 volatile organic chemicals in
child's bedroom
Trichloroethylene exposure not associated with percentages of
IL-4 CD3+ or IFN-y CD8+ T-cells
Lehmann et al. (2001)
Germany. 1995-1999. n = 200
36-mo old children
Cytokine concentration—serum
urine sample (trichloroacetic acid
concentration), blood sample,
questionnaire (smoking history,
age, residence), workplace TCE
measures (personal samples, 4
exposed and 4 nonexposed
workers)
Nonexposed workers similar to office controls for all cytokine
measures. Compared to nonexposed workers, the
trichloroethylene exposed workers had
decreased IL-4 (mean 3.9 vs. 8.1 pg/mL)
increased IL-2 (mean 798 vs. 706 pg/mL)
increased IFN-y (mean 37.1 vs. 22.9 pg/mL)
Iavicoli et al. (2005)
Italy. n = 35 printers using TCE,
30 nonexposed workers (in
same factory, did not use or
were not near TCE), 40 office
worker controls. All men.
Mean age -33 yr
3^
rs
sj
3
a
>!
&
fs*
2^
Grj
>!
o
a
a
§•
>!
o
rs
o
a
>!
o
H
o
o
2!
a
"T3
H S.
W|
O •
p
o
c
o
H
W
H
o;
s
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
skin disorders (Kamijima et al., 2007). Six of the patients were from the United States or
Europe, with the remainder occurring in China, Singapore, Philippines, and other Asian
countries. One study in Guangdong province, in southeastern China, included more than 100 of
these cases in a single year (Huang et al., 2002). Kamijima et al. categorized the case
descriptions as indicative of hypersensitivity syndrome (n = 124) or a variation of erythema
multiforme, Stevens-Johnson syndrome, and toxic epiderma necrolysis (// = 115), with 21 other
cases unclassified in either category. The fatality rate, approximately 10%, was similar in the
two groups, but the prevalence of fever and lymphadenopathy was higher in the hypersensitivity
syndrome patients. Hepatitis was seen in 92-94% of the multiforme, Stevens-Johnson
syndrome, and toxic epiderma necrolysis patients, but the estimates within the hypersensitivity
syndrome group were more variable (46-94%) (Kamijima et al., 2007).
Some of the case reports reviewed by Kamijima et al. provided information on the total
number of exposed workers, working conditions, and measures of exposure levels. From the
available data, generalized skin disease within a worksite occurred in 0.25-13%) of workers in
the same location, doing the same type of work (Kamijima et al., 2007). The measured
3 3
concentration of trichloroethylene ranged from <50 mg/m to more than 4,000 mg/m , and
exposure scenarios included inhalation only and inhalation with dermal exposures. Disease
manifestation generally occurred within 2-5 weeks of initial exposure, with some intervals up to
3 months. Most of the reports were published since 1995, and the geographical distribution of
cases reflects the newly industrializing areas within Asia.
Kamijima and colleagues recently conducted an analysis of urinary measures of
trichloroethylene metabolites (trichloroacetic acid and trichloroethanol) in 25 workers
hospitalized for hypersensitivity skin disease in 2002 (Kamijima et al., 2008). Samples taken
within 15 days of the last exposure to trichloroethylene exposure were available for 19 of the
25 patients, with a mean time of 8.4 days. Samples from the other patients were not used in the
analysis because the half life of urinary trichloroacetic acid is 50-100 hours. In addition,
3-6 healthy workers doing the same type of work in the factories of the affected worker, and
2 control workers in other factories not exposed to trichloroethylene were recruited in
2002-2003 for a study of breathing zone concentration of volatile organochlorines and urinary
measures of trichloroethylene metabolites. Worksite measures of trichloroethylene concentration
were also obtained. Adjusting for time between exposure and sample collection, mean urinary
concentration at the time of last exposure among the 19 patients was 206 mg/mL for
trichloroacetic acid. Estimates for trichloroethanol were not presented because of the shorter
half-life for this compound. Urinary trichloroacetic acid levels in the healthy exposed workers
varied among the 4 factories, with means (±standard deviation [SD]) of 41.6 (±18.0),
This document is a draft for review purposes only and does not constitute Agency policy.
402 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
131 (±90.2), 180 (±92), and 395 (±684). The lower values were found in a factory in which the
degreasing machine had been partitioned from the workers after the illnesses had occurred.
Trichloroethylene concentrations (personal time-weighted averages) at the factories of the
affected workers ranged from 164-2,330 mg/m (30-431 ppm). At the two factories with no
affected workers in the past 3 years, the mean personal time-weighted average trichloroethylene
3 3
concentrations were 44.9 mg/m (14 ppm) and 1,803 mg/m (334 ppm). There was no
commonality of additives or impurities detected among the affected factories that could explain
the occurrence of the hypersensitivity disorder.
To examine genetic influences on disease risk, Dai et al. conducted a case-control study
of 111 patients with trichloroethylene-related severe generalized dermatitis and
152 trichloroethylene-exposed workers who did not develop this disease (Dai et al., 2004).
Patients were recruited from May 1999 to November 2003 in Guangdong Province, and were
employed in approximately 80 electronic and metal-plating manufacturing plants. Initial
symptoms occurred within 3 months of exposure. The comparison group was drawn from the
same plants as the cases, and had worked for more than 3 months without development of skin or
other symptoms. Mean age in both groups was approximately 23 years. A blood sample was
obtained from study participants for genotyping of TNF-a, TNF-P, and interleukin (IL)-4
genotypes. The genes were selected based on the role of TNF and of interleukin-4 in
hypersensitivity and inflammatory responses. The specific analyses included two
polymorphisms in the promoter region of TNF-a (G —~ A substitution at position -308)
designated as TNF All, with wild-type designated TNFAI; and aG^A substitution at position
-238), a polymorphism at the first intron on TNF-P, and a polymorphism in the promoter region
of IL-4 (C —~ T substitution at -590). There was no difference in the frequency of the
—238
TNF-a , TNF-P, or IL-4 polymorphisms between cases and controls, but the wild-type
—308
TNF-a genotype was somewhat more common among cases (TNF A I/I genotype 94% in
cases and 86% in controls).
Kamijima et al. note the similarities, particular with respect to specific skin
manifestations, of the case presentations of trichloroethylene-related generalized skin diseases to
conditions that have been linked to specific medications (e.g., carbamazepine, allopurinol,
antibacterial sulfonamides), possibly in conjunction with reactivation of specific latent herpes
viruses (Kamijima et al., 2007). A previous review by these investigators discusses insights with
respect to drug metabolism that may be useful in developing hypotheses regarding susceptibility
to trichloroethylene-related generalized skin disorders (Nakajima et al., 2003). Based on
consideration of metabolic pathways and intermediaries, variability in CYP2E1,
UDP-glucoronyltransferase, glutathione-S transferase, and N-acetyl transferase (NAT) activities
This document is a draft for review purposes only and does not constitute Agency policy.
403 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
could be hypothesized to affect the toxicity of trichloroethylene. NAT2 is most highly expressed
in liver, and the "slow" acetylation phenotype (which arises from a specific mutation) has been
associated with adverse effects of medications, including drug-induced lupus (Lemke and
McQueen, 1995) and hypersensitivity reactions (Spielberg, 1996). There are limited data
pertaining to genetic or other sources of variability in these enzymes on risk of trichloroethylene-
related generalized skin diseases, however. In a study in Guangdong province, CYP1A1,
GSTM1, GSTP1, GSTT1, and NAT2 genotypes in 43 cases of trichloroethylene-related
generalized skin disease were compared to 43 healthy trichloroethylene-exposed workers (Huang
et al., 2002). The authors reported that the NAT2 slow acetylation genotype was associated with
disease, but the data pertaining to this finding were not presented.
4.6.1.1.4. Cytokine profiles and lymphocyte subsets
Cytokines are produced by many of the immune regulatory cells (e.g., macrophages,
dendritic cells), and have many different effects on the immune system. The T-helper Type 1
(Thl) cytokines, are characterized as "pro-inflammatory" cytokines, and include TNF-a and
interferon (IFN)-y. Although this is a necessary and important part of the innate immune
response to foreign antigens, an aberrant pro-inflammatory response may result in a chronic
inflammatory condition and contribute to development of scarring or fibrotic tissue, as well as to
autoimmune diseases. Th2 cytokines are important regulators of humoral (antibody-related)
immunity. IL-4 stimulates production of IgE and thus influences IgE-mediated effects such as
allergy, atopy, and asthma. Th2 cytokines can also act as "brakes" on the inflammatory
response, so the balance between different types of cytokine production is also important with
respect to risk of conditions resulting from chronic inflammation. Several studies have examined
cytokine profiles in relation to occupational or environmental TCE exposure (see Table 4-69).
The 2001 Lehmann et al. study of 36-month old children (described above) also included
a blood sample taken at the 3-year study visit, which was used to determine the percentages of
specific cytokine producing T-cells in relation to the indoor volatile organic compounds
exposures measured at birth. There was no association between trichloroethylene exposure and
either IL-4 CD3+ or IFN-y CD8+ T-cells (Lehmann et al., 2001).
Another study by Lehmann et al. examined the relationship between indoor exposures to
volatile organic compounds and T-cell subpopulations measured in cord blood of newborns
(Lehmann et al., 2002). The study authors randomly selected 85 newborns (43 boys and
42 girls) from a larger cohort study of 997 healthy, full-term babies, recruited between 1997 and
1999 in Germany. Exclusion criteria included a history in the mother of an autoimmune disease
This document is a draft for review purposes only and does not constitute Agency policy.
404 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
or infectious disease during the pregnancy. Twenty-eight volatile organic compounds were
measured via passive indoor sampling in the child's bedroom for a period of 4 weeks after the
birth (a period which is likely to reflect the exposures during the prenatal period close to the time
of delivery). The levels were generally similar or slightly higher than the levels seen in the
previous study using samples from the bedrooms of the 36-month-old children. The highest
3 3
levels of exposure were seen for limonene (median 24.3 (J,g/m ), a-pinene (median 19.3 (J,g/m )
3 3
and toluene (median 18.3 |ig/m ), and the median exposure of trichloroethylene was 0.6 (J,g/m
(0.2 (J,g/m3 and 1.0 (J,g/m3 for the 25th and 75th percentiles, respectively). Flow cytometry was
used to measure the presence of CD3 T-cells obtained from the cord blood labeled with
antibodies against IFN-y, tumor necrosis factor-a, IL-2, and IL-4. There was some evidence of a
decreased level of IL-2 with higher trichloroethylene exposure in the univariate analysis, with
median percentage of IL-2 cells of 46.1 and 33.0% in the groups that were below the 75th
percentile and above the 75th percentile of trichloroethylene exposure, respectively. In analyses
adjusting for family history of atopy, gender and smoking history of the mother during
pregnancy, there was little evidence of an association with either IL-2 or IFN-y, but there was a
trend of increasing trichloroethylene levels associated with decreased IL-4 and increased IFN-y.
Iavicoli et al. examined cytokine levels in 35 trichloroethylene-exposed workers (Group
A) from a printing area of a factory in Italy. Their work involved use of trichloroethylene in
degreasing (Iavicoli et al., 2005). Two comparison groups were included. Group B consisted of
30 other factory workers who were not involved in degreasing activities and did not work near
this location, and Group C consisted of 40 office workers at the factory. All study participants
were male and had worked at their present position for at least 3 years, and all were considered
healthy. Personal breathing zone air samples from four workers in Group A and four workers in
Group B were obtained in three consecutive shifts (24 total samples) to determine air
concentration of trichloroethylene. A urine sample was obtained from each Group A and Group
B worker (end of shift at end of work week) for determination of trichloroacetic acid
concentrations (corrected for creatinine), and blood samples were collected for assessment of
IL-2, IL-4, and IFN-y concentrations in serum using enzyme-linked immunosorbent assays.
"3
Among exposed workers, the mean trichloroethylene concentration was approximately 35 mg/m
(30.75 ± SD 9.9, 37.75 ± 23.0, and 36.5 ± 8.2 mg/m3 in the morning, evening, and night shifts,
respectively). The urinary trichloroacetic acid concentrations were much higher in exposed
workers compared with nonexposed workers (mean ± SD, Group A 13.3 ± 5.9 mg/g creatinine;
Group B 0.02 ± 0.02 mg/g creatinine). There was no difference in cytokine levels between the
two control groups, but the exposed workers differed significantly (all ^-values < 0.01 using
Dunnett's test for multiple comparisons) from each of the two comparison groups. The observed
This document is a draft for review purposes only and does not constitute Agency policy.
405 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
differences were a decrease in IL-4 levels (mean 3.9, 8.1, and 8.1 pg/mL for Groups A, B, and C,
respectively), and an increase in IL-2 levels (mean 798, 706, and 730 pg/mL for Groups A, B,
and C, respectively) and in IFN-y levels (mean 37.1, 22.9, and 22.8 pg/mL for Groups A, B, and
C, respectively).
The available data from these studies (Iavicoli et al., 2005; Lehmann et al., 2001;
Lehmann et al., 2002) provide some evidence of an association between increased
trichloroethylene exposure and modulation of immune response involving an increase in pro-
inflammatory cytokines (IL-2, IFN-y) and a decrease in Th2 (allergy-related) cytokines (e.g., IL-
4). These observations add support to the influence of trichloroethylene in immune-related
conditions affected by chronic inflammation.
Lan et al. (2010) examined lymphocyte subsets among 80 TCE-exposed workers and
96 controls in Guangdong, China. Six factories using TCE for cleaning metals, optical lenses, or
circuit boards were included in this study. These factories did not use other solvents (benzene,
styrene, ethylene oxide, formaldehyde, or epichlorohydrin), based on an exposure screening
using Drager tubes and 3M Badges. Eighty workers from these factories and 96 unexposed
controls (frequency matched by sex and 5-year age groups to controls) from clothes
manufacturers, a food production factory, and a hospital, were included in the study. The study
was conducted in 2006. Study participants provided a blood sample, buccal cells, postshift and
overnight urine samples, and completed a questionnaire with demographic, alcohol and smoking
history, and occupational history data. A blood sample was used for a complete blood count and
differential lymphocyte subset analysis. At the time of the blood draw, a clinical examination,
including measurement of height and weight, and symptoms of recent respiratory infection
(which could affect the differential blood cell counts) was conducted. TCE monitoring was
conducted using full-shift personal air exposure measurements. The median level of exposure,
based on the mean of two measurements taken for each participant in the month before the blood
draw, among the 80 TCE-exposed workers was 12 ppm. The analysis used this level to
categorize workers into high (>12 ppm; mean 38 ppm) and low (<12 ppm; mean 5 ppm)
exposures. Among the controls, the mean TCE exposure was <0.03 ppm. The total number of
lymphocytes, T cells, CD4+ T cells, CD8+ T cells, B cells and natural killer (NK) cells was
significantly lower among TCE-exposed workers compared with controls, with the largest
decrease seen in the higher exposure group. For example, the age- and sex-adjusted lymphocyte
count was 2,154, 2,012, and 1,671 cells/|iL blood in the controls, <12 and >12 ppm groups,
respectively (trendp = <0.0001). Plasma concentrations of soluble CD27 and CD30, two
costimulators involved in the regulation of T cells, were also decreased in both exposure groups
compared with controls. Similar patterns were seen when limited to the 77 workers with
This document is a draft for review purposes only and does not constitute Agency policy.
406 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
exposure levels <100 ppm, and when limited to the 60 workers with exposure levels <25 ppm.
Granulocytes, monocytes and platelet counts did not differ by exposure. The authors note that
the immunosuppression and decreased lymphocyte activation seen in this study provide support
the biological plausibility of a role of TCE exposure in NHL.
4.6.1.1.5. Autoimmune disease
4.6.1.1.5.1.Disease clusters and geographic-based studies
Reported clusters of diseases have stimulated interest in environmental influences on
systemic autoimmune diseases. These descriptions include investigations into reported clusters
of systemic lupus erythematosus (Balluz et al., 2001; Dahlgren et al., 2007) and Wegener
granulomatosis (Albert et al., 2005). Wegener granulomatosis, an autoimmune disease involving
small vessel vasculitis, usually with lung or kidney involvement, is a very rare condition, with an
incidence rate of 3-14 per million per year (Mahr et al., 2006). Trichloroethylene was one of
several ground water contaminants identified in a recent study investigating a cluster of seven
cases of Wegener granulomatosis around Dublin, Pennsylvania. Because of the multiple
contaminants, it is difficult to attribute the apparent disease cluster to any one exposure.
In addition to the study of asthma and infectious disease history among residents of
Woburn, Massachusetts (Lagakos et al., 1986) (see Section 4.6.1.1.1), Byers et al. provide data
pertaining to immune function from 23 family members of leukemia patients in Woburn,
Massachusetts (Byers et al., 1988). Serum samples were collected in May and June of 1984 and
in November of 1985 (several years after 1979, when the contaminated wells had been closed).
Total lymphocyte counts and lymphocyte subpopulations (CD3, CD4, and CD8) and the
CD4/CD8 ratio were determined in these samples, and in samples from a combined control
group of 30 laboratory workers and 40 residents of Boston selected through a randomized
probability area sampling process. The study authors also assessed the presence of antinuclear
antibodies (ANA) or other autoantibodies (antismooth muscle, antiovarian, antithyroglobulin,
and antimicrosomal antibodies) in the family member samples and compared the results with
laboratory reference values. The age distribution of the control group, and stratified analyses by
age, are not provided. The lymphocyte subpopulations (CD3, CD4, and CD8) were higher and
the CD4/CD8 ratio was lower in the Woburn family members compared to the controls in both
of the samples taken in 1984. In the 1985 samples, however, the lymphocyte subpopulation
levels had decreased and the CD4/CD8 ratio had increased; the values were no longer
statistically different from the controls. None of the family member serum samples had
This document is a draft for review purposes only and does not constitute Agency policy.
407 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
antithyroglobulin or antimicrosomal antibodies, but 10 family-member serum samples (43%) had
ANA (compared to <5% expected based on the reference value). Because the initial blood
sample was taken in 1984, it is not possible to determine the patterns at a time nearer to the time
of the exposure. The coexposures that occurred also make it difficult to infer the exact role of
trichloroethylene in any alterations of the immunologic parameters.
Kilburn and Warshaw reported data from a study of contamination by metal-cleaning
solvents (primarily trichloroethylene) and heavy metals (e.g., chromium) of the aquifer of the
Santa Cruz River in Tucson, Arizona (Kilburn and Warshaw, 1992b). Exposure concentrations
above 5 ppb (6-500 ppb) had been documented in some of the wells in this area. A study of
neurological effects was undertaken between 1986 and 1989 (Kilburn and Warshaw, 1993b), and
two of the groups within this larger study were also included in a study of symptoms relating to
systemic lupus erythematosus. Residents of Tucson (n = 362) were compared to residents of
southwest Arizona (n = 158) recruited through a Catholic parish. The Tucson residents were
selected from the neighborhoods with documented water contamination (>5 ppb
trichloroethylene for at least 1 year between 1957 and 1981). Details of the recruitment strategy
are not clearly described, but the process included recruitment of patients with lupus or other
rheumatic diseases (Kilburn and Warshaw, 1992b, 1993b). The prevalence of some self-reported
symptoms (malar rash, arthritis/arthralgias, Raynaud syndrome, skin lesions, and seizure or
convulsion was significantly higher in Tucson, but there was little difference between the groups
in the prevalence of oral ulcers, anemia, low white blood count or low platelet count, pleurisy,
alopecia, or proteinuria. The total number of symptoms reported was higher in Tucson than in
the other southwest Arizona residents (14.3 vs. 6.4% reported four or more symptoms,
respectively). Low-titer (1:80) ANA were seen in 10.6 and 4.7% of the Tucson and other
Arizona residents, respectively (p = 0.013). However, since part of the Tucson study group was
specifically recruited based on the presence of rheumatic diseases, it is difficult to interpret these
results.
4.6.1.1.5.2.Case-control studies
Interest in the role of organic solvents, including trichloroethylene, in autoimmune
diseases was spurred by the observation of a scleroderma-like disease characterized by skin
thickening, Raynaud's phenomenon, and acroosteolysis and pulmonary involvement in workers
exposed to vinyl chloride (Gama and Meira, 1978). A case report in 1987 described the
occurrence of a severe and rapidly progressive case of systemic sclerosis in a 47-year-old woman
This document is a draft for review purposes only and does not constitute Agency policy.
408 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
who had cleaned X-ray tubes in a tank of trichloroethylene for approximately 2.5 hours (Lockey
etal., 1987).
One of the major impediments to autoimmune disease research is the lack of disease
registries, which make it difficult to identify incident cases of specific diseases. There are no
cohort studies of the incidence of autoimmune diseases in workers exposed to trichloroethylene.
Most of the epidemiologic studies of solvents and autoimmune disease rely on general measures
of occupational exposures to solvents, organic solvents, or chlorinated solvents exposures. A
two- to threefold increased risk of systemic sclerosis (scleroderma) (Aryal et al., 2001; Garabrant
et al., 2003; Maitre et al., 2004), rheumatoid arthritis (Lundberg et al., 1994; Sverdrup et al..
2005). undifferentiated connective tissue disease (Lacey et al., 1999), and antineutrophil-
cytoplasmic antibody (ANCA)-related vasculitis (Beaudreuil et al., 2005; Lane et al., 2003) has
generally been seen in these studies, but there was little evidence of an association between
solvent exposure and systemic lupus erythematosus in two recent case-control studies (Cooper et
al., 2004; Finckh et al., 2006).
Two case-control studies of scleroderma (Bovenzi et al., 2004; Maitre et al., 2004) and
two of rheumatoid arthritis (Olsson et al., 2004; Olsson et al., 2000) provide data concerning
solvent exposure that occurred among metal workers or in jobs that involved cleaning metal (i.e.,
types of jobs which were likely to use trichloroethylene as a solvent). There was a twofold
increased risk among male workers in the two studies of rheumatoid arthritis from Sweden
(Olsson et al., 2004; Olsson et al., 2000). The results from the smaller studies of scleroderma
were more variable, with no exposed cases seen in one study with 93 cases and 206 controls
(Maitre et al., 2004), and an odds ratio of 5.2 (95% CI: 0.7, 37) seen in a study with 56 cases and
171 controls (Bovenzi et al., 2004).
Five other case-control studies provide data specifically about trichloroethylene exposure,
based on industrial hygienist review of job history data (see Table 4-70). Three of these studies
are of scleroderma (Diot et al., 2002; Garabrant et al., 2003; Nietert et al., 1998), one is of
undifferentiated connective tissue disease (Lacey et al., 1999), and one is of small vessel
vasculitidies involving ANCAs (Beaudreuil et al., 2005).
These studies included some kind of expert review of job histories, but only two studies
included a quantification of exposure (e.g., a cumulative exposure metric, or a "high" exposure
group) (Diot et al., 2002; Nietert et al., 1998). Most of the studies present data stratified by sex,
and as expected, the prevalence of exposure (either based on type of job or on industrial
hygienist assessment) is considerably lower in women compared with men. In men the studies
generally reported odds ratios between 2.0 and 8.0, and in women, the odds ratios were between
1.0 and 2.0. The incidence rate of scleroderma in the general population is approximately
This document is a draft for review purposes only and does not constitute Agency policy.
409 DRAFT—DO NOT CITE OR QUOTE
-------
1 5-10 times higher in women compared with men, which may make it easier to detect large
2 relative risks in men.
This document is a draft for review purposes only and does not constitute Agency policy.
410 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-70. Case-control studies of autoimmune diseases with measures of trichloroethylene exposure
to
Disease, source of data
Results:
exposure prevalence, OR, 95% CI
Reference, location,
sample size, age
Scleroderma
Structured interview
(specific jobs and
materials; jobs held 1 or
more years). Exposure
classified by self-report
and by expert review (job
exposure matrix).
Men
Maximum intensity 30% cases, 10% controls OR: 3.3 (95% CI: 1.0,
10.3)
Cumulative intensity 32% cases, 21% controls OR: 2.0 (95% CI: 0.7, 5.3)
Maximum probability 16% cases, 3% controls OR: 5.1 (95% CI: not
calculated)
Women
Maximum intensity 6% cases, 7% controls OR: 0.9 (95% CI: 0.3, 2.3)
Cumulative intensity 10% cases, 9% controls OR: 1.2 (95% CI: 0.5, 2.6)
Maximum probability 4% cases, 5% controls OR: 0.7 (95% CI: 0.2, 2.2)
Nietert et al. (1998)
South Carolina. Prevalent
cases, 178 cases (141
women, 37 men), 200
hospital-based controls.
Mean age at onset 45.2 yr
Structured interview
(specific jobs and
materials; jobs held 6 or
more months). Exposure
classified by expert
review.
Men and women
Any exposure: cases 16%, controls 8% OR: 2.4 (95% CI: 1.0, 5.4)
High exposure:51 cases 9%, controls 1% OR: 7.6 (95% CI: 1.5, 37.4)
Men
Any exposure: cases 64%, controls 27% OR: 4.7 (95% CI: 0.99, 22.0)
Women
Any exposure: cases 9%, controls 4% OR: 2.1 (95% CI: 0.65, 6.8)
Diot et al. (2002)
France. Prevalent cases, 80
cases (69 women, 11
men), 160 hospital
controls. Mean age at
diagnosis 48 yr
Structured interview
(specific jobs and
materials; jobs held 3 or
more months). Exposure
classified by self-report
and by expert review.
Women
Self report: cases 1.3%, controls 0.7% OR: 2.0 (95% CI: 0.8, 4.8)
Expert review: cases 0.7%, controls 0.4% OR: 1.9 (95% CI: 0.6, 6.6)
Garabrant et al. (2003)
Michigan and Ohio.
Prevalent cases, 660 cases
(all women), 2,227
population controls.13
Ages 18 and older
Undifferentiated connective tissue disease
cs
sj
3
a
>!
&
fs*
si
O
Grj
>!
o
a
a
§•
>!
o
rs
o
a
>!
o
H
o
o
2;
a
"13
H £.
W|
O •
p
o
c
o
H
W
H
o;
s
-------
Structured interview
(specific jobs and
materials; jobs held 3 or
more months). Exposure
classified by self-report
and by expert review.
Women
Self report: cases 0.5%, controls 0.7% OR: 0.88 (95% CI: 0.11, 6.95)
Expert review: cases 0.5%, controls 0.4% OR: 1.67 (95% CI: 0.19, 14.9)
Lacey et al. (1999),
Michigan and Ohio.
Prevalent cases, 205 cases
(all women), 2,095
population controls.
Ages 18 and older
Table 4-70. Case-control studies of autoimmune diseases with measures of trichloroethylene exposure (continued)
Disease, source of data
Results:
exposure prevalence, OR, 95% CI
Reference, location, sample
size, age
ANCA-related diseases0
Structured interview (specific
jobs and materials; jobs held 6
or more months). Exposure
classified by expert review.
Men and women (data not presented separately by sex)
cases 18.3%, controls 17.5% OR: 1.1 (95% CI: 0.5, 2.4)
Beaudreuil et al. (2005)
France. Incident cases, 60
cases (~50%> women), 120
hospital controls. Mean age
61 yr
aCumulative exposure defined as product of probability x intensity x frequency x duration scores, summed across all jobs; scores of >1 classified as "high."
bTotal ri: n with TCE data: self -report 606 cases, 2,138 control; expert review 606 cases, 2,137 controls.
Diseases included Wegener glomerulonephritis (n = 20), microscopic polyangiitis (n = 8), pauci-immune glomerulonephritis (n = 10), uveitis (n = 6),
Churg-Strauss syndrome (n = 4), stroke (n = 4) and other diseases (no more than 2 each).
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
The EPA conducted a meta-analysis of the three scleroderma studies with specific
measures of trichloroethylene (Diot et al., 2002; Garabrant et al., 2003; Nietert et al., 1998),
examining separate estimates for males and for females. The resulting combined estimate for
"any" exposure, using a random effects model to include the possibility of nonrandom error
between studies (DerSimonian and Laird, 1986), was OR: 2.5 (95% CI: 1.1, 5.4) for men and
OR: 1.2 (95% CI: 0.58, 2.6) in women. (Because the "any" exposure variable was not included
in the published report, Dr. Paul Nietert provided the EPA with a new analysis with these results,
e-mail communication from Paul Nietert to Glinda Cooper, November 28, 2007.)
Specific genes may influence the risk of developing autoimmune diseases, and genes
involving immune response (e.g., cytokines, major histocompatibility complex, B- and T-cell
activation) have been the focus of research pertaining to the etiology of specific diseases. The
metabolism of specific chemical exposures may also be involved (Cooper et al., 1999).
Povey et al. (2001) examined polymorphisms of two cytochrome CYP genes, CYP2E1 and
CYP2C19, in relation to solvent exposure and risk of developing scleroderma. These specific
genes were examined because of their hypothesized role in metabolism of many solvents,
including trichloroethylene. Seven scleroderma patients who reported a history of solvent
exposure were compared to 71 scleroderma patients with no history of solvent exposure and to
106 population-based controls. The CYP2E1*3 allele and the CYP2E1*4 allele were more
common in the 7 solvent-exposed patients (each seen in 2 of the 7 patients; 29%) than in either
of the comparison groups (approximately 5% for CYP2E1*3 and 14% for CYP2E1*4). The
authors present these results as observations that require a larger study for corroboration and
further elucidation of specific interactions.
4.6.1.1.6. Cancers of the Immune System, Including Childhood Leukemia
4.6.1.1.7. Description of studies
Human studies have reported cancers of the immune system resulting from TCE
exposure. Lymphoid tissue neoplasms arise in the immune system and result from events that
occur within immature lymphoid cells in the bone marrow or peripheral blood (leukemias), or
more mature cells in the peripheral organs (NHL). As such, the distinction between lymphoid
leukemia and NHL is largely distributional with overlapping entities, such that a particular
lymphoid neoplasm may manifest both lymphomatous and leukemic features during the course
of the disease (Weisenburger, 1992). The broad category of lymphomas can be divided into
specific types of cancers, including non-Hodgkin lymphoma, Hodgkin lymphoma, multiple
This document is a draft for review purposes only and does not constitute Agency policy.
413 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
myeloma, and various types of leukemia (e.g., acute and chronic forms of lymphoblastic and
myeloid leukemia). The classification criteria for these cancers has changed over the past 30
years, reflecting improved understanding of the underlying stem cell origins of these specific
subtypes. Lymphomas are grouped according to the World Health Organization (WHO)
classification as B-cell neoplasms, T-cell/NK-cell neoplasms, and Hodgkin lymphoma, formerly
known as Hodgkin disease (Harris et al., 2000). For example, hairy cell leukemia, chronic
lymphocytic leukemia, non-Hodgkin lymphoma, and multiple myeloma may arise from mature
B cells and are types of NHLs according to the WHO's lymphoma classification system (Morton
et al., 2007, 2006). Most of the studies of TCE exposure evaluate NHL defined as
lymphosarcoma, reticulum-cell sarcoma, and other lymphoid tissue neoplasms with recently
published studies reporting on total B-cell or specific B-cell neoplasms.
Numerous studies are found in the published literature on NHL and either broad exposure
categories or occupational title. The NHL studies generally report positive associations with
organic solvents or job title as aircraft mechanic, metal cleaner or machine tool operator, and
printers, although associations are not observed consistently across all studies, specific solvents
are not identified, and different lymphoma classifications are adopted ('t Mannetje et al., 2008;
Alexander et al., 2007b; Blair et al., 1993; Boffetta and de Vocht, 2007; Chiu and Weisenburger,
2003; Cocco et al., 2010; Dryver et al., 2004; Figgs et al., 1995; Karunanayake et al., 2008;
Lynge et al., 1997; Orsi et al., 2010; Richardson et al., 2008; Schenk et al., 2009; Seidler et al.,
2007; Tatham et al., 1997; Vineis et al., 2007; Wang et al., 2009). A major use of TCE is the
degreasing, as vapor or cold state solvent, of metal and other products with potential exposure in
jobs in the metal industry, printing industry and aircraft maintenance or manufacturing industry
(Bakke et al., 2007). The recent NHL case-control study of Purdue et al. (2009) examined
degreasing tasks, specifically, and reported an increasing positive trend between NHL risk in
males and three degreasing exposure surrogates: average frequency (hours/year) (p = 0.02),
maximal frequency (hours/year), (p = 0.06), or cumulative number of hours (p = 0.04).
As described in Appendix B, the EPA conducted a thorough and systematic search of
published epidemiological studies of cancer risk and trichloroethylene exposure using the
PubMed, TOXNET®, and EMBASE® bibliographic database. The EPA also requested
unpublished data pertaining to trichloroethylene from studies that may have collected these data
but did not include it in their published reports. ATSDR and state health department peer-
reviewed studies were also reviewed. Information from each of these studies relating to
specified design and analysis criteria was abstracted. These criteria included aspects of study
design, representativeness of study subjects, participation rate/loss to follow-up, latency
considerations, potential for biases related to exposure misclassification, disease
This document is a draft for review purposes only and does not constitute Agency policy.
414 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
misclassification, and surrogate information, consideration of possible confounding, and
approach to statistical analysis. All studies are considered for hazard identification but those
studies more fully meeting the objective criteria provided the greater weight for identifying a
cancer hazard.
The body of evidence on NHL and trichloroethylene is comprised of occupational cohort
studies, population-based case-control studies and geographic studies. Four case-control studies
and four geographic studies also examine childhood leukemia and trichloroethylene. Most
studies report observed risk estimates and associated confidence intervals for NHL and overall
TCE exposure. The studies included a broad but sometimes slightly different group of
lymphosarcoma, reticulum-cell sarcoma, and other lymphoid tissue neoplasms, with the
exception of the Nordstrom et al. (1998) case-control study, which examined hairy cell leukemia,
now considered a NHL, the Zhao et al. (2005) cohort study, which reported only results for all
lymphohematopoietic cancers, including nonlymphoid types and excluding chronic lymphocytic
leukemia, and the Greenland et al. (1994) nested case-control study which reported results for
NHL and Hodgkin lymphoma combined. Persson and Fredrikson (1999) do not identify the
classification system for defining NHL, and Hardell et al. (1994) define NHL using the
Rappaport classification system. Miligi et al. (2006) used an NCI classification system and
considered chronic lymphocytic leukemias and NHL, classified as lymphosarcoma,
reticulosarcoma, and other lymphoid tissue neoplasms, together, while Cocco et al. (2010), used
the World Health Organization classification system, which reclassifies lymphocytic leukemias
and NHLs as lymphomas of B-cell or T-cell origin. EPA staff, additionally, was able to obtain
results generally consistent with the traditional NHL definition from Dr. Cocco, although
lymphomas not otherwise specified were excluded (personal communication, Pierluigi Cocco to
Cheryl Siegel Scott). The cohort studies (except for Zhao et al., 2005) and the nested case-
control study of Greenland et al. (1994) have some consistency in coding NHL, with NHL
defined as lymphosarcoma and reticulum-cell sarcoma (ICD code 200) and other lymphoid tissue
neoplasms (ICD code 202) using the ICD Revisions 7, 8, or 9. Revisions 7 and 8 are essentially
the same with respect to NHL; under Revision 9, the definition of NHL was broadened to
include some neoplasms previously classified as Hodgkin lymphomas (Banks, 1992). Wang et
al. (2009) refer to their cases as "NHL" cases and according to the ICD-0 classification system
that they used, their cases are more specifically NHL subtypes such as diffuse, lymphosarcoma,
or follicular lymphoma (9590-9642, 9690-9701) or mast cell tumors (9740-9750) which is
consistent with the traditional definition of NHL (i.e., ICD-7, -8, -9 codes 200 + 202) (Morton et
al., 2003). NHL cases in Purdue et al. (2011) were also classified according to ICD-0 (2nd
rd
Edition converted to ICO-O 3 Edition codes), included diffuse, follicular T-cell and all other
This document is a draft for review purposes only and does not constitute Agency policy.
415 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
NHL subtypes, which is generally consistent with the traditional definition of NHL, although this
grouping does not include the malignant lymphomas of unspecified type coded as M-9590-9599.
Fewer studies presented in published papers this information for cell-specific lymphomas,
leukemia, leukemia cell type, or multiple myeloma (Anttila et al., 1995; Axelson et al., 1994;
Boice et al., 1999; Boice et al., 2006b; Cocco et al., 2010; Costantini et al., 2008; Gold et al., In
Press; Hansen et al., 2001; Morgan et al., 1998; Raaschou-Nielsen et al., 2001; Radican et al.,
2008).
The seven cohort studies with data on the incidence of lymphopoietic and hematopoietic
cancer in relation to trichloroethylene exposure range in size (803 (Hansen et al., 2001) to 86,868
(Chang et al., 2005)), and were conducted in Denmark, Sweden, Finland, Taiwan and the United
States (see Table 4-71; for additional study descriptions, see Appendix B). Some subjects in the
Hansen et al. study are also included in a study reported by Raaschou-Nielsen et al. (2003);
however, any contribution from the former to the latter are minimal given the large differences in
cohort sizes of these studies (Hansen et al., 2001; Raaschou-Nielsen et al., 2003). The exposure
assessment techniques used in all studies except Chang et al. (2005) and Sung et al. (2007)
included a detailed job exposure matrix (Blair et al., 1998; Zhao et al., 2005), biomonitoring data
(Anttila et al., 1995; Axelson et al., 1994; Hansen et al., 2001), or reference to industrial hygiene
records on TCE exposure patterns and factors that affected exposure, indicating a high
probability of TCE exposure potential (Raaschou-Nielsen et al., 2003) with high probability of
TCE exposure to individual subjects. Subjects in Chang et al. (2005) and Sung et al. (2007), two
studies with overlapping subjects employed at an electronics plant in Taiwan, have potential
exposure to several solvents including TCE; all subjects are presumed as "exposed" because of
employment in the plant although individual subjects would be expected to have differing
exposure potentials. The lack of attribution of exposure intensity to individual subjects yields a
greater likelihood for exposure misclassification compared to the six other studies with exposure
assessment approaches supported by information on job titles, tasks, and industrial hygiene
monitoring data. Incidence ascertainment in two cohorts began 21 (Blair et al., 1998) and
38 years (Zhao et al., 2005) after the inception of the cohort. Specifically, Zhao et al. (2005)
noted that their results may not accurately reflect the effects of carcinogenic exposure that
resulted in nonfatal cancers before 1988. Because of the issues concerning case ascertainment
raised by this incomplete coverage, observations must be interpreted in light of possible bias
reflecting incomplete ascertainment of incident cases.
Eighteen cohort or PMR studies describing mortality risks from lymphopoietic and
hematopoietic cancer are summarized in Table 4-72 (for additional study descriptions, see
Appendix B). Two studies examined cancer incidence, Radican et al. (2008), who updated
This document is a draft for review purposes only and does not constitute Agency policy.
416 DRAFT—DO NOT CITE OR QUOTE
-------
1 mortality in Blair et al. (1998) cohort, and Zhao et al. (2005), and are identified above. In 10 of
2 the 18 studies presenting mortality risks (ATSDR, 2004a; Blair et al., 1989; Chang et al., 2003b;
3 Clapp and Hoffman, 2008; Costa et al., 1989; Garabrant et al., 1988; Henschler et al., 1995;
4 Sinks et al., 1992; Sung et al., 2007; Wilcosky et al., 1984)
This document is a draft for review purposes only and does not constitute Agency policy.
417 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-71. Incidence cohort studies of TCE exposure and lymphopoietic and hematopoietic cancer risk
Population
exposure group
Lymphopoietic
cancer
non-Hodgkin
lymphoma
Leukemia
Multiple
myeloma
Reference(s) and study
description13
Relative risk
(95% CI)a
wa
Relative risk
(95% CI)a
tf
Relative risk
(95% CI)a
wa
Relative risk
(95% CI)a
na
Aerospace workers (Rocketdyne), CA
Zhao et al. (2005)
Any TCE exposure
Not reported
Not reported
n = 5,049 (2,689 with high
cumulative TCE exposure),
began work before 1980,
worked at least 2 yr, alive with
no cancer diagnosis in 1988,
follow-up from 1988-2000,
job exposure matrix
(intensity), internal referents
(workers with no TCE
exposure). Leukemia and
multiple myeloma
observations included in non-
Hodgkin lymphoma category.
Low cumulative TCE
score
1.0 (referent)
28
Medium cumulative
TCE score
0.88 (0.47, 1.65)
16
High cumulative TCE
score
0.20 (0.03, 1.46)
1
(p for trend)
(0.097)
Electronic workers, Taiwan
Chang et al. (2005); Sung et
al. (2007)
All employees
0.67 (0.42, 1.01)
22
n = 88,868 (n = 70,735
female), follow-up
1979-1997, does not identify
TCE exposure to individual
subjects (Chang et al., 2005).
Males
0.73 (0.27, 1.60)
6
Not reported
Not reported
Not reported
Females
0.65 (0.37, 1.05)
16
Not reported
Not reported
Not reported
Females
0.78(0.49, 1.17)
23
Not reported
n = 63,982 females, follow-up
1979-2001, does not identify
TCE exposure to individual
subjects (Sung et al., 2007).
-------
Table 4-71. Incidence cohort studies of TCE exposure and lymphopoietic and hematopoietic cancer risk
(continued)
Population
exposure group
Lymphopoietic
cancer
non-Hodgkin
lymphoma
Leukemia
Multiple
myeloma
Reference(s) and study
description13
Relative risk
(95% CI)a
tf
Relative risk
(95% CI)
M3
Relative risk
(95% CIf
wa
Relative risk
(95% CI)a
na
Blue-collar workers, Denmark
Raaschou-Nielsen et al. (2003)
Any exposure
1.1 (1.0, 1.6)
229
1.2 (1.0, 1.5)
96
1.2 (0.9, 1.4)
82
1.03 (0.70, 1.47)
31
n = 40,049 (14,360 with
presumed higher level
exposure to TCE), worked for
at least 3 mo, follow-up from
1968-1997, documented TCE
use0. EPA based the
lymphopoietic cancer category
on summation of ICD codes
200-204.
Subcohort w/higher
exposured
Not reported
1.5 (1.2,2.0)
65
Not reported
Not reported
Employment duration
1-4.9 yr
1.5 (1.1,2.1)
35
>5 yr
1.6 (1.1,2.2)
30
Biologically-monitored workers, Denmark
Hansen etal. (2001)
Any TCE exposure
2.0(1.1,3.3)
15
3.1 (1.3,6.1)
8
2.0 (0.7, 4.4)
6
0.71 (0.02, 3.98)
1
n = 803, U-TCA or air TCE
samples, follow-up
1968-1996 (subset of
Raaschlou-Nielsen et al.
(2003) cohort). EPA based the
lymphopoietic cancer category
on summation of ICD codes
200-204.
Cumulative exposure
(Ikeda), males
Not reported
Not reported
Not reported
<17 ppm-yr
3.9(0.8, 11)
3
>17 ppm-yr
3.1 (0.6, 9.1)
3
Mean concentration
(Ikeda), males
Not reported
Not reported
Not reported
<4 ppm
3.9(1.1, 10)
4
4+ppm
3.2(1.1, 10)
4
Employment duration,
males
Not reported
Not reported
Not reported
<6.25 yr
2.5 (0.3, 9.2)
2
>6.25 yr
4.2(1.1, 11)
4
-------
Table 4-71. Incidence cohort studies of TCE exposure and lymphopoietic and hematopoietic cancer risk
(continued)
Population
exposure group
Lymphopoietic
cancer
non-Hodgkin
lymphoma
Leukemia
Multiple
myeloma
Reference(s) and study
description13
Relative risk
(95% CI)a
tf
Relative risk
(95% CI)
M3
Relative risk
(95% CIf
wa
Relative risk
(95% CI)a
na
Aircraft maintenance workers, Hill Air Force Base, UT
Blair etal. (1998)
TCE Subcohort
Not reported
Not reported
Not reported
Not reported
n = 10,461 men and
3,605 women (total
n = 14,066, n = 7,204 with
TCE exposure), employed at
least 1 yrfrom 1952-1956,
follow-up 1973-1990, job
exposure matrix (intensity),
internal referent (workers with
no chemical exposures).
Males, cumulative
exposure
0
1.0 (referent)
1.0 (referent)
1.0 (referent)
1.0 (referent)
9
<5 ppm-yr
0.8 (0.4, 1.7)
12
0.9 (0.3, 2.6)
8
0.4(0.1,2.0)
2
0.8 (0.1, 12.7)
1
5-25 ppm-yr
0.7(0.3, 1.8)
7
0.7 (0.2, 2.6)
4
0
3.8 (0.4, 37.4)
3
>25 ppm-yr
1.4(0.6,2.9)
17
1.0 (0.4, 2.9)
7
0.9 (0.2, 3.7)
4
5.1 (0.6, 43.7)
5
Females, cumulative
exposure
0
1.0 (referent)
1.0 (referent)
1.0 (referent)
1.0 (referent)
<5 ppm-yr
1.2(0.3,4.4)
3
0.6 (0.1, 5.0)
1
0
Not reported
2
5-25 ppm-yr
1.9 (0.4, 8.8)
2
0
2.4(0.3,21.8)
1
Not reported
1
>25 ppm-yr
0.9(9.2,3.3)
3
0.9 (0.2, 4.5)
2
0
Not reported
1
Biologically-monitored workers, Finland
Anttila et al. (1995)
Any TCE exposure
1.51 (0.92,2.33)
20
1.81 (0.78, 3.56)
8
1.08 (0.35, 2.53)
5
1.62 (0.44, 4.16)
4
n = 3,089 men and women, U-
TCA samples, follow-up
1967-1992.
Mean air-TCE (Ikeda extrapolation)
<6 ppm
1.36 (0.65, 2.49)
10
2.01 (0.65, 4.69)
5
0.39(0.01,2.19)
1
1.48 (0.18, 5.35)
2
6+ ppm
2.08 (0.95, 3.95)
9
1.40 (0.17, 5.04)
2
2.65 (0.72, 6.78)
4
2.41 (0.29, 6.78)
2
-------
Table 4-71. Incidence cohort studies of TCE exposure and lymphopoietic and hematopoietic cancer risk
(continued)
Population
exposure group
Lymphopoietic
cancer
non-Hodgkin
lymphoma
Leukemia
Multiple
myeloma
Reference(s) and study
description13
Relative risk
(95% CI)a
tf
Relative risk
(95% CI)
M3
Relative risk
(95% CIf
wa
Relative risk
(95% CI)a
na
Biologically-monitored workers, Sweden
Axelsonetal. (1994)
Males
1.17(0.47,2.40)
1
1.56 (0.51,3.64)
5
Not reported
0.57 (0.01,3.17)
1
n= 1,421 men and 249 women
(total 1,670), U-TCA samples,
follow-up 1958-1987. EPA
based the lymphopoietic
cancer category includes ICD-
7 200-203.
0-17 ppm (Ikeda
extrapolation)
Not reported
1.44 (0.30, 4.20)
3
Not reported
18-35 ppm (Ikeda
extrapolation)
(0, 8.58)
0
>36 ppm (Ikeda
extrapolation)
6.25 (0.16, 34.8)
1
Females
Not reported
Not reported
Not reported
"n = number of observed cases.
Standardized incidence ratios using an external population referent group unless otherwise noted.
°Exposure assessment based on industrial hygiene data on TCE exposure patterns and factors that affect such exposure (Raaschou-Nielsen et al., 2002), with high
probability of TCE exposure potential to individual subjects. Companies included iron and metal (48%), electronics (11%), painting (11%), printing (8%),
chemical (5%), dry cleaning (5%), and other industries.
dDefined as at least 1 yr duration and first employed before 1980.
-------
Table 4-72. Mortality cohort and PMR studies of TCE exposure and lymphopoietic and hematopoietic
cancer risk
Population,
exposure group
Lymphopoietic cancer
non-Hodgkin
lymphoma
Leukemia
Multiple myeloma
Reference(s) and study
description13
Relative risk
(95% CI)
tf
Relative risk
(95% CI)
wa
Relative risk
(95% CI)
tf
Relative risk
(95% CI)
na
Computer manufacturing workers (IBM), NY
Clapp and Hoffman (2008)
Males
2.24(1.01,4.19)
9
Not reported
Not reported
Not reported
3
n= 115 cancer deaths from
1969-2001, proportional
cancer mortality ratio, does
not identify TCE exposure
to individual subjects. EPA
based the lymphopoietic
cancer category on "all
lymphatic cancers."
Females
Not reported
0
Not reported
Not reported
Not reported
0
Aerospace workers (Rocketdyne), CA
Boice et al. (2006b)
Any TCE (utility/eng
flush)
0.74 (0.34, 1.40)
9
0.21 (0.01, 1.18)
1
1.08 (0.35,2.53)
5
0.50 (0.01, 2.77)
1
n = 41,351 (1,111 Santa
Susana workers with any
TCE exposure), employed
on or after 1948-1999,
worked >6 mo, follow-up to
1999, job exposure matrix
without quantitative
estimate of TCE intensity.
-------
Table 4-72. Mortality cohort and PMR studies of TCE exposure and lymphopoietic and hematopoietic cancer
risk (continued)
Population,
exposure group
Lymphopoietic cancer
non-Hodgkin
lymphoma
Leukemia
Multiple myeloma
Reference(s) and study
description13
Relative risk
(95% CI)
tf
Relative risk
(95% CI)
tf
Relative risk
(95% CI)
rf
Relative risk
(95% CI)
na
Aerospace workers (Rocketdyne), CA (continued)
Zhao et al. (2005)
Any TCE exposure
Not reported
Not reported
60
Not reported
Not reported
n = 6,044 (n = 2,689 with
high cumulative level
exposure to TCE), began
work and worked at least 2
yr in 1950 or later - 1993,
follow-up to 2001, job
exposure matrix (intensity),
internal referents (workers
with no TCE exposure).
Leukemia and multiple
myeloma observations
included in non-Hodgkin
lymphoma category.
Low cumulative
TCE score
1.0 (referent)
27
Medium
cumulative TCE
score
1.49 (0.86, 2.57)
27
High TCE score
1.30 (0.52,3.23)
6
(p for trend)
(0.370)
View-Master employees, OR
ATSDR (2004a)
Males
0.58 (0.11, 1.69)
3
0.69 (0.08, 2.49)
2
0.50 (0.01, 2.79)
1
n = 616 deaths from
1989-2001, proportional
mortality ratio, does not
identify TCE exposure to
individual subjects. EPA
based the non-Hodgkin
lymphoma cancer category
on "other lymphopoietic
tissue" which included NHL
and multiple myeloma.
Females
0.64 (0.28, 1.26)
8
0.52 (0.14, 1.33)
4
0.67(0.14, 1.96)
3
-------
Table 4-72. Mortality cohort and PMR studies of TCE exposure and lymphopoietic and hematopoietic cancer
risk (continued)
Population,
exposure group
Lymphopoietic cancer
non-Hodgkin
lymphoma
Leukemia
Multiple myeloma
Reference(s) and study
description13
Relative risk
(95% CI)
tf
Relative risk
(95% CI)
tf
Relative risk
(95% CI)
rf
Relative risk
(95% CI)
na
Electronic workers, Taiwan
Chang et al. (2003b)
All employees
n = 88,868 (« = 70,735
female), began work
1978-1997, follow-up
1985-1997, does not
identify TCE exposure to
individual subjects.
Males
Not reported
1.27 (0.41, 2.97)
5
0.44 (0.05, 1.59)
2
Not reported
Females
Not reported
1.14(0.55,2.10)
10
0.54 (0.23, 1.07)
8
Not reported
Aerospace workers (Lockheed), CA
Boice et al. (1999)
Routine TCE
n = 77,965 (« = 2,267 with
routine TCE exposure and
n = 3.016 with intermittent-
routine TCE exposure),
began work >1960, worked
at least 1 yr, follow-up from
1960-1996, job exposure
matrix without quantitative
estimate of TCE intensity.
Any TCE exposure
1.5 (0.81, 1.60)
36
1.19 (0.65, 1.99)
14
1.05 (0.54, 1.84)
12
0.91 (0.34, 1.99)
6
Routine -intermittent
Any TCE exposure
Not reported
Not reported
Not reported
Exposure duration
Not reported
Not reported
0 yr
1.0 (referent)
32
1.0 (referent)
24
<1 yr
0.74 (0.32, 1.72)
7
0.45 (0.13, 1.54)
3
1-4 yr
1.33 (0.64, 2.78)
10
1.48 (0.64, 3.41)
8
>5 yr
1.62 (0.82, 3.22)
14
0.51 (0.15, 1.76)
3
p for trend
0.20
>0.20
-------
Table 4-72. Mortality cohort and PMR studies of TCE exposure and lymphopoietic and hematopoietic cancer
risk (continued)
Population,
exposure group
Lymphopoietic cancer
non-Hodgkin
lymphoma
Leukemia
Multiple myeloma
Reference(s) and study
description13
Relative risk
(95% CI)
tf
Relative risk
(95% CI)
tf
Relative risk
(95% CI)
rf
Relative risk
(95% CI)
na
Uranium-processing workers (Fernald), OH
Ritz (1999a)
Any TCE exposure
Not reported
Not reported
Not reported
Not reported
n = 3,814 (n = 2,971 with
TCE), began work
1951-1972, worked >3 mo,
follow-up to 1989, internal
referents (workers with no
TCE exposure).
No TCE exposure
1.0 (referent)
Not reported
Not reported
Not reported
Light TCE
exposure, >2 yr
1.45 (0.68, 3.06)°
18
Not reported
Not reported
Not reported
Moderate TCE
exposure, >2 yr
1.17(0.15, 9.00)°
1
Not reported
Not reported
Not reported
Aerospace workers (Hughes), CA
Morgan et al. (1998)
TCE subcohort
0.99 (0.64, 1.47)
25
0.96 (0.20, 2.81)d
3
1.05 (0.50, 1.93)
10
1.08 (0.35, 2.53)e
5
n = 20,508 (4,733 with TCE
exposure), worked >6 mo
1950-1985, follow-up to
1993, external and internal
(all non-TCE exposed
workers) workers referent,
job exposure matrix
(intensity).
TCE subcohort
1.01 (0.46, 1.92)e
9
Low intensity
(<50 ppm)
1.07 (0.51, 1.96)
10
1.79 (0.22, 6.46)d
2
0.85 (0.17, 2.47)
3
High intensity
(>50 ppm)
0.95 (0.53, 1.57)
15
0.50 (0.01, 2.79)d
1
1.17(0.47, 2.41)
7
TCE subcohort (Cox Analysis)
Never exposed
1.0 (referent)
82
1.0 (referent)
8
1.0 (referent)
32
Ever exposed
1.05 (0.67, 1.65)f
25
1.36 (0.35, 5.22)4 f
3
0.99 (0.48, 2.03)f
10
Peak
No/Low
1.0 (referent)
90
1.0 (referent)
9
1.0 (referent)
35
Medium/High
1.08 (0.64, 1.82)
17
1.31 (0.28, 6.08)d
2
1.10(0.49,2.49)
7
-------
Table 4-72. Mortality cohort and PMR studies of TCE exposure and lymphopoietic and hematopoietic cancer
risk (continued)
Lymphopoietic cancer
non-Hodgkin
lymphoma
Leukemia
Multiple myeloma
Population,
exposure group
Relative risk
(95% CI)
tf
Relative risk
(95% CI)
tf
Relative risk
(95% CI)
rf
Relative risk
(95% CI)
na
Reference(s) and study
description13
Aerospace workers (Hughes), CA (continued)
Cumulative
Referent
1.0 (referent)
82
1.0 (referent)
8
1.0 (referent)
32
Low
1.09 (0.56, 2.14)
10
2.25 (0.46, ll.l)d
2
0.69 (0.21, 2.32)
3
High
1.03 (0.59, 1.79)
15
0.81 (0.10, 6.49)d
1
1.14(0.5,2.60)
7
Aircraft maintenance workers, Hill Air Force Base, UT
Blair et al. (1998); Radican
et al. (2008)
TCE subcohort
1.1 (0.7, 1.8)8
66
2.0 (0.9, 4.6)8
28
0.6 (0.3, 1.2)8
16
1.3 (0.5,3.4)
14
n = 14,066 (n = 7,204 ever
Males, cumulative exposure
exposed to TCE), employed
at least 1 yr from
1952-1956, follow-up to
1990 (Blair et al., 1998) or
to 2000 (Radican et al.,
0
1.0 (referent)
1.0 (referent)
1.0 (referent)
1.0 (referent)
<5 ppm-yr
1.1 (0.6,2.1)
21
1.8 (0.6, 5.4)
10
1.0 (0.3, 3.2)
7
1.0(0.2,4.2)
4
5-25 ppm-yr
1.0 (0.4,2.1)
11
1.9 (0.6,6.3)
6
0
0.8(0.1,4.4)
2
2008), job exposure matrix,
internal referent (workers
with no chemical
>25 ppm-yr
1.3 (0.7,2.5)
21
1.1 (0.3,3.8)
5
1.2 (0.4, 3.6)
7
1.2(0.3,4.7)
4
Females, cumulative exposure
exposures).
0
1.0 (referent)
1.0 (referent)
1.0 (referent)
<5 ppm-yr
1.5 (0.6,4.0)
6
3.8 (0.8, 18.9)
3
0.4(0.1,3.2)
1
3.2 (0.5, 19.8)
2
5-25 ppm-yr
0.7 (0.1,4.9)
1
0
0
4.3 (0.4, 23.4)
1
>25 ppm-yr
1.1 (0.4,3.0)
6
3.6 (0.8, 16.2)
4
0.3 (0.1,2.4)
1
1.3 (0.1, 13.2)
1
TCE subcohort
1.06 (0.75, 1.51) h
106
1.36 (0.77, 2.39)h
46
0.64 (0.35, 1.18)h
27
1.35 (0.62, 2.93)
25
-------
Table 4-72. Mortality cohort and PMR studies of TCE exposure and lymphopoietic and hematopoietic cancer
risk (continued)
Population,
exposure group
Lymphopoietic cancer
non-Hodgkin
lymphoma
Leukemia
Multiple myeloma
Reference(s) and study
description13
Relative risk
(95% CI)
tf
Relative risk
(95% CI)
tf
Relative risk
(95% CI)
rf
Relative risk
(95% CI)
na
Aircraft maintenance workers, Hill Air Force Base, UT (continued)
Males, cumulative
exposure
1.12(0.72, 1.73)
88
1.56 (0.79, 4.21)
37
0.77 (0.37, 1.62)
24
1.08 (0.43,2.71)
19
0
1.0 (referent)
1.0 (referent)
1.0 (referent)
1.0 (referent)
<5 ppm-yr
1.04 (0.63, 1.74)
34
1.83 (0.79, 4.21)
18
0.86 (0.36, 2.02)
11
0.69 (0.21, 2.27)
5
5-25 ppm-yr
1.06 (0.49, 1.88)
21
1.17(0.42,3.24)
7
0.51 (0.16, 1.63)
4
1.58 (0.53,4.71)
7
>25 ppm-yr
1.25 (0.75, 2.09)
33
1.50 (0.61,3.69)
12
0.87 (0.35,2.14)
9
1.19 (0.40,3.54)
7
Females, cumulative
exposure
1.00 (0.55, 1.83)
18
1.18 (0.49, 2.85)
9
0.36 (0.10, 1.32)
3
2.37 (0.67, 8.44)
6
0
1.0 (referent)
1.0 (referent)
1.0 (referent)
1.0 (referent)
<5 ppm-yr
1.10(0.48, 2.54)
7
1.48 (0.47, 4.66)
4
0.35 (0.05, 2.72)
1
2.20 (0.40, 12.02)
2
5-25 ppm-yr
0.38 (0.05, 2.79)
1
0
0
2.79 (0.31,25.05)
1
>25 ppm-yr
1.11 (0.53,2.31)
10
1.30 (0.45, 3.77)
5
0.48 (0.10, 2.19)
2
2.38 (0.53, 10.67)
3
Cardboard manufacturing workers, Arnsburg, Germany
Henschler et al. (1995)
TCE-exposed
subjects
1.10(0.03,6.12)
1
n = 169 TCE exposed and
n = 190 unexposed men,
employed >1 yr from
1956-1975, follow-up to
1992, local population
referent, qualitative
exposure assessment.
Unexposed subjects
from same factory
1.11 (0.03,6.19)
1
-------
Table 4-72. Mortality cohort and PMR studies of TCE exposure and lymphopoietic and hematopoietic cancer
risk (continued)
Population,
exposure group
Lymphopoietic cancer
non-Hodgkin
lymphoma
Leukemia
Multiple myeloma
Reference(s) and study
description13
Relative risk
(95% CI)
tf
Relative risk
(95% CI)
tf
Relative risk
(95% CI)
rf
Relative risk
(95% CI)
na
General Electric plant, Pittsfield, MA
0.76 (0.24, 2.42)''J
15
1.1 (0.46,2.66)'
22
Greenland et al. (1994)
Nested case-control study,
n = 512 cancer (cases) and
1,202 noncancer (controls)
male deaths reported to
pension fund between
1969-1984 among workers
employed <1984 and with
job history record, job
exposure matrix-ever held
job with TCE exposure.
Hodgkin lymphoma in NHL
grouping.
Cardboard manufacturing workers, Atlanta, GA
Sinks et al. 1992
0.3 (0.0, 1.6)
1
Not reported
Not reported
Not reported
n = 2,050, employed on or
before 1957-1988, follow-
up to 1988, Material Data
Safety Sheets used to
identify chemicals used in
work areas.
-------
Table 4-72. Mortality cohort and PMR studies of TCE exposure and lymphopoietic and hematopoietic cancer
risk (continued)
Population,
exposure group
Lymphopoietic cancer
non-Hodgkin
lymphoma
Leukemia
Multiple myeloma
Reference(s) and study
description13
Relative risk
(95% CI)
tf
Relative risk
(95% CI)
tf
Relative risk
(95% CI)
rf
Relative risk
(95% CI)
na
U.S. Coast Guard employees
Blair et al. (1989)
Marine inspectors
1.57 (0.91,2.51)
17
1.75 (0.48, 4.49)
4
1.55 (0.62, 3.19)
1
Not reported
n = 3,781 males (1,767
marine inspectors),
employed 1942-1970,
follow-up to 1980. TCE and
nine other chemicals
identified as potential
exposures; no exposure
assessment to individual
subjects.
Noninspectors
0.60 (0.24, 1.26)
7
0.41 (0.01,2.30)
1
0.66(0.14, 1.94)
3
Not reported
Aircraft manufacturing employees, Italy
Costa et al. (1989)
All male subjects
0.80 (0.41, 1.40)
12
Not reported
Not reported
Not reported
n = 7,676 males, employed
on or before 1954-1981,
followed to 1981, job titles
of white- and blue-collar
workers, technical staff, and
administrative clerks, does
not identify TCE exposure
to individual subjects.
Aircraft manufacturing, San Diego, CA
Garabrant et al. (1988)
All employees
0.82 (0.56, 1.15)
32
0.82 (0.44, 1.4l)d
13
0.82 (0.47, 1.32)
10
Not reported
n = 14,067, employed at
least 4 yr with company and
>1 d at San Diego plant
from 1958-1982, followed
to 1982, does not identify
TCE exposure to individual
subjects.
0.65 (0.21, 1.52)k
5
-------
L/l
K>
Table 4-72. Mortality cohort and PMR studies of TCE exposure and lymphopoietic and hematopoietic cancer
risk (continued)
Lymphopoietic cancer
non-Hodgkin
lymphoma
Leukemia
Multiple myeloma
Population,
exposure group
Relative risk
(95% CI)
tf
Relative risk
(95% CI)
tf
Relative risk
(95% CI)
rf
Relative risk
(95% CI)
na
Reference(s) and study
description13
Solvent-exposed rubber workers
Wilcosky et al. (1984)
2.41
3
0.81
3
Nested case-control study,
n = 9 lymphosarcoma and
10 leukemia (cases) and
20% random sample of all
other deaths (controls)
between 1964-1973 in
cohort of n = 6,678,
exposure assessment by
company record for use in
work area.
3^
cs
sj
3
a
>!
&
rv «
<=> §
to
>!
o
a
a
§•
>!
o
rs
o
a
>!
o
H
o
o
2;
a
"T3
H S.
W|
O •
p
o
c
o
H
W
H
o;
s
"n = number of observed cases.
bUnless otherwise noted, all studies reported standardized mortality ratios using an external population referent group.
°Logistic regression analysis with 15 lag for TCE exposure (Ritz, 1999a).
dIn Morgan et al. (1998) and Garabrant et al. (1988), this category was based on lymphosarcoma and reticulosarcoma.
eAs presented in Mandel et al. (2006) for NHL, this category defined as ICD-7, ICDA-8, and ICD-9 codes of 200 and 202. As presented in Alexander et al.
(2006) for multiple myeloma.
fRisk ratio from Cox Proportional Hazard Analysis, stratified by age and sex, from Environmental Health Strategies (1997) Final Report to Hughes Corporation
(Communication from Paul A. Cammer, President, Trichloroethylene Issues Group to Cheryl Siegel Scott, U.S. EPA, December 22, 1997).
8 Estimated relative risks from Blair et al. (1998) from Poisson regression models adjusted for date of hire, calendar year of death and sex.
h Estimated relative risks from Radican et al. (2008) from Cox proportional hazard models adjusted for age and sex.
1 Odds ratio from nested case-control analysis.
lymphomas, lymphosarcomas, reticulosarcomas, and Hodgkin lymphoma (ICDA-8 200-202) in Greenland et al. (1994).
kOther lymphatic and hematopoietic tissue neoplasms (Garabrant et al., 1988).
IBM = International Business Machines Corporation.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
a relatively limited exposure assessment methodology was used, study participants may not
represent the underlying population, or there was a low exposure prevalence of TCE exposure.
For reasons identified in the systematic review, these studies are given less weight in the overall
evaluation of the literature than the eight other cohort studies that better met the ideals of
evaluation criteria (Blair et al., 1998 and extended follow-up by; Boice et al., 1999; Boice et al.,
2006b; Greenland et al., 1994; Morgan et al., 1998; Radican et al., 2008; Ritz, 1999a; Zhao et al.,
2005).
Case-control studies of NHL from United States (Connecticut), Germany, Italy, Sweden,
and Canada were identified, and are summarized in Table 4-73 (for additional study descriptions,
see Appendix B). These studies identified cases from hospital records (Cocco et al., 2010;
Costantini et al., 2008; Hardell et al., 1994; Mester et al., 2006; Miligi et al., 2006; Persson and
Fredrikson, 1999; Seidler et al., 2007; Siemiatycki, 1991); the Surveillance, Epidemiology, and
End Results Cancer Registry—Connecticut residents (Wang et al., 2009), Iowa, Los Angeles
County, and Seattle and Detroit metropolitan area residents (Purdue et al., 2011), or Seattle and
Detroit metropolitan area residents (Gold et al., In Press); or the Swedish Cancer Registry
(Nordstrom et al., 1998), and hospital or population controls. These studies assign potential
occupational TCE exposure to cases and controls using self-reported information obtained from a
mailed questionnaire (Hardell et al., 1994; Nordstrom et al., 1998; Persson and Fredrikson, 1999)
or from direct interview with study subjects, with industrial hygienist ratings of exposure
potential and a job exposure matrix (Cocco et al., 2010; Costantini et al., 2008; Miligi et al.,
2006; Purdue et al., 2011; Seidler et al., 2007; Siemiatycki, 1991; Wang et al., 2009).
Additionally, large multiple center lymphoma case-control studies examine specific types of
NHL (Cocco et al., 2010; Miligi et al., 2006; Purdue et al., 2011; Wang et al., 2009), leukemia
(Costantini et al., 2008), or multiple myeloma (Cocco et al., 2010; Costantini et al., 2008; Purdue
etal.,2011).
Four geographic based studies on NHL in adults are summarized in Table 4-74 (for
additional study descriptions, see Appendix B) and subjects in three studies are identified based
upon their residence in a community where TCE was detected in water serving the community
(ATSDR, 2006b; Cohn et al., 1994b; Vartiainen et al., 1993). Both Cohn et al. (1994b) and
ATSDR (2006b) also present estimates for childhood leukemia and these observations are
discussed below with other studies reporting on childhood leukemia. A subject is assumed to
have a probability of exposure due to residence likely receiving water containing TCE. Most
studies do not include statistical models of water distribution networks, which may influence
TCE concentrations delivered to a home, nor a subject's ingestion rate to estimate TCE exposure
This document is a draft for review purposes only and does not constitute Agency policy.
431 DRAFT—DO NOT CITE OR QUOTE
-------
1 to individual study subjects. ATSDR (2006b) adopts exposure modeling of soil vapor
2 contamination to define
This document is a draft for review purposes only and does not constitute Agency policy.
432 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-73. Case-control studies of TCE exposure and lymphopoietic cancer, leukemia or multiple myeloma
Population
Cancer type and exposure group
Odds ratio
(95% CI)
n exposed
cases
Reference(s)
Men and women aged 20-74 in
Non-Hodgkin lymphoma
Gold et al. (In Press); Purdue et al. (2011)
Iowa, Los Angeles County (CA),
Seattle and Detroit metropolitan
areas
Any TCE exposure
Possible
1.1 (0.9, 1.3)
545
Probable
1.4 (0.8, 2.4)
45
Average weekly exposure3
0 ppm-h per week
1.0
341
1-60 ppm-h per week
1.6 (0.7, 3.8)
15
61-150 ppm-h per week
0.5 (0.2, 1.4)
7
>150 ppm-h per week
2.5 (1.1,6.1)
23
(p for linear trend)
0.02
Cumulative exposure3
0
1.0
341
1-46,800 ppm-h
1.4 (0.6, 3.3)
14
46,801-112,320 ppm-h
0.6 (0.2, 1.7)
7
>112,320 ppm-h
2.3 (1.0, 5.0)
24
(p for linear trend)
0.08
Non-Hodgkin lymphoma types
Probable TCE exposure
Diffuse
0.9 (0.5, 2.0)
155
Follicular
2.1 (1.0,4.2)
13
Chronic lymphocytic leukemia
2.7 (1.2, 5.8)
11
-------
Table 4-73. Case-control studies of TCE exposure and lymphopoietic cancer, leukemia or multiple myeloma
(continued)
Population
Cancer type and exposure group
Odds ratio
(95% CI)
n exposed
cases
Reference(s)
Men and women aged 20-74
(continued)
Multiple myeloma
Gold et al. (In Press); Purdue et al. (2011)
(continued)
Any TCE exposure
1.4(0.9,2.1)
66
High confidence exposure13
1.7 (1.0,2.7)
43
Cumulative exposure13
0
1.0
139
1-471 ppm-h
1.1 (0.4,2.9)
6
472-3,000 ppm-h
1.6 (0.7,3.5)
11
3,001-7,644 ppm-h
1.5 (0.6,3.9)
7
7,645-570,000 ppm-h
2.3 (1.1, 5.0)
17
(p for linear trend)
0.03
Men and women aged >17 yr in
Czech Republic, Finland, France,
Germany, Ireland, Italy, and Spain
(Epilymph study)
All Centers:
Cocco et al. (2010)
B-cell NH1
Any TCE exposure
0.8 (0.6, 1.1)
71
Cumulative Exposure
Low
0.9 (0.6, 1.6)
26
Medium
0.5 (0.3, 0.9),
16
High
1.0 (0.6, 1.6)
29
(p for linear trend)
0.16
-------
Table 4-73. Case-control studies of TCE exposure and lymphopoietic cancer, leukemia or multiple myeloma
(continued)
Population
Cancer type and exposure group
Odds ratio
(95% CI)
n exposed
cases
Reference(s)
Men and women aged >17 yr in
Non-Hodgkin lymphoma types0
Cocco et al. (2010) (continued)
Czech Republic, Finland, France,
Germany, Ireland, Italy, and Spain
(Epilymph study) (continued)
Diffuse large B-cell
0.7(0.4, 1.1)
17
Follicular
1.2(0.6,2.3)
11
Chronic lymphocytic leukemia
0.9(0.5, 1.5)
18
Multiple myeloma
0.6(0.3, 1.2)
9
T-cell lymphoma
0.9 (0.4, 2.2)
6
German Centers:
Seidler et al. (2007); Mester et al. (2006)
Non-Hodgkin lymphoma
Any TCE exposure
Not reported
Cumulative TCE
0 ppm-yr
1.0
610
>0- <4 ppm-yr
0.7 (0.4, 1.1)
40
4.4- <35 ppm-yr
0.7 (0.5, 1.2)
32
High exposure, >35 ppm-yr
2.1 (1.0,4.8)
21
(p for linear trend)
0.14
>35 ppm-yr, 10 yr lag
2.2 (1.0, 4.9)
Women aged 21-84 in
Non-Hodgkin lymphoma
Wang et al. (2009)
Connecticut, U.S.
Any TCE exposure
1.2 (0.9, 1.8)
77
Low intensity TCE exposure
1.1 (0.8, 1.6)
64
Medium-high intensity TCE exposure
2.2 (0.9, 5.4)
13
(p for linear trend)
0.06
Low probability TCE exposure
1.1 (0.7, 1.8)
43
Medium-high probability TCE exposure
1.4 (0.9, 2.4)
34
(p for linear trend)
0.37
-------
Table 4-73. Case-control studies of TCE exposure and lymphopoietic cancer, leukemia or multiple myeloma
(continued)
Population
Cancer type and exposure group
Odds ratio
(95% CI)
n exposed
cases
Reference(s)
Women aged 21-84 in
Low intensity TCE exposure/low probability
0.9 (0.6, 1.5)
30
Wang et al. (2009) (continued)
Connecticut, U.S. (continued)
Low intensity/medium-high probability
1.4 (0.9, 2.4)
34
Medium-high intensity/low probability
2.2 (0.9, 5.4)
13
Medium-high intensity/medium-high probability
0
Population in 8 Italian regions
Non-Hodgkin lymphoma
Miligi et al. (2006); Costantini et al. (2008)
Any TCE exposure
Not reported
TCE exposure intensity
Very low/low
0.8 (0.5, 1.3)
35
Medium/high
1.2 (0.7, 2.0)
35
(p for linear trend)
0.8
Duration exposure, medium/high TCE intensity
<15 yr
1.1 (0.6,2.1)
22
>15 yr
1.0 (0.5, 2.6)
12
(p for linear trend)
0.72
Other non-Hodgkin lymphoma
TCE exposure intensity, medium/high
Small lymphocytic NHL
0.9(0.4,2.1)
7
Follicular NHL
Not presented
3
Diffuse NHL
1.9 (0.9, 3.7)
13
Other NHL
1.2 (0.6,2.4)
11
Multiple myeloma
0.9 (0.3, 2.4)
27
-------
Table 4-73. Case-control studies of TCE exposure and lymphopoietic cancer, leukemia or multiple myeloma
(continued)
Population
Cancer type and exposure group
Odds ratio
(95% CI)
n exposed
cases
Reference(s)
Population in 8 Italian regions
(continued)
Leukemia
Miligi et al. (2006); Costantini et al. (2008)
(continued)
Any TCE exposure
Not reported
TCE exposure intensity
Very low/low
1.0 (0.5, 1.8)
17
Medium/high
0.7 (0.4, 1.5)
11
Chronic lymphocytic leukemia
Any TCE exposure
Not reported
TCE exposure intensity
Very low/low
1.2 (0.5, 2.7)
8
Medium/high
0.9 (0.3, 2.6)
4
Population of Orebro and
Linkoping, Sweden
B-cell non-Hodgkin lymphoma
Persson and Fredrikson (1999)
Any TCE exposure
1.2 (0.5, 2.4)
16
Population of Sweden
Hairy cell lymphoma
Nordstrom et al. (1998)
Any TCE exposure
1.5 (0.7, 3.3
9
Population of Umea, Sweden
Non-Hodgkin lymphoma
Hardell etal. (1994)
Any exposure to TCE
7.2 (1.3,42)
4
Population of Montreal, Canada
Non-Hodgkin lymphoma
Siemiatycki et al. (1991)
Any TCE exposure
1.1 (0.6, 2.3)d
6
Substantial TCE exposure
0.8(0.2, 2.5)d
2
"For Purdue et al. (2011), OR for subjects interviewed using computer-assisted personal interview with job modules and includes subjects assessed as unexposed
or with probably exposure, defined as holding one or more jobs with an assigned probability of TCE exposure of >50%.
bFor Gold et al. (In Press) subjects with jobs assessed with low confidence considered as unexposed.
Tor Cocco et al. (2010), OR for subjects with high confidence assessment of TCE exposure.
d90% CI.
-------
Table 4-74. Geographic-based studies of TCE and non-Hodgkin lymphoma or leukemia in adults
Population
Exposure group
non-Hodgkin lymphoma
Leukemia
Reference3
Relative risk
(95% CI)
n exposed
cases
Relative risk
(95% CI)
n exposed
cases
Two study areas in Endicott, NY
0.54(0.22, 1.12)
7
0.79 (0.34, 1.55)
8
ATSDR (2006b)
Residents of 13 census tracts inRedlands, CA
1.09 (0.84, 1.38)
111
1.02 (0.74, 1.35)
77
Morgan and Cassady (2002)
Population in New
Jersey
Males, maximum estimated TCE concentration (ppb) in municipal drinking water
Cohnetal. (1994b)
<0.1
1.00
493
1.00
438
0.1-0.5
1.28(1.10, 1.48)
272
0.85 (0.71, 1.02)
162
>5.0
1.20 (0.94, 1.52)
78
1.10(0.84, 1.90)
63
Females, maximum estimated TCE concentration (ppb) in municipal drinking water
<0.1
1.00
504
1.00; 315
0.1-0.5
1.02 (0.87, 1.2)
26
1.13 (0.93, 1.37)
156
>5.0
1.36 (1.08, 1.70)
87
1.43 (1.43, 1.90)
56
Population in Finland
Residents of Hausjarvi
0.6(0.3, 1.1)
14
1.2(0.8, 1.7)
33
Vartiainen et al. (1993)
Residents of Huttula
1.4(1.0,2.0)
13
0.7(0.4, 1.1)
19
aNo geographic-based study reported a relative risk estimate for multiple myeloma except Vartiainen et al. (1993) who observed standardized incidence ratios of
0.7 (95% CI: 0.3, 1.3) and 0.6 (95% CI; 0.2, 1.3) for residents of Hausjarvi and Huttula, respectively.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
study area boundaries and to identify census tracts with a higher probability of exposure to
volatile organic solvents without identifying exposure concentrations to TCE and other solvents.
In these studies, one level of exposure to all subjects in a geographic area is assigned, although
there is some inherent measurement error and misclassification bias because not all subjects are
exposed uniformly.
NHL risk is statistically significantly elevated in three studies in which there is a high
likelihood of TCE exposure in individual study subjects (e.g., based on job-exposure matrices or
biomarker monitoring) and which met, to a sufficient degree, the standards of epidemiologic
design and analysis in a systematic review (3.1, 95% CI: 1.3, 6.1 (Hansen et al., 2001); 1.5, 95%
CI: 1.2, 2.0, subcohort with higher exposure (Raaschou-Nielsen et al., 2003), 2.3, 95% CI: 1.0,
5.0, >112,320-ppm hours cumulative TCE exposure, 2.5, 95% CI: 1.1, 6.1, >150-ppm hours
average weekly TCE exposure (Purdue et al., 2011)). Two of these incidence studies report
statistically significantly associations for NHL for subjects with longer employment duration as a
surrogate of TCE exposure (>6.25 year, 4.2, 95% CI: 1.1, 11 (Hansen et al., 2001); >5 year, 1.6,
95% CI: 1.1, 2.2, (Raaschou-Nielsen et al., 2003)) and Purdue et al. (2011) report a positive
trend with NHL and cumulative TCE exposure (p = 0.08) or average weekly TCE exposure (p =
0.02). Hansen et al. (2001) also examined two other exposure surrogates, cumulative exposure
and exposure intensity, with estimated risk larger in low exposure groups than for high exposure
groups. A fourth study from Sweden reports a large and imprecise risk with TCE (7.2, 95% CI:
1.3, 42 (Hardell et al., 1994)) based on four exposed cases. Cohort mortality studies and other
case-control studies, except Cocco et al. (2010), observed a 10—50% increased risk between
NHL and any TCE exposure (1.2, 95% CI: 0.65, 1.99 (Boice et al., 1999); 1.36, 95% CI: 0.28,
6.08 (Morgan et al., 1998); 1.5, 95% CI: 0.7, 3.3 (Nordstrom et al., 1998); 1.2, 95% CI: 0.5, 2.4
(Persson and Fredrikson, 1999); 1.36, 95% CI: 0.77, 2.39 (Radican et al., 2008); 1.1,
95% CI: 0.6, 2.3 (Siemiatycki, 1991); 1.2, 95% CI: 0.9, 1.8 (Wang et al., 2009)).
Odds ratios are higher for diffuse or follicular NHL, primarily B-cell lymphomas, than
for all non-Hodgkin lymphomas in both studies which examine forms of lymphoma, although
based on few exposed cases and inconsistently reported (Cocco et al., 2010; Miligi et al., 2006;
Purdue et al., 2011) (see Table 4-74). Observations in the two other studies of B-cell lymphomas
(Persson and Fredrikson, 1999; Wang et al., 2009) appear consistent with Miligi et al. (2006) and
Purdue et al. (2011). Together, these observations suggest that the associations between
trichloroethylene and specific NHL types are stronger than the associations seen with other
forms of NHL, and that disease misclassification may be introduced in studies examining
trichloroethylene and NHL as a broader category. Mortality observations in other occupational
cohorts (Costa et al., 1989; Garabrant et al., 1988; Greenland et al., 1994; Wilcosky et al., 1984)
This document is a draft for review purposes only and does not constitute Agency policy.
439 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
(ATSDR, 2004a; Boice et al., 2006b; Chang et al., 2003b; Henschler et al., 1995; Ritz, 1999a;
Sung et al., 2007) included a risk estimate of 1.0 in 95% CIs; these studies neither add to nor
detract from the overall weight of evidence given their lower likelihood for TCE exposure due to
inferior exposure assessment approaches, lower prevalence of exposure, lower statistical power,
and fewer exposed deaths.
Seven studies presented estimated risks for leukemia and overall TCE exposure
(Anttila et al., 1995; Blair et al., 1998 and its update by Radican et al., 2008; Morgan et al., 1998;
Boice et al., 1999, 2006; Hansen et al., 2001; Raachou-Nielsen et al., 2003); only three studies
also presented estimated risks for a high exposure category (Anttila et al., 1995; Blair et al.,
1998; Morgan et al., 1998). Three case-control studies presented estimated risk for leukemia
categories and overall TCE exposure or low or high TCE exposure category (Cocco et al., 2010;
Costantini et al., 2008; Purdue et al., 2011). Risk estimates in these cohort studies ranged from
0.64 (95% CI: 0.35, 1.18) (Radican et al., 2008) to 2.0 (95% CI: 0.7, 4.44) (Hansen et al., 2001).
The largest study, with 82 observed incident leukemia cases, reported a relative risk estimate of
1.2 (95% CI: 0.9, 1.4) (Raaschou-Nielsen et al., 2003). Case-control studies which examined all
leukemias (Costantini et al., 2008) or chronic lymphocytic leukemia (Cocco et al., 2010;
Costantini et al., 2008; Purdue et al., 2011), and TCE exposure are quite limited in statistical
power. Risk estimates in the four case-control studies ranged from 0.7 (95% CI: 0.4, 1.5) for all
leukemias and medium to high exposure intensity (Costantini et al. (2008) to 2.7 (95% CI: 1.2,
5.8) for chronic lymphocytic leukemia and probable TCE exposure (Purdue et al., 2011).
Eight cohort studies presented estimated risks for multiple myeloma and overall TCE
exposure (Anttila et al., 1995; Axelson et al., 1994) (Blair et al., 1998; and its update by Radican
et al., 2008) (Boice et al., 1999; Boice et al., 2006b; Hansen et al., 2001; Morgan et al., 1998;
Raaschou-Nielsen et al., 2003); only three studies also presented estimated risks for a high
exposure category (Anttila et al., 1995; Boice et al., 1999; Radican et al., 2008). Three case-
control studies presented estimated risk for multiple myeloma and overall TCE exposure or low
or high TCE exposure category (Cocco et al., 2010; Costantini et al., 2008; Gold et al., In Press).
Risk estimates in these cohort studies ranged from 0.57 (95% CI: 0.01, 3.17) (Axelson et al.,
1994) to 1.62 (95% CI: 0.44, 4.16) (Anttila et al., 1995). The largest cohort study, with 31
observed incident multiple myeloma cases, reported a relative risk estimate of 1.03 (95% CI:
0.70, 1.47) (Raaschou-Nielsen et al., 2003). The largest case-control study of 43 exposed
multiple myeloma cases with high confidence TCE exposure reported an odds ratio of 1.7 (95%
CI: 1.0, 2.7) and a positive trend with increasing cumulative TCE exposure (p = 0.03) (Gold et
al., In Press).
This document is a draft for review purposes only and does not constitute Agency policy.
440 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
The number of studies of childhood lymphoma including acute lymphatic leukemia and
trichloroethylene is much smaller than the number of studies of trichloroethylene and adult
lymphomas, and consists of four case-control studies (Costas et al., 2002; Lowengart et al., 1987;
McKinney et al., 1991; Shu et al., 1999) and four geographic based studies (ADHS, 1990, 1995;
Aickin et al., 1992; ATSDR, 2006a, 2008; Cohn et al., 1994b), (see Table 4-75). An additional
publication, focusing on ras mutations, based on one of the case-control studies is also available
(Shu et al., 2004). All four case-control studies evaluate maternal exposure, and three studies
also examine paternal occupational exposure (Lowengart et al., 1987; McKinney et al., 1991;
Shu et al., 2004; Shu et al., 1999). There are relatively few cases with maternal exposure (range
0-16) in these case-control studies, and only Shu et al. have a large number (n = 136) of cases
with paternal exposure (Shu et al., 2004; Shu et al., 1999). The small numbers of exposed case
parents limit examination of possible susceptibility time windows. Overall, evidence for
association between parental trichloroethylene exposure and childhood leukemia is not robust or
conclusive.
The results from the studies of Costas et al. (2002) and Shu et al. (2004; 1999) suggest a
fetal susceptibility to maternal exposure during pregnancy, with relative risks observed for this
time period equal or higher than the relative risks observed for periods before conception or after
birth (see Table 4-75). The studies by Lowengart et al. (1987) and McKinney et al. (1991) do
not provide informative data pertaining to this issue due to the small number (n = <3) of exposed
case mothers. A recent update of a cohort study of electronics workers at a plant in Taiwan
(Chang et al., 2003b; 2005) reported a fourfold increased risk (3.83; 95% CI: 1.17, 12.55 (Sung
et al., 2008)) for childhood leukemia risk among the offspring of female workers employed
during the three months before to three months after conception. Exposures at this factory
included trichloroethylene, perchloroethylene, and other organic solvents (Sung et al., 2008).
The lack of TCE assignment to individual subjects in this study decrease its weight in the overall
analysis.
The evidence for an association between childhood leukemia and paternal exposure to
solvents is quite strong (Colt and Blair, 1998); however, for studies of TCE exposure, the small
numbers of exposed case fathers in two studies (Lowengart et al., 1987; McKinney et al., 1991)
and, for all three studies, likelihood of misclassification resulting from a high percentage of
paternal occupation information obtained from proxy interviews, limits observation
interpretations. Both Lowengart et al. (1987) and McKinney et al. (1991) provide some evidence
for a two- to fourfold increase of childhood leukemia risk and paternal occupational exposure
although the population study of Shu et al. (2004; 1999), with 13% of case father's occupation
reported by proxy respondents, does not appear to support the earlier and smaller studies.
This document is a draft for review purposes only and does not constitute Agency policy.
441 DRAFT—DO NOT CITE OR QUOTE
-------
1
This document is a draft for review purposes only and does not constitute Agency policy.
442 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-75. Selected results from epidemiologic studies of TCE exposure and
2 childhood leukemia
3
Relative risk
(95% CI)
n observed
events
Reference(s)
Cohort studies (solvents)
Childhood leukemia among offspring of electronic workers
Sung et al. (2008)
Nonexposed
1.0a
9
Exposed pregnancy to organic solvents
3.83 (1.17, 12.55)
6
Case-control studies
Children's Cancer Group Study (children <15 yr age)
Acute lymphocytic leukemia
Maternal occupational exposure to TCE
Shuetal. (1999)
Anytime
1.8(0.8,4.1)
15
Preconception
1.8 (0.8, 5.2)
9
During pregnancy
1.8 (0.5, 6.4)
6
Postnatal
1.4(0.5,4.1)
9
Paternal occupational exposure to TCE
Anytime
1.1 (0.8, 1.5)
136
Preconception
1.1 (0.8, 1.5)
100
During pregnancy
0.9 (0.6, 1.4)
56
Postnatal
1.0 (0.7, 1.3)
77
K-ras + acute lymphocytic leukemia
Shu et al., (2004)
Maternal occupational exposure to TCE
Anytime
1.8 (0.6, 4.8)
5
Preconception
2.0 (0.7, 6.3)
4
During pregnancy
3.1 (1.0, 9.7)
4
Postnatal
0
Paternal occupational exposure to TCE
Anytime
0.6 (0.3, 1.4)
9
Preconception
0.6 (0.3, 1.5)
8
During pregnancy
0.3 (0.1, 1.2)
2
Postnatal
0.4 (0.1, 1.4)
3
4
This document is a draft for review purposes only and does not constitute Agency policy.
443 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-75. Selected results from epidemiologic studies of TCE exposure and
childhood leukemia (continued)
Relative risk
(95% CI)
n observed
events
Reference(s)
Residents of ages <19 in Woburn, MA
Costas et al. (2002)
Maternal exposure 2 yr before conception to diagnosis
Never
1.00
3
Least
5.00 (0.75, 33.5)
9
Most
3.56 (0.51,24.8)
7
(p for linear trend)
>0.05
Maternal exposure 2 yr before conception
Never
1.00
11
Least
2.48 (0.42, 15.2)
4
Most
2.82 (0.30, 26.4)
4
(p for linear trend)
>0.05
Birth to diagnosis
Never
1.00
7
Least
1.82 (0.31, 10.8)
7
Most
0.90 (0.18, 4.56)
5
(p for linear trend)
>0.05
Maternal exposure during pregnancy
Never
1.00
9
Least
3.53 (0.22, 58.1)
3
Most
14.3 (0.92, 224)
7
(p for linear trend)
<0.05
Population <14 yr of age in 3 areas north England, United Kingdom
McKinney et al. (1991)
Acute lymphocytic leukemia and NHL
Maternal occupation exposure to TCE
Preconception
1.16(0.13,7.91)
2
Paternal occupational exposure to TCE
Preconception
2.27 (0.84, 6.16)
9
Periconception and gestation
4.49(1,15,21)
7
Postnatal
2.66 (0.82, 9.19)
7
This document is a draft for review purposes only and does not constitute Agency policy.
444 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-75. Selected results from epidemiologic studies of TCE exposure and
childhood leukemia (continued)
Relative risk
(95% CI)
n observed
events
Reference(s)
Los Angeles Cancer Surveillance Program
Lowengart et al. (1987)
Acute lymphocytic and nonlymphocytic leukemia, <10 yr of age
Maternal occupational exposure to TCE
0
Paternal occupational exposure to TCE
One year before pregnancy
2.0 (p = 0.16)
6/3 b
During pregnancy
2.0 (p = 0.16)
6/3 b
After delivery
2.7 (0.64, 15.6)
8/3 b
Geographic based studies
Two study areas in Endicott, NY
ATSDR (2006a)
Leukemia, <19 yr of age
Not reported
<6
Population in New Jersey
Acute lymphocytic leukemia
Maximum estimated TCE concentration (ppb)
in municipal drinking water
Cohnetal. (1994b)
Males
<0.1
1.00
45
0.1-0.5
0.91 (0.53, 1.57)
16
>5.0
0.54 (0.17, 17.7)
3
Females
<0.1
1.00
25
0.1-0.5
1.85 (1.03, 3.70)
22
>5.0
2.36 (1.03,5.45)
7
Resident of Tucson Airport Area, AZ
ADHS (1990, 1995)
Leukemia, <19 yr of age
1970-1986
1.48 (0.74, 2.65)
11
1987-1991
0.80 (0.31,2.05)
3
Resident of West Central Phoenix, AZ
Aickinetal. (1992)
Leukemia, <19 yr of age
1.95 (1.43,2.63)
38
1
2 internal referents, live born children among female workers not exposed to organic solvents.
3 bDiscordant pairs.
4
5
This document is a draft for review purposes only and does not constitute Agency policy.
445 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
The geographic based studies for adult lymphopoietic (see Table 4-74) or childhood
leukemias (see Table 4-75) do not greatly contribute to the overall weight of evidence. While
some studies observed statistically significantly elevated risks for NHL or childhood cancer,
these studies generally fulfilled only the minimal of evaluation criteria with questions raised
about subject selection (Morgan and Cassady, 2002), their use of less sophisticated exposure
assessment approaches and associated assumption of an average exposure to all subjects (all
studies), and few cases with high level parental exposure (all studies).
4.6.1.1.8. Meta-analysis of NHL risk
Meta-analysis is adopted as a tool for examining the body of epidemiologic evidence on
NHL and TCE exposure and to identify possible sources of heterogeneity. The meta-analysis of
NHL examines 17 cohort and case-control studies identified through a systematic review and
evaluation of the epidemiologic literature on TCE exposure (Anttila et al., 1995; Axelson et al.,
1994; Boice et al., 1999; Cocco et al., 2010; Greenland et al., 1994; Hansen et al., 2001; Hardell
et al., 1994; Miligi et al., 2006; Morgan et al., 1998; Nordstrom et al., 1998; Persson and
Fredrikson, 1999; Purdue et al., 2011; Raaschou-Nielsen et al., 2003; Radican et al., 2008;
Siemiatycki, 1991; Wang et al., 2009; Zhao et al., 2005) and two studies as alternatives (Blair et
al., 1998; Boice et al., 2006b). These 19 studies of NHL and TCE had high likelihood of
exposure, were judged to have met, to a sufficient degree, the criteria of epidemiologic design
and analysis, and reported estimated risks for overall TCE exposure; 13 of these studies, also,
presented estimated NHL risk with high level TCE exposure (Anttila et al., 1995; Axelson et al.,
1994; Boice et al., 1999; Cocco et al., 2010; Hansen et al., 2001; Miligi et al., 2006; Morgan et
al., 1998; Purdue et al., 2011; Raaschou-Nielsen et al., 2003; Radican et al., 2008; Siemiatycki,
1991; Wang et al., 2009; Zhao et al., 2005). Full details of the systematic review, criteria to
identify studies for including in the meta-analysis, and meta-analysis methodology and findings
are discussed in Appendices B and C.
The meta-analyses of the overall effect of TCE exposure on NHL suggest a small, robust,
and statistically significant increase in NHL risk. The summary estimate from the primary
random effect meta-analysis (RRm) was 1.23 (95% CI: 1.07, 1.42) (see Figure 4-16). This result
and its statistical significance were not influenced by individual studies. Removal of individual
studies resulted in RRm estimates between 1.18 (with the removal of Hansen et al., 2001) to 1.27
(with the removal of Miligi et al. (2006) or Cocco et al. (2010)), and lower 95% CIs excluded 1.0
(all ^-values werep < 0.02). The result is similarly not sensitive to individual risk ratio estimate
This document is a draft for review purposes only and does not constitute Agency policy.
446 DRAFT—DO NOT CITE OR QUOTE
-------
1 selections. Use of six alternative selections, individually, resulted in RRm estimate that ranged
2 from 1.20 (95% CI: 1.03, 1.39) (with estimated overall RR for incidence in Zhao et al., 2005) to
This document is a draft for review purposes only and does not constitute Agency policy.
447 DRAFT—DO NOT CITE OR QUOTE
-------
L/l
K>
5^
§•
rs
sj
3
a
>!
ca
fs*
rv «
4^ >8
4^ ^
00 §
>!
o
a
a
§•
>!
5
o
rs
o
a
>!
o
fe
H
o
o
z;
o^
St
H
o;
a
"T3
H S.
W|
O •
p
o
c
o
H
W
Study
Anttila (1005)
Axelsori (1094)
Boice (1099)
Greenland (1094)
Hansen (2001)
Morgan ( 1008)
R a asc h o u- N i e Ise n (2003)
Radio an (2008)
Zhao (2005)
Ceeco (2010)
Hardeli .1004)
Miligi (2006)
Nordstrom (1008)
Persson&F rederikson (1000)
Purdue (2011)
Siemiatycki (1001)
Wang (2000)
OVERALL
TCE Expostne ami Non-Hoilgkin Lymphoma
u. I
Relative Risk and 95% CI
-O
"O"
n
RR
LCL
UCL
1.81
0.78
3.56
1.52
0.4©
3.53
1.10
0.83
'1.65
o
: i
0.24
2 42
3.10
1.30
6.10
1.01
0.46
1.92
1.24
1.01
1.52
1.36
0.77
2.39
1.44
0.90
2.30
0.80
0.50
1.10
7.20
1.30
42.00
0.90
0.70
1.30
1.50
0.70
3.30
1.20
0.50
2.40
1.40
0.80
2.40
1.10
0.50
2.50
1.20
0.00
1.80
1.23
1.07
1.42
10
Figure 4-16. Meta-analysis of NHL and overall TCE exposure. The summary estimate is in the bottom row.
Symbol sizes reflect relative weights of the studies. The horizontal midpoint of the bottom diamond represents the
RRm estimate.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
1.28 (95% CI: 1.09, 1.49) (with Raaschou-Nielsen et al. (2003) subgroup). Nor was the RRm
estimate highly sensitive to restriction of the meta-analysis to only those studies for which RR
estimates for the traditional definition of NHL were available. An alternate analysis which
omitted Miligi (which included CLLs), Nordstrom (which was a study of hairy cell leukemias),
Persson and Frederikson (for which the classification system not specified), and Greenland
(which included Hodgkin lymphomas) and which included Boice (2006b) instead of Zhao
(which included all lymphohematopoietic cancers) yielded an RRm estimate of 1.27 (95% CI:
1.05, 1.55). Meta-analysis of the highest exposure groups, either duration, intensity, or their
product, cumulative exposure, results in an RRm of 1.43 (95% CI: 1.13, 1.82), which is greater
than the RRm from the overall exposure analysis, and provides additional support for an
association between NHL and TCE (see Figure 4-17). No single study was overly influential;
removal of individual studies resulted in RRm estimates that were all statistically significant (all
with p < 0.025) and that ranged from 1.38 (with the removal of Purdue et al. (2011)) to 1.57
(with the removal of Cocco et al. (2010)). In addition, the RRm estimate was not highly
sensitive to alternate RR estimate selections. Use of the 9 alternate selections, individually,
resulted in RRm estimates that were all statistically significant (all withp < 0.025) and all in the
narrow range from 1.40 (95% CI: 09, 1.80) (with Blair et al. (1998) incidence relative risk
instead of Radican et al. (2008) mortality hazard ratio) to 1.49 (95% CI: 1.14, 1.93) (with Hansen
et al. [2001] duration). The highest exposure category groups have a reduced likelihood for
exposure misclassification because they are believed to represent a greater differential TCE
exposure compared to people identified with overall TCE exposure. Observation of greater risk
associated with higher exposure category compared to overall (typically any versus none)
exposure comparison additionally suggests an exposure-response gradient between NHL and
TCE, although estimation of a level of exposure associated with the summary or meta-relative
risk is not possible.
Low-to-moderate heterogeneity in RRm is observed across the results of the 17 studies in
the meta-analysis of the overall effect of TCE and the 13 studies with highest exposure groups,
but it was not statistically significant (p = 0.16 and p = 0.30, respectively. The / -values were
26%) for overall exposure and 14% for highest exposure groups, suggesting low-to-moderate and
low heterogeneity, respectively. To investigate the heterogeneity, subgroup analyses were done
examining the cohort and case-control studies separately. Difference between cohort and case-
control studies could explain much of the observed heterogeneity. In the subgroup analysis of
overall exposure and of highest exposure groups, increased risk of NHL was strengthened in
analysis limited to cohort studies and reduced in the case-control study analysis. Examination of
heterogeneity in cohort and case-control studies of overall exposure separately was not
This document is a draft for review purposes only and does not constitute Agency policy.
449 DRAFT—DO NOT CITE OR QUOTE
-------
L/l
K>
5^
§•
rs
sj
3
a
>!
ca
fs*
rv «
u\ *5
<=> §
to
>!
o
a
a
§•
>!
5
o
rs
o
a
>!
o
fe
H
w ^
z; ^
o^
H J
1—1 s
H S.
W|
O •
o
c
o
H
W
:I:tutly
- nttiki •; Ivv? •
Axelson (1994)
Boice (1999)
Hansen (2001)
Morgan (1998)
Ra3schoy-Nielsen (2003)
Radio an (2008)
Zhao (2005)
Cocco (2010)
Miligi (2006)
P indue (20 1 I j
Siemiat/cki (1991)
Wang (2009)
OVERALL
0.1
<|Kin Lymphoma - Highest Exposure Groups
Relative Risk and 85% CI RR LCL
10
UCL
-to
0
17
5.04
25
0
10
34.83
82
0
0^1
70
0
56
8.00
81
0
10
6.46
60
1
10
-40
0
71
2.80
30
0
52
3.23
70
0
40
1.30
20
0
70
2.00
30
1
10
10.10
80
0
20
3.30
20
0
00
5.40
43
1
13
1.82
Figure 4-17. Meta-analysis of NHL and TCE exposure—highest exposure groups. The summary estimate is in the
bottom row. Symbol sizes reflect relative weights of the studies. The horizontal midpoint of the bottom diamond
represents the RRm estimate.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
2
statistically significant in either case (I -values for the cohort studies were 12%, suggesting low
heterogeneity and 27% for the case-control studies, suggesting low-to-moderate heterogeneity)
although some may be present given that statistical tests of heterogeneity are generally
insensitive in cases of minor heterogeneity. Subgroup analyses examining the cohort and case-
control studies highest exposure groups, separately, showed no residual heterogeneity in the
1 1
cohort subgroup (I = 0%) and moderate heterogeneity in the case-control subgroup (I -value
was 53%>) that was not statistically significant (p = 0.08). Although no further attempt was made
to quantitatively investigate potential sources of heterogeneity, the removal of the Cocco et al.
(2010) study, an influential study, eliminates all of the heterogeneity, suggesting that the RR
estimate for the highest exposure group from that study is a relative outlier.
In general, sources of heterogeneity are uncertain and may reflect several features known
to influence epidemiologic studies. Study design itself is unlikely to be an underlying cause of
heterogeneity and, to the extent that it may explain some of the differences across studies, is
more probably a surrogate for some other difference(s) across studies that may be associated
with study design. Furthermore, other potential sources of heterogeneity may be masked by the
broad study design subgroupings. The true source(s) of heterogeneity across these studies is an
uncertainty.
One reason may be differences in exposure assessment and in overall TCE exposure
concentration between cohort and case-control studies. Several cohort and case-control studies
included TCE assignment from information on job and task exposures, e.g., a JEM (Boice et al.,
1999; Boice et al., 2006b; Cocco et al., 2010; Miligi et al., 2006; Morgan et al., 1998; Purdue et
al., 2011; Radican et al., 2008; Siemiatycki, 1991; Wang et al., 2009; Zhao et al., 2005), or from
an exposure biomarker in either breath or urine (Anttila et al., 1995; Axelson et al., 1994; Hansen
et al., 2001). Three case-control studies (Hardell et al., 1994; Nordstrom et al., 1998; Persson
and Fredrikson, 1999) relied on self-reported TCE exposure. No information is available to
judge the degree of possible misclassification bias associated with a particular exposure
assessment approach; it is quite possible that in some cohort studies, in which past exposure is
inferred from various data sources, exposure misclassification may be as great as in population-
based or hospital-based case-control studies. In addition, a low overall TCE exposure prevalence
is anticipated in population case-control studies which would typically assess a large number of
workplaces and operations, where exposures are less well defined, and where case and control
subjects identified as exposed to TCE probably have minimal contact (NRC, 2006). Observed
higher risk ratios with higher exposure categories in NHL case-control studies support exposure
differences as a source of heterogeneity.
This document is a draft for review purposes only and does not constitute Agency policy.
451 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Diagnostic inaccuracies are likely another source of heterogeneity in the meta-analysis
through study differences in NHL groupings and in lymphoma classification schemes, although
restricting the meta-analysis to only those studies for which RR estimates based on the traditional
NHL definition were available did not eliminate all heterogeneity. All studies include a broad
but slightly different group of lymphosarcoma, reticulum-cell sarcoma, and other lymphoid
tissue neoplasms (Codes 200 and 202), except Nordstrom et al. (1998), Zhao et al. (2005), and
Greenland et al. (1994). Cohort studies have some consistency in coding NHL, with NHL
defined as lymphosarcoma and reticulum-cell sarcoma (200) and other lymphoid tissue
neoplasms (202) using the ICD, Revision 7, 200 and 202—four studies (Anttila et al., 1995;
Axelson et al., 1994; Hansen et al., 2001; Raaschou-Nielsen et al., 2003), ICD-Adapted,
Revision 8 (Blair et al., 1998), and ICD-7, -8, -9, and -10, per the version in use at the time of
death Morgen et al., 1998 (as presented in Boice et al., 1999; Mandel et al., 2006; Radican et al.,
2008), as does the case-control study of Siemiatycki (1991) whose coding scheme for NHL is
consistent with ICD 9, 200 and 202. Case-control studies, on the other hand, have adopted other
classification systems for defining NHL including the NCI Working Formulation (Miligi et al.,
2006), Rappaport (Hardell et al., 1994), or else do not identify the classification system for
defining NHL (Persson and Fredrikson, 1999). Cocco et al. (2010) used the WHO/Revised
European-American Lymphoma (REAL) classification system, which reclassifies lymphocytic
leukemias and NHLs as lymphomas of B-cell or T-cell origin and considers CLLs and multiple
myelomas as (non-Hodgkin) lymphomas; however, we were able to obtain results generally
consistent with the traditional NHL definition from Dr. Cocco, although lymphomas not
otherwise specified were excluded. Wang et al. (2009) defined NHL using ICD-O-2 codes
(M-9590-9595, 9670-9688, 9690-9698, 9700-9723), which is consistent with the traditional
definition of NHL (i.e., ICD-7, -8, -9 codes 200 + 202). Purdue et al. (2011) used ICD-O-3
codes 967-972, which is generally consistent with the traditional definition of NHL, although
this grouping does not include the malignant lymphomas of unspecified type coded as
M-9590-9599.
There is some evidence of potential publication bias in this data set; however, it is
uncertain that this is actually publication bias rather than an association between standard error
and effect size resulting for some other reason, e.g., a difference in study populations or
protocols in the smaller studies. Furthermore, if there is publication bias in this data set, it does
not appear to account completely for the finding of an increased NHL risk.
NRC (2006) deliberations on trichloroethylene commented on two prominent evaluations
of the then-current epidemiologic literature using meta-analysis techniques. These studies were
by Wartenberg et al. (2000), and by Kelsh et al. (2005), submitted by Exponent-Health Sciences
This document is a draft for review purposes only and does not constitute Agency policy.
452 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
to NRC during their deliberations and subsequently published in a paper on NHL (Mandel et al.,
2006) and a paper on multiple myeloma and leukemia (Alexander et al., 2006). The NRC found
weaknesses in the techniques used in each of these studies, and suggested that EPA conduct a
new meta-analysis of the epidemiologic data on trichloroethylene using objective and transparent
criteria so as to improve on the past analyses. EPA staff conducted their analysis according to
NRC (2006) suggestions for transparency, systematic review criteria, and examination of both
cohort and case-control studies. The EPA analysis of NHL analysis considered a larger number
of studies than in the previous analyses (Mandel et al., 2006; Wartenberg et al., 2000), includes
recently published studies (Boice et al., 2006b; Cocco et al., 2010; Miligi et al., 2006; Purdue et
al., 2011; Radican et al., 2008; Wang et al., 2009; Zhao et al., 2005), and combines both cohort
and case-control studies.
4.6.2. Animal Studies
The immunosuppressive and immunomodulating potential of TCE has not been fully
evaluated in animal models across various exposure routes, over various relevant durations of
exposure, across representative life stages, and/or across a wide variety of endpoints.
Nevertheless, the studies that have been conducted indicate a potential for TCE-induced
immunotoxicity, both following exposures in adult animals and during immune system
development (i.e., in utero and preweaning exposures).
4.6.2.1.1. Immunosuppression
A number of animal studies have indicated that moderate to high concentrations of TCE
over long periods have the potential to result in immunosuppression in animal models, dependant
on species and gender. These studies are described in detail below and summarized in
Table 4-76.
4.6.2.1.2. Inhalation exposures
Mature cross-bred dogs (5/group) were exposed to 0-, 200-, 500-, 700-, 1,000-, 1,500-, or
2,000-ppm TCE for 1-hour or to 700 ppm TCE for 4 hours, by tracheal intubation under
intravenous sodium pentobarbital anesthesia. An additional group of dogs was exposed by
venous injection of 50 mg/kg TCE administered at a rate of 1 mL/minute (Hobara et al., 1984).
Blood was sampled pre- and postexposure for erythrocyte and leukocyte counts. Marked,
This document is a draft for review purposes only and does not constitute Agency policy.
453 DRAFT—DO NOT CITE OR QUOTE
-------
1 transient decreases in leukocyte counts were observed at all exposure levels 30 minutes after
2 initiation of exposure. At the end of the exposure period, all types of leukocytes were decreased
3 (by 85%); neutrophils were decreased 33%, and lymphocytes were increased 40%. There were
4 no treatment-related changes in erythrocyte counts, hematocrit values, or thrombocyte counts.
This document is a draft for review purposes only and does not constitute Agency policy.
454 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-76. Summary of TCE immunosuppression studies
Exposure route/vehicle,
duration, dose
NOAEL; LOAEL3
Results
Reference, species/strain
sex/number
Inhalation exposure studies
Single 1-h exposure to all dose
groups; plus single 4-h exposure
at 700 ppmb
0, 200, 500, 700, 1,000, 1,500, or
2,000 ppm
LOAEL: 200 ppm
Marked transient 1 leukocyte counts at all exposure levels 30
min after initiating exposure. At end of exposure, 85% 1
leukocyte counts (33% 1 neutrophils, 40% 1 lymphocytes).
Hobara et al. (1984)
Dog, cross-bred, both sexes,
5/group
Single 3-h exposure. Also, 3 h/d
on 5 d at lowest dose
0, 2.6, 5.2, 10.6, 25.6, or 48 ppm
NOAEL: 2.6 ppm
LOAEL: 5.2 ppm
Challenged with Streptococcus zooepidemicus to assess
susceptibility to infection and Klebsiella pneumoniae to
assess bacterial clearance. For single exposure: dose-related
sig. t mortality at >5.2 ppm over 14 d. Sig. { in bactericidal
activity at 10.6 ppm.
Aranyi et al. (1986)
Mouse, CD-I females, 4-5 wk old,
approx. 30 mice/group, 5-10
replications; for pulmonary
bactericidal activity assay, 17-24
mice/group
Single 3-h exposure.
0, 5, 10, 25, 50, 100, 200 ppm
NOAEL: 25 ppm
LOAEL: 50 ppm
Challenged with Streptococcus zooepidemicus to assess
susceptibility to infection and bacterial clearance. For single
exposure: dose-related sig. t mortality at >50 ppm over 20 d.
Dose dependent responses also observed in the clearance of
bacteria from the lung at >50 ppm, the number of mice with
delayed bacterial clearance at various postinfection time
points at >50 ppm, and the phagocytic function of alveolar
macrophages at 200 ppm.
Selgrade and Gilmour (2010)
Mouse, CD-I females, 5-6 wk old,
at least 38 mice/group
Single 3-h exposure,
50-200 ppm°
Challenged with Streptococcus zooepidemicus. Dose-related
t mortality, bacterial antiphagocytic capsule formation, and
bacterial survival. Dose-related impairment of alveolar
macrophages; increased neutrophils in bronchoalveolar fluid
at 3 d postinfection.
Park et al. (1993) (abstract)
Mouse, CD-I, (sex and #/group not
specified)
4-wk, 6 h/d, 5 d/wk
0,100,300, or 1,000 ppm
NOAEL: 300 ppm
LOAEL: 1,000 ppm
At 1,000 ppm, 64% i plaque-forming cell assay response.
Woolhiser et al. (2006)
Rat, Sprague-Dawley, female,
16/group
-------
Table 4-76. Summary of TCE immunosuppression studies (continued)
Exposure route/vehicle,
duration, dose
NOAEL; LOAELa
Results
Reference, species/strain
sex/number
Oral exposure studies
Gavage in 10% emulphor, 14 d,
daily, 0, 24, or 240 mg/kg-day
LOAEL: 24 mg/kg-
day
Sig. 1 cell-mediated immune response to SRBC at both dose
levels.
Sanders et al. (1982b)
Mouse, CD-I, male, 9-12/group
Drinking water with 1%
emulphor, 4-6 mo
0,0.1,1.0,2.5, or 5.0 mg/mL
LOAEL: 0.1 mg/kg-
day
In females, humoral immunity at 2.5 and 5 mg/mL
TCE, whereas cell-mediated immunity and bone
marrow stem cell colonization 4- at all four
concentrations. The males were relatively unaffected
after both 4 and 6 mo.
Sanders et al. (1982b)
Mouse, CD-I, male and female,
7-25/group
Gavage, 14 d, 0, 14.4, or 144
mg/kg-day chloral hydrate
NOAEL: 144 mg/kg-
day
No treatment-related effects.
Kauffmann et al. (1982)
Mouse, CD-I, male, 12/group
Drinking water, 90 d, 0, 0.07, or
0.7 mg/mL chloral hydrate. (M:
0, 16, or 160 mg/kg-day; F: 0, 18,
or 173 mg/kg-day)
NOAEL: 0.07 mg/mL
LOAEL: 0.7 mg/mL
Sig. 1 cell-mediated immune response (plasma
hemagglutination titers and spleen antibody-producing cells
of mice sensitized to SRBC) in females at 0.7 mg/mL.
Kauffmann et al. (1982)
Mouse, CD-I, male and female,
15-20/group
Drinking water, From mating to
PND 21 orPND 56, (emulphor
concentration not provided)
0 (emulphor), 1, or 10 ppm
LOAEL: 1 ppm
At 10 ppm, I body weight and length at PND 21. IgM
antibody response to SRBC challenge suppressed in both (51
and $ pups at 10 ppm, and ; pups at 1 ppm, J, in splenic
CD4+CD8-T-cells. At 56 PND, striking f in natural killer
cell activity seen at both doses.
Adams et al. (2003) (abstract)
Mouse, B6C3F1, both sexes,
numbers of pups not stated
Drinking water, from GD 0-3 or
8 wk of age, 0, 1,400, or 14,000
PPb
LOAEL: 1,400 ppb
Suppressed PFC responses in both sexes and ages at 14,000
ppb, in males at both ages at 1,400 ppb, and in females at 8
wk at 1,400 ppb. Numbers of spleen B220+ cells 1 at 3 wk
at 14,000 ppb. Pronounced t thymus T-cell populations at 8
wk.
Peden-Adams et al. (2006)
Mouse, B6C3F1, dams and both
sexes offspring, 5 litters/group;
5-7 pups/group at 3 wk;
4-5 pups/sex/group at 8 wk
-------
L/l
K>
Table 4-76. Summary of TCE immunosuppression studies (continued)
Exposure route/vehicle,
duration, dose
NOAEL; LOAEL3
Results
Reference, species/strain
sex/number
Drinking water, from GD 0 to
7-8 wk of age; 0, 0.5, or
2.5 mg/mL
LOAEL: 0.5 mg/mL
At 0.5 mg/mL: Sig j postweaning weight; sig.t IFNy
produced by splenic CD4+ cells at 5-6 wk; sig j splenic
CD8+ and B220+ lymphocytes; sig. t IgG2a and histone; sig.
altered CD4-/CD8- and CD4+/CD8+ thymocyte profile
At 2.5 mg/mL: Sig j postweaning weight; sig. t IFNy
produced by splenic CD4+ and CD8+ cells at 4-5 and 5-6
wk; sig i splenic CD4+, CD8+, and B220+ lymphocytes;
sig. altered CD4+/CD8+ thymocyte profile.
Blossom and Doss (2007)
Mouse, MRL +/+, dams and both
sexes offspring, 3 litters/group;
8-12 pups/group
Drinking water, from GD 0 to
PND 42; 0 or 0.1 mg/mL;
maternal dose = 25.7 mg/kg-day;
offspring PND 24-42 dose =
31.0 mg/kg-day
LOAEL: 0.1 mg/mL
At 0.1 mg/mL: at PND 20, sig. t thymocyte cellularity and
distribution, associated with sig. t in thymocyte subset
distribution; sig. t reactive oxygen species generation in total
thymocytes; sig. t in splenic CD4+ T-cell production of IFN-
y and IL-2 in females and TNF-a in males at PND 42.
Blossom et al. (2008)
Mouse, MRL +/+, dams and both
sexes offspring, 8 litters/group; 3-8
pups/group
Drinking water, from GD 0 to 12
mo of age; 0 (1% emulphor),
1,400, or 14,000 ppb
LOAEL: 1,400 ppb
At 1,400 ppb: splenic CD4-/CD8- cells sig. t in females;
thymic CD4+/CD8+ cells sig. 1 in males; 18% f in male
kidney weight.
At 14,000 ppb: thymic T-cell subpopulations (CD8+,
CD4/CD8-, CD4+) sig. i in males.
Peden-Adams et al. (2008)
Mouse, MRL +/+, dams and both
sexes offspring, unknown #
litters/group, 6-10
offspring/sex/group
Intraperitoneal injection exposure studies
3 d, single daily injection, 0, 0.05,
0.5, or 5 mmol/kg/day
NOAEL: 0.05
mmol/kg/day
LOAEL: 0.5
mmol/kg/day
1 natural killer cell activity at 0.5 and 5 mmol/kg/day. 1
splenocyte counts at 5 mmol/kg/day.
Wright et al. (1991)
Rat, Sprague-Dawley
3 d, single daily injection, 0 or 10
mmol/kg/day
LOAEL: 10
mmol/kg/day
1 natural killer cell activity and 1 spleen weights at 10
mmol/kg/day.
Wright etal. (1991)
Mouse, B6C3F1
3^
rs
sj
3
a
>!
&
<5"
K St
O
>!
>!
o
a
a
s
§•
>!
a
o
rs
o
5
Grj
-J
O
V
fe
H
o
o
2;
s
f?
a
"T3
H S.
W
P
o
c
o
H
W
H
o;
Bolded studies carried forward for consideration in dose-response assessment (see Section 5).
aNOAEL and LOAEL are based upon reported study findings.
bInhalation, tracheal intubation under anesthesia.
°Exact dose levels not specified.
I, t = decreased, increased; GD = gestation day; PFC = plaque-forming cell; sig. = statistically significant; SRBC = sheep red blood cells.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
In a study that examined the effects of a series of inhaled organic chemical air
contaminants on murine lung host defenses, Aranyi et al. exposed female CD-I mice to single
3-hour exposures of TCE at time-weighted concentrations of 0, 2.6, 5.2, 10.6, 25.6, or 48 ppm
(Aranyi et al., 1986). Additionally, at the dose at which no adverse treatment-related effect
occurred with a single exposure (i.e., 2.6 ppm), a multiple exposure test (5 days, 3 hours/day)
was conducted. Susceptibility to Streptococcus zooepidemicus aerosol infection and pulmonary
bactericidal activity to inhaled Klebsiella pneumoniae were evaluated. There was a significant
(p < 0.0001) treatment by concentration interaction for mortality, with the magnitude of the
effect increasing with concentration. A significant (p < 0.0001) treatment by concentration
interaction was also found for bactericidal activity. Single 3-hour exposures at 10.6, 25.6, and
48 ppm resulted in significant increases in mortality, although increases observed after single
exposures at 5.2 or 2.6 ppm or five exposures at 2.6 ppm were not significant. Pulmonary
bactericidal activity was significantly decreased after a single exposure at 10.6 ppm, but single
exposures to 2.6 or 5.2 ppm resulted in significant increases.
Suppression of pulmonary host defenses and enhanced susceptibility to respiratory
bacterial infection was studied in female CD-I mice by Selgrade and Gilmour (2010). The mice
(5-6 weeks of age; at least 38 per exposure group) were exposed via inhalation for 3 hours to
concentrations of 0, 5, 10, 25, 50, 100, or 200 ppm TCE. The mice were then challenged by
aerosol doses of Streptococcus zooepidemicus. Bacterial clearance (based upon organisms
present in lung lavage fluid) and a phagocytic index (percentage of phagocytic cells in lung
lavage fluid and the number of bacteria ingested per phagocytic cell) were assessed. Mortality
due to infection was significantly increased with TCE exposure concentration at exposures of
50 ppm and higher (NOAEL = 25 ppm). Dose-dependent responses were also observed for the
clearance of bacteria from the lung at >25 ppm, the number of mice with delayed bacterial
clearance at various postinfection time points at >25 ppm, and the phagocytic function of
alveolar macrophages at 200 ppm. The higher NOAEL for mortality observed in this study
compared to Aranyi et al. (1986), i.e., 25 ppm versus 5 ppm, was attributed to the use of
unencapsulated bacteria in this study; the study authors suggested that this may be more
representative of the human condition.
In a host-resistance assay, CD-I mice (sex and number/group not specified) exposed to
TCE by inhalation for 3 hours at 50-200 ppm were found to be more susceptible to increased
infection following challenge with Streptococcus zooepidemicus administered via aerosol (Park
et al., 1993). Dose-related increases in mortality, bacterial antiphagocytic capsule formation, and
bacterial survival were observed. Alveolar macrophage phagocytosis was impaired in a dose-
This document is a draft for review purposes only and does not constitute Agency policy.
458 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
responsive manner, and an increase in neutrophils in bronchoalveolar lavage fluid was observed
in exposed mice 3 days post infection.
A guideline (OPPTS 870.3800) 4-week inhalation immunotoxicity study was conducted
in female Sprague-Dawley rats (Woolhiser et al., 2006). The animals (16/group) were exposed
to TCE at nominal levels of 0, 100, 300, or 1,000 ppm for 6 hours/day, 5 days/week. Effects on
the immune system were assessed using an antigen response assay, relevant organs weights,
histopathology of immune organs, and hematology parameters. Four days prior to study
termination, the rats were immunized with sheep red blood cells (SRBC), and within 24 hours
following the last exposure to TCE, a plaque forming cell assay was conducted to determine
effects on splenic anti-SRBC IgM response. Minor, transient effects on body weight and food
consumption were noted in treated rats for the first 2 weeks of exposure. Mean relative liver and
kidney weights were significantly (p = 0.05) increased at 1,000 ppm as compared to control,
while lung, spleen, and thymus weights were similar to control. No treatment-related effects
were observed for hematology, white blood cell differential counts, or histopathological
evaluations (including spleen, thymus, and lung-associated lymph nodes). At 1,000 ppm, rats
demonstrated a 64% decrease in plaque forming cell assay response. Lactate dehydrogenase,
total protein levels, and cellular differentiation counts evaluated from bronchoalveolar lavage
(BAL) samples were similar between control and treated groups. A phagocytic assay using BAL
cells showed no alteration in phagocytosis, although these data were not considered fully reliable
since (1) the number of retrieved macrophage cells was lower than expected and pooling of
samples was conducted and (2) samples appear to have been collected at 24 hours after the last
exposure (rather than within approximately 2 hours of the last exposure), thereby allowing for
possible macrophage recovery. The NOAEL for this study was considered by the study authors
to be 300 ppm, and the LOAEL was 1,000 ppm; however, the effect level may have actually
been lower. It is noted that the outcome of this study does not agree with the studies by Aranyi
et al. (1986) and Park et al. (1993), both of which identified impairment of macrophage
phagocytic activity in BAL following inhalation TCE exposures.
4.6.2.1.3. Oral exposures
In a study by Sanders et al., TCE was administered to male and female CD-I mice for 4
or 6 months in drinking water at concentrations of 0, 0.1, 1, 2.5, or 5 mg/mL (Sanders et al.,
1982b). In females, humoral immunity was suppressed at 2.5 and 5 mg/mL, while cell-mediated
immunity and bone marrow stem cell activity were inhibited at all dose levels. Male mice were
relatively unaffected either at 4 or 6 months, even though a preliminary study in male CD-I mice
This document is a draft for review purposes only and does not constitute Agency policy.
459 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
(exposed to TCE for 14 days by gavage at 0, 24, or 240 mg/kg-day) had demonstrated a decrease
in cell-mediated immune response to SRBC in male mice at both treatment levels.
A significant decrease in humoral immunity (as measured by plasma hemagglutination
titers and the number of spleen antibody producing cells of mice sensitized to sheep
erythrocytes) was observed by Kaufmann et al. (1982) in female CD-I mice (15-20/group)
following a 90-day drinking water exposure to 0, 0.07, or 0.7 mg/mL (equivalent to 0, 18, or
173 mg/kg) chloral hydrate, a metabolite of TCE. Similar responses were not observed in male
CD-I mice exposed for 90 days in drinking water (at doses of 0, 16, or 160 mg/kg-day), or when
administered chloral hydrate by gavage to 12/group for 14 days at 14.4 or 144 mg/kg-day.
The potential for developmental immunotoxicity was assessed in B6C3F1 mice
administered TCE in drinking water at dose levels of 0, 1,400 or 14,000 ppb from gestation day
(GD) 0 to either 3 or 8 weeks of age (Adams et al., 2003 [preliminary data]; Peden-Adams et al.,
2006). At 3 and 8 weeks of age, offspring lymphocyte proliferation, NK cell activity, SRBC-
specific IgM production (plaque-forming cell [PFC] response), splenic B220+ cells, and thymus
and spleen T-cell immunophenotypes were assessed. Delayed-typed hypersensitivity and
autoantibodies to double-stranded DNA (dsDNA) were evaluated in offspring at 8 weeks of age.
Observed positive responses consisted of suppressed PFC responses in males at both ages and
both TCE treatment levels, and in females at both ages at 14,000 ppb and at 8 weeks of age at
1,400 ppb. Spleen numbers of B220+ cells were decreased in 3-week old pups at 14,000 ppb.
Pronounced increases in all thymus T-cell subpopulations (CD4+, CD8+, CD4+/CD8+, and
CD4-/CD8-) were observed at 8 weeks of age. Delayed hypersensitivity response was
increased in 8-week old females at both treatment levels and in males at 14,000 ppb only. No
treatment-related increase in serum anti-dsDNA antibody levels was found in the offspring at 8
weeks of age.
In a study designed to examine potential susceptibility of the young (Blossom and Doss,
2007), TCE was administered to groups of pregnant MRL +/+ mice in drinking water at
occupationally-relevant levels of 0, 0.5, or 2.5 mg/mL. A total of 3 litters per treatment group
were maintained following delivery (i.e., a total of 11 pups at 0 mg/mL TCE, 8 pups at
0.5 mg/mL TCE, and 12 pups at 2.5 mg/mL TCE), and TCE was continuously administered to
the offspring until young adulthood (i.e., 7-8 weeks of age). Although there were no effects on
reproduction, offspring postweaning body weights were significantly decreased in both treated
groups. Additionally, TCE exposure was found to modulate the immune system following
developmental and early life exposures. Decreased spleen cellularity and reduced numbers of
CD4+, CD8+, and B220+ lymphocyte subpopulations were observed in the postweaning
offspring. Thymocyte development was altered by TCE exposures, as evidenced by significant
This document is a draft for review purposes only and does not constitute Agency policy.
460 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
alterations in the proportions of double-negative subpopulations and inhibition of in vitro
apoptosis in immature thymocytes. TCE was also shown to induce a dose-dependent increase in
CD4+ and CD8+ T-lymphocyte IFNy in peripheral blood by 4-5 weeks of age, although these
effects were no longer observed at 7-8 weeks of age. Serum antihistone autoantibodies and total
IgG2a were significantly increased in treated offspring; however, no histopathological signs of
autoimmunity were observed in the liver and kidneys at sacrifice.
This increase in T-cell hyperactivity was further explored in a study by Blossom et al.
(2008). In this study, MRL +/+ mice were treated with 0 or 0.1 mg/mL TCE in the drinking
water. Based on drinking water consumption data, average maternal doses of TCE were
25.7 mg/kg-day, and average offspring (PND 24-42) doses of TCE were 31.0 mg/kg-day.
Treatment was initiated at the time of mating, and continued in the females (8/group) throughout
gestation and lactation. Pups were weaned at PND 24, and the offspring were continued on
drinking water treatment in a group-housed environment until study termination (PND 42).
Subsets of offspring were sacrificed at PND 10 and 20, at which time developmental and
functional endpoints in the thymus were evaluated (i.e., total cellularity, CD4+/CD8+ ratios,
CD24 differentiation markers, and double-negative subpopulation counts). Indicators of
oxidative stress were measured in the thymus at PND 10 and 20, and in the brain at PND 42.
Mitogen-induced intracellular cytokine production by splenic CD4+ and CD8+ T-cells was
evaluated in juvenile mice and brain tissue was examined at PND 42 for evidence of
inflammation. Behavioral testing was also conducted; these methods and results are described in
Section 4.3. TCE treatment did not affect reproductive capacity, parturition, or ability of dams to
maintain litters. The mean body weight of offspring was not different between the control and
treated groups. Evaluation of the thymus identified a significant treatment-related increase in
cellularity, accompanied by alterations in thymocyte subset distribution, at PND 20 (sexes
combined). TCE treatment also appeared to promote T-cell differentiation and maturation at
PND 42, and ex vivo evaluation of cultured thymocytes indicated increased reactive oxygen
species (ROS) generation. Evaluation of peripheral blood indicated that splenic CD4+ T-cells
from TCE-exposed PND 42 mice produced significantly greater levels of IFN-y and IL-2 in
males and TNF-a in both sexes. There was no effect on cytokine production on PND 10 or 20.
The dose of TCE that resulted in adverse offspring outcomes in this study (i.e., 0.1 mg/mL,
equivalent to 25.7-31.0 mg/kg-day) is comparable to that which has been previously
demonstrated to result in immune system alterations and autoimmunity in adult MRL +/+ mice
(i.e., 0.1 mg/mL, equivalent to 21 mg/kg-day; (Griffin et al., 2000b).
Another study that examined the effects of developmental exposure to TCE on the
MRL+/+ mouse was conducted by Peden-Adams et al. (2008). In this study, MRL/MpJ (i.e.,
This document is a draft for review purposes only and does not constitute Agency policy.
461 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
MRL +/+) mice (unspecified number of dams/group) were exposed to TCE (solubilized with 1%
emulphor) in drinking water at levels of 0, 1,400, or 14,000 ppb from GD 0 and continuing until
the offspring were 12 months of age. TCE concentrations in the drinking water were reported to
be analytically confirmed. Endpoints evaluated in offspring at 12 months of age included final
body weight; spleen, thymus, and kidney weights; spleen and thymus lymphocyte
immunophenotyping (CD4 or CD8); splenic B-cell counts; mitogen-induced splenic lymphocyte
proliferation; serum levels of autoantibodies to dsDNA and glomerular antigen (GA),
periodically measured from 4-12 months of age; and urinary protein measures. Reported sample
sizes for the offspring measurements varied from 6-10 per sex per group; the number of source
litters represented within each sample was not specified. The only organ weight alteration was
an 18% increase in kidney weight in the 1,400 ppb males. Splenic CD4-/CD8- cells were
altered in female mice (but not males) at 1,400 ppm only. Splenic T-cell populations, numbers
of B220+ cells, and lymphocyte proliferation were not affected by treatment. Populations of
thymic T-cell subpopulations (CD8+, CD4-/CD8-, and CD4+) were significantly decreased in
male but not female mice following exposure to 14,000-ppb TCE, and CD4+/CD8+ cells were
significantly reduced in males by treatment with both TCE concentrations. Autoantibody levels
(anti-dsDNA and anti-GA) were not increased in the offspring over the course of the study,
indicating that TCE did not contribute to the development of autoimmune disease markers
following developmental exposures that continued into adult life.
Overall, the studies by Peden-Adams et al. (2006; 2008), Blossom and Doss (2007), and
Blossom et al. (2008), which examined various immunotoxicity endpoints following exposures
that spanned the critical periods of immune system development in the rodent, were generally
not designed to assess issues such as posttreatment recovery, latent outcomes, or differences in
severity of response that might be attributed to the early life exposures.
4.6.2.1.4. Intraperitoneal administration
Wright et al. reported that following 3 days of single intraperitoneal injections of TCE in
Sprague-Dawley rats at 0, 0.05, 0.5, or 5 mmol/kg/day and B6C3F1 mice at 0 or 10
mmol/kg/day, NK cell activity was depressed in the rats at the mid- and high-dose levels, and in
the mice at the high dose level (Wright et al., 1991). Also at the highest dose levels tested,
decreased splenocyte counts and relative spleen weight were observed in the rats and mice,
respectively. In vitro assays demonstrated treatment-related decreases in splenocyte viability,
inhibition of lipopolysaccharide-stimulated lymphocyte mitogenesis, and inhibited NK cell
This document is a draft for review purposes only and does not constitute Agency policy.
462 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
activity suggesting the possibility that compromised immune function may play a role in
carcinogenic responses of experimental animals treated with TCE.
4.6.2.1.5. Hypersensitivity
Evidence of a treatment-related increase in delayed hypersensitivity response has been
observed in guinea pigs following dermal exposures with TCE and in mice following exposures
that occurred both during development and postnatally (see Table 4-77).
In a modified guinea pig maximization test, Tang et al. (2002) evaluated the contact
allergenicity potential of TCE and three metabolites (trichloroacetic acid, trichloroethanol, and
chloral hydrate) in 4 animals (FMMU strain, sex not specified) per group (Tang et al., 2002).
Edema and erythema indicative of skin sensitization (and confirmed by histopathology) were
observed. Sensitization rates were reported to be 71.4% for TCE and 58.3% for trichloroacetic
acid, as compared to a reference positive control response rate (i.e., 100% for 2,4-
dinitrochlorobenzene). In this study, the mean response scores for TCE, trichloroacetic acid, and
2,4-dinitrochlorobenzene were 2.3, 1.1, and 6.0, respectively. TCE was judged to be a strong
allergen and TCA was a moderate allergen, according to the criteria of Magnusson and Kligman
(Magnusson and Kligman, 1969). Trichloroethanol and chloral hydrate were not found to elicit a
dermal hypersensitivity response.
Immune-mediated hepatitis associated with dermal hypersensitivity reactions in the
guinea pig following TCE exposures was characterized by Tang et al. (2008). In this study,
FMMU strain female guinea pigs (5-6/group) were treated with intradermal injection of 0, 167,
500, 1,500, or 4,500 mg/kg TCE or with a dermal patch containing 0 or 900 mg/kg TCE and
sacrificed at 48 hours posttreatment. At the intradermal dose of 1,500 mg/kg, a significant
increase (p < 0.05) in serum AST level was observed. At 4,500 mg/kg, significantly (p < 0.01)
increased ALT and AST levels were reported, and total protein and globulin decreased
significantly (p < 0.05). Histopathological examination of the liver revealed fatty degeneration,
hepatic sinusoid dilation, and inflammatory cell infiltration. No changes were observed at the
intradermal doses of 500 mg/kg or below, or the dermal patch dose of 900 mg/kg. A Guinea Pig
Maximization Test was also conducted according to the procedures of Magnusson and Kligman
on 10 FMMU females/group, in which the total TCE dosage from induction through challenge
phases was below 340 mg/kg. TCE treatment resulted in dermal erythema and edema, and the
sensitization rate was 66% (i.e., classified as a strong sensitizer). Significant increases (p < 0.05)
in ALT, AST, lactate dehydrogenase, and relative liver weight, and significant decreases
(p < 0.05) in albumin, IgA, and GGT were observed. Additionally, hepatic lesions (diffuse
ballooning changes without lymphocyte infiltration and necrotic hepatocytes) were noted. It was
This document is a draft for review purposes only and does not constitute Agency policy.
463 DRAFT—DO NOT CITE OR QUOTE
-------
1 concluded that TCE exposure to guinea pigs resulted in delayed type hypersensitivity reactions
2 with hepatic injury that was similar to occupational medicamentosa-like dermatitis disorders
3 observed in human occupational studies.
This document is a draft for review purposes only and does not constitute Agency policy.
464 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-77. Summary of TCE hypersensitivity studies
to
Exposure route/vehicle,
duration, dose
NOAEL; LOAELa
Results
Reference, species/strain
sex/number
Induction by single intradermal
injection, then challenge by
dermal application at 21 d
0 or 0.1 mL induction; 0 or 0.2
mL challenge
TCE, TCA, TCOH, and chloral
hydrate
Edema and erythema (confirmed by
histopathology) indicative of skin
sensitization for TCE (strong sensitizer)
and TCA (moderate sensitizer)
Tang et al. (2002)
Guinea pig, FMMU strain, sex
not specified, 4/group
Intradermal injection, 0, 167,
500, 1,500, or 4,500 mg/kg
Dermal patch, 0 or 900 mg/kg
Hypersensitivity: total dose from
induction through challenge
<340 mg/kg
Intradermal NOAEL:
500 mg/kg
Intradermal LOAEL:
1,500 mg/kg
Dermal patch NOAEL:
900 mg/kg
Intradermal injection: At 1,500 mg/kg:
Sig. | AST; at 4,500 mg/kg, sig. | ALT
and AST, sig. [ total protein and
globulin; fatty degeneration of liver
Dermal patch: no effects of treatment
Hypersensitivity: sensitization rate of
66% (strong sensitizer), with edema and
erythema; sig. | ALT, AST, and lactate
dehydrogenase; sig. | relative liver
weight; sig. [ albumin, IgA, and GGT;
hepatic lesions (ballooning changes)
Tang et al., (2008)
Guinea pig, FMMU strain,
female, 5-6/group for
intradermal/dermal patch
study, 10/group for
hypersensitivity study, female
Drinking water, from GD 0 to
8 wk of age
0,1,400, or 14,000 ppb
LOAEL: 1,400 ppb
Sig. | swelling of foot pad in females
at 1,400 and in both sexes at 14,000
ppb
Peden-Adams et al. (2006)
Mouse, B6C3F1, both sexes,
5 litters/group; 4-5
pups/sex/group at 8 wksb
cs
sj
3
a
>!
&
"^s
fs*
rv «
G\ ^
^ §
>!
o
a
a
§•
>!
o
rs
o
a
>!
o
fe
H
o
o
2;
St
a
"T3
H S.
W|
O •
p
o
c
o
H
W
H
o;
Bolded study carried forward for consideration in dose-response assessment (see Section 5).
'NOAEL and LOAEL are based upon reported study findings.
bSubset of immunosuppression study.
-------
e
cd
o
*a
§
T3
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Also, as indicated in Section 4.6.2.1.2 above, in a developmental immunotoxicity-type
study in B6C3F1 mice, administration of TCE in drinking water at dose levels of 0, 1,400, or
14,000 ppb from GD 0 through to 8 weeks of age resulted in an increased delayed
hypersensitivity response in 8-week old female offspring at both treatment levels and in males at
the high dose of 14,000 ppb (Peden-Adams et al., 2006).
In an in vitro study that evaluated a number of chlorinated organic solvents, nonpurified
rat peritoneal mast cells (NPMC) and rat basophilic leukemia (RBL-2H3) cells were sensitized
with anti-DNP (dinitrophenol) monoclonal IgE antibody and then stimulated with
DNP-conjugated bovine serum albumin plus TCE (Seo et al., 2008). TCE enhanced antigen-
induced histamine release from NPMC and RBL-2H3 cells in a dose-related manner, and
increased IL-4 and TNF-a production from the RBL-2H3 cells. In an in vivo study, i.p.-injected
TCE was found to markedly enhance passive cutaneous anaphylaxis reaction in antigen-
challenged rats. These results suggest that TCE increases histamine release and inflammatory
mediator production from antigen-stimulated mast cells via the modulation of immune
responses; TCE exposure may lead to the enhancement of allergic disease through this response.
4.6.2.1.6. Autoimmunity
A number of studies have been conducted to examine the effects of TCE exposure in
mouse strains (i.e., MRL +/+, MRL -lpr, or NZB x NZW) which are all known to be genetically
susceptible to autoimmune disease. The studies have demonstrated the potential for TCE to
induce autoimmune disease (as demonstrated in Table 4-78 which summarizes those studies
which assessed serology, ex vivo assays of cultured splenocytes, and/or clinical or
histopathology). These and other studies conducted in susceptible mouse strains have proven to
be useful tools in exploring various aspects of the mode of action for this response.
Khan et al. used the MRL +/+ mouse model to evaluate the potential for TCE and one of
its metabolites, dichloroacetyl chloride (DCAC) to elicit an autoimmune response (Khan et al.,
1995). Female mice (4-5/group) were dosed by intraperitoneal injection with 10 mmol/kg TCE
or 0.2 mmol/kg DCAC every 4th day for 6 weeks and then sacrificed. Spleen weights and IgG
were increased. ANA and anti-ssDNA (single-stranded DNA) antibodies were detected in the
serum of TCE- and DC AC-treated mice; anticardiolipin antibodies were detected in the serum of
DCAC-treated mice. A greater magnitude of response observed with DCAC treatment suggested
that the metabolite may be important to the mechanism of TCE-induced autoimmunity.
Other studies in female MRL +/+ mice (8/group) examined exposure via drinking water.
In one of these studies, mice were treated with 2.5 or 5.0 mg/mL (455 or 734 mg/kg-day) TCE in
This document is a draft for review purposes only and does not constitute Agency policy.
467 DRAFT—DO NOT CITE OR QUOTE
-------
1 drinking water for up to 22 weeks (Gilbert et al., 1999; Griffin et al., 2000a). Serial sacrifices
2 were conducted at Weeks 4, 8, and 22. Significant increases in ANA and total serum
This document is a draft for review purposes only and does not constitute Agency policy.
468 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-78. Summary of autoimmune-related studies of TCE and metabolites in mice and rats (by sex, strain,
and route of exposure)3
Number/group, vehicle, dose,
duration
NOAEL;
LOAELb
Results
Reference
Serology
Ex vivo assays of cultured
splenocytes
Clinical and
histopathology
Autoimmune-prone: female MRL +/+ mice, drinking water
8 per group, 0, 2.5, or
5 mg/mL TCE (average 0, 455,
or 734 mg/kg-day), 4, 8, or 22
wk
LOAEL:
2.5 mg/mL
Increased ANA at 4 and
8 wk, no difference
between groups at 22 wk
Increased activated CD4+
T-cells and IFN-y secretion
across doses at 4 wk, these
effects were reversed at 22 wk;
decreased IL-4 secretion (4
and 22 wk)
No evidence of liver or renal
damage, based on serum
alanine aminotransferase,
sorbitol dehydrogenase, and
blood urea nitrogen.
Griffin et al.,
(2000a)
8 per group, 0, 0.1, 0.5, or 2.5
mg/mL TCE (0, 21, 100, or
400 mg/kg-day), 4 or 32 wk
LOAEL:
0.1 mg/mL
Increased ANA in all
treated groups at 4 wk, but
not at 32 wk
Increased activated CD4+
T-cells (32 wk), IFN-y
secretion (4 and 32 wk), no
effect on IL-4 secretion
Extensive hepatic
mononuclear cellular infiltrate
in 0.5 and 2.5 mg/mL groups,
and hepatocyte reactive
changes in all treated groups
at 32 wk.
Griffin et al.,
(2000b)
6-8 per group, 0, 0.1, or 0.9
mg/mL trichloroacetaldehyde
hydrate (0, 24, or 220 mg/kg-
day) or trichloroacetic acid (0,
27, or 205 mg/kg-day), 4 wk
LOAEL:
0.1 mg/mL
Increased ANA and
antihistone antibodies at
0.9 mg/mL
trichloroacetaldehyde
hydrate0
Increased activated CD4+
T-cells at 0.1 and 0.9 g/mL
doses of both metabolites.
At 0.9 mg/mL, increased
IFN-y secretion, no effect on
IL-4 secretion
No evidence of liver of kidney
damage, based on serum
alanine aminotransferase, liver
and kidney histology..
Blossom et al.
(2004)
8 per group, 0, 0.1, 0.3, or
0.9 mg/mL
trichloroacetaldehyde hydrate (
0, 13, 46, or 143 mg/kg-day),
40 wk
LOAEL:
0.9 mg/mL
Slightly suppressed anti-
ssDNA, anti-dsDNA, and
antihistone antibody
expression; differences not
statistically significant
Increased activated CD4+
T-cells and increased INF-y
secretion, no effect on IL-4
secretion
Diffuse alopecia, skin
inflammation and ulceration,
mononuclear cell infiltration,
mast cell hyperplasia, dermal
fibrosis. Statistically
significant increase at 0.9
mg/mL dose group, but also
increased at lower doses. No
liver or kidney histopathology
effects seen.
Blossom et al.
(2007)
-------
Table 4-78. Summary of autoimmune-related studies of TCE and metabolites (by sex, strain, and route of
exposure) (continued)
Number/group, vehicle, dose,
duration
NOAEL;
LOAELb
Results
Reference
Serology
Ex vivo assays of cultured
splenocytes
Clinical and
histopathology
5 per group, 0 or 0.5 mg/mL
TCE (mean 60 ^g/g-d), 48 wk
LOAEL:
0.5 mg/mL
Increased ANA after 24 wk
but not statistically
significant
Increased INF-y secretion after
36 wk but not statistically
significant
Hepatic necrosis; hepatocyte
proliferation; leukocyte
infiltrate in the liver, lungs,
and kidneys; no difference in
serum aminotransferase liver
enzymes
Cai et al.
(2008)
Autoimmune-prone: male and female offspring MRL +/+ mice, drinking water
3 litters/group,
8-12 offspring/group; 0, 0.5,
or 2.5 mg/mL, GD 0 to 7-8 wk
of age
LOAEL:
0.5 mg/mL
Increased antihistone
antibodies and total IgG2(l
in treated groups
Dose-dependent increase in
IFN-y secretion at 4-5 wk of
age but not 7-8 wk of age
No histopathological effects in
liver or kidneys
Blossom and
Doss (2007)
8 litters/group,
8-12 offspring/group; 0 or
0.1 mg/mL; maternal dose =
25.7 mg/kg-day; offspring
PND 24-42 dose = 31.0
mg/kg-day; GD 0 to PND 42
LOAEL:
0.1 mg/mL
Not evaluated
Increased IFN-y and IL-2 in
females, increased TNF-a in
both sexes
Not evaluated
Blossom et al.
(2008)
Unknown # litters/group,
6-10 offspring/sex/group; 0
(1% emulphor), 1,400, or
14,000 ppb; GD 0 to 12 mo of
age
NOAEL: 1,400 ppb
No increase in
autoantibody levels
Not evaluated
Not evaluated
Peden-Adams et
al. (2008)
-------
Table 4-78. Summary of autoimmune-related studies of TCE and metabolites (by sex, strain, and route of
exposure) (continued)
Number/group, vehicle, dose,
duration
NOAEL;
LOAELb
Results
Reference
Serology
Ex vivo assays of cultured
splenocytes
Clinical and
histopathology
Autoimmune-prone: female MRL +/+ mice, intraperitoneal injection
4-5 per group, 0 (corn oil),
10 mmol/kg TCE, or
0.2 mmol/kg dichloroacetyl
chloride, every 4th day for
6 wk
LOAEL:
10 mmol/kg TCE,
0.2 mmol/kg
dichloroacetyl
chloride
In both groups, increased
ANA and anti-ssDNA
antibodies. In
dichloroacetyl chloride
group, anticardiolipin
antibodies. No difference
in antihistone, -Sm, or
-DNA antibodies
Not evaluated
Not evaluated
Khan et al.
(1995)
6 per group, 0 (corn oil),
0.2 mmol/kg dichloroacetyl
chloride, or 0.2 mmol/kg
dichloroacetic anhydride,
2 times per wk for 6 wk
LOAEL:
0.2 mmol/kg TCE,
0.2 mmol/kg
dichloroacetic
anhydride
In both treated groups,
increased ANA
In both treated groups,
increased IL-la, IL-lfi, IL-3,
IL-6, IFN-y, G-CSF and KC
secretion; decreased IL-5. In
dichloroacetyl chloride group,
increased IL-17 and INF-ad
In both treated groups,
increased lymphocytes in
spleen, thickening of alveolar
septa with lymphocytic
interstitial infiltration
Cai et al. (2006)
Autoimmune-prone: female NZB x NZW mice, drinking water
6 per group, 0, 1,400, or
14,000 ppb TCEe f, 27 wk
exposure
LOAEL: 1,400 ppb
Increased anti-dsDNA
antibodies at 19 wk and at
32-32 wk in the 1,400 ppb
group
Not evaluated
At 14,000 ppb, proteinuria
increased beginning at 20 wk;
renal pathology scores
increased, no evidence of liver
disease
Gilkeson et al.
(2004)
10 per group, 0,1,400, or
14,000 ppb TCEf, 27 wk
exposure
LOAEL: 1,400
ppb
Increased anti-dsDNA
antibodies at 19 wk and at
32-32 wk in the 1,400 ppb
group
No effect on splenocyte NK
activity
No effect on renal pathology
score; liver disease not
examined
Keil et al.
(2009)
-------
Table 4-78. Summary of autoimmune-related studies of TCE and metabolites (by sex, strain, and route of
exposure) (continued)
Number/group, vehicle, dose,
duration
NOAEL;
LOAELb
Results
Reference
Serology
Ex vivo assays of cultured
splenocytes
Clinical and
histopathology
Autoimmune-prone: male MRL—IprApr mice, inhalation
5 per group, 0,500,1,000, or
2,000 ppm TCE, 4 h/d,
6 d/wk, 8 wk
LOAEL: 500 ppm
At >500 ppm, dose-related
liver inflammation,
splenomegaly and
hyperplasia of lymphatic
follicles; at 1,000 ppm,
immunoblastic cell
formation in lymphatic
follicles, no changes in
thymus
Kaneko et al.
(2000)
Autoimmune-inducible: female brown Norway Rat, gavage
6-8 per group, 0, 100, 200,
400 mg/kg, 5 d/wk, 6 wk
followed by 1 mg/kg HgCl2
challenge
NOAEL 500 mg/kg
Not reported8
Not evaluated
Not evaluated
White et al.
(2000)
Nonautoimmune-prone: female B6C3F1 mice, drinking water
6 per group, 0, 1,400, or
14,000 ppb TCE,ef 30 wk
exposure
LOAEL: 1,400 ppb
Anti-dsDNA increased in
1,400 ppb group beginning
at age 32 wk and in the
14,000 ppb group
beginning at age 26 wk
No effect on splenocyte NK
activity
No renal disease observed
Gilkeson et al.
(2004)
-------
Table 4-78. Summary of autoimmune-related studies of TCE and metabolites (by sex, strain, and route of
exposure) (continued)
Number/group, vehicle, dose,
duration
NOAEL;
LOAELb
Results
Reference
Serology
Ex vivo assays of cultured
splenocytes
Clinical and
histopathology
10 per group, 0,1,400, or
14,000 ppb TCE,f 30 wk
exposure
LOAEL: 1,400
ppb
Anti-dsDNA increased
beginning at 26 wk in the
14,000 ppb group and at
32 wk of age in the
1,400 ppb group;
increases in anti-ssDNA
antibodies seen in both
groups at 32 wk.
Anti-GA were not
affected
No effect on splenocyte NK
activity
Increased renal pathology
scores in 1,400 ppb group;
Significant decrease in
thymus weight in both
groups
Keil et al.
(2009)
Bolded studies carried forward for consideration in dose-response assessment (see Section 5).
Selected endpoints, based on those reported across the majority of studies. Lupus-prone mouse strains develop lupus-like condition spontaneously, with
virtually complete penetrance. The autoimmune-inducible (Brown Norway) rat has been used as a model of mercuric chloride induced glomerulonephritis and
experimental autoimmune myasthenia gravis.
bNOAEL and LOAEL are based upon reported study findings.
°No difference reported in anti-dsDNA, -ssDNA, -ribonucleosome, -SSA, -SSB, -Sm, -Jo-1, or -Scl-70 antibodies.
dNo difference reported in secretion of other cytokines measured: IL-2, IL-4, IL-10, IL-12, TNF-a, granulocyte monocyte colony stimulating factor, macrophage
inflammatory protein-la, and RANTES (CCL-5).
Dose levels cited in the report (Gilkeson et al., 2004) were incorrect; corrections provided by personal communication from Margie Peden-Adams (Medical
University of South Carolina) to Glinda Cooper (U.S. EPA) on 13 August 2008; dose levels in this table are correctly report.
fDose in mg/kg-day not given.
8Anti-dsDNA tests were described in the methods section; no effect of TCE on serum IgE levels was seen, and it is not clear if the additional serological tests
were conducted in the TCE portion of this study or if they were conducted but not reported because no effect was seen.
G-CSF = granulocyte colony stimulating factor, KC = keratinocyte-derived chemokine.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
immunoglobulin were found at 4 weeks of TCE treatment (indicating an autoimmune response),
but not at 32 weeks. Increased expression of the activation marker C44 on splenic CD4+ cells
was observed at 32 weeks. In addition, at 4 and 32 weeks, splenic T-cells from treated mice
secreted more IFN-y than control T-cells (significant at 0.5 and 2.5 mg/mL), consistent with a
Thl immune or inflammatory response. By 22 weeks of TCE treatment, a specific immune
serum antibody response directed against dichloroacetylated proteins was activated in hepatic
tissues, indicating the presence of protein adducts. There was a slight but statistically significant
increase in serum alanine aminotransferase levels at 32 weeks at 0.5 mg/mL. Histopathological
evaluation at 32 weeks revealed extensive hepatic lymphocytic cell infiltration at 0.5 and
2.5 mg/mL; all treated groups contained significantly more hepatocyte reactive changes (i.e.,
presence of multinucleated hepatocytes, variations in hepatocyte morphology, and hepatocytes in
mitosis) than controls.
In a subsequent study which assessed occupationally relevant concentrations, TCE was
administered to female MRL +/+ mice (8/group) in drinking water at treatment levels of 0.1, 0.5,
or 2.5 mg/mL (21, 100, or 400 mg/kg-day) for 4 and 32 weeks (Griffin et al., 2000b). At 4
weeks, significant increases in serum antinuclear antibody levels were observed at 0.1 and
0.5 mg/kg-day; at 32 weeks, the effects were observed at all three treatment levels. A dose-
related increase in the percentage of activated CD4+ T-cells in spleens and lymph nodes of
treated mice was observed at 32 weeks, and the CD4+ T-cells were found to secrete Thl-type
cytokines at 4 and 32 weeks.
A similar response was observed by Cai et al. following chronic (48 weeks) exposure of
TCE to female MRL +/+ mice (5/group) in drinking water at 0 or 0.5 mg/mL (approximately
60 (j,g/g/day) (Cai et al., 2008). After 11 weeks of treatment, a statistically significant decrease
in body weight gain was observed. After 24 weeks of exposure, serum ANA were consistently
elevated in treated mice as compared to control, although statistical significance was not
achieved. Apparent treatment-related effects on serum cytokines included decreased IL-6 after
36 and 48 weeks, decreased TNF-a after 48 weeks, and increased granulocyte colony stimulating
factor (G-CSF) after 36 weeks of treatment. After 36 weeks of treatment, ex vivo cultured
splenocytes secreted higher levels of IFN-y than control splenocytes. Although there were no
observed effects on serum aminotransferase liver enzymes at termination, statistically significant
incidences of hepatocytic necrosis and leukocyte infiltration (including CD3+ T lymphocytes)
into liver lobules were observed in treated mice after 48 weeks of exposure. Hepatocyte
proliferation was also increased. TCE treatment for 48 weeks also induced necrosis and
extensive infiltration of leukocytes in the pancreas, infiltration of leukocytes into the perivascular
and peribronchial regions of the lungs, and thickening of the alveolar septa in the lungs. At 36
This document is a draft for review purposes only and does not constitute Agency policy.
474 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
and 48 weeks of exposure, massive perivascular infiltration of leukocytes (including CD3+
T lymphocytes) was observed in the kidneys, and immunoglobulin deposits were found in the
glomeruli.
To examine the role of metabolic activation in the autoimmune response, Griffin et al.
(2000c) treated MRL +/+ mice with 2.5 mg/mL (300 mg/kg-day) TCE in drinking water for
4 weeks (Griffin et al., 2000c). Immune responses were examined in the presence or absence of
subcutaneous doses of 200 mg/kg-day diallyl sulfide, a specific inhibitor of CYP2E1 which is
known to be a primary CYP that is active in TCE metabolism. With diallyl sulfide cotreatment
that resulted in a decreased level of CYP2E1 apoprotein in liver microsomes, the enhanced
mitogen-induced proliferative capacity of T-cells was inhibited and the reduction in IL-4 levels
secreted by CD4+ T-cells was reversed for TCE-treated MRL +/+ mice. This study suggests that
metabolism of TCE by CYP2E1 is responsible, at least in part, for the treatment-related CD4+ T-
cell alterations.
The TCE metabolite, trichloroacetaldehyde (TCAA) or trichloroacetaldehyde hydrate
(TCAH), was also evaluated in MRL +/+ mice (Blossom et al., 2007; Blossom and Gilbert,
2006; Gilbert et al., 2004) in order to determine if outcomes similar to the immunoregulatory
effects of TCE would be observed, and to attempt to further characterize the role of metabolism
in the mode of action for TCE. At concentrations ranging from 0.04-1 mM, TCAA stimulated
proliferation of murine Thl cells treated with anti-CD3 antibody or antigen in vitro. At similar
concentrations, TCAA induced phenotypic alterations consistent with upregulation of CD28 and
downregulation of CD62L in cloned memory Thl cells and DC4+ T-cells from untreated MRL
+/+ mice. Phosphorylation of activating transcription factor 2 (ATF-2) and c-Jun (two
components of the activator protein-a transcription factor) was also observed with
TCAA-induced Thl cell activation. Higher concentrations of TCAA formed a Schiff base on
T-cells, which suppressed the ability of TCAA to phosphorylate ATF-2. These findings
suggested that TCAA may promote T-cell activation by stimulating the mitogen-activated
protein kinase pathway in association with Schiff base formation on T-cell surface proteins
(Gilbert et al., 2004).
In order to determine whether metabolites of TCE could mediate the immunoregulatory
effects previously observed with TCE treatment (i.e., the generation of lupus and autoimmune
hepatitis, associated with activation of IFN-y-producing CD4+ T-cells), Blossom et al. (2004)
administered TCE metabolites, TCAH and trichloroacetic acid (TCA), to MRL +/+ mice
(6-8/group) in drinking water for 4 weeks. Drinking water concentrations were 0, 0.1, or
0.9 mg/mL; average daily doses were calculated as 0, 24, or 220 mg/kg-day for TCAH and 0, 27,
or 205 mg/kg-day for TCA. These treatment levels were considered to be physiologically
This document is a draft for review purposes only and does not constitute Agency policy.
475 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
relevant and to reflect occupational exposure. A phenotypic analysis of splenic and lymph node
cells, cytokine profile analysis, evaluation of apoptosis in CD4+ T-cells, and examination of
serum markers of autoimmunity (anti-ssDNA, antihistone, or ANA) were conducted. Exposure
to TCAH or TCA at both treatment levels was found to promote CD4+ T-cell activation, as
shown by significant (p < 0.05) increases in the percentage of CD62L10 CD4+ T-cells in the
spleens and lymph nodes of the MRL +/+ mice. Increased levels of IFN-y were secreted by
CD4+ T-cells from mice treated by TCAH and TCA. No significant changes in body weight
were observed; spleen weights were similar between control and treated mice with the exception
of a significant decrease in spleen weight from mice treated with 0.9 mg/mL TCA. Liver and
kidney histology were not affected, and serum alanine aminotransferase levels were similar for
control and treated mice. A generalized trend towards an increase in serum autoantibodies
(anti-ssDNA) was observed in TCAH-treated mice, and slight but significant increases in
antihistone and antinuclear antibody production were observed in mice treated with
0.9 mg/mL-day TCAH.
The autoimmune response of female MRL +/+ mice to DC AC, a metabolite of TCE, and
to dichloroacetic anhydride (DCAA) a similar acylating agent, was evaluated by Cai et al.
(2006). Six mice/group were injected intraperitoneally, twice weekly for 6 weeks, with
0.2 mmol/kg DCAC or DCAA in corn oil. Body weight gain was significantly decreased after 5
or 6 weeks treatment with DCAC and DCAA. DCAC treatment resulted in significant increases
in total serum IgG (77% increase over control) and IgGl (172% increase over control), as well as
the induction of DC AC-specific IgG and IgGl. Serum IgM levels were significantly decreased
by 25 and 18% in DCAC and DCAA-treated mice, respectively. IgE levels were increased
100%) over controls in DCAC-treated mice. Of eight Thl/Th2 cytokines measured, only IL-5
was decreased in DCAC- and DCAA-treated mice. Serum ANA were detected in both DCAC-
and DCAA-treated mice. Treatment-related increases in cytokine and chemokine secretion in
cultured splenocytes were observed for DCAC and DCAA (IL-1, G-CSF, keratinocyte-derived
chemokine, IL-3, and IL-6). DCAC-treated splenocytes also secreted more IL-17 and IFN-a
than controls. Histopathological changes were observed in the spleens of DCAC and DCAA-
treated mice (lymphocyte population increases in the red pulp). With both DCAC and DCAA
treatment, the alveolar septa were thickened in the lungs, moderate levels of lymphocytic
interstitial infiltrates were present in tissues, and alveolar capillaries were clogged with
erythrocytes. These findings were attributed both to the predisposition of the MRL +/+ mice
towards autoimmune disease, and to the treatment-related induction of autoimmune responses.
Fas-dependant activation-induced cell death leading to autoimmune disease has been
shown to be related to impaired Fas or FasL ligand expression in humans and mice, and defects
This document is a draft for review purposes only and does not constitute Agency policy.
476 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
in the Fas-signaling pathways have been described in autoimmune disease models. The study by
Blossom and Gilbert examined the effects of TCAH on Fas-dependent autoimmune cell death
(Blossom and Gilbert, 2006). In this study, TCAH (1) inhibited apoptosis of antigen-activated
cells, (2) did not protect CD4+ T-cells from Fas-independent apoptosis, (3) did not inhibit
autoimmune cell death induced by direct engagement of the Fas receptor, (4) inhibited the
expression of FasL but not Fas on the surface of activated CD4+ T-cell, (5) increased release of
FasL from CD4+ cells in a metalloprotein-dependent manner, and (6) increased metalloprotein
MMP-7 expression.
Gilbert et al. (2006) studied the effect of treatment on apoptosis in CD4+ T-lymphocytes
isolated from MRL +/+ female mice that had been exposed to TCE (0, 0.1, 0.5, or 2.5 mg/mL) in
the drinking water for 4 or 32 weeks or to TCAH (0.1, 0.3, or 0.9 mg/mL) in drinking water for 4
or 40 weeks. After only 4 weeks, decreased activation-induced apoptosis was associated with
decreased FasL expression in the CD4+ T-cells, suggesting that TCE- and TCAH-induced
autoimmune disease was promoted through suppression of the process that would otherwise
delete activated self-reactive T-lymphocytes. By 32 weeks of treatment, TCE had induced
autoimmune hepatitis, which was associated with the promotion of oxidative stress, the
formation of liver protein adducts, and the stimulated production of antibodies to those adducts.
TCAH-treated mice did not exhibit autoimmune hepatitis by 40 weeks, but developed a dose-
dependent alopecia and skin inflammation (Blossom et al., 2007). TCAH appeared to modulate
the CD4+ T-cell subset by promoting the expression of an activated/effector phenotype with an
increased capacity to secrete the proinflammatory cytokine IFN-y. A 4-week exposure to TCAH
attenuated activation-induced cell death and the expression of the death receptor Fas in CD4+
cells; these effects were not seen after a 40-week exposure period. Differences in response were
tentatively attributed to higher levels of metalloproteinases (specifically MMP-7) at 4-weeks of
treatment, suggesting a possible mechanism for the promotion of skin pathology by TCAH.
The role of protein adduct formation in autoimmune response has been pursued by
various researchers. Halmes et al. administered a single i.p. dose of TCE in corn oil to male
Sprague-Dawley rats (2/group) at 0 or 1,000 mg/kg (Halmes et al., 1997). Using antiserum that
recognizes TCE covalently bound to protein, a single 50 kDa microsomal adduct was detected by
Western blot in livers of treated rats. Using affinity chromatography, a 50 kDa dichloroacetyl
protein was also isolated from rat plasma. The protein was reactive immunochemically with
anti-CYP2El antibodies. The data suggest that the protein adduct may be CYP2E1 that has been
released from TCE-damaged hepatocytes.
Cai et al. examined the role of protein haptenization in the induction of immune
responses (Cai et al., 2007). In this study, MRL +/+ mice were immunized with albumin adducts
This document is a draft for review purposes only and does not constitute Agency policy.
477 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
of various TCE reactive intermediates of oxidative metabolism. Serum immunoglobulins and
cytokine levels were measured to evaluate immune responses against the haptenized albumin.
Antigen-specific IgG responses (subtypes: IgGl, IgG2a, and IgG2b) were found. Serum levels
of G-CSF were increased in immunized mice, suggesting macrophage activation. Following
immunization with formyl-albumin, lymphocyte infiltration in the hepatic lobule and portal area
was increased. This study suggests that proteins that are haptenized by metabolites of TCE may
act as antigens to induce humoral immune responses and T-cell-mediated hepatitis.
A possible role for oxidative stress in inflammatory autoimmune disease was proposed by
Khan et al. (2001). A study was performed in which female MRL +/+ mice were treated with
10 mmol/kg TCE or 0.2 mmol/kg DCAC via intraperitoneal injection every 4th day for 2, 4, 6, or
8 weeks. Antimalondialdehyde serum antibodies, a marker of lipid peroxidation and oxidative
stress, were measured and were found to increase by 4 weeks of treatment, marginally for TCE
and significantly for DCAC. It was reported that antimalondialdehyde antibodies has also been
found to be present in the serum of systemic lupus erythematosus-prone MRL-lpr/lpr mice.
In another study that addressed the association of oxidative and nitrosative stress, and the
role of lipid peroxidation and protein nitration, in TCE-mediated autoimmune response,
Wang et al. treated female MRL +/+ mice with 0.5 mg/mL TCE in drinking water for 48 weeks
(Wang et al., 2007b). The formation of antibodies in the serum to lipid peroxidation-derived
aldehyde protein adducts was evaluated. With TCE treatment, the serum levels of
antimalondialdehyde and anti-4-hydroxynonenal protein adduct antibodies, inducible nitric oxide
synthase, and nitrotyrosine were increased. These were associated with increases in antinuclear-,
anti-ssDNA- and anti-dsDNA antibodies. The involvement of lipid peroxidation-derived
aldehyde protein adducts in TCE autoimmunity was further explored, using female MRL +/+
mice that were administered by i.p. injections of TCE at 10 mmol/kg, either every 4th day for 6 or
12 weeks (Wang et al., 2007a) or once per week for 4 weeks (Wang et al., 2008). Significant
increases in malondialdehyde and 4-hydroxynonenal protein adducts, as well as significant
induction of specific antibodies directed against these antigens were observed in both studies.
Wang et al. also demonstrated a significant proliferation of CD4+ T-cells in TCE-treated mice,
and splenic lymphocytes from TCE-treated mice released more IL-2 and IFN-y when stimulated
with MDA- or 4-hydroxynonenal-adducted mouse serum albumin (Wang et al., 2008). Overall,
the result of these studies suggest a role for lipid peroxidation aldehydes in the induction and/or
exacerbation of autoimmune response in the MRL +/+ animal model, and the involvement of
Thl cell activation.
In studies conducted in other rodent strains, less consistent outcomes have been observed.
Inhalation exposure of an autoimmune-prone strain of male mice (MRL-lpr/lpr) to 0-, 500-,
This document is a draft for review purposes only and does not constitute Agency policy.
478 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
1,000-, or 2,000-ppm TCE for 4 hours/day, 6 days/week, for 8 weeks resulted in depressed serum
IgG levels and increased numbers of lymphoblastoid cells (Kaneko et al., 2000). Also at
2,000 ppm, changes in T-cell helper to suppressor cell ratios were observed. At
histopathological evaluation, dose-dependent inflammation and associated changes were noted in
the liver at >500 ppm, hyperplasia of the lymphatic follicles of the spleen and splenomegaly
were observed at >500 ppm, and the spleen exhibited the development of an immunoblastic-cell-
like structure at 1,000 ppm.
A 26-week drinking water study of TCE in NZB x NZW (NZBWF1) autoimmune-prone
mice demonstrated an increase in anti-dsDNA antibodies at 19 weeks and at 32 and 34 weeks in
the 1,400 ppb group, and increased kidney disease at 14,000 ppb (i.e., increased proteinuria at
20 weeks; increased renal pathology scores at termination, based upon glomerular proliferation,
inflammation, and necrosis) (Gilkeson et al., 2004). 11 Also in that study, a small increase in
anti-dsDNA antibody production, without kidney disease, was observed in B6C3F1 mice, with
statistically significant (p < 0.05) or borderline (p = 0.07) effects seen in the 1,400-ppb group at
observations between 32 and 39 weeks of age, and in the 14,000 ppb group at observations
between 26 and 39 weeks of age.
Keil et al. (2009) also assessed the effects of TCE exposure on NZWBF1 mice,
comparing the responses to those of TCE-exposed B6C3F1 mice, which are not autoimmune
prone (Keil et al., 2009). In this study, groups of NZWBF1 and B6C3F1 female mice (10/dose
level) were administered 0, 1,400, or 14,000 ppb TCE in the drinking water. Treatment was
initiated at 9 weeks of age and continued until 36 weeks of age for the NZBWF1 and until
39 weeks of age for the B6C3F1 mice. Body weight; spleen, thymus, liver, and kidney weight;
spleen and thymus cellularity; and renal pathology were assessed. Splenic lymphocyte
proliferation, autoantibody production (anti-dsDNA, anti-ssDNA, and antiglomerular), total
serum IgG, NK cell activity, and mitogen-induced lymphocyte proliferation were conducted.
Administration of TCE did not result in alterations in NK cell activity or T- or B-cell
proliferation in either strain of mice. In the NZBWF1 mice, there was little evidence of an
increase or of an acceleration in ssDNA antibody production with TCE exposure, but as was seen
in the earlier study by these investigators (Gilkeson et al., 2004), dsDNA antibodies were
increased at 19 weeks and at 32-34 weeks in the 1,400 ppb group. However, antiglomerular
antibody levels were increased in NZBWF1 mice early in the study, returning to control levels
by 23 weeks of age. In the B6C3F1 mice the number of activated T-cells (CD4++/CD44+) was
1 IThe study was reported in symposium proceedings. Dose levels cited in the proceedings were incorrect; however,
corrections were provided by personal communication from Margie Peden-Adams (Medical University of South
Carolina) to Glinda Cooper (U.S. EPA) on 13 August 2008, and dose levels are correctly reported here.
This document is a draft for review purposes only and does not constitute Agency policy.
479 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
increased (significantly at 14,000 ppm; p < 0.05) and thymus weights were significantly
decreased (p < 0.05) in a dose-responsive manner. Renal pathology (as indicated by renal score
based on assessment of glomerular inflammation, proliferation, crescent formation and necrosis)
was significantly increased (p < 0.05) at 1,400 ppm. Also in the B6C3F1 mice, autoantibodies to
dsDNA were increased relative to controls beginning at 26 weeks in the 14,000-ppb group and at
32 weeks of age in the 1,400 ppb group; increases in anti-ssDNA antibodies were seen in both
groups at 32 weeks. Antiglomerular antibodies were not affected in B6C3F1 mice. In summary,
the authors concluded that this study showed that 27-30 weeks of TCE drinking water
administration to NZBWF1 (autoimmune-prone) mice did not contribute to the progression of
autoimmune disease, while similar administration to B6C3F1 (nonautoimmune-prone) mice
increased the expression of a number of markers that are associated with autoimmune disease.
This study is important in that it demonstrates that autoimmune responses to TCE exposure in
animal models are not solely dependent upon a genetic predisposition to autoimmune disease.
White et al. conducted a study in female Brown Norway rats, which have been shown to
be susceptible to development of chemically-induced IgE mediated glomerulonephritis that is
similar to the nephritic damage seen in systemic lupus erythematosus (White et al., 2000). TCE
administered by gavage 5 days/week at 100, 200, or 400 mg/kg did not increase in IgE levels
after 6 weeks exposure, or after an additional challenge with 1 mg/kg HgCh.
Several studies have examined the potential for autoimmune response following oral
exposures during pre- and postnatal immune system development, as described in
Section 4.6.2.1.2 above. Peden-Adams et al. conducted two such studies. In the first study,
B6C3F1 mice were treated with either 1,400 or 14,000 ppb TCE in drinking water from GD 0 to
postnatal Week 8 (Peden-Adams et al., 2006). No treatment-related increases in serum anti-
dsDNA antibody levels were observed in the 8-week old offspring, although it is noted that the
mouse strain used in the experiment is not an autoimmune-prone animal model. A more recent
study (Peden-Adams et al., 2008) exposed pregnant MRL +/+ mice to TCE in drinking water at
levels of 0, 1,400, or 14,000 ppb from GD 0 and continued the exposures until the offspring were
12 months of age. Consistent with the findings of the 2006 publication, autoantibody levels
(anti-dsDNA and antiglomerular) were not increased in the offspring over the course of the
study. Contrasting with these negative studies, the lupus-prone MRL +/+ mouse model was
utilized in two additional drinking water studies with developmental exposures in which there
was some indication of a positive association between developmental exposures to TCE and the
initiation of autoimmune disease. Blossom and Doss (2007) administered TCE to pregnant MRL
+/+ mice in drinking water at levels of 0, 0.5, or 2.5 mg/mL and continued administration to the
offspring until approximately 7-8 weeks of age. TCE exposure induced a dose-dependent
This document is a draft for review purposes only and does not constitute Agency policy.
480 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
increase in T-lymphocyte IFN-y in peripheral blood at 4-5 weeks of age, but this effect was not
observed in splenic T-lymphocytes at 7-8 weeks of age. Serum antihistone autoantibodies and
total IgG2a were significantly increased in the TCE-treated offspring; however, histopathological
evaluation of the liver and kidneys did not reveal any treatment-related signs of autoimmunity.
In a study by Blossom et al. (2008), pregnant MRL +/+ mice were administered TCE in the
drinking water at levels of 0 or 0.1 mg/mL from GD 0 through lactation, and continuing
postweaning in the offspring until GD 42. Significant treatment-related increases in pro-
inflammatory cytokines (IFN-y and 11-2 in males and TNF-a in both sexes) produced by splenic
CD4+ T-cells were observed in PND 42 offspring.
In summary, TCE treatment induces and exacerbates autoimmune disease in genetically
susceptible strains of mice, and has also been shown to induce signs of autoimmune disease in a
nongenetically predisposed strain. Although the mechanism for this response is not fully
understood, a number of studies have been conducted to examine this issue. The primary
conclusion to date is that metabolism of the TCE to its chloral or dichloroacetic acid metabolites
is at least partially responsible for activating T-cells or altering T-cell regulation and survival
associated with polyclonal disease in susceptible mice strains.
4.6.2.1.7. Cancers of the Immune System
Cancers of the immune system that have been observed in animal studies and are
associated with TCE exposure are summarized in Tables 4-79 and 4-80. The specific tumor
types observed are malignant lymphomas, lymphosarcomas, and reticulum cell sarcomas in mice
and leukemias in rats.
In the NCI (1976) study, the results for Osborne-Mendel rats were considered
inconclusive due to significant early mortality, but exposure to B6C3F1 mice were also
analyzed. Limited increases in lymphomas over controls were observed in both sexes of mice
exposed (see Table 4-79). The NCI study (1976) used technical grade TCE which contained
two known carcinogenic compounds as stabilizers (epichlorohydrin and 1,2-epoxybutane). A
later study (Henschler et al., 1984) in which mice were given TCE that was pure, industrial, and
stabilized with one or both of these stabilizers did not find significant increases in lymphomas
over historical controls. A gavage study by NTP (1988), which used TCE stabilized with
diisopropylamine, did not see an increase in lymphomas in all four strains of rats (ACI, August,
Marshall, and Osborne-Mendel). The final NTP study (1990) in male and female F344 rats and
B6C3F1 mice, using epichlorohydrin-free TCE, again experienced early mortality in male rats.
This study did not observe significant increase in lymphomas over that of controls. Henschler et
This document is a draft for review purposes only and does not constitute Agency policy.
481 DRAFT—DO NOT CITE OR QUOTE
-------
1 al. (1980) tested NMRI mice, WIST rats and Syrian hamsters of both sexes, and observed a
2 variety of tumors in both sexes (Henschler et al., 1980), consistent with the spontaneous tumor
3 Table 4-79. Malignant lymphomas incidence in mice exposed to TCE in
4 gavage and inhalation exposure studies
5
Cancer type, species, and sex
Prevalence in exposure groups:
n affected//? total (% affected)
Reference
Gavage exposure
Malignant lymphomas
Vehicle control
1,000 mg/kg-day
NTP (1990)
B6C3F1 mice, male
11/50 (22%)
13/50 (26%)
B6C3F1 mice, female
7/48 (15%)
13/49 (27%)
Lymphosarcomas and reticulum
cell sarcomas
Vehicle control
Low dose
High dose
NCI (1976)b
B6C3F1 mice, male
1/20 (5%)
4/50 (8%)
2/48 (4%)
B6C3F1 mice, female
1/20 (5%)
5/50 (10%)
5/47(11%)
Malignant lymphomas
Control
TCE-
pure
TCE-
indust
TCE-
EPC
TCE-
BO
TCE-
EPC-BO
Henschler et al.
(1984)c
Swiss (ICR/HA) mice, male
19/50
(38%)
16/50
(32%)
17/49
(35%)
11/49
(22%)
11/49
(22%)
12/49
(24%)
Swiss (ICR/HA) mice, female
28/50
(56%)
21/50
(42%)
19/50
(38%)
20/50
(40%)
23/48
(48%)
18/50
(36%)
Inhalation exposure
Malignant lymphomas
Control
96
480
Henschler et al.
(1980)d
Han:NMRI mice, male
7/30 (23%)
7/29 (24%)
6/30 (20%)
Han:NMRI mice, female6
9/29 (31%)
17/30 (57%)
18/28 (64%)
6
7 aAfter 103 wk gavage exposure, beginning at 8 wk of age.
8 bAfter 90 wk gavage exposure, beginning at 5 wk of age. Low dose is 1,200 mg/kg-day for male mice,
9 900 mg/kg-day for female mice (5 d/wk). High dose is 2,400 mg/kg-day for male mice, 1,800 mg/kg-day for
10 female mice (5 d/wk).
11 0 After 72 wk gavage exposure (corn oil), beginning at 5 wk of age. Male mice received 2,400 mg/kg-day, female
12 mice received 1,800 mg/kg-day. Stabilizers were added in the percentage w/w: TCE-EPC, 0.8%, TCE-BO, 0.8%,
13 TCE-EPC-BO, 0.25 and 0.25%.
14 dAfter 78 wk inhalation exposure. Administered daily concentration: low dose is 96 (mg/m3) and high dose is 480
15 (mg/m3), equivalent to 100 and 500 ppm (100 ppm = 540 mg/m3), adjusted for 6 h/d, 5 d/wk exposure.
16 "Statistically significant by Cochran-Armitage trend test (p < 0.05).
17
18 Sources: NTP (1990) Tables 8, 9; NCI (1976) Table XXXa; Henschler et al. (1980) Table 3a.
19
This document is a draft for review purposes only and does not constitute Agency policy.
482 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-80. Leukemia incidence in rats exposed to TCE in gavage and
2 inhalation exposure studies
3
Species and sex
Prevalence in exposure groups:
n affected/« total (% affected)
Reference
Gavage exposure
Control
50 mg/kg
250 mg/kg
Maltoni et al. (1986)a
Sprague-Dawley rats, male
0/30
(0%)
2/30 (6.7%)
3/30
(10.0%)
Sprague-Dawley rats, female
1/30
(3.3%)
0/30
(0%)
0/30
(0%)
Control
500 mg/kg
1,000 mg/kg
NTP (1988)b
August rats, female
0/50
(0%)
1/50
(2%)
5/50
(10%)
Inhalation exposure
Control
100 ppm
300 ppm
600 ppm
Maltoni etal. (1988)c
Sprague-Dawley rats, male
9/135
(6.7)
13/130
(10.0)
14/130
(10.8)
15/130
(11.5)
Sprague-Dawley rats, female
7/145
(4.8)
9/130
(6.9)
2/130
(1.5)
11/130
(8.5)
4
5 "After 52 wk gavage exposure, beginning at 13 wk of age, olive oil vehicle. Percentage affected and starting n given
6 in reported; EPA calculated n affected.
7 bAfter 104 wk gavage exposure, beginning at 6.5-8 wk of age, corn oil vehicle.
8 0 After 104 wk inhalation exposure, BT304 and BT304bis. Percentage affected and starting n given in reported; EPA
9 calculated n affected.
10
11
12 incidence in this strain (Deerberg and Miiller-Peddinghaus, 1970; Deerberg et al., 1974).
13 Henschler et al. did not show an increase in lymphomas in rats or hamsters of either sex
14 (Henschler et al., 1980). Background levels of lymphomas in this mouse strain are high, making
15 it difficult to determine if the increased lymphomas in female mice is a treatment effect. In a
16 follow-up study, Henschler et al. (1984) examined the role of stabilizers of TCE in the
17 lymphomas demonstrated in female mice in the 1980 paper. Each exposure group had
18 -50 SPF-bred ICR/HA-Swiss mice and exposure was for 18 months. Background incidence of
19 tumors was high in all groups. Focusing just on malignant lymphomas (see Table 4-79), the high
20 background incidence in unexposed animals again makes it difficult to determine if there is TCE
21 and/or stabilizer-related incidence of lymphomas. There are no data at any other timepoint than
22 18 months. A high mortality rate in all animals as well as the increased incidence of
23 'background' lymphomas in that report was also a problem and may have been related to the
24 shorter time frame.
25 Maltoni et al. reported a nonsignificant increase in leukemias in male rats exposed via
26 inhalation (Matoni et al., 1988, 1986). Maltoni et al. (1986) demonstrates a borderline higher
This document is a draft for review purposes only and does not constitute Agency policy.
483 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
frequency of leukemias in male Sprague-Dawley rats following exposure by ingestion for
52 weeks, believed by the authors to be related to an increase in lymphoblastic lymphosarcomas
(see Table 4-80). The gavage study by NTP (1988), which used TCE stabilized with
diisopropylamine, observed leukemia in female August rats with a positive trend, but was not
significantly greater than the vehicle controls.
In summary, overall there is limited available data in animals on the role of TCE in
lymphomas and leukemias. There are few studies that analyze for lymphomas and/or leukemias.
Lymphomas were described in four studies (Henschler et al., 1984; Henschler et al., 1980; NCI,
1976; NTP, 1990) but study limitations (high background rate) in most studies make it difficult
to determine if these are TCE-induced. Three studies found positive trends in leukemia in
specific strains and/or gender (Maltoni et al., 1988; Maltoni et al., 1986; NTP, 1988). Due to
study limitations, these trends cannot be determined to be TCE-induced.
4.6.3. Summary
4.6.3.1.1. Noncancer Effects
The human and animal studies of TCE and immune-related effects provide strong
evidence for a role of TCE in autoimmune disease and in a specific type of generalized
hypersensitivity syndrome. The data pertaining to immunosuppressive effects is weaker. It
should also be noted that immune-related and inflammatory effects, particularly cell-mediated
immunity involving cytokine production and activation of macrophages and natural killer cells,
may influence a variety of other conditions of considerable public health importance, including
cancer (tumor surveillance) and atherosclerosis. Thus the relevance of immune-related effects of
TCE should are not only limited to diseases affecting organs and tissues within the immune
system. The relation between systemic autoimmune diseases, such as scleroderma, and
occupational exposure to TCE has been reported in several recent studies. A meta-analysis of
scleroderma studies (Diot et al., 2002; Garabrant et al., 2003; Nietert et al., 1998) conducted by
the EPA resulted in a statistically significant combined odds ratio for any exposure in men
(OR: 2.5, 95% CI: 1.1, 5.4), with a lower relative risk seen in women in women (OR: 1.2,
95% CI: 0.58, 2.6). The incidence of systemic sclerosis among men is very low (approximately
1 per 100,000 per year), and is approximately 10 times lower than the rate seen in women
(Cooper and Stroehla, 2003). Thus, the human data at this time do not allow us to determine if
the difference in effect estimates between men and women reflects the relatively low background
risk of scleroderma in men, gender-related differences in exposure prevalence or in the reliability
This document is a draft for review purposes only and does not constitute Agency policy.
484 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
of exposure assessment (Messing et al., 2003), a gender-related difference in susceptibility to the
effects of TCE, or chance. Changes in levels of inflammatory cytokines were reported in an
occupational study of degreasers exposed to TCE (Iavicoli et al., 2005) and a study of infants
exposed to TCE via indoor air (Lehmann et al., 2001; Lehmann et al., 2002). Experimental
studies support the biological plausibility of these effects. Numerous studies have demonstrated
accelerated autoimmune responses in autoimmune-prone mice (Blossom et al., 2007; Blossom et
al., 2004; Cai et al., 2008; Griffin et al., 2000a; Griffin et al., 2000b). With shorter exposure
periods, effects include changes in cytokine levels similar to those reported in human studies.
More severe effects, including autoimmune hepatitis, inflammatory skin lesions, and alopecia,
were manifest at longer exposure periods, and interestingly, these effects differ somewhat from
the "normal" expression in these mice. Immunotoxic effects, including increases in anti-dsDNA
antibodies in adult animals and decreased plaque forming cell response with prenatal and
neonatal exposure, have been also reported in B6C3F1 mice, which do not have a known
particular susceptibility to autoimmune disease (Gilkeson et al., 2004; Peden-Adams et al.,
2006). Recent mechanistic studies have focused on the roles of various measures of oxidative
stress in the induction of these effects by TCE (Wang et al., 2007b; Wang et al., 2008).
There have been a large number of case reports of a severe hypersensitivity skin disorder,
distinct from contact dermatitis and often accompanied by hepatitis, associated with occupational
exposure to TCE, with prevalences as high as 13% of workers in the same location (Kamijima et
al., 2007; Kamijima et al., 2008). Evidence of a treatment-related increase in delayed
hypersensitivity response accompanied by hepatic damage has been observed in guinea pigs
following intradermal injection (Tang et al., 2008), 2002), and hypersensitivity response was also
seen in mice exposed via drinking water pre- and postnatally (GD 0 through to 8 weeks of age)
(Peden-Adams et al., 2006).
Human data pertaining to TCE-related immunosuppression resulting in an increased risk
of infectious diseases is limited to the report of an association between reported history of
bacteria of viral infections in Woburn, MA (Lagakos et al., 1986). Evidence of localized
immunosuppression, as measured by pulmonary response to bacterial challenge (i.e., risk of
Streptococcal pneumonia-related mortality and clearance of Klebsiella bacteria) was seen in an
acute exposure study in CD-I mice (Aranyi et al., 1986). A 4-week inhalation exposure in
Sprague-Dawley rats reported a decrease in plaque forming cell response at exposures of
1,000 ppm (Woolhiser et al., 2006).
This document is a draft for review purposes only and does not constitute Agency policy.
485 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
4.6.3.1.2. Cancer
Associations observed in epidemiologic studies of lymphoma and TCE exposure suggest
a causal relation between trichloroethylene exposure and NHL. Issues of study heterogeneity,
potential publication bias, and weaker exposure-response results contribute uncertainty to the
evaluation of the available data.
In a review of the NHL studies, studies in which there is a high likelihood of TCE
exposure in individual study subjects (e.g., based on job-exposure matrices, biomarker
monitoring, or industrial hygiene data on TCE exposure patterns and factors that affect such
exposure) and which met, to a sufficient degree, the standards of epidemiologic design and
analysis were identified. These studies generally reported excess relative risk estimates for NHL
between 0.8 and 3.1 for overall TCE exposure. Statistically significant elevated relative risk
estimates with NHL and overall TCE exposure were observed in two cohort (Hansen et al., 2001;
Raaschou-Nielsen et al., 2003) and one case-control (Hardell et al., 1994) study. Both cohort
studies reported statistically significant associations with NHL for subjects with longer
employment duration as a surrogate of TCE exposure as does a second case-control study with
high-quality exposure-assessment methodology reported statistically significant associations
with highest cumulative TCE exposure or highest average-weekly TCE exposure (Purdue et al.,
2011). Hardell et al. (1994) reported a strong but imprecise association, in part reflecting
possible bias from subject-reported exposure history and few exposed cases. Other identified
studies reported a 10-50% elevated relative risk estimate with overall TCE exposure that were
not statistically significant, except for two population case-control studies of NHL, one of which
did not report relative risk estimates with overall TCE exposure but did for medium-high
intensity or cumulative TCE exposure (Anttila et al., 1995; Axelson et al., 1994; Boice et al.,
1999; Cocco et al., 2010; Greenland et al., 1994; Miligi et al., 2006; Morgan et al., 1998;
Nordstrom et al., 1998; Persson and Fredrikson, 1999; Purdue et al., 2011; Radican et al., 2008;
Siemiatycki, 1991; Wang et al., 2009; Zhao et al., 2005). Fifteen additional studies were given
less weight because of their lesser likelihood of TCE exposure and other design limitations that
would decrease study power and sensitivity (ATSDR, 2004a, 2006a; Blair et al., 1989; Chang et
al., 2003b; Chang et al., 2005; Clapp and Hoffman, 2008; Cohn et al., 1994b; Costa et al., 1989;
Garabrant et al., 1988; Henschler et al., 1995; Morgan and Cassady, 2002; Ritz, 1999a;
Vartiainen et al., 1993; Wilcosky et al., 1984) Sinks et al., 1992. The observed lack of
association with NHL in these studies likely reflects study design and exposure assessment
limitations and is not considered inconsistent with the overall evidence on TCE and NHL.
Consistency of the association between TCE exposure and NHL is further supported by
the results of meta-analyses of 17 studies reporting risk estimates for overall TCE exposure that
This document is a draft for review purposes only and does not constitute Agency policy.
486 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
met the meta-analysis inclusion criteria. These meta-analyses found a statistically significant
increased summary relative risk estimate for NHL of 1.23 (95% CI: 1.07, 1.42) for overall TCE
exposure. The analysis of NHL was robust to the removal of individual studies and the use of
alternate relative risk estimates from individual studies, and in only one cases was the resulting
summary relative risk no longer statistically significant (lower 95% confidence bounds of 1.00).
Some evidence heterogeneity was observed, particularly between cohort and case-control
studies, but it was not statistically significant (p = 0.10); and, in addition, there was some
evidence of potential publication bias. Analyzing the cohort and case-control studies separately
resolved most of the heterogeneity, but the result for the summary case-control studies was only
a 7% increased relative risk estimate and was not statistically significant. The sources of
heterogeneity are uncertain but may be the result of some bias associated with exposure
assessment and/or disease classification, or from differences between cohort and case-control
studies in average TCE exposure.
Exposure-response relationships are examined in the TCE epidemiologic studies only to a
limited extent. Many studies examined only overall "exposed" versus "unexposed" groups and
did not provide exposure information by level of exposure. Others do not have adequate
exposure assessments to confidently distinguish between levels of exposure. The NHL case-
control study of Purdue et al. (2011) reported a statistically significant trend with TCE exposure
(p = 0.02 for average-weekly TCE exposure), and NHL risk in Boice et al. (1999) appeared to
increase with increasing exposure duration (p = 0.20 for routine-intermittent exposed subjects).
The borderline statistically significant trend with TCE intensity in the case-control study of
Wang et al. (2009 \p = 0.06]) and with cumulative TCE exposure in the case-control study of
Purdue et al. (2011 \p = 0.08] is consistent with that observed with average weekly TCE
exposure in Purdue et al. (2011). Further support was provided by meta-analyses using only the
highest exposure groups, which yielded a higher summary relative risk estimate (1.57 [95% CI:
1.27, 1.94]) than for overall TCE exposure (1.27 [95% CI: 1.04, 1.44]).
Few risk factors are recognized for NHL, with the exception of viruses,
immunosuppression or smoking, which are associated with specific NHL subtypes (Besson et al.,
2006). Associations between NHL and TCE exposure are based on groupings of several
subtypes. Two of the seven NHL case-control studies adjusted for age, sex and smoking in
statistical analyses (Miligi et al., 2006; Wang et al., 2009), two others adjusted for age and sex
(Cocco et al., 2010; Purdue et al., 2011), and the other three case-control studies presented only
unadjusted estimates of the odds ratio (Hardell et al., 1994; Nordstrom et al., 1998; Persson and
Fredrikson, 1999).
This document is a draft for review purposes only and does not constitute Agency policy.
487 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Animal studies describing rates of lymphomas and/or leukemias in relation to TCE
exposure (Henschler et al., 1984; Henschler et al., 1980; Maltoni et al., 1988; Maltoni et al.,
1986; NCI, 1976; NTP, 1988, 1990) are available. Henschler et al. (1980) reported statistically
significant increases in lymphomas in female Han:NMRI mice treated via inhalation. While
Henschler et al. (1980) suggested these lymphomas were of viral origin specific to this strain,
subsequent studies reported increased lymphomas in female B6C3F1 mice treated via corn oil
gavage (NTP, 1990) and leukemias in male Sprague-Dawley and female August rats (Maltoni et
al., 1986; NTP, 1988). However, these tumors had relatively modest increases in incidence with
treatment, and were not reported to be increased in other studies.
4.7. RESPIRATORY TRACT TOXICITY AND CANCER
4.7.1. Epidemiologic Evidence
4.7.1.1.1. Chronic Effects: Inhalation
Two reports of a study of 1,091 gun-manufacturing workers are found on noncancer
pulmonary toxicity (Cakmak et al., 2004; Saygun et al., 2007). A subset of these workers
(n = 411) had potential exposure to multiple organic solvents including toluene, acetone, butanol,
xylene, benzene and TCE used to clean gun parts; however, both papers lacked information on
exposure concentration. Mean exposure duration in Cakmak et al. (2004) was 17 years
(SD = 7.9) for nonsmokers and 16 years (SD = 7.1) for smokers. Cakmak et al. (2004) indicated
effects of smoking and exposure to solvents, with smoking having the most important effect on
asthma-related symptoms (smoking, OR = 2.8, 95% CI: 2.0, 3.8; solvent exposure, OR = 1.4,
95% CI: 1.1, 1.9). Similarly, smoking, but not solvent exposure, was shown as a statistically
significantly predictor of lung function decrements. Saygun et al. (2007) reported on a 5-year
follow-up of 393 of the original 1,091 subjects, 214 of who were exposed to solvents. Of the
393 original subjects, the prevalence of definitive asthma symptoms, a more rigorous definition
than used by Cakmak et al. (2004), was 3.3% among exposed and 1.1% among nonexposed
subjects,/* > 0.05. Saygun et al. (2007) presents observations on lung function tests for 697
current workers, a group which includes the 393 original study subjects. Smoking, but not
solvent exposure, was a predictor of mean annual forced expiratory volume (FEVi) decrease.
This document is a draft for review purposes only and does not constitute Agency policy.
488 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
4.7.1.1.2. Cancer
Cancers of the respiratory tract including the lung, bronchus, and trachea are examined in
25 cohort, community studies and case-control studies of TCE. Twelve of the 25 studies
approached standards of epidemiologic design and analysis identified in the review of the
epidemiologic body of literature on TCE and cancer (Anttila et al., 1995; see Appendix B;
Axelson et al., 1994; Blair et al., 1998; Boice et al., 1999; Boice et al., 2006b; Greenland et al.,
1994; Hansen et al., 2001; Morgan et al., 1998; Raaschou-Nielsen et al., 2003; Radican et al.,
2008; Siemiatycki, 1991; Zhao et al., 2005). Cancers at other sites besides lung, bronchus, and
trachea in the respiratory system are more limitedly reported in these studies. Some information
is available on laryngeal cancer; however, only 9 of the 16 occupational cohort studies providing
information on lung cancer also reported findings for this site. Case-control studies of lung or
laryngeal cancers and occupational title or organic solvent exposure were found in the literature.
Two case-control studies of lung cancer, one population-based and the other nested within a
cohort, were of TCE exposure specifically. Lung and laryngeal cancer risk ratios reported in
cohort, community and case-control studies are found in Table 4-81.
Lung cancer relative risks were reported in 11 of 12 cohort studies of aircraft
manufacturing, aircraft maintenance, aerospace, and metal workers, with potential exposure to
TCE as a degreasing agent, and in occupational cohort studies employing biological markers of
TCE exposures. All 11 studies had a high likelihood of TCE exposure in individual study
subjects and were judged to have met, to a sufficient degree, the standards of epidemiologic
design and analysis (Anttila et al., 1995; Axelson et al., 1994; Blair et al., 1998; Boice et al.,
1999; Boice et al., 2006b; Greenland et al., 1994; Hansen et al., 2001; Morgan et al., 1998;
Raaschou-Nielsen et al., 2003; Radican et al., 2008; Zhao et al., 2005). Lung cancer risks were
not reported for Fernald uranium processing workers with potential TCE exposure (Ritz, 1999a),
a study of less weight than the other 11 studies.. The incidence study of Raaschou-Nielsen et al.
(2003) was the largest cohort, with 40,049 subjects identified as potentially exposed to TCE in
several industries (primarily, in the iron/metal and electronic industries), including 14,360 of
whom had presumably higher level exposures to TCE. The study included 632 lung cancer cases
and reported a 40% elevated incidence in TCE exposed males and females combined (95% CI:
1.32, 1.55), with no exposure duration gradient. The 95% CIs in other studies of lung cancer
incidence included a risk ratio of 1.0 (Anttila et al., 1995; Axelson et al., 1994; Blair et al., 1998;
Hansen et al., 2001; Zhao et al., 2005). Lung cancer mortality risks in studies of TCE exposure
to aircraft manufacturing, aircraft maintenance, and aerospace workers included a relative risk of
1.0 in their 95% CIs (Blair et al., 1998; Boice et al., 2006b; Morgan et al., 1998; Radican et al.,
2008; Zhao et al., 2005). Boice et al. (1999) observed a 24% decrement (95% CI: 0.60, 0.95) for
This document is a draft for review purposes only and does not constitute Agency policy.
489 DRAFT—DO NOT CITE OR QUOTE
-------
1 subjects with routine TCE exposure. Exposure-response analyses using internal controls
2 (unexposed subjects at the same company) showed a statistically significant decreasing trend
3 between lung cancer risk and routine or intermittent TCE exposure duration. The routine or
4 intermittent category is broader and includes more subjects with potential TCE exposure. Five
5 other studies with internal controls do not provide evidence of either an increasing or decreasing
6 pattern between TCE and lung cancer incidence or mortality (Blair et al., 1998; Boice et al.,
7 2006b; Morgan et al., 1998; Radican et al., 2008; Zhao et al., 2005).
8
This document is a draft for review purposes only and does not constitute Agency policy.
490 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-81. Selected results from epidemiologic studies of TCE exposure and
2 lung cancer
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Cohort studies—incidence
Aerospace workers (Rocketdyne)
Zhao et al. (2005)
Any exposure to TCE
Not reported
Low cumulative TCE score
1.00a
43
Medium cumulative TCE score
1.36 (0.86,2.14)
35
High TCE score
1.11 (0.60,2.06)
14
P for trend
0.60
All employees at electronics factory (Taiwan)
1.07 (0.72, 1.52)
30
Chang et al. (2005)
Danish blue-collar worker with TCE exposure
Raaschou-Nielsen et al.
Any exposure, all subjects
1.4 (1.32, 1.55)
632
(2003)
Any exposure, males
1.4 (1.28, 1.51)
559
Any exposure, females
1.9 (1.48,2.35)
73
Employment duration
<1 yr
1.7 (1.46, 1.93)
209
1-4.9 yr
1.3 (1.16, 1.52)
218
>5 yr
1.4 (1.23, 1.63)
205
Biologically-monitored Danish workers
Hansen etal. (2001)
Any TCE exposure, males
0.8 (0.5, 1.3)
16
Any TCE exposure, females
0.7(0.01,3.8)
1
Cumulative exposure (Ikeda)
Not reported
<17 ppm-yr
>17 ppm-yr
Mean concentration (Ikeda)
Not reported
<4 ppm
4+ppm
Employment duration
Not reported
<6.25 yr
>6.25 yr
Aircraft maintenance workers (Hill Air Force Base, UT)
Blair etal. (1998)
TCE subcohort
Not reported
Males, cumulative exposure
0
1.0a
<5 ppm-yr
1.0 (0.6, 2.0)
24
5-25 ppm-yr
0.8 (0.4, 1.6)
11
>25 ppm-yr
0.8 (0.4, 1.7)
15
3
This document is a draft for review purposes only and does not constitute Agency policy.
491 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-81. Selected results from epidemiologic studies of TCE exposure and
lung cancer (continued)
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Females, cumulative exposure
Blair et al. (1998) (continued)
0
1.0a
<5 ppm-yr
1
5-25 ppm-yr
1
>25 ppm-yr
1
Biologically-monitored Finnish workers
Anttila et al. (1995)
All subjects
0.92 (0.59, 1.35)
25
Mean air-TCE (Ikeda extrapolation)
<6 ppm
1.02 (0.58, 1.66)
16
6+ppm
0.83 (0.33, 1.71)
7
Biologically-monitored Swedish workers
Axelsonetal. (1994)
Any TCE exposure, males
0.69 (0.31, 1.30)
9
Any TCE exposure, females
Not reported
Cohort and PMR-mortality
Computer manufacturing workers (IBM), NY
Clapp and Hoffman (2008)
Males
1.03 (0.71, 1.42)
35
Females
0.95 (0.20, 2.77)
3
Aerospace workers (Rocketdyne)
Any TCE (utility or engine flush workers)
1.24 (0.92, 1.63)
51
Boice et al. (2006b)
Engine flush—duration of exposure
Referent
1.0a
472
0 yr (utility workers with TCE exposure)
0.5 (0.22, 1.00)
7
<4 yr
0.8 (0.50, 1.26)
27
>4 yr
0.8 (0.46, 1.41)
24
Any exposure to TCE
Not reported
Zhao et al. (2005)
Low cumulative TCE score
1.00a
99
Medium cumulative TCE score
1.05 (0.76, 1.44)
62
High TCE score
1.02 (0.68, 1.53)
33
p for trend
0.91
View-Master employees
ATSDR (2004a)
Males
0.81 (0.42, 1.42)b
12
Females
0.99 (0.71, 1.35)b
41
This document is a draft for review purposes only and does not constitute Agency policy.
492 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-81. Selected results from epidemiologic studies of TCE exposure and
lung cancer (continued)
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
United States uranium-processing workers (Fernald)
Ritz (1999a)
Any TCE exposure
Not reported
Light TCE exposure, >2 yr duration0
Not reported
Moderate TCE exposure, >2 yr duration0
Not reported
Aerospace workers (Lockheed)
Boice et al. (1999)
Routine exposure
0.76 (0.60, 0.95)
78
Routine-intermittent exposure3
Not reported
173
Duration of exposure
0 yr
1.0
288
<1 yr
0.85 (0.65, 1.13)
66
1-4 yr
0.98 (0.74, 1.30)
63
>5 yr
0.64 (0.46, 0.89)
44
Trend test
p < 0.05
Aerospace workers (Hughes)
Morgan et al. (1998)
TCE subcohort
1.10 (0.89, 1.34)
97
Low intensity (<50 ppm)
1.49 (1.09, 1.99)
45
High intensity (>50 ppm)
0.90 (0.67, 1.20)
52
TCE subcohort (Cox Analysis)13
Never exposed
1.00a
291
Ever exposed
1.14(0.90, 1.44)
97
Peak
No/Low
1.00a
324
Medium/High
1.07 (0.82, 1.40)
64
Cumulative
Referent
1.00a
291
Low
1.47 (1.07, 2.03)
45
High
0.96 (0.72, 1.29)
52
Aircraft maintenance workers (Hill Air Force Base, UT)
Blair etal. (1998)
TCE subcohort
Any TCE exposure
0.9 (0.6, 1.3)a
109
This document is a draft for review purposes only and does not constitute Agency policy.
493 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-81. Selected results from epidemiologic studies of TCE exposure and
lung cancer (continued)
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Males, cumulative exposure
Blair et al. (1998) (continued)
0
1.0a
51
<5 ppm-yr
1.0 (0.7, 1.6)
43
5-25 ppm-yr
0.9 (0.5, 1.6)
23
>25 ppm-yr
1.1 (0.7, 1.8)
38
Females, Cumulative exp
0
1.0a
2
<5 ppm-yr
0.6(0.1,2.4)
2
5-25 ppm-yr
0.6(0.1,4.7)
11
>25 ppm-yr
0.4 (0.1, 1.8)
2
TCE subcohort
Radican et al. (2008)
Any TCE exposure
0.83 (0.63, 1.08)
166
Males, cumulative exposure
0.91 (0.67, 1.24)
155
0
1.0a
66
<5 ppm-yr
0.96 (0.67, 1.37)
5-25 ppm-yr
0.71 (0.46, 1.11)
31
>25 ppm-yr
1.00 (0.69, 1.45)
58
Females, cumulative exposure
0.53 (0.27, 1.07)
11
0
1.0a
<5 ppm-yr
0.69 (0.27, 1.77)
5
5-25 ppm-yr
0.65 (0.16,2.73)
2
>25 ppm-yr
0.39(0.14, 1.11)
4
Cardboard manufacturing workers in Arnsburg, Germany
Henschler et al. (1995)
TCE-exposed workers
1.38 (0.55, 2.86)
7
Unexposed workers
1.06 (0.34, 2.47)
5
Deaths reported to GE pension fund (Pittsfield, MA)
1.01 (0.69, 1.47)d
139
Greenland et al. (1994)
U.S. Coast Guard employees
Blair etal. (1998)
Marine inspectors
0.52 (0.31,0.82)
18
Noninspectors
0.81 (0.55, 1.16)
30
Aircraft manufacturing employees (Italy)
Costa et al. (1989)
All employees
0.99 (0.73, 1.32)
99
Aircraft manufacturing plant employees (San Diego, CA)
Garabrant et al. (1988)
All subjects
0.80 (0.68, 0.95)
138
This document is a draft for review purposes only and does not constitute Agency policy.
494 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-81. Selected results from epidemiologic studies of TCE exposure and
lung cancer (continued)
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Lamp manufacturing workers (GE)
0.58 (0.27, 1.27)
6
Shannon et al. (1988)
Rubber industry workers (Ohio)
0.64 (p > 0.05)°
11
Wilcosky et al. (1984)
Case-control studies
Population of Montreal, Canada
Siemiatycki et al. (1991)
Any TCE exposure
0.9 (0.6, 1.5)e
21
Substantial TCE exposure
0.6 (0.3, 1.2)e
9
Geographic based studies
Two study areas in Endicott, NY
1.28 (0.99, 1.62)
68
ATSDR (2006a)
Residents of 13 census tracts
Morgan and Cassidy (2002)
In Redlands, CA
0.71 (0.61, 0.81)f
356
Iowa residents with TCE in water supply
Isacsonetal. (1985)
Males
<0.15 (ig/L
343.1s
1,181
>0.15 (ig/L
345.7s
299
Females
<0.15 (ig/L
58.7s
289
>0.15 (ig/L
47.8s
59
1
2 internal referents, workers not exposed to TCE.
3 bRisk ratio from Cox Proportional Hazard Analysis, stratified by age, sex, and decade (EHS, 1997).
4 0 Odds ratio from nested case-control study.
5 dOdds ratio from nested case-control analysis.
6 e90% CI.
7 f99% CI.
8 8Average annual age-adjusted incidence (per 100,000).
9
10 GE = General Electric, IBM = International Business Machines Corporation, No. obs. events = number of observed
11 events.
12
13
14 The population studied by Garabrant et al. (1988), ATSDR (2004a) and Chang et al.
15 (2005) are all employees (white- and blue-collar) at a manufacturing facility or plant with
16 potential TCE exposures. Garabrant et al. (1988) observed a 20% deficit in lung cancer
17 mortality (95% CI: 0.68, 0.95) in their study of all employees working for 4 or more years at an
18 aircraft manufacturing company. Blair et al. (1998), a study of Coast Guard marine inspectors
19 with potential for TCE exposure but lacking assessment to individual subjects, observed a 48%
20 deficit in lung cancer mortality (95% CI: 0.31, 0.82). Confidence intervals (95% CI) in Costa et
This document is a draft for review purposes only and does not constitute Agency policy.
495 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
al. (1989), Chang et al. (2005) and ATSDR (2004a) included a risk of 1.0. TCE exposure was
not known for individual subjects in these studies. A wide potential for TCE exposure is likely
ranging from subjects with little to no TCE exposure potential to those with some TCE exposure
potential. Exposure misclassification bias, typically considered as a negative bias, is likely
greater in these studies compared to studies adopting more sophisticated exposure assessment
approaches, which are able to assign quantitative exposure metrics to individual study subjects.
All three studies were of lower likelihood for TCE exposure, in addition to limited statistical
power and other design limitations, and these aspects, in addition to potential exposure
misclassification bias were alternative explanations of observed findings.
One population case-control study examined the relationship between lung cancer and
TCE exposure (Siemiatycki, 1991) with risk ratios of 0.9 (95% CI: 0.6, 1.5) for any TCE
exposure and 0.6 (95% CI: 0.3, 1.2) for substantial TCE exposure after adjustment for cigarette
smoking. TCE exposure prevalence in cases in this study was 2.5% for any exposure. Only 1%
had "substantial" (author's term) exposure, limiting the sensitivity of this study. Relative risks
above 2.0 could only be detected with sufficient (80%) statistical power. The finding of no
association of lung cancer with TCE exposure, therefore, is not surprising. One nested case-
control study of rubber workers observed a smoking unadjusted risk of 0.64 (95% CI: not
presented in paper) in those who had >1 year cumulative exposure to TCE (Wilcosky et al.,
1984).
Three geographic based studies reported lung cancer incidence or mortality risks for
drinking water contamination with TCE (ATSDR, 2006a; Isacson et al., 1985; Morgan and
Cassady, 2002). Morgan and Cassidy (2002) observed a relative risk of 0.71 (99% CI: 0.61,
0.81) for lung cancer among residents of Redlands County, CA, whose drinking water was
contaminated with TCE and perchlorate. However, ATSDR (2006a) reported a 28% increase
(95%) CI: 0.99, 1.62) in lung cancer incidence among residents living in a area in Endicott, NY,
whose drinking water was contaminated with TCE and other solvents. No information on
smoking patterns is available for individual lung cancer cases as identified by the New York
State Department of Health (NYS DOH) for other cancer cases in this study (ATSDR, 2008).
Isacson et al. (1985) presented lung cancer age-adjusted incidence rates for Iowa residents by
TCE level in drinking water supplies and did not observe an exposure-response gradient.
Exposure information is inadequate in all three of these studies, with monitoring data, if
available, based on few samples and for current periods only, and no information on water
distribution, consumption patterns, or temporal changes. Thus, TCE exposure potential to
individual subjects was not known with any precision, introducing misclassification bias, and
greatly limiting their ability to inform evaluation of TCE and lung cancer.
This document is a draft for review purposes only and does not constitute Agency policy.
496 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Laryngeal cancer risks are presented in a limited number of cohort studies involving TCE
exposure. No case-control or geographic based studies of TCE exposure were found in the
published literature. All but one of the cohort studies providing information on laryngeal cancer
observed less that five incident cases or deaths. Accordingly, these studies are limited for
examining the relationship between TCE exposure and laryngeal cancer. Risk ratios for
laryngeal cancer are found in Table 4-82.
In summary, studies in humans examining lung and laryngeal cancer and TCE exposure
are inconclusive and do not support either a positive or a negative association between TCE
exposure and lung cancer or laryngeal cancer. Raaschou-Nielsen et al. (2003), with the largest
numbers of lung cancer cases of all studies, was the only one to observe a statistically
significantly elevated lung cancer risk with TCE exposure. Raaschou-Nielsen et al. (2003) also
noted several factors that may have confounded or biased their results in either a positive or
negative direction. This study and other cohort studies, as with almost any occupational study,
were not able to control confounding by exposure to chemicals other than TCE (although no
such chemical was apparent in the reports). Information available for factors related to
socioeconomic status (e.g., diet, smoking, alcohol consumption) was also not available. Such
information may positively confound smoking-related cancers such as lung cancer, particularly
in those studies, which adopted national rates to derive expected numbers of site-specific cancer,
if greater smoking rates were over-represented in blue-collar workers or residents of lower socio-
economic status. The finding of a larger risk among subjects with shortest exposure also argues
against a causal interpretation for the observed association for all subjects (NRC, 2006).
Four studies reported a statistically significant deficit in lung cancer incidence (Blair et
al., 1998; Boice et al., 1999; Garabrant et al., 1988; Morgan and Cassady, 2002). Absence of
smoking information in these studies would introduce a negative bias if the studied population
smoked less than the referent population and may partially explain the lung cancer decrements
observed in these studies. Morgan and Cassidy (2002) noted the relatively high education high
income levels, and high access to health care of subjects in this study compared to the averages
for the county as a whole, likely leading to a lower smoking rate compared to their referent
population. Garabrant et al. (1988) similarly attributed their observations to negative selection
bias introduced when comparison is made to national mortality rates, also known as a "healthy
worker effect." The statistically significant decreasing trend in Boice et al. (1999) with exposure
duration to intermittent or routine exposure may reflect a protective effect between TCE and lung
cancer. The use of internal controls in this analysis reduces bias associated with use of an
external population who may have different smoking patterns than an employed population.
However, the exposure assessment approach in this study is limited due to inclusion of subjects
This document is a draft for review purposes only and does not constitute Agency policy.
497 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-82. Selected results from epidemiologic studies of TCE exposure and
2 laryngeal cancer
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Cohort studies—incidence
Aerospace workers with TCE exposure
Not reported
Zhao et al. (2005)
Danish blue-collar worker with TCE exposure
Raaschou-Nielsen et al. (2003)
Any exposure, males
1.2 (0.87, 1.52)
53
Any exposure, females
1.7 (0.33, 4.82)
3
Employment duration
Not reported
<1 yr
1-4.9 yr
>5 yr
Biologically-monitored Danish workers
Hansen etal. (2001)
Any TCE exposure, males
1.1 (0.1,3.9)
2
Any TCE exposure, females
0
(0.1 exp)
Cumulative exposure (Ikeda)
Not reported
<17 ppm-yr
>17 ppm-yr
Mean concentration (Ikeda)
Not reported
<4 ppm
4+ ppm
Employment duration
Not reported
<6.25 yr
>6.25 yr
Aircraft maintenance workers (Hill Air Force Base, Utah)
Blair etal. (1998)
TCE subcohort
Any exposure
Not reported
Males, cumulative exposure
Not reported
0
<5 ppm-yr
5-25 ppm-yr
>25 ppm-yr
Females, cumulative exposure
Not reported
0
<5 ppm-yr
5-25 ppm-yr
>25 ppm-yr
3
This document is a draft for review purposes only and does not constitute Agency policy.
498 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-82. Selected results from epidemiologic studies of TCE exposure and
laryngeal cancer (continued)
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Biologically-monitored Finnish workers
Not reported
Anttila et al. (1995)
Mean air-TCE (Ikeda extrapolation from
U-TCA)
Not reported
<6 ppm
6+ppm
Biologically-monitored Swedish workers
Axelsonetal. (1994)
Any TCE exposure, males
1.39 (0.17, 5.00)
2
Any TCE exposure, females
Not reported
Cohort and PMR-mortality
Computer manufacturing workers (IBM), NY
Not reported
Clapp and Hoffman (2008)
Aerospace workers (Rocketdyne)
Any TCE (utility or engine flush workers)
1.45 (0.18, 5.25)
2
Boice et al. (2006b)
Engine flush—duration of exposure
Not reported
Referent
0 yr (utility workers with TCE exposure)
<4 yr
>4 yr
Any exposure to TCE
Not reported
Zhao et al. (2005)
View-Master employees
Not reported
ATSDR (2004a)
Males
Females
All employees at electronic factory (Taiwan)
Chang et al. (2003b)
Males
0
(0.90 exp)
Females
0
0
(0.23 exp)
United States uranium-processing workers (Fernald)
Ritz (1999a)
Any TCE exposure
Not reported
Light TCE exposure, >2 yr duration4
Not reported
Moderate TCE exposure, >2 yr duration
Not reported
Aerospace workers (Lockheed)
Boice et al. (1999)
Routine exposure
1.10 (0.30, 2.82)
4
Routine-intermittent exposure
Not reported
This document is a draft for review purposes only and does not constitute Agency policy.
499 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-82. Selected results from epidemiologic studies of TCE exposure and
laryngeal cancer (continued)
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Aerospace workers (Hughes)
Morgan et al. (1998)
TCE subcohort
Not reported
Low intensity (<50 ppm)
High intensity (>50 ppm)
Peak
Not reported
No/low
Medium/high
Cumulative
Not reported
Referent
Low
High
Aircraft maintenance workers (Hill Air Force Base, UT)
Blair et al. (1998)
TCE subcohort
Not reported
Males, cumulative exposure
Not reported
0
<5 ppm-yr
5-25 ppm-yr
>25 ppm-yr
Females, cumulative exposure
Not reported
0
<5 ppm-yr
5-25 ppm-yr
>25 ppm-yr
Cardboard manufacturing workers in Arnsburg,
Germany
Not reported
Henschler et al. (1995)
Deaths reported to GE pension fund (Pittsfield, MA)
Not examined
Greenland et al. (1994)
U. S. Coast Guard employees
Blair et al. (1998)
Marine inspectors
0.57 (0.01,3.17)
1
Noninspectors
0.58 (0.01, 3.20)
1
Aircraft manufacturing employees (Italy)
Costa et al. (1989)
All employees
0.27 (0.03, 0.98)
2
Aircraft manufacturing plant employees (San Diego, CA)
Garabrant et al. (1988)
All subjects
0
(7.41 exp)
1 GE = General Electric, IBM = International Business Machines Corporation, No. obs. events = number of observed
2 events.
This document is a draft for review purposes only and does not constitute Agency policy.
500 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
identified with intermittent TCE exposure (i.e., workers who would be exposed only during
particular shop runs or when assisting other workers during busy periods) (Boice et al., 1999).
The Boice et al. (1999) analysis is based on twice as many lung cancer deaths (i.e., 173 lung
cancer deaths) among subjects with routine or intermittent TCE exposure compared to only
routinely exposed subjects (78 deaths). Subjects identified as intermittently exposed are
considered as having a lower exposure potential than routinely exposed subject and their
inclusion in exposure-response analyses may introduce exposure misclassification bias. Such
bias is a possible explanation for the decreasing trend observation, particularly if workers with
lower potential for TCE exposure have longer exposure (employment) durations.
Thus, a qualitative assessment suggests the epidemiological literature on respiratory
cancer and TCE, although limited and of sufficient power to detect only large relative risks, does
not provide strong evidence for any association between TCE exposure and lung cancer. These
studies can only rule out risks of a magnitude of 2.0 or greater for lung cancer and relative risks
greater than 3.0 or 4.0 for laryngeal cancer for exposures to studied populations.
4.7.2. Laboratory Animal Studies
4.7.2.1.1. Respiratory Tract Animal Toxicity
Limited studies are available to determine the effects of TCE exposure on the respiratory
tract (summarized in Table 4-83). Many of these studies in mice have examined acute effects
following intraperitoneal administration at relatively high TCE doses. However, effects on the
bronchial epithelium have been noted in mice and rats with TCE administered via gavage, with
doses 1,000 mg/kg-day and higher reported to cause rales and dyspnea (Narotsky et al., 1995)
and pulmonary vasculitis (NTP, 1990) in rats. Mice appear to be more sensitive than rats to
histopathological changes in the lung via inhalation; pulmonary effects are also seen in rats with
gavage exposure. It is difficult to compare intraperitoneal to oral and inhalation routes of
exposure given the risk of peritonitis and paralytic ileus. Any inflammatory response from this
route of administration can also affect the pulmonary targets of TCE exposure such as the Clara
cells.
This section reviews the existing literature on TCE, and the role of the various TCE
metabolites in TCE-induced lung effects. The most prominent toxic effect reported is damage to
Clara cells in mouse lung. The nonciliated, columnar Clara cells comprise the majority of the
bronchiolar and terminal bronchiolar epithelium in mice, and alveolar Type I and Type II cells
constitute the alveolar epithelium. These cells have been proposed as a progenitor of lung
This document is a draft for review purposes only and does not constitute Agency policy.
501 DRAFT—DO NOT CITE OR QUOTE
-------
1 adenocarcinomas in both humans and mice (Kim et al., 2005). Long-term studies have not
2 focused on the detection of pulmonary adenoma carcinomas but have shown a consistently
This document is a draft for review purposes only and does not constitute Agency policy.
502 DRAFT—DO NOT CITE OR QUOTE
-------
to
Table 4-83. Animal toxicity studies of trichloroethylene
Reference
Animals (sex)
Exposure route
Dose/exposure concentration
Exposed
Results
Green et al.
(1997b)
CD-I mice (F)
Inhalation
450-ppm, 6 h/d, 5 d with 2 d break
then 5 more days; sacrificed 18 h
after 1, 5, 6, or 10 exposures
5/group
Increased vacuolation and proliferation of Clara
cells caused by accumulation of chloral.
Forkert and
Forkert (1994)
CD-I mice (M)
Intraperitoneal
injection
2,000 mg/kg in corn oil (0.01 mL/g
BW); sacrificed 15, 30, 60 and 90 d
after single exposure
10/group
Increased fibrotic lesions, with early signs
visible at 15 d postexposure.
Villaschi et al.
(1991)
BC3F1 mice (M)
Single inhalation
30 min 500, 1,000, 2,000, 3,500, and
7,000 ppm; sacrificed 2 h, 24 h, 2, 5,
or 7 d post exposure
3/group
Increased vacuolation and proliferation of
nonciliated bronchial cells. Injury was maximal
at 24 h with some repair occurring between 24 h
and 48 h.
Odum et al.
(1992)
CD-I mice (F)
Inhalation
6 h/d; separate repeated study in
mice: 450 ppm for 6 h/d, 5 d/wk for 2
wk; sacrificed 24 h after exposure;
repeat study sacrificed at 2, 5, 6, 8, 9,
12, or 13 d; mice: 20, 100, 200, 450,
1,000, or 2,000 ppm
4/group
Dose-dependent increase in Clara cell
vacuolation in mice after a single exposure,
resolved after 5 d repeated exposures but
recurred following a 2-d break from exposure.
Changes accompanied by decrease in CYP
activity in mice. Exposure to chloral alone
demonstrated similar response as TCE exposure
in mice. No changes were seen in rats.
Alpk APfSD rats
(F)
Inhalation
6 h/d; repeat study sacrificed at 2, 5,
6, 8, 9, 12, or 13 d; rats: 500, or 1,000
ppm
4/group
Kurasawa (1988)
(translation)
Ethanol-treated
(130) and
nontreated (110)
Wistar rats (M)
Inhalation
500, 1,000, 2,000, 4,000, and 8,000
ppm for 2 h; sacrificed 22 h after
exposure
10/group
TCE exposure resulted in highly selective
damage to Clara cells that occurred between 8
and 22 h after the highest exposure with repair
by 4 wk post exposure.
Forkert et al.
(2006)
CD-I mice (M);
wild-type (mixed
129/Sv and
C57BL) and
CYP2El-null
mice (M)
Intraperitoneal
injection
500, 750, and 1,000 mg/kg in corn
oil; for inhibition studies mice
pretreated with 100 mg/kg diallyl
sulfone; for immunoblotting, 250,
500, 750, and 1,000 mg/kg; for PNP
hydroxylation, 50, 100, 250, 500,
750, and 1,000 mg/kg; sacrificed 4 h
after exposure
4/group
TCE bioactivationby CYP2E1 and/or 2F2
correlated with bronchiolar cytotoxicity in mice.
3^
rs
sj
3
a
>!
&
fs*
, si
o ^
U> p
>!
>!
o
a
a
HH
0 &
M ^
'/O >!
fel
T" §
1 ^
o £.
TO
2; ^
O^
H J
1—1 s
H S.
W|
O •
-------
Table 4-83. Animal toxicity studies of trichloroethylene (continued)
Reference
Animals (sex)
Exposure route
Dose/exposure concentration
Exposed
Results
Forkert et al.
(1985)
CD-I mice (M)
Intraperitoneal
injection
2,000, 2,500 or 3,000 mg/kg in
mineral oil; sacrificed 24 h
postexposure for dose response; time
course sacrificed 1, 2, 12, and 24 h
postexposure
10/group
Clara cell injury was increased following
exposure at all doses tested; time course
demonstrated a rapid and marked reduction in
pulmonary microsomal cytochrome P450
content and aryl hydrocarbon hydroxylase
activity. Alveolar Type II cells were also
affected.
Forkert and Birch
(1989)
CD-I mice (M)
Intraperitoneal
injection
2,000 mg/kg in corn oil; sacrificed 1,
2, 4, 8, 12, and 24 h postexposure
10/group
Necrotic changes seen in Clara cells as soon as 1
h postexposure; increased vacuolation was seen
by 4 h postexposure; covalent binding of TCE to
lung macro molecules peaked at 4 h and reached
a plateau at 12 and 24 h post exposure.
Stewart et al.
(1979)
Le Mesurier et al.,
(1980)
Wistar Rats (F)
Inhalation
(whole-body
chamber)
30 min, 48.5 g/m3 (9,030 ppm);
sacrificed at 5 and 15 d postexposure
5/group
Decreased recovery of pulmonary surfactant
(dose-dependent).
Lewis et al.
(1984)
Mice
Inhalation (Pyrex
bell jars)
10,000 ppm, 1-4 h daily for 5
consecutive days; sacrificed 24 h
after last exposure
~28/group
Increased vacuolation and reduced activity of
pulmonary mixed function oxidases.
Scott etal. (1988)
CD-I mice (M)
Intraperitoneal
injection
single injection of 2,500-3,000
mg/kg, sacrificed 24 h postexposure
4/group
Clara cells were damaged and exfoliated from
the epithelium of the lung.
NTP (1990)
F344 rats (M,F)
B6C3F1 mice
(M,F)
Gavage
Male rats: 0, 125, 250, 500, 1,000,
and 2,000 mg/kg BW (corn oil);
female rats: 0, 62.5, 125, 250, 500 or
1,000 mg/kg BW (corn oil); Mice: 0,
375, 750, 1,500, 3,000, and 6,000
mg/kg BW (corn oil); dosed 5d/w for
13 wk
10/group
Increased pulmonary vasculitis in the high-dose
groups of male and female rats (6/10 group as
compared to 1/10 in controls). No pulmonary
effects described in mice at this time point.
-------
Table 4-83. Animal toxicity studies of trichloroethylene (continued)
Reference
Animals (sex)
Exposure route
Dose/exposure concentration
Exposed
Results
Prendergast et al.
(1967)
Sprague-Dawley
or Long-Evans
rats; Hartley
Guinea pigs;
New Zealand
albino rabbits;
beagle dogs;
squirrel monkeys
(sex not given for
any species)
Inhalation
730 ppm for 8 h/d, 5 d/w, 6 wk or 35
ppm for 90 d constant
Rats (15);
guinea pigs
(15); rabbit
(3); dog (2);
monkey (3)
No histopathological changes observed,
although rats were described to show a nasal
discharge in the 6 wk study. No quantification
was given.
Narotsky et al.
(1995)
F344 rats (F)
Gavage
0, 1,125, 1,500 mg/kg-day
21, 16, or
17 per
group
Rales and dyspnea were observed in the TCE
high-dose group; two females with dyspnea
subsequently died.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
positive response in mice but not rats. However, chronic toxicity data on noncancer effects are
very limited.
4.7.2.1.2. Acute and short-term effects: inhalation
Relatively high-dose single and multiple inhalation exposures to TCE result in dilation of
endoplasmic reticulum and vacuolation of nonciliated (Clara) cells throughout the bronchial tree
in mice. A single study in rats reported similar findings. In mice, single exposure experiments
show vacuolation at all dose levels tested with the extent of damage increasing with dose.
Villaschi et al. (1991) reported similar degrees of vacuolation in B6C3F1 mice (3/group) at
24 hours after the start of exposure across all tested doses (500, 1,000, 2,000, 3,500, and
7,000 ppm, 30 minutes), with the percentage of the nonciliated cells remaining vacuolated at
48 hours increasing with dose. Clara cell vacuolation was reported to be resolved 7 days after
single 30 minute exposure to TCE. Odum et al. (1992) reported that, when observed 24 hours
after the start of 6 hours exposure, the majority of Clara cells in mice were unaffected at the
lowest dose of 20 ppm exposures, while marked vacuolation was observed at 200 ppm (no
quantitative measures of damage given and only 3 animals per group were examined).
In rats, Odum et al. (1992) reported no morphological changes in the female Alpk APfSD
rat epithelium after 6 hours exposure (500 or 1,000 ppm) when observed 24 hours after the start
of exposure (n = 3/group). However, Kurasawa reported pronounced dose-related morphological
changes in Clara cells at the highest dose (8,000 ppm) for 2 hours in Wistar rats (n = 10 per
group). At 500 and 1,000 ppm, slight dilation of the apical surface was reported, but
morphological measurements (the ratio of the lengths of the apical surface to that of the base line
of apical cytoplasm) were not statistically-significantly different from controls. From
2,000-8,000 ppm, a progressively increasing flattening of the apical surface was observed. In
addition, at 2,000 ppm, slight dilation of the smooth endoplasmic reticulum was also observed,
with marked dilation and possible necrosis at 8,000 ppm. Kurasawa (1988) also examined the
time-course of Clara cell changes following a single 8,000-ppm exposure, reporting the greatest
effects at 1 day to 1 week, repair at 2 weeks, and nearly normal morphology at 4 weeks. The
only other respiratory effect that has been reported from one study in rats exposed via inhalation
is a reduction in pulmonary surfactant yield following 30 minute exposures at 9,030 ppm for 5 or
15 days (Stewart et al., 1979). Therefore, single inhalation experiments (Kurasawa, 1988; Odum
et al., 1992; Villaschi et al., 1991) suggest that the Clara cell is the target for TCE exposure in
both rats and mice and that mice are more susceptible to these effects. However, the database is
This document is a draft for review purposes only and does not constitute Agency policy.
506 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
limited in its ability to discern quantitative differences in susceptibility or the nature of the dose-
response after a single dose of TCE.
Other experiments examined the effects of several days of TCE inhalation exposure in
mice and potential recovery. While single exposures require 1-4 weeks for complete recovery,
after short-term repeated exposure, the bronchial epithelium in mice appears to either adapt to or
become resistant to damage Odum et al. (1992) and Green et al. (1997b) observed Clara cells in
mice to be morphologically normal at the end of exposures 6 hours/day for 4 or 5 days. As with
single dose experiments, the extent of recovery in multidose exposures may be dose-dependent.
Using a very high dose, Lewis et al. (1984) report vacuolation of bronchial epithelial cells after
4 hours/day, but not 1 hours/day, (10,000 ppm) for 5 days in mice. In addition, Odum et al.
(1992) reported that the damage to Clara cells that resolved after repeated exposures of 5 days, a
sign of adaptation to TCE exposure, returned when exposure was resumed after 2 days.
In rats, only one inhalation study reported in two published articles (Le Mesurier et al.,
1980; Stewart et al., 1979) using repeated exposures examined pulmonary histopathology.
Interestingly, this study reported vacuolation in Type 1 alveolar cells, but not in Clara cells, after
5 days of exposure to approximately 9,030 ppm for 30 minutes/day (only dose tested). In
addition, abnormalities were observed in the endothelium (bulging of thin endothelial segments
into the microcirculatory lumen) and minor morphological changes in Type 2 alveolar cells.
Although exposures were carried out for 5 consecutive days, histopathology was recorded up to
15 days post exposure, giving cell populations time to recover. Because earlier time points were
not examined, it is not possible to discern whether the lack of reported Clara cell damage in rats
following repeated exposure is due to recovery or lack of toxicity in this particular experiment.
Although recovery of individual damaged cells may occur, cell proliferation, presumed
from labeling index data suggestive of increased DNA synthesis, contributes, at least in part, to
the recovery of the bronchial epithelium in mice. Villaschi et al. (1991) observed a dose-
dependent increase in labeling index as compared to controls in the mouse lung at 48 hours after
a single TCE exposure (30 minutes; 500, 1,000, 2,000, 3,500, or 7,000 ppm), which decreased to
baseline values at 7 days postexposure. Morphological analysis of cells was not performed,
although the authors stated the dividing cells had the appearance of Clara cells. Interestingly,
Green et al. (1997b) reported no increase in BrdU labeling 24 hours after a single exposure
(6 hours 450 ppm), but did see increased BrdU labeling at the end of multiple exposures
"3
(1/day, 5 days) while Villaschi et al. (1991) reported increased [ HJThymidine labeling 2, 5, and
7 days after single 30 minute exposures to 500-7,000 ppm. Therefore, the data for single
exposures at 450-500 ppm may be consistent if increased cell proliferation occurred only for a
short period of time around 48 hours postexposure, and was thereby effectively washed-out by
This document is a draft for review purposes only and does not constitute Agency policy.
507 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
the longer "averaging time" in the experiments by Green et al. (1997b). Also, these
contradictory results may be due to differences in methodology. Green et al. (1997b) and
Villaschi et al. (1991) reported very different control labeling indices (6 and 0%, respectively)
while reporting similar absolute labeling indices at 450-500 ppm (6.5 and 5.2%, respectively).
The different control values may be a result of substantially-different times over which the label
was incorporated: the mice in Green et al. (1997b) were given BrdU via a surgically-implanted
osmotic pump over 4 days prior to sacrifice, while the mice in Villaschi et al. (1991) were given
"3
a single intraperitoneal dose of [ HJThymidine 1 hour prior to sacrifice. Stewart et al. (1979)
observed no stimulation of thymidine incorporation after daily exposure to TCE (9,000 ppm) for
up to 15 days. This study did, however, report a nonstatistically significant reduction in orotate
incorporation, an indicator of RNA synthesis, after 15 days, although the data was not shown.
At the biochemical level, changes in pulmonary metabolism, particularly with respect to
CYP activity, have been reported following TCE exposure via inhalation or intraperitoneal
administration in mice. Odum et al. (1992) reported reduced enzyme activity in Clara cell
sonicates of ethoxycoumarin O-deethylase, aldrin epoxidation, and nicotinamide adenine
dinucleotide phosphate-oxidase (NADPH) cytochrome c reductase after 6 hour exposures to
20-2,000 ppm TCE, although the reduction at 20 ppm was not statistically significant. No
reduction of GST activity as determined by chlorodinitrobenzene as a substrate was detected.
With repeated exposure at 450 ppm, the results were substrate-dependent, with ethoxycoumarin
(9-deethylase activity remaining reduced, while aldrin epoxidation and NADPH cytochrome c
reductase activity showing some eventual recovery by 2 weeks. The results reported by Odum et
al. (1992) for NADPH cytochrome c reductase were consistent with those of Lewis et al. (1984),
who reported similarly reduced NADPH cytochrome c reductase activity following a much
larger dose of 10,000 ppm for 1 and 4 hours/day for 5 days in mice (strain not specified). TCE
exposure has also been associated with a decrease in pulmonary surfactant. Repeated exposure
of female Wistar rats to TCE (9,000 ppm, 30 minutes/day) for 5 or 15 days resulted in a
significant decrease in pulmonary surfactant as compared to unexposed controls (Le Mesurier et
al., 1980).
4.7.2.1.2.1.Acute and short-term effects: intraperitoneal injection and gavage exposure
As stated above the intraperitoneal route of administration is not a relevant paradigm for
human exposure. A number of studies have used this route of exposure to study the effects of
acute TCE exposure in mice. In general, similar lung targets are seen following inhalation or
intraperitoneal treatment in mice (Forkert et al., 2006; Forkert and Birch, 1989; 1985; Scott et al.,
This document is a draft for review purposes only and does not constitute Agency policy.
508 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
1988). Inhalation studies generally reported the Clara cell as the target in mice. No lung
histopathology from intraperitoneal injection studies in rats is available. Forkert et al. (1985) and
Forkert and Birch (1989) reported vacuolation of Clara cells as soon as 1 hour following
intraperitoneal administration of a single dose of 2,000 mg/kg in mice. At 2,500 mg/kg, both
Forkert et al. (1985) and Scott et al. (1988) reported exfoliation of Clara cells and parenchymal
changes, with morphological distortion in alveolar Type II cells and inconsistently observed
minor swelling in Type I cells at 24 hours postexposure. Furthermore, at 3,000 mg/kg,
Scott et al. (1988) also reported a significant (85%) decrease in intracellularly stored surfactant
phospholipids at 24 hours postexposure. These data indicate that both Clara cells and alveolar
Type I and II cells are targets of TCE toxicity at these doses and using this route of
administration. Recently, Forkert et al. (2006) reported Clara cell toxicity that showed increased
severity with increased dose (pyknotic nuclei, exfoliation) at 500-1,000 mg/kg intraperitoneal
doses as soon as 4 hours postexposure in mice. Even at 500 mg/kg, a few Clara cells were
reported with pyknotic nuclei that were in the process of exfoliation. Damage to alveolar Type II
cells was not observed in this dose range. The study by Scott et al. (1988) examined surfactant
phospholipids and phospholipase A2 activity in male CD-I mice exposed by intraperitoneal
injection of TCE (2,500 or 3,000 mg/kg, 24 hours). The lower concentration led to damage to
and exfoliation of Clara cells from the epithelial lining into the airway lumen, while only the
higher concentration led to changes in surfactant phospholipids. This study demonstrated an
increase in total phospholipid content in the lamellar body fractions in the mouse lung.
The study by Narotsky et al. (1995) exposed F344 timed-pregnant rats to TCE (0, 1,125,
and 1,500 mg/kg BW) by gavage and examined both systemic toxicity and developmental effects
at 14 days postexposure. Rales and dyspnea in the dams were observed in the high-dose group,
with two of the animals with dyspnea subsequently dying. The developmental effects observed
in this study are discussed in more detail in Section 4.8.
4.7.2.1.2.2.Subchronic and chronic effects
There are a few reports of the subchronic and chronic noncancer effects of TCE on the
respiratory system from intraperitoneal exposure in mice and from gavage exposure in rats.
Forkert and Forkert (1994) reported pulmonary fibrosis in mice 90 days after intraperitoneal
administration of a single 2,000 mg/kg dose of TCE. The effects were in the lung parenchyma,
not the bronchioles where Clara cell damage has been observed after acute exposure. It is
possible that fibrotic responses in the alveolar region occur irrespective of where acute injury
occurs. Effects upon Clara cells can also impact other areas of the lung via cytokine regulation
This document is a draft for review purposes only and does not constitute Agency policy.
509 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
(Elizur et al., 2008). Alternatively, the alveolar and/or capillary components of the lung may
have been affected by TCE in a manner that was not morphologically apparent in short-term
experiments. In addition effects from a single or a few short-term exposures may take longer to
manifest. The latter hypothesis is supported by the alveolar damage reported by Odum et al.
(1992) after chloral administration by inhalation, and by the adducts reported in alveolar Type II
cells by Forkert et al. (2006) after 500-1,000 mg/kg TCE intraperitoneal administration.
As noted previously, rats have responded to short-term inhalation exposures of TCE with
Clara cell and alveolar Type I and II effects. After repeated inhalation exposures over 6 weeks
(8 hours/day, 5 days/week, 730 ppm) and continuous exposures over 90 days (35 ppm),
Prendergast et al. (1967) noted no histopathologic changes in rats, guinea pigs, rabbits, dogs, or
monkeys after TCE exposure, but did describe qualitatively observing some nasal discharge in
the rats exposed for 6 weeks. The study details in Prendergast et al. (1967) are somewhat
limited. Exposed animals are described as "typically" 15 Long-Evans or Sprague-Dawley rats,
15 Hartley guinea pigs, 3 squirrel monkeys, 3 New Zealand albino rabbits, and 2 beagle dogs.
Controls were grouped between studies. In a 13-week NTP study in F344/N rats (n = 10/group)
exposed to TCE (0-2,000 mg/kg-day 5 days/week) by gavage, pulmonary vasculitis was
observed in 6/10 animals of each sex of the highest dose group (2,000 mg/kg-day), in contrast
tol/10 in controls of each sex (NTP, 1990).
4.7.2.1.3. Respiratory Tract Cancer
Limited studies have been performed examining lung cancer following TCE exposure
(summarized in Table 4-84). TCE inhalation exposure was reported to cause statistically
significant increase in pulmonary tumors (i.e., pulmonary adenocarcinomas) in some studies in
mice, but not in studies in rats and hamsters. Oral administration of TCE frequently resulted in
elevated lung tumor incidences in mice, but not in any tested species was there a statistically
significant increase. This section will describe the data regarding TCE induction of pulmonary
tumors in rodent models. The next sections will consider the role of metabolism and potential
MO As for inhalation carcinogenicity, primarily in mice.
4.7.2.1.4. Inhalation
There are three published inhalation studies examining the carcinogenicity of TCE at
exposures from 0-600 ppm, two of which reported statistically significantly increased lung
tumor incidence in mice at the higher concentrations (Fukuda et al., 1983; Henschler et al., 1980;
This document is a draft for review purposes only and does not constitute Agency policy.
510 DRAFT—DO NOT CITE OR QUOTE
-------
1 1988; Maltoni et al., 1986). Rats and hamsters did not show an increase in lung tumors
2 following exposure.
3
This document is a draft for review purposes only and does not constitute Agency policy.
511 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-84. Animal carcinogenicity studies of trichloroethylene
to
Reference
Animals (sex)
Exposure route
Dose/exp cone
(stabilizers, if any)
Pulmonary tumor incidences
Benign + malignant
Malignant only
Fukuda et al.
(1983)
ICR mice (F)
S-D rats (F)
Inhalation, 7 h/d, 5
d/wk, 104 wk, hold
until 107 wk
0, 50, 150, or 450 ppm
(epichlorohydrin)
Mice: 6/49, 5/50, 13/50, 11/46
Rats: 0/50, 0/50, 1/47, 1/51
Mice: 1/49, 3/50, 8/50a, 7/46a
Rats: none
Maltoni et al.
(1988; 1986)
S-D rats (M, F)
Swiss mice
(M, F)
B6C3F1 mice
(M, F)
Inhalation, 7 h/d, 5
d/wk, 104 wk, hold
until death
0, 100, 300, or 600 ppm
Rats: 0/280, 0/260, 0/260, 0/260
Swiss Mice: M: 10/90, 11/90,
23/903, 27/90b;F: 15/90, 15/90,
13/90, 20/90
B6C3F1 Mice: M: 2/90, 2/90,
3/90, 1/90; F: 4/90, 6/90, 7/90,
15/903
Rats: 0/280, 0/260, 0/260,
0/260
Swiss Mice: M: 0/90, 0/90,
0/90, 1/90;F: 2/90, 0/90, 0/90,
2/90
B6C3F1 Mice M: 0/90, 0/90,
0/90, 0/90; F: 0/90, 1/90,
0/90, 0/90
Henschler et al.
(1980)
Wistar rats (M, F)
Syrian hamsters
(M, F)
NMRI mice
Inhalation, 6 h/d, 5
d/wk, 78 wk, hold
until 130 wk (mice
and hamsters) or 156
wk (rats)
0, 100, or 500 ppm
(triethanolamine)
Rats: M: 1/29, 1/30, 1/30; F: 0/28;
1/30; 0/30
Hamsters: 0/60, 0/59, 0/60
Mice: M: 1/30, 3/29, 1/30;
F: 3/29, 0/30,1/28
Rats: M: 1/29, 1/30, 1/30; F:
0/28; 1/30; 0/30
Hamsters: 0/60, 0/59, 0/60
Mice: M: 5/30, 3/29, 1/30;
F: 1/29, 3/30,0/28
Henschler et al.
(1984)
Swiss mice
(M, F)
Gavage, 5/wk, 72 wk
hold 104 wk
2.4 g/kg BW (M), 1.8 g/kg
BW (F) all treatments;
(control, triethanolamine,
industrial, epichlorohydrin,
1,2-epoxybutane, both)
M: 18/50, 17/50, 14/50, 21/50,
15/50, 18/50; F: 12/50, 20/50,
21/50, 17/50, 18/50, 18/50
M: 8/50, 6/50, 7/50, 5/50, 7/50,
7/50; F: 5/50, 11/50, 8/50,
3/50, 7/50, 7/50
Van Duuren et
al. (1979)
Swiss mice
(M, F)
Gavage, 1/wk, 89 wk
0 or 0.5 mg (unknown)
0/30 for all groups
0/30 for all groups
NCI (1976)
Osborne-Mendel
rats (M, F)
B6C3F1 mice
(M, F)
Gavage, 5/wk, 78
wk, hold until 110
wk (rats) or 90 wk
(mice)
Rats: TWA: 0, 549, or 1,097
mg/kg
Mice: TWA: M: 0, 1,169, or
2,339 mg/kg; F: 0, 869, or
1,739 mg/kg (epoxybutane,
epichlorohydrin)
Rats: M: 1/20, 0/50, 0/50; F: 0/20,
1/47, 0/50
Mice: M: 0/20, 5/50, 2/48; F:
1/20, 4/50, 7/47
Rats: M: 0/20, 0/50, 0/50; F:
0/20, 1/47, 0/50
Mice: M: 0/20, 0/50, 1/48; F:
0/20, 2/50, 2/47
rs
sj
3
a
>!
&
fs*
si
O
Grj
>!
o
a
a
§•
>!
¦ g
rs
o
a
>!
o £
w
!z! ^
H J
1—1 s
H £.
W|
O •
-------
L/l
K>
Table 4-84. Animal carcinogenicity studies of trichloroethylene (continued)
Reference
Animals (sex)
Exposure route
Dose/exp cone
(stabilizers, if any)
Pulmonary tumor incidences
Benign + malignant
Malignant only
NTP (1988)
ACI, August,
Marshall,
Osborne-
Mendel rats
Gavage, 1/d, 5 d/wk,
103 wk
0, 500, or 1,000 mg/kg
(diisopropylamine)
ACI M: 1/50, 4/47, 0/46; F: 0/49,
2/47, 2/42
August M: 1/50, 1/50, 0/49; F:
1/50, 1/50, 0/50
Marshall M: 3/49, 2/50, 2/47; F:
3/49, 3/49, 1/46
Osborne-Mendel M: 2/50, 1/50,
1/50; F: 0/50, 3/50, 2/50
ACI M: 1/50, 2/47, 0/46; F:
0/49, 1/47, 2/42
August M: 0/50, 1/50, 0/49; F:
1/50, 0/50, 0/50
Marshall M: 3/49, 2/50, 2/47;
F: 3/49, 3/49, 1/46
Osborne-Mendel M: 1/50,
1/50, 0/50; F: 0/50, 3/50, 1/50
NTP (1990)
F344 rats
(M, F)
B6C3F1 mice
(M,F)
Gavage, 1/d, 5 d/wk,
103 wk
Mice: 0 or 1,000 mg/kg
Rats: 0, 500, 1,000 mg/kg
Mice: M: 7/49, 6/50; F: 1/48,
4/49
Rats: M: 4/50, 2/50, 3/49; F:
1/50, 1/49, 4/50
Mice: M: 3/49, 1/50; F: 1/48,
0/49
Rats: M: 3/50, 2/50, 3/49; F:
0/50, 0/49, 2/50
Maltoni et al.
(1986)
S-D rats (M, F)
Gavage, 1/d, 4-5
d/wk, 56 wk; hold
until death
0, 50 or 250 mg/kg
M: 0/30, 0/30, 0/30; F: 0/30,
0/30, 0/30
M: 0/30, 0/30, 0/30; F: 0/30,
0/30, 0/30
3^
rs
sj
3
a
>!
&
fs*
si
*3
O
Grj
>!
o
a
a
§•
>!
o
rs
o
a
>!
o
H
o
o
2!
a
"T3
H S.
W|
O •
p
o
c
o
H
W
H
o;
Statistically-significantly different from controls by Fisher's exact test (p < 0.05).
' Statistically-significantly different from controls by Fisher's exact test (p < 0.01).
St
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
The inhalation studies by Fukuda et al. (1983), which involved female ICR mice and
Sprague-Dawley rats, observed a threefold increase in lung tumors per mouse in those exposed
to the two higher concentrations (150-450 ppm) but reported no increase in lung tumors in the
rats. Maltoni et al. (1988; 1986) reported statistically-significantly increased pulmonary tumors
in male Swiss and female B6C3F1 mice at the highest dose of 600 ppm, but no significant
increases in any of the other species/strains/sexes tested. Henschler et al. (1980) tested NMRI
mice, Wistar rats and Syrian hamsters of both sexes, and reported no observed increase in
pulmonary tumors any of the species tested (see Section 4.4 and Appendix E for details of the
conduct of these studies).
4.7.2.1.5. Gavage
None of the six chronic gavage studies (Henschler et al., 1984; Maltoni et al., 1986; NCI,
1976; NTP, 1988, 1990; Van Duuren et al., 1979), which exposed multiple strains of rats and
mice to 0-3,000 mg/kg TCE for at least 56 weeks, reported a statistically-significant excess in
lung tumors, although nonstatistically-significant increases were frequently observed in mice.
The study by Van Duuren et al. (1979) examined TCE along with 14 other halogenated
compounds for carcinogenicity in both sexes of Swiss mice. While no excess tumors were
observed, the dose rate of 0.5 mg once per week is equivalent to an average dose rate of
approximately 2.4 mg/kg-day for a mouse weighing 30 g, which is about 400-fold smaller than
that in the other gavage studies. In the NCI (1976) study, the results for Osborne-Mendel rats
were considered inconclusive due to significant early mortality, but female B6C3F1 mice
(though not males) exhibited a nonstatistically-significant elevation in pulmonary tumor
incidence. The NCI study (1976) used technical grade TCE which contained two known
carcinogenic compounds as stabilizers (epichlorohydrin and 1,2-epoxybutane), but a later study
by Henschler et al. (1984) in which mice were given TCE that was either pure, industrial, and
stabilized with one or both of these stabilizers found similar pulmonary tumors regardless of the
presence of stabilizers. In this study, female mice (n = 50) had elevated, but again not
statistically-significant, increases in pulmonary tumors. A later gavage study by NTP (1988),
which used TCE stabilized with diisopropylamine, observed no pulmonary tumors, but chemical
toxicity and early mortality rendered this study inadequate for determining carcinogenicity. The
final NTP study (1990) in male and female F344 rats and B6C3F1 mice, using epichlorohydrin-
free TCE, again showed early mortality in male rats. Similar to the other gavage studies, a
nonstatistically significant elevation in (malignant) pulmonary tumors was observed in mice, in
this case in both sexes. These animal studies show that while there is a limited increase in lung
This document is a draft for review purposes only and does not constitute Agency policy.
514 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
tumors following gavage exposure to TCE in mice, the only statistically significant increase in
lung tumors occurs following inhalation exposure in mice.
4.7.3. Role of Metabolism in Pulmonary Toxicity
TCE oxidative metabolism has been demonstrated to play a main role in TCE pulmonary
toxicity in mice. However, data are not available on the role of specific oxidative metabolites in
the lung. The Clara cell is thought to be the cell type responsible for much of the CYP
metabolism in the lung. Therefore, damage to this cell type would be expected to also affect
metabolism. More direct measures of CYP and isozyme-specific depression following TCE
exposure have been reported following intraperitoneal administration in mice. Forkert et al.
(1985) reported significant reduction in microsomal aryl hydrocarbon hydroxylase activity as
well as CYP content between 1 and 24 hours after exposure (2,000-3,000 mg/kg i.p. TCE).
Maximal depression occurred between 2 and 12 hours, with aryl hydrocarbon hydroxylase
activity (a function of CYP) less than 50% of controls and CYP content less than 20% of
controls. While there was a trend towards recovery from 12-24 hours, depression was still
significant at 24 hours. Forkert et al. (2005) reported decreases in immunoreactive CYP2E1,
CYP2F2, and CYP2B1 in the 4 hours after TCE treatment with 750 mg/kg intraperitoneal
injection in mice. The amount and time of maximal reduction was isozyme dependent
(CYP2E1: 30% of controls at 2 hours; CYP2F2: abolished at 30 minutes; CYP2B1: 43% of
controls at 4 hours). Catalytic markers for CYP2E1, CYP2F2, and CYP2B enzymes showed
rapid onset (15 minutes or less after TCE administration) of decreased activity, and continued
depression through 4 hours. Decrease in CYP2E1 and CYP2F2 activity (measured by PNP
hydroxylase activity) was greater than that of CYP2B (measured by pentoxyresorufin
(9-dealkylase activity). Forkert et al. (2006) reported similar results in which 4 hours after
treatment, immunodetectable CYP2E1 protein was virtually abolished at doses 250-1,000 mg/kg
and immunodetectable CYP2F2 protein, while still detectable, was reduced. PNP hydroxylase
activity was also reduced 4 hours after treatment to 37% of controls at the lowest dose tested of
50 mg/kg, with further decreases to around 8% of control levels at doses of 500 mg/kg and
higher. These results correlate with previously described increases in Clara cell cytotoxicity, as
well as dichloroacetyl lysine (DAL) protein adduct formation. DAL adducts were observed in
the bronchiolar epithelium of CD-I mice and most prominent in the cellular apices of Clara cells
(Forkert et al., 2006). This study also examined the effect of TCE in vitro exposure on the
formation of chloral hydrate in lung microsomes from male CD-I mice and CYP2E1 knock-out
mice. The rates of CH formation were the same for lysosomes from both CD-I and CYP2E1
This document is a draft for review purposes only and does not constitute Agency policy.
515 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
knockout mice from 0.25 mM to 0.75 mM, but the CH formation peaked earlier for in the wild-
type lysosomes (0.75 mM) as compared to CYP2El-null lysosomes (1 mM).
The strongest evidence for the necessary role of TCE oxidation is that pretreatment of
mice with diallyl sulfone (DASO2), an inhibitor of CYP2E1 and CYP2F2, protected against
TCE-induced pulmonary toxicity. In particular, following an intraperitoneal TCE dose of
750 mg/kg, Clara cells and the bronchiolar epithelium in mice pretreated with the
CYP2E1/CYP2F2 inhibitor appeared normal. In naive mice given the same dose, the epithelium
was attenuated due to exfoliation and there was clear morphological distortion of Clara cells
(Forkert et al., 2005). In addition, the greater susceptibility of mouse lungs relative to rat lungs
is consistent with their larger capacity to oxidize TCE, as measured in vitro in lung microsomal
preparations (Green et al., 1997b). Analysis by immunolocalization also found considerably
higher levels of CYP2E1 in the mouse lung, heavily localized in Clara cells, as compared to rat
lungs, with no detectable CYP2E1 in human lung samples (Green et al., 1997b). In addition,
both Green et al. (1997b) and Forkert et al. (2006) report substantially lower metabolism of TCE
in human lung microsomal preparations than either rats or mice. It is clear that CYP2E1 is not
the only CYP enzyme involved in pulmonary metabolism, as lung microsomes from CYP2E1-
null mice showed greater or similar rates of CH formation compared to those from wild-type
mice. Recent studies have suggested a role for CYP2F2 in TCE oxidative metabolism, although
more work is needed to make definitive conclusions. In addition, there may be substantial
variability in human lung oxidative metabolism, as Forkert et al. (2006) reported that in
microsomal samples from eight individuals, five exhibited no detectable TCE oxidation (<0.05
pmol/mg protein/20 minutes), while others exhibited levels well above the limit of detection
(0.4-0.6 pmol/mg protein/minute).
In terms of direct pulmonary effects of TCE metabolites, Odum et al. (1992) reported that
mice exposed to 100 ppm via inhalation of chloral for 6 hours resulted in bronchiolar lesions
similar to those seen with TCE, although with a severity equivalent to 1,000 ppm TCE
exposures. In addition, some alveolar necrosis, alveolar oedema, and desquamation of the
epithelium were evident. In the same study, TCOH (100 and 500 ppm) also produced Clara cell
damage, but with lower incidence than TCE, and without alveolar lesions, while TCA treatment
produced no observable pulmonary effects. Therefore, it has been proposed that chloral is the
active metabolite responsible for TCE pulmonary toxicity, and the localization of damage to
Clara cells (rather than to other cell types, as seen with direct exposure to chloral) is due to the
localization of oxidative metabolism in that cell type (Green, 2000; Green et al., 1997b; Odum et
al., 1992). However, the recent identification by Forkert et al. (2006) of DAL adducts, also
localized with Clara cell, suggests that TCE oxidation to dichloroacetyl chloride, which is not
believed to be derived from chloral, may also contribute to adverse health effects.
This document is a draft for review purposes only and does not constitute Agency policy.
516 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Due to the histological similarities between TCE- and chloral-induced pulmonary
toxicity, consistent with chloral being the active moiety, it has been proposed that the limited or
absent capacity for reduction of chloral (rapidly converted to CH in the presence of water) to
TCOH and glucuronidation of TCOH to TCOG in mouse lungs leads to "accumulation" of
chloral in Clara cells. However, the lack of TCOH glucuronidation capacity of Clara cells
reported by Odum et al. (1992), while possibly an important determinant of TCOH
concentrations, should have no bearing on CH concentrations, which depend on the production
and clearance of CH only. While isolated mouse Clara cells form smaller amounts of TCOH
relative to CH (Odum et al., 1992), the cell-type distribution of the enzymes metabolizing CH is
not clear. Indeed, cytosolic fractions of mouse, rat and human whole lungs show significant
activity for CH conversion to TCOH (Green et al., 1997b). In particular, in mouse lung
subcellular fractions, 1 micromole of TCE in a 1.3 mL reactivial was converted to CH at a rate of
1 nmol/minute/mg microsomal protein, while 10 nmol CH in a 1.3 mL reactivial was converted
to TCOH at a rate of 0.24 nmol/minute/mg cytosolic protein (Green et al., 1997b). How this
fourfold difference in activity would translate in vivo is uncertain given the 100-fold difference
in substrate concentrations, lack of information as to the concentration-dependence of activity,
and uncertain differences between cytosolic and microsomal protein content in the lung. It is
unclear whether local pulmonary metabolism of chloral is the primary clearance process in vivo,
as in the presence of water, chloral rapidly converts to chloral hydrate, which is soluble in water
and hence can rapidly diffuse to surrounding tissue and to the blood, which also has the capacity
to metabolize chloral hydrate (Lipscomb et al., 1996). Nonetheless, experiments with isolated
perfused lungs of rats and guinea pigs found rapid appearance of TCOH in blood following TCE
inhalation exposure, with no detectable chloral hydrate or TCOG (Dalbey and Bingham, 1978).
Therefore, it appears likely that chloral in the lung either is rapidly metabolized to TCOH, which
then diffuses to blood, or diffuses to blood as CH and is rapidly metabolized to TCOH by
erythrocytes (Lipscomb et al., 1996).
This hypothesis is further supported by in vivo data. No in vivo data in rats on CH after
TCE administration were located, and Fisher et al. (1998) reported CH in blood of human
volunteers exposed to TCE via inhalation were below detection limits. In mice, however, after
both inhalation and oral gavage exposure to TCE, CH has been reported in whole lung tissue at
concentrations similar to or somewhat greater than that in blood (Abbas and Fisher, 1997;
Greenberg et al., 1999). A peak concentration (1.3 jug/g) of pulmonary CH was reported after
inhalation exposure to 600 ppm—at or above exposures where Clara cell toxicity was reported in
acute studies (Green et al., 1997b; Odum et al., 1992). However, this was fivefold less than the
reported pulmonary CH concentration (6.65 jug/g) after gavage exposures of 1,200 mg/kg.
Specifically, a 600-ppm exposure or 450-ppm exposure reported in the Maltoni et al. and
This document is a draft for review purposes only and does not constitute Agency policy.
517 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Fukuda et al. studies results in a greater incidence in lung tumors than the
1,000-1,200 mg/kg-day exposures in the NTP (1990) and NCI (1976) bioassays. However, the
peak CH levels measured in whole lung tissues after inhalation exposure to TCE at 600 ppm
were reported to be about fivefold lower than that at 1,200 mg/kg by gavage, therefore, showing
the opposite pattern (Abbas and Fisher, 1997; Greenberg et al., 1999). No studies of Clara cell
toxicity after gavage exposures were located, but several studies in mice administered TCE via
intraperitoneal injection did show Clara cell toxicity at around a dose of 750 mg/kg (Forkert et
al., 2006) or above (e.g., Forkert and Birch, 1989; Forkert and Forkert, 1994). However, as
noted previously, i.p. exposures are subject to an inflammatory response, confounding direct
comparisons of dose via other routes of administration.
Although, whole lung CH concentrations may not precisely reflect the concentrations
within specific cell types, as discussed above, the water solubility of CH suggests rapid
equilibrium between cell types and between tissues and blood. Both Abbas and Fisher (1997)
and Greenberg et al. (1999) were able to fit CH blood and lung levels using a PBPK model that
did not include pulmonary metabolism, suggesting that lung CH levels may be derived largely by
systemic delivery, i.e., from CH formed in the liver. However, a more detailed PBPK model-
based analysis of this hypothesis has not been performed, as CH is not included in the PBPK
model developed by Hack et al. (2006) that was updated in Section 3.5.
Two studies have reported formation of reactive metabolites in pulmonary tissues as
assessed by macromolecular binding after TCE intraperitoneal administration. Forkert and Birch
(1989) reported temporal correlations between the severity of Clara cell necrosis with increased
levels of covalent binding macromolecules in the lung of TCE or metabolites with a single
2,000 mg/kg dose of [14C]TCE. The amount of bound TCE or metabolites per gram of lung
tissue, DNA, or protein peaked at 4 hours and decreased progressively at 8, 12, and 24 hours.
The fraction of radioactivity in lung tissue macromolecules that was covalently bound reached a
plateau of about 20% from 4-24 hours, suggesting that clearance of total and covalently bound
TCE or metabolites was similar. The amount of covalent binding in the liver was three- to
10-fold higher than in the lung, although hepatic cytotoxicity was not apparent. This tissue
difference could either be due to greater localization of metabolism in the lung, so that
concentrations reactive metabolites in individual Clara cells are greater than both the lung as a
whole and hepatocytes, or because of greater sensitivity of Clara cells as compared to
hepatocytes to reactive metabolites. More recently, Forkert et al. (2006) examined DAL adducts
resulting from metabolism of TCE to dichloroacetyl chloride as an in vivo marker of production
of reactive metabolites. Following intraperitoneal administration of 500-1,000 mg/kg TCE in
CD-I mice, they found localization of DAL adducts believed to be from oxidative metabolism
within Clara cell apices, with dose-dependent increase in labeling with a polyclonal anti-DAL
This document is a draft for review purposes only and does not constitute Agency policy.
518 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
antibody that correlated with increased Clara cell damage. Dose-dependent DAL adducts were
also found in alveolar Type II cells, although no morphologic changes in those cells were
observed. Both Clara cell damage (as discussed above) and DAL labeling were abolished in
mice pretreated with DASO2, an inhibitor of CYP2E1 and CYP2F2. However, Clara cell
damage in treated CYP2El-null mice was more severe than in CD-I mice. Although DAL
labeling was less pronounced in CYP2El-null mice as compared to CD-I mice, this was due in
part to the greater histopathologic damage leading to attenuation of the epithelium and loss of
Clara cells in the null mice. In addition, protein immunoblotting with anti-DAL, anti-CYP2El
and anti-CYP2F2 antibodies suggested that a reactive TCE metabolite including dichloroacetyl
chloride was formed that is capable of binding to CYP2E1 and CYP2F2 and changing their
protein structures. Follow-up studies are needed in the lung and other target tissues to determine
the potential role of the DAL adducts in TCE-induced toxicity.
Finally, although Green (2000) and others have attributed species differences in
pulmonary toxicity to differences in the capacity for oxidative metabolism in the lung, it should
be noted that the concentration of the active metabolite is determined by both its production and
clearance (Clewell et al., 2000). Therefore, while the maximal pulmonary capacity to produce
oxidative metabolites is clearly greater in the mouse than in rats or humans, there is little
quantitative information as to species differences in clearance, whether by local chemical
transformation/metabolism or by diffusion to blood and subsequent systemic clearance. In
addition, existing in vitro data on pulmonary metabolism are at millimolar TCE concentrations
where metabolism is likely to be approaching saturation, so the relative species differences at
lower doses has not been characterized. Studies with recombinant CYP enzymes examined
species differences in the catalytic efficiencies of CYP2E1, CYP2F, and CYP2B1, but the
relative contributions of each isoform to pulmonary oxidation of TCE in vivo remains unknown
(Forkert et al., 2005). Furthermore, systemic delivery of oxidative metabolites to the lung may
contribute, as evidenced by respiratory toxicity reported with i.p. administration. Therefore,
while the differences between mice and rats in metabolic capacity are correlated with their
pulmonary sensitivity, it is not clear that differences in capacity alone are accurate quantitative
predictors of toxic potency. Thus, while it is likely that the human lung is exposed to lower
concentrations of oxidative metabolites, quantitative estimates for differential sensitivity made
with currently available data and dosimetry models are highly uncertain.
In summary, it appears likely that pulmonary toxicity is dependent on in situ oxidative
metabolism, however, the active agent has not been confidently identified. The similarities in
histopathologic changes in Clara cells between TCE and chloral inhalation exposure, combined
with the wider range of cell types affected by direct chloral administration relative to TCE, led
some to hypothesize that chloral is the toxic moiety in both cases, but with that generated in situ
This document is a draft for review purposes only and does not constitute Agency policy.
519 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
from TCE in Clara cells "accumulating" in those cells (Green, 2000). However, chemical and
toxicokinetic data suggest that such "accumulation" is unlikely for several reasons. These
include the rapid conversion of chloral to chloral hydrate in the presence of water, the water
solubility of CH leading to rapid diffusion to other cell types and blood, the likely rapid
metabolism of chloral hydrate to TCOH either in pulmonary tissue or in blood erythrocytes, and
in vivo data showing lack of correlation across routes of exposure between whole-lung CH
concentrations and pulmonary carcinogenicity and toxicity. However, additional possibilities for
the active moiety exist, such as dichloroacetyl chloride, which is derived through a TCE
oxidation pathway independent of chloral and which appears to result in adducts with lysine
localized in Clara cells.
4.7.4. Mode of Action for Pulmonary Carcinogenicity
A number of effects have been hypothesized to be key events in the pulmonary
carcinogenicity of TCE, including cytotoxicity leading to increased cell proliferation, formation
of DAL protein adducts, and mutagenicity. As stated previously, the target cell for pulmonary
adenocarcinoma formation has not been established. Much of the hazard and MOA information
has focused on Clara cell effects from TCE which is a target in both susceptible and
nonsusceptible rodent species for lung tumors. However, the role of Clara cell susceptibility to
TCE-induced lung toxicity or to other potential targets such as lung stem cells that are activated
to repopulate both Clara and Type II alveolar cells after injury, has not been determined for
pulmonary carcinogenicity. While all of the events described above may be plausibly involved
in the MOA for TCE pulmonary carcinogenicity, none have been directly shown to be necessary
for carcinogenesis.
4.7.4.1.1. Mutagenicity via Oxidative Metabolism
The hypothesis is that TCE acts by a mutagenic MOA in TCE- induced lung tumors.
According to this hypothesis, the key events leading to TCE-induced lung tumor formation
constitute the following: the oxidative metabolism of TCE producing chloral/chloral hydrate
delivered to pulmonary tissues, causes direct alterations to DNA (e.g., mutation, DNA damage,
and/or micronuclei induction). Mutagenicity is a well-established cause of carcinogenicity.
This document is a draft for review purposes only and does not constitute Agency policy.
520 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
4.7.4.1.2. Experimental support for the hypothesized mode of action
Pulmonary toxicity has been proposed to be dependent on in situ oxidative metabolism,
however, the active agent has not been confidently identified. The similarities in histopathologic
changes in Clara cells between TCE and chloral inhalation exposure, combined with the wider
range of cell types affected by direct chloral administration relative to TCE, led some to
hypothesize that chloral is the toxic moiety. Chloral that is formed from the metabolism of TCE
is quickly converted to CH upon hydration under physiological conditions. As discussed in
Section 4.2.4, CH clearly induces aneuploidy in multiple test systems, including bacterial and
fungal assays in vitro (Crebelli et al., 1991; Kafer, 1986; Kappas, 1989), mammalian cells in
vitro (Sbrana et al., 1993; Vagnarelli et al., 1990), and mammalian germ-line cells in vivo (Miller
and Adler, 1992; Russo et al., 1984). Conflicting results were observed in in vitro and in vivo
mammalian studies of micronuclei formation (Beland, 1999a; Degrassi and Tanzarella, 1988;
Giller et al., 1995; Nesslany and Marzin, 1999; Russo and Levis, 1992a; 1992b) with positive
results in germ-line cells (Nutley et al., 1996)(Allen et al., 1994). In addition, it is mutagenic in
the Ames bacterial mutation assay for some strains (Beland, 1999a; Giller et al., 1995; Haworth
et al., 1983; Ni et al., 1994). Structurally related chlorinated aldehydes 2-chloroacetyaldehyde
and 2,2-dichloroacetaldehyde are both alkylating agents, are both positive in a genotoxic assay
(Bignami et al., 1980), and both interact covalently with cellular macromolecules (Guengerich et
al., 1979).
As discussed in the section describing the experimental support for the mutagenic MOA
for liver carcinogenesis (see Section 4.5.7.1), it has been argued that CH mutagenicity is unlikely
to be the cause of TCE carcinogenicity because the concentrations required to elicit these
responses are several orders of magnitude higher that achieved in vivo (Moore and Harrington-
Brock, 2000). Similar to the case of the liver, it is not clear how much of a correspondence is to
be expected from concentrations in genotoxicity assays in vitro and concentrations in vivo, as
reported in vivo CH concentrations are in whole lung homogenate while in vitro concentrations
are in culture media. None of the available in vivo genotoxicity assays used the inhalation route
that elicited the greatest lung tumor response under chronic exposure conditions, so direct in vivo
comparisons are not possible. Finally, as discussed in Section 4.5.7.1, the use of i.p.
administration in many other in vivo genotoxicity assays complicates the comparison with
carcinogenicity data.
As discussed above (see Section 4.7.3), chemical and toxicokinetic data are not
supportive of CH being the active agent of TCE-induced pulmonary toxicity, and directly
contradict the hypothesis of chloral "accumulation." Nonetheless, CH has been measured in the
mouse lung following inhalation and gavage exposures to TCE (Abbas and Fisher, 1997;
Greenberg et al., 1999), possibly the result of both in situ production and systemic delivery.
This document is a draft for review purposes only and does not constitute Agency policy.
521 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Therefore, in principle, CH could cause direct alterations in DNA in pulmonary tissue.
However, as discussed above, the relative amounts of CH measured in whole lung tissue from
inhalation and oral exposures do not appear to correlate with sensitivity to TCE lung tumor
induction across exposure routes. While these data cannot rule out a role for mutagenicity
mediated by CH due to various uncertainties, such as whether whole lung CH concentrations
accurately reflect cell-type specific concentrations and possible confounding due to strain
differences between inhalation and oral chronic bioassays, they do not provide support for this
MOA.
Additional possibilities for the active moiety exist, such as dichloroacetyl chloride, which
is derived through a TCE oxidation pathway independent of chloral and which appears to result
in adducts with lysine localized in Clara cells (Forkert et al., 2006). DCA, which has some
genotoxic activity, is, also, presumed to be formed through this pathway (see Section 3.3).
Currently, however, there are insufficient data to support a role for these oxidative metabolites in
a mutagenic MOA.
4.7.4.1.3. Cytotoxicity Leading to Increased Cell Proliferation
The hypothesis is that TCE acts by a cytotoxicity MOA in TCE-induced pulmonary
carcinogenesis. According to this hypothesis, the key events leading to TCE-induced lung tumor
formation constitute the following: TCE oxidative metabolism in situ leads to currently unknown
reactive metabolites that cause cytotoxicity, leading to compensatory cellular proliferation and
subsequently increased mutations and clonal expansion of initiated cells.
4.7.4.1.4. Experimental support for the hypothesized mode of action
Evidence for the hypothesized MOA consists primarily of (1) the demonstration of acute
cytotoxicity and transient cell proliferation following TCE exposure in laboratory mouse studies;
(2) toxicokinetic data supporting oxidative metabolism being necessary for TCE pulmonary
toxicity; (3) the association of lower pulmonary oxidative metabolism and lower potency for
TCE-induced cytotoxicity with the lack of observed pulmonary carcinogenicity in laboratory
rats. However, there is a lack of experimental support linking TCE acute pulmonary cytotoxicity
to sustained cellular proliferation of chronic exposures or clonal expansion of initiated cells.
As discussed above, a number of acute studies have shown that TCE is particularly
cytotoxic to Clara cells in mice, which has been suggested to be involved in the development of
mouse lung tumors (Buckpitt et al., 1995; Forkert and Forkert, 1994; Kim et al., 2005). In
This document is a draft for review purposes only and does not constitute Agency policy.
522 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
addition, studies examining cell labeling by either BrdU (Green et al., 1997b) or 3H-thymidine
incorporation (Villaschi et al., 1991) suggest increased cellular proliferation in mouse Clara cells
following acute inhalation exposures to TCE. Moreover, in short-term studies, Clara cells appear
to become resistant to cytotoxicity with repeated exposure, but regain their susceptibility after
2 days without exposure. This observation led to the hypothesis that the 5 day/week inhalation
dosing regime (Fukuda et al., 1983; Henschler et al., 1980; Maltoni et al., 1988; 1986) in the
chronic mouse studies leads to periodic cytotoxicity in the mouse lung at the beginning of each
week followed by cellular regeneration, and that the increased rate of cell division leads to
increased incidence of tumors by increasing the overall mutation rate and by increasing the
division rate of already initiated cells (Green, 2000). However, longer-term studies to test this
hypothesis have not been carried out.
As discussed above (see Section 4.7.3), there is substantial evidence that pulmonary
oxidative metabolism is necessary for TCE-induced pulmonary toxicity, although the active
moiety remains unknown. In addition, the lower capacity for pulmonary oxidative metabolism
in rats as compared to mice is consistent with studies in rats not reporting pulmonary cytotoxicity
until exposures higher than those in the bioassays, and the lack of reported pulmonary
carcinogenicity in rats at similar doses to mice. However, rats also have a lower background rate
of lung tumors (Green, 2000), and so would be less sensitive to carcinogenic effects in that tissue
to the extent that relative risks is the important metric across species. In addition, this MOA
hypothesis requires a number of additional key assumptions for which there are currently no
direct evidence. First, the cycle of cytotoxicity, repair, resistance to toxicity, and loss of
resistance after exposure interruption, has not been documented and under the proposed MOA
should continue under chronic exposure conditions. This cycle has thus, far only been observed
in short term (up to 13-day) studies. In addition, although Clara cells have been identified as the
target of toxicity whether they or endogenous stem cells in the lung are the cells responsible for
mouse lung tumors has not been established. There is currently no data as to the cell type of
origin for TCE-induced lung tumors.
This hypothesized MOA has been proposed for other compounds that induce mouse lung
tumors, such as coumarin, naphthalene, and styrene (e.g., Cruzan et al., 2009). Among these,
only for styrene have there been studies of chronic duration linking cytotoxicity with
hyperplasia, and no studies appear to provide experimental linkage to clonal expansion of
initiated cells.
This document is a draft for review purposes only and does not constitute Agency policy.
523 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
4.7.4.1.5. Additional Hypothesized Modes of Action with Limited Evidence or Inadequate
Experimental Support
4.7.4.1.6. Role of formation of DAL protein adducts
As discussed above, Forkert et al. (2006) recently observed dose-dependent formation of
DAL protein adducts in the Clara cells of mice exposed to TCE via intraperitoneal injection.
While adducts were highly localized in Clara cells, they were also found in alveolar Type II
cells, though these cells did not show signs of cytotoxicity in this particular experimental
paradigm. In terms of the MOA for TCE-induced pulmonary carcinogenicity, these adducts may
either be causally important in and of themselves, or they may be markers of a different causal
effect. For instance, it is possible that these adducts are a cause for the observed Clara cell
toxicity, and Forkert et al. (2006) suggested that the lack of toxicity in alveolar Type II cells may
indicate that "there may be a threshold in adduct formation and hence bioactivation at which
toxicity is manifested." In this case, they are an additional precursor event in the same causal
pathway proposed above. Alternatively, these adducts may be indicative of effects related to
carcinogenesis but unrelated to cytotoxicity. In this case, the Clara cell need not be the cell type
of origin for mouse lung tumors.
Because of their recent discovery, there is little additional data supporting, refuting, or
clarifying the potential role for DAL protein adducts in the MOA for TCE-induced pulmonary
carcinogenesis. For instance, the presence and localization of such adducts in rats has not been
investigated, and could indicate the extent to which the level of adduct formation is correlated
with existing data on species differences in metabolism, cytotoxicity, and carcinogenicity. In
addition, the formation of these adducts has only been investigated in a single dose study using
i.p. injection. As stated above, i.p. injection may involve the initiation of a systemic
inflammatory response that can activate lung macrophages or affect Clara cells. Experiments
with repeated exposures over chronic durations and by inhalation or oral of administration would
be highly informative. Finally, the biological effects of these adducts, whether cytotoxicity or
something else, have not been investigated.
This document is a draft for review purposes only and does not constitute Agency policy.
524 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
4.7.4.1.7. Conclusions About the Hypothesized Modes of Action
4.7.4.1.8. (1) Is the hypothesized mode of action sufficiently supported in the test animals?
4.7.4.1.8.1 .Mutagenicity
Chloral hydrate is clearly genotoxic, as there are substantial data from multiple in vitro
and in vivo assays supporting its ability induce aneuploidy, with more limited data as to other
genotoxic effects, such as point mutations. Chloral hydrate is also clearly present in pulmonary
tissues of mice following TCE exposures similar to those inducing lung tumors in chronic
bioassays. However, chemical and toxicokinetic data are not supportive of CH being the
predominant metabolite for TCE carcinogenicity. Such data include the water solubility of CH
leading to rapid diffusion to other cell types and blood, its likely rapid metabolism to TCOH
either in pulmonary tissue or in blood erythrocytes, and in vivo data showing lack of correlation
across routes of exposure between whole lung CH concentrations and pulmonary
carcinogenicity. Therefore, while a role for mutagenicity via CH in the MOA of TCE-induced
lung tumors cannot be ruled about, available evidence is inadequate to support the conclusion
that direct alterations in DNA caused by CH produced in or delivered to the lung after TCE
exposure constitute a MOA for TCE-induced lung tumors.
4.7.4.1.8.2. Cytotoxicity
The MOA hypothesis for TCE-induced lung tumors involving cytotoxicity is supported
by relatively consistent and specific evidence for cytotoxicity at tumorigenic doses in mice.
However, the majority of cytotoxicity-related key events have been investigated in studies less
than 13 days, and none has been shown to be causally related to TCE-induced lung tumors. In
addition, the cell type (or types) of origin for the observed lung tumors in mice has not been
determined, so the contribution to carcinogenicity of Clara cell toxicity and subsequent
regenerative cell division is not known. Similarly, the relative contribution from recently
discovered dichloroacetyl-lysine protein adducts to the tumor response has not been investigated
and has currently only been studied in i.p. exposure paradigms of short duration. In summary,
while there are no data directly challenging the hypothesized MOA described above, the existing
support for their playing a causal role in TCE-induced lung tumors is largely associative, and
based on acute or short term studies. Therefore, there are inadequate data to support a cytotoxic
MOA based on the TCE-induced cytotoxicity in Clara cells in the lungs of test animals.
This document is a draft for review purposes only and does not constitute Agency policy.
525 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
4.7.4.1.8.3 Additional hypothesis
Inadequate data are available to develop a MOA hypothesis based on recently discovered
DAL adducts induced by TCE inhalation and i.p. exposures. It will therefore, not be considered
further in the conclusions below.
Overall, therefore, the MOA for TCE-induced lung tumors is considered unknown at this
time.
4.7.4.1.9. (2) Is the hypothesized mode of action relevant to humans?
4.7.4.1.9.1 .Mutagenicity
The evidence discussed above demonstrates that CH is mutagenic in microbial as well as
test animal species. There is, therefore, the presumption that they would be mutagenic in
humans. Therefore, this MOA is considered relevant to humans.
4.7.4.1.9.2. Cytotoxicity
No data from human studies are available on the cytotoxicity of TCE and its metabolites
in the lung, and no causal link between cytotoxicity and pulmonary carcinogenicity has been
demonstrated in animal or human studies. Nonetheless, in terms of human relevance, no data
suggest that the proposed key events are not biologically plausible in humans, therefore,
qualitatively, TCE-induced lung tumors are considered relevant to humans. This conclusion that
this hypothesized MOA is qualitatively relevant has also been reached for other compounds for
which the MOA has been postulated (Cruzan et al., 2009). Information about the relative
pharmacodynamic sensitivity between rodents and humans is absent, but information on
pharmacokinetic differences in lung oxidative metabolism does exist and will be considered in
dose-response assessment when extrapolating between species (see Section 5.2.1.2).
This document is a draft for review purposes only and does not constitute Agency policy.
526 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
4.7.4.1.10. (3) Which populations or lifestages can be particularly susceptible to the
hypothesized mode of action?
4.7.4.1.10.1. Mutagenicity
The mutagenic MOA is considered relevant to all populations and lifestages. According
to EPA's Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005c) and Supplemental
Guidance for Assessing Susceptibility From Early-Life Exposure to Carcinogens (U.S. EPA,
2005d), there may be increased susceptibility to early-life exposures for carcinogens with a
mutagenic mode of action. However, because the weight of evidence is inadequate to support a
mutagenic MOA for TCE pulmonary carcinogenicity, and in the absence of chemical-specific
data to evaluate differences in susceptibility, the ADAFs should not be applied, in accordance
with the Supplemental Guidance.
4.7.4.1.10.2. Cytotoxicity
No information based is available as to which populations or lifestages may be
particularly susceptible to TCE-induced lung tumors. However, pharmacokinetic differences in
lung oxidative metabolism among humans do exist, and because of the association between lung
oxidative metabolism and toxicity, will be considered in dose-response assessment when
extrapolating within species.
4.7.5. Summary and Conclusions
The studies described here show pulmonary toxicity found mainly in Clara cells in mice
(Forkert and Birch, 1989; Forkert et al., 1985; Green et al., 1997b; Odum et al., 1992; Villaschi
et al., 1991) and rats (Kurasawa, 1988). The most convincing albeit limited data regarding this
type of toxicity was demonstrated predominantly in mice exposed via inhalation, although some
toxicity was shown in intraperitoneal injection studies. Increased vacuolation of Clara cells was
often seen within the first 24-hours-of-exposure, depending on dose, but with cellular repair
occurring within days or weeks of exposure. Continued exposure led to resistance to
TCE-induced Clara cell toxicity, but damage recurred if exposure was stopped after 5 days and
then resumed after 2 days without exposure. However, Clara cell toxicity has only been
observed in acute and short-term studies, and it is unclear whether they persist with subchronic
or chronic exposure, particularly in mice, which are the more sensitive species. With respect to
pulmonary carcinogenicity, statistically-significantly increased incidence of lung tumors from
This document is a draft for review purposes only and does not constitute Agency policy.
527 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
chronic inhalation exposures to TCE was observed female ICR mice (Fukuda et al., 1983), male
Swiss mice, and female B6C3F1 mice (Maltoni et al., 1986), though not in other sex/strain
combinations, nor in rats (Henschler et al., 1980; Maltoni et al., 1986). However, lung toxicity
and Clara cell effects have also been observed in rats. Overall, the limited carcinogenesis studies
described above are consistent with TCE causing mild increases in pulmonary tumor incidence
in mice, but not in other species tested such as rats and hamsters.
The epidemiologic studies are quite limited for examining the role of TCE in cancers of
the respiratory system, with no studies found on TCE exposure specifically examining toxicity of
the respiratory tract. The two studies found on organic solvent exposure which included TCE
suggested smoking as a primary factor for observed lung function decreases among exposed
workers. Animal studies have demonstrated toxicity in the respiratory tract, particularly damage
to the Clara cells (nonciliated bronchial epithelial cells), as well as decreases in pulmonary
surfactant following both inhalation and intraperitoneal exposures, especially in mice. Dose-
related increases in vacuolation of Clara cells have been observed in mice and rats as early as
24 hours postexposure (2006; Forkert and Birch, 1989; Forkert et al., 1985; Kurasawa, 1988;
Odum et al., 1992; Scott et al., 1988). Mice appear to be more sensitive to these changes, but
both species show a return to normal cellular morphology at four weeks postexposure (Odum et
al., 1992). Studies in mice have also shown an adaptation or resistance to this damage after only
4-5 days of repeated exposures (Green et al., 1997b; Odum et al., 1992). The limited
epidemiological literature on lung and laryngeal cancer in TCE-exposed groups is inconclusive
due to study limitations (low power, null associations, confidence intervals on relative risks that
include 1.0). These studies can only rule out risks of a magnitude of 2.0 or greater for lung
cancer and relative risks greater than 3.0 or 4.0 for laryngeal cancer for exposures to studied
populations and thus, may not detect a level of response consistent with other endpoints. Animal
studies demonstrated a statistically significant increase in pulmonary tumors in mice following
chronic inhalation exposure to TCE (Fukuda et al., 1983; 1988; Maltoni et al., 1986). These
results were not seen in other species tested (rats, hamsters; (Fukuda et al., 1983; Henschler et
al., 1980; 1988; Maltoni et al., 1986). By gavage, elevated, but not statistically significant,
incidences of benign and/or malignant pulmonary tumors have been reported in B6C3F1 mice
(Henschler et al., 1984; NCI, 1976; NTP, 1990). No increased pulmonary tumor incidences have
been reported in rats exposed to TCE by gavage (NCI, 1976; NTP, 1988, 1990), although all the
studies suffered from early mortality in at least one sex of rat.
Although no epidemiologic studies on the role of metabolism of TCE in adverse
pulmonary health effects have been published, animal studies have demonstrated the importance
of the oxidative metabolism of TCE by CYP2E1 and/or CYP2F2 in pulmonary toxicity.
Exposure to DASO2, an inhibitor of both enzymes protects against pulmonary toxicity in mice
This document is a draft for review purposes only and does not constitute Agency policy.
528 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
following exposure to TCE (Forkert et al., 2005). The increased susceptibility in mice correlates
with the greater capacity to oxidize TCE based on increased levels of CYP2E1 in mouse lungs
relative to lungs of rats and humans (Forkert et al., 2006; Green et al., 1997b), but it is not clear
that these differences in capacity alone are accurate quantitative predictors of sensitivity to
toxicity. In addition, available evidence argues against the previously proposed hypothesis (e.g.,
Green, 2000) that "accumulation" of chloral in Clara cells is responsible for pulmonary toxicity,
since chloral is first converted the water-soluble compounds chloral hydrate and TCOH that can
rapidly diffuse to surrounding tissue and blood. Furthermore, the observation of DAL protein
adducts, likely derived dichloroacetyl chloride and not from chloral, that were localized in Clara
cells suggests an alternative to chloral as the active moiety. While chloral hydrate has shown
substantial genotoxic activity, chemical and toxicokinetic data on CH as well as the lack of
correlation across routes of exposure between in vivo measurements of CH in lung tissues and
reported pulmonary carcinogenicity suggest that evidence is inadequate to conclude that a
mutagenic MOA mediated by CH is operative for TCE-induced lung tumors. Another MOA for
TCE-induced lung tumors has been plausibly hypothesized to involve cytotoxicity leading to
increased cell proliferation, but the available evidence is largely associative and based on short-
term studies, so a determination of whether this MOA is operative cannot be made. The recently
discovered formation of DAL protein adducts in pulmonary tissues may also play a role in the
MOA of TCE-induced lung tumors, but an adequately defined hypothesis has yet to be
developed. Therefore, the MOA for TCE-induced lung tumors is currently considered unknown,
and this endpoint is thus, considered relevant to humans. Moreover, none of the available data
suggest that any of the currently hypothesized mechanisms would be biologically precluded in
humans.
4.8. REPRODUCTIVE AND DEVELOPMENTAL TOXICITY
4.8.1. Reproductive Toxicity
An assessment of the human and experimental animal data, taking into consideration the
overall weight of the evidence, demonstrates a concordance of adverse reproductive outcomes
associated with TCE exposures. Effects on male reproductive system integrity and function are
particularly notable and are discussed below. Cancers of the reproductive system in both males
and females have also been identified and are discussed below.
This document is a draft for review purposes only and does not constitute Agency policy.
529 DRAFT—DO NOT CITE OR QUOTE
-------
4.8.1.1.1. Human Reproductive Outcome Data
1 A number of human studies have been conducted that examined the effects of TCE on
2 male and female reproduction following occupational and community exposures. These are
3 described below and summarized in Table 4-85. Epidemiological studies of female human
4 reproduction examined infertility and menstrual cycle disturbances related to TCE exposure.
5 Other studies of exposure to pregnant women are discussed in the section on human
6 developmental studies (see Section 4.8.2.1). Epidemiological studies of male human
7
This document is a draft for review purposes only and does not constitute Agency policy.
530 DRAFT—DO NOT CITE OR QUOTE
-------
1
Table 4-85. Human reproductive effects
Subjects
Exposure
Effect
Reference
Female and male combined effects
Reproductive behavior
75 men and 71 women
living near Rocky
Mountain Arsenal,
Colorado
Low: <5.0 ppb
Medium: >5.0-<10.0 ppb
High: <10.0 ppb
Highest: <15 ppb
Altered libido3
Low: referent
Med: ORadj: 0.67 (95% CI: 0.18-2.49)
High: ORadj: 1.65 (95% CI: 0.54-5.01)
Highest: OR^,: 2.46 (95% CI: 0.59-10.28)
ATSDR
(2001)
Female effects
Infertility
197 women
occupationally exposed to
solvents in Finland
1973-1983
U-TCA (nmol/L)b
Median: 48.1
Mean: 96.2 ± 19.2
Reduced incidence of fecundability in the
high exposure group0 as measured by time to
pregnancy
Low: IDR = 1.21 (95%CI: 0.73-2.00)
High: IDR = 0.61 (95%CI: 0.28-1.33)
Sallmen et
al. (1995)
71 women living near
Rocky Mountain Arsenal,
Colorado
Low: <5.0 ppb
Med: >5.0 to <10.0 ppb
High: <10.0 ppb
No effect on lifetime infertility3
Low: referent
Med: OR^: 0.45 (95% CI: 0.02-8.92)
High: ORadl: 0.88 (95% CI: 0.13-6.22)
ATSDR
(2001)
Menstrual cycle disturbance
71 women living near
Rocky Mountain Arsenal,
Colorado
Low: <5.0 ppb
Med: >5.0 to <10.0 ppb
High: <10.0 ppb
Increase in abnormal menstrual cycle
(defined as <26 d or >30 d)
Low: referent
Med: OR^: 4.17 (95% CI: 0.31-56.65)
High: ORadl: 2.39 (95% CI: 0.41-13.97)
ATSDR
(2001)
184 women working in a
factory assembling small
electrical parts in Poland
Mean indoor air TCE: 200
mg/m3
18% reporting increase in amenorrhea in
exposed group (n = 140), compared to 2%
increase in unexposed group (n = 44)
Zielinski
(1973)
32 women working in dry
cleaning or metal
degreasing in
Czechoslovakia11
0.28-3.4 mg/L TCE for
0.5-25 yr
31% reporting increase in menstrual
disturbances 3
Bardodej and
Vyskocil
(1956)
20-yr-old woman was
occupationally exposed to
TCE via inhalation
Urine total trichloro-
compounds 3.2 ng/mL
(21-25 d after exposure)
Amenorrhea, followed by irregular
menstruation and lack of ovulation
Sagawa et al.
(1973)
Male effects
Reproductive behavior
43 men working in dry
cleaning or metal
degreasing in
Czechoslovakia
0.28-3.4 mg/L TCE for
0.5-25 yr
30% reporting decreased potency3
Bardodej and
Vyskocil
(1956)
30 male workers in a
money printing shop in
Egypt
38-172 ppm TCE
Decreased libido reported in 10 men (33%),
compared to 3 men in the control group
(10%)
El Ghawabi
et al. (1973)
2
This document is a draft for review purposes only and does not constitute Agency policy.
531 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-85. Human reproductive effects (continued)
Subjects
Exposure
Effect
Reference
42 yr-old male aircraft
mechanic in UK
TCE exposure reported but
not measured; exposure for
25 yr
Gynaecomastia, impotence
Saihan et al.
(1978)
Altered sperm quality
15 men working as metal
degreasers in Denmark
TCE exposure reported but
not measured
Nonsignificant increase in percentage of two
YFF in spermatozoa; no effect on sperm
count or morphology
Rasmussen
et al. (1988)
85 men of Chinese
descent working in an
electronics factory
Mean personal air TCE:
29.6 ppm; MeanU-TCA:
22.4 mg/g creatinine
Decreased normal sperm morphology and
hyperzoospermia
Chia et al.
(1996)
Altered endocrine function
85 men of Chinese
descent working in an
electronics factory
Mean personal air TCE:
29.6 ppm; MeanU-TCA:
22.4 mg/g creatinine
Increased DHEAS and decreased FSH,
SHBG and testosterone levels; dose-response
observed
Chia et al.
(1997)
85 men of Chinese
descent working in an
electronics factory
Mean personal air TCE:
29.6 ppm; MeanU-TCA:
22.4 mg/g creatinine
Decreased serum levels of testosterone and
SHBG were significantly correlated with
years of exposure to TCE; increased insulin
levels for exposure <2 yr
Goh et al.
(1998)
Infertility
282 men occupationally
exposed to solvents in
Finland 1973-1983
U-TCA (nmol/L):
High exposure:0
Mean: 45 (SD 42)
Median 31
Low exposure:0
Mean: 41 (SD 88)
Median: 15
No effect on fecundability0 (as measured by
time to pregnancy)
Low: FDR: 0.99 (95% CI: 0.63-1.56)
Intermediate/High: FDR:° 1.03 (95%
CI: 0.60-1.76)
Sallmen et
al. (1998)
8 male mechanics seeking
treatment for infertility in
Canada
Urine (|imol/):
TCA: <0.30-4.22
TCOH: <0.60-0.89
Seminal fluid (pg/extract):
TCE: 20.4-5,419.0
Chloral: 61.2-1,739.0
TCOH 2.7-25.5
TCA: <100-5,504
DCA: <100-13,342
Infertility could not be associated with TCE
as controls were 5 men also in treatment for
infertility
Forkert et al.
(2003)
75 men living near Rocky
Mountain Arsenal,
Colorado
Low: <5.0 ppb
Med: >5.0 to <10.0 ppb
High: <10.0 ppb
No effect on lifetime infertility (not defined)
Low: referent
Med: n/a
High: ORad|: 0.83 (95% CI: 0.11-6.37)
ATSDR
(2001)
1 aNot defined by the authors.
2 bAs reported in Lindbohm et al. (1990).
3 °Low/intermediate exposure indicated use of TCE <1 or 1-4 d/wk, and biological measures indicated high exposure.
4 High exposure indicated daily use of TCE, or if biological measures indicated high exposure.
5 dNumber inferred from data provided in Tables 2 and 3 in Bardodej and Vyskocil (1956).
6 UK = United Kingdom.
This document is a draft for review purposes only and does not constitute Agency policy.
532 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
reproduction examined reproductive behavior, altered sperm morphology, altered endocrine
function, and infertility related to TCE exposure.
4.8.1.1.2. Female and male combined human reproductive effects
Reproductive behavior
A residential study of individuals living near the Rocky Mountain Arsenal in Colorado
examined the reproductive outcomes in 75 men and 71 women exposed to TCE in drinking water
(ATSDR, 2001). TCE exposure was classified as high (>10.0 ppb), medium (>5.0 to <10.0 ppb),
and low (<5.0 ppb). Altered libido for men and women combined was observed in a dose-
response fashion, although the results were nonsignificant. The results were not stratified by
gender.
4.8.1.1.3. Female human reproductive effects
4.8.1.1.3.1.Infertility
Sallmen et al. (1995) examined maternal occupational exposure to organic solvents and
time-to-pregnancy. Cases of spontaneous abortion and controls from a prior study of maternal
occupational exposure to organic solvents in Finland during 1973-1983 and pregnancy outcome
(Lindbohm et al., 1990) were used to study time-to-pregnancy of 197 couples. Exposure was
assessed by questionnaire during the first trimester and confirmed with employment records.
Biological measurements of TCA in urine in 64 women who held the same job during pregnancy
and measurement (time of measurement not stated) had a median value of 48.1 (amol/L (mean:
96.2 ± 19.2 |imol/L) (Lindbohm et al., 1990). Nineteen women had low exposure to TCE (used
<1 or 1-4 times/week), and 9 had high exposure to TCE (daily use). In this follow-up study, an
additional questionnaire on time-to-pregnancy was answered by the mothers (Sallmen et al.,
1995). The incidence density ratio (IDR) was used in this study to estimate the ratio of average
incidence rate of pregnancies for exposed women compared to nonexposed women; therefore, a
lower IDR indicates infertility. For TCE, a reduced incidence of fecundability was observed in
the high exposure group (IDR: 0.61, 95% CI: 0.28-1.33) but not in the low exposure group
(IDR: 1.21, 95% CI: 0.73-2.00). A similar study of paternal occupational exposure (Sallmen et
al., 1998) is discussed in Section 4.2.1.2.
The residential study in Colorado discussed above did not observe an effect on lifetime
infertility infertility in the medium (ORadj: 0.45; 95% CI: 0.02-8.92) or high exposure groups
This document is a draft for review purposes only and does not constitute Agency policy.
533 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
(ORadj: 0.88; 95% CI: 0.13-6.22) (ATSDR, 2001). Curiously, exposed women had more
pregnancies and live births than controls.
4.8.1.1.3 .2.Menstrual cycle disturbance
The ATSDR (2001) study discussed above also examined effects on the menstrual cycle
(ATSDR, 2001). Nonsignificant associations without a dose-response were seen for abnormal
menstrual cycle in women (ORadj: 2.23, 95% CI: 0.45-11.18).
Other studies have examined the effect of TCE exposure on the menstrual cycle. One
study examined women working in a factory assembling small electrical parts (Zielinski, 1973,
translated). The mean concentration of TCE in indoor air was reported to be 200 mg/m . Of the
140 exposed women, eighteen percent suffered from amenorrhea, compared to only 2% of the 44
nonexposed workers. The other study examined 75 men and women working in dry cleaning or
metal degreasing (Bardodej and Vyskocil, 1956). Exposures ranged from 0.28-3.4 mg/L, and
length of exposure ranged from 0.5-25 years. This study reported that many women
experienced menstrual cycle disturbances, with a trend for increasing air concentrations and
increasing duration of exposure.
An additional case study of a 20-year-old woman was occupationally exposed to TCE via
inhalation. The exposure was estimated to be as high as 10 mg/mL or several thousand ppm,
based on urine samples 21-25 days after exposure of 3.2 ng/mL of total trichloro-compounds.
The primary effect was neurological, although she also experienced amenorrhea, followed by
irregular menstruation and lack of ovulation as measured by basal body temperature curves
(Sagawa et al., 1973).
4.8.1.1.4. Male human reproductive effects
4.8.1.1.4.1 Reproductive behavior
One study reported on the effect of TCE exposure on the male reproductive behavior in
75 men working in dry cleaning or metal degreasing (Bardodej and Vyskocil, 1956). Exposures
ranged from 0.28-3.4 mg/L, and length of exposure ranged from 0.5-25 years. This study found
that men experienced decreased potency or sexual disturbances; the authors speculated that the
effects on men could be due to the CNS effects of TCE exposure. This study also measured
serial neutral 17-ketosteroid determinations but they were found to be not statistically significant
(Bardodej and Vyskocil, 1956).
This document is a draft for review purposes only and does not constitute Agency policy.
534 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
An occupational study of 30 men working in a money printing shop were exposed to
TCE for <1 year to 5 years (El Ghawabi et al., 1973). Depending on the job description, the
exposures ranged from 38-172-ppm TCE. Ten (33%) men suffered from decreased libido,
compared to three (10%) of unexposed controls. However, these results were not stratified by
exposure level or duration. The authors speculate that decreased libido was likely due to the
common symptoms of fatigue and sleepiness.
A case study described a 42 year-old man exposed to TCE who worked as an aircraft
mechanic for approximately 25 years (Saihan et al., 1978). He suffered from a number of health
complaints including gynaecomastia and impotence, along with neurotoxicity and
immunotoxicity. In addition, he drank alcohol daily which could have increased his response to
TCE.
4.8.1.1.4.2.Altered sperm quality
Genotoxic effects on male reproductive function were examined in a study evaluating
occupational TCE exposure in 15 male metal degreasers (Rasmussen et al., 1988). No
measurement of TCE exposure was reported. Sperm count, morphology, and spermatozoa
Y-chromosomal nondisjunction during spermatogenesis were examined, along with
chromosomal aberrations in cultured lymphocytes. A nonsignificant increase in percentage of
two fluorescent Y-bodies (YFF) in spermatozoa were seen in the exposed group (p > 0.10), and
no difference was seen in sperm count or morphology compared to controls.
An occupational study of men using TCE for electronics degreasing (Chia et al., 1996,
1997; Goh et al., 1998) examined subjects (n = 85) who were offered a free medical exam if they
had no prior history related to endocrine function, no clinical abnormalities, and normal liver
function tests; no controls were used. These participants provided urine, blood, and sperm
samples. The mean urine TCA level was 22.4 mg/g creatinine (range: 0.8-136.4 mg/g
creatinine). In addition, 12 participants provided personal 8-hour air samples, which resulted in
a mean TCE exposure of 29.6 ppm (range: 9-131 ppm). Sperm samples were divided into two
exposure groups; low for urine TCE less than 25 mg/g creatinine, and high for urine TCA greater
than or equal to 25 mg/g creatinine. A decreased percentage of normal sperm morphology was
observed in the sperm samples in the high exposure group (n = 48) compared to the low
exposure group (n = 37). However, TCE exposure had no effect on semen volume, sperm
density, or motility. There was also an increased prevalence of hyperzoospermia (sperm density
of >120 million sperm per mL ejaculate) with increasing urine TCA levels (Chia et al., 1996).
This document is a draft for review purposes only and does not constitute Agency policy.
535 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
4.8.1.1.4.3.Altered endocrine function
Two studies followed up on the study by Chia et al. (1996) to examine endocrine function
(Chia et al., 1997; Goh et al., 1998). The first examined serum testosterone, follicle-stimulating
hormone (FSH), dehydroepiandrosterone sulphate (DHEAS), and sex-hormone binding globulin
(SHBG) (Chia et al., 1997). With increased years of exposure to TCE, an increase in DHEAS
levels were seen, from 255 ng/mL for <3 years to 717.8 ng/mL >7 years exposure. Also with
increased years of exposure to TCE, decreased FSH, SHBG and testosterone levels were seen.
The authors speculated these effects could be due to decreased liver function related to TCE
exposure (Chia et al., 1997).
The second follow-up study of this cohort studied the hormonal effects of chronic low-
dose TCE exposure in these men (Goh et al., 1998). Because urine TCE measures only indicate
short-term exposure, long-term exposure was indicated by years of exposure. Hormone levels
examined include androstenedione, Cortisol, testosterone, aldosterone, SHBG, and insulin.
Results show that a decrease in serum levels of testosterone and SHBG were significantly
correlated with years of exposure to TCE, and an increase in insulin levels were seen in those
exposed for less than 2 years. Androstenedione, Cortisol, and aldosterone were in normal ranges
and did not change with years of exposure to TCE.
4.8.1.1.4.4.Infertility
Sallmen et al. (1998) examined paternal occupational exposure and time-to-pregnancy
among their wives. Cases of spontaneous abortion and controls from a prior study of pregnancy
outcome (Taskinen et al., 1989) were used to study time-to-pregnancy of 282 couples. Exposure
was determined by biological measurements of the father who held the same job during
pregnancy and measurement (time of measurement not stated) and questionnaires answered by
both the mother and father. An additional questionnaire on time-to-pregnancy was answered by
the mother for this study six years after the original study (Sallmen et al., 1998). The level of
exposure was determined by questionnaire and classified as "low/intermediate" if the chemical
was used <1 or 1-4 days/week and biological measures indicated high exposure (defined as
above the reference value for the general population), and "high" if used daily or if biological
measures indicated high exposure. For 13 men highly exposed, mean levels of urine TCA were
45 (amol/L (SD 42 |imol/L; median 31 |imol/L); for 22 men low/intermediately exposed, mean
levels of urine TCA were 41 |amol/L (SD 88 |imol/L; median 15 |imol/L). The terminology IDR
was replaced by fecundability density ratio (FDR) in order to reflect that pregnancy is a desired
outcome; therefore, a high FDR indicates infertility. No effect was seen on fertility in the low
This document is a draft for review purposes only and does not constitute Agency policy.
536 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
exposure group (FDR: 0.99, 95% CI: 0.63-1.56) or in the intermediate/high exposure group
(FDR: 1.03, 95% CI: 0.60-1.76). However, the exposure categories were grouped by
low/intermediate versus high, whereas the outcome categories were grouped by low versus
intermediate/high, making a dose-response association difficult.
A small occupational study reported on eight male mechanics exposed to TCE for at least
two years who sought medical treatment for infertility (Forkert et al., 2003). The wives were
determined to have normal fertility. Samples of urine from two of the eight male mechanics
contained TCA and/or TCOH, demonstrating the rapid metabolism in the body. However,
samples of seminal fluid taken from all eight individuals detected TCE and the metabolites
chloral hydrate and TCOH, with two samples detecting DCA and one sample detecting TCA.
Five unexposed controls also diagnosed with infertility did not have any TCE or metabolites in
samples of seminal fluid. There was no control group that did not experience infertility.
Increased levels of TCE and its metabolites in the seminal fluid of exposed workers compared to
lower levels found in their urine samples was explained by cumulative exposure and
mobilization of TCE from adipose tissue, particularly that surrounding the epididymis. In
addition, CYP2E1 was detected in the epididymis, demonstrating that metabolism of TCE can
occur in the male reproductive tract. However, this study could not directly link TCE to the
infertility, as both the exposed and control populations were selected due to their infertility.
The ATSDR (2001) study discussed above on the reproductive effects from TCE in
drinking water of individuals living near the Rocky Mountain Arsenal in Colorado did not
observe infertility or other adverse reproductive effects for the high exposure group compared to
the low exposure group (ORadj: 0.83; 95% CI: 0.11-6.37). Curiously, exposed men had more
pregnancies and live births than controls.
4.8.1.1.5. Summary of human reproductive toxicity
Following exposure to TCE, adverse effects on the female reproductive system observed
include reduced incidence of fecundability (as measured by time-to-pregnancy) and menstrual
cycle disturbances. Adverse effects on the male reproductive system observed include altered
sperm morphology, hyperzoospermia, altered endocrine function, decreased sexual drive and
function, and altered fertility. These are summarized in Table 4-85.
This document is a draft for review purposes only and does not constitute Agency policy.
537 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
4.8.1.1.6. Animal Reproductive Toxicity Studies
A number of animal studies have been conducted that examined the effects of TCE on
reproductive organs and function following either inhalation or oral exposures. These are
described below and summarized in Tables 4-86 and 4-87. Other animal studies of offspring
exposed during fetal development are discussed in the section on animal developmental studies
(see Section 4.8.2.2).
4.8.1.1.7. Inhalation exposures
Studies in rodents exposed to TCE via inhalation are described below and summarized in
Table 4-86. These studies focused on various aspects of male reproductive organ integrity,
spermatogenesis, or sperm function in rats or mice. In the studies published after the year 2000,
the effects of either 376 or 1,000-ppm TCE were studied following exposure durations ranging
from 1-24 weeks, and adverse effects on male reproductive endpoints were observed.
This document is a draft for review purposes only and does not constitute Agency policy.
538 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-86. Summary of mammalian in vivo reproductive toxicity studies—
2 inhalation exposures
3
Reference
Species/strain/
sex/number
Exposure
level/duration
NOAEL;
LOAELa
Effects
Forkert et al.
(2002)
Mouse, CD-I,
male, 6/group
0 or 1,000 ppm
(5,374 mg/m3)b
6 h/d, 5 d/wk,
19 d over
4 wk
LOAEL: 1,000
ppm
U-TCA and U-TCOH increased by 2nd and
3rd wk, respectively. Cytochrome P450 2E1
and /7-nitrophenol hydroxylation in
epididymal epithelium > testicular Leydig
cells. Choral also generated from TCE in
epididymis > testis. Sloughing of epididymal
epithelial cells after 4 wk exposure.
Kan et al.
(2007)
Mouse, CD-I,
male, 4/group
0 or 1,000 ppm
6 h/d,5 d/wk,
1-4 wk
LOAEL: 1,000
ppm
Light microscopy findings: degeneration and
sloughing of epididymal epithelial cells as
early as 1 wk into exposure; more severe by 4
wk. Ultrastructional findings: vesiculation in
cytoplasm, disintegration of basolateral cell
membranes, sloughing of epithelial cells.
Sperm found in situ in cytoplasm of
degenerated epididymal cells. Abnormalities
of the head and tail in sperm located in the
epididymal lumen.
Kumar et al.
(2000b)
Rat, Wistar,
male,
12-13/group
0 or 376 ppm
4 h/d, 5 d/wk,
2-10 wk
exposure, 2-8
wk rest period
LOAEL: 376
ppm
Alterations in testes histopathology (smaller,
necrotic spermatogenic tubules), T sperm
abnormalities, and sig. T pre- and/or
postimplantation loss in litters observed in
the groups with 2 or 10 wk of exposure, or 5
wk of exposure with 2 wk rest.
Kumar et al.
(2000a)
Rat, Wistar,
males,
12-13/group
0 or 376 ppm
4 h/d, 5 d/wk,
12 and 24 wk
LOAEL: 376
ppm
Sig. ^ in total epididymal sperm count and
sperm motility, with sig. -1 in serum
testosterone, sig. T in testes cholesterol, sig. -1
of glucose 6-phosphate dehydrogenase and
17-p-hydroxy steroid dehydrogenase at 12
and 24 wk exposure.
Kumar et al.
(2000a)
Rat, Wistar,
male, 6/group
0 or 376 ppm
4 h/d, 5 d/wk,
12 and 24 wk
LOAEL: 376
ppm
BW gain sig. J,. Testis weight, sperm count
and motility sig. j, effect stronger with
exposure time. After 12 wk, numbers of
spermatogenic cells and spermatids j, some
of the spermatogenic cells appeared necrotic.
After 24 wk testes were atrophied, tubules
were smaller, had Sertoli cells and were
almost devoid of spermatocytes and
spermatids. Leydig cells were hyperplastic.
SDH, G6PDH sig. j, GGT and
P-glucuronidase sig. f; effects stronger with
exposure time.
4
This document is a draft for review purposes only and does not constitute Agency policy.
539 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-86. Summary of mammalian in vivo reproductive toxicity studies—
inhalation exposures (continued)
Reference
Species/strain/
sex/number
Exposure
level/duration
NOAEL;
LOAELa
Effects
Land et al.
(1981)
Mouse,
C57BlxC3H
(Fl), male, 5
or 10/group
0,0.02%, or
0.2%
4 h/d, 5 d, 23 d
rest
NOAEL:
0.02%
LOAEL: 0.2%
Sig. | percentage morphologically abnormal
epididymal sperm.
Xu et al.
(2004)
Mouse, CD-I,
male,
4-27/group
0 or 1,000 ppm
(5.37 mg/L)b
6 h/d, 5 d/wk,
1-6 wk
LOAEL: 1,000
ppm
Sig. 1 in vitro sperm-oocyte binding and in
vivo fertilization
Bolded studies carried forward for consideration in dose-response assessment (see Section 5).
1
2 aNOAEL and LOAEL are based upon reported study findings.
3 bDose conversion calculations by study author(s).
4
5 G6PDH = glucose 6-p dehydrogenase.
6
This document is a draft for review purposes only and does not constitute Agency policy.
540 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-87. Summary of mammalian in vivo reproductive toxicity studies—
2 oral exposures
3
Reference
Species/strain/
sex/number
Dose
level/exposure
duration
Route/vehicle
NOAEL;
LOAELa
Effects
Studies assessing male reproductive outcomes
DuTeaux et al.
(2003)
Rat, Sprague-
Dawley, male,
3/group
0, 0.2, or 0.4%
(0, 143, or 270
mg/kg-day)
Drinking
water; 3%
ethoxylated
castor oil
vehicle
LOEL: 0.2%
TCE metabolite-protein
adducts formed by a
cytochrome P450-mediated
pathway were detected by
fluorescence
imunohistochemistry in the
epithelia of corpus epididymis
and in efferent ducts.
DuTeaux et
al. (2004a)
Rat, Sprague-
Dawley, male,
3/group, or
Simonson
albino
(UC-Davis),
male, 3/group
0,0.2, or 0.4%
(0,143, or 270
mg/kg-day)
14 d
Drinking
water, 3%
ethoxylated
castor oil
vehicle
LOAEL:
0.2%
Dose-dependent -l in ability
of sperm to fertilize oocytes
collected from untreated $ s.
Oxidative damage to sperm
membrane in head and mid-
piece was indicated by dose-
related T in oxidized
proteins and lipid
peroxidation.
Veeramachane
ni et al. (2001)
Rabbit, Dutch
belted, females
and offspring;
7-9
offspring/group
9.5- or 28.5-ppm
TCEb
GD 20 thru
lactation, then to
offspring thru
postnatal wk 15
Drinking
water
LOAEL:
9.5 ppm
Decreased copulatory
behavior; acrosomal
dysgenesis, nuclear
malformations; sig. 1LH and
testosterone.
Zenick et al.
(1984)
Rat, Long-
Evans, male,
10/group
0,10,100, or
1,000 mg/kg-day
6 wk, 5 d/wk;
4 wk recovery
Gavage, corn
oil vehicle
NOAEL: 100
mg/kg-day
LOAEL:
1,000 mg/kg-
day
At 1,000 mg/kg, BW I,
liver/BW ratiosT, and
impaired copulatory
behavior. Copulatory
performance returned to
normal by 5th wk of
exposure. At wk 6, TCE
and metabolites
concentrated to a significant
extent in male reproductive
organs.
Studies assessing female reproductive outcomes
Berger and
Horner (2003)
Rat, Simonson
(S-D derived),
female,
(5-6) x 3/group
0 or 0.45%
2 wk
Drinking
water, 3%
Tween
vehicle
LOAEL:
0.45%
In vitro fertilization and
sperm penetration of oocytes
sig. -1 with sperm harvested
from untreated males.
4
This document is a draft for review purposes only and does not constitute Agency policy.
541 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-87. Summary of mammalian in vivo reproductive toxicity studies—
oral exposures (continued)
Reference
Species/strain/
sex/number
Dose
level/exposure
duration
Route/vehicle
NOAEL;
LOAEL'
Effects
Cosby and
Dukelow
(1992)
Mouse,
B6D2F1,
female,
7-12/group
0, 24, or 240
mg/kg-day
GD 1-5, 6-10, or
11-15
Gavage, corn
oil vehicle
NOAEL: 240
mg/kg-day
No treatment-related effects
on in vitro fertilization in
dams or offspring.
Manson et al.
(1984)
Rat, Long-
Evans, female,
23-25/group
0,10,100, or
1,000 mg/kg-day
6 wk: 2 wk
premating, 1 wk
mating period,
GD 1-21
Gavage, corn
oil vehicle
NOAEL: 100
mg/kg-day
LOAEL:
1,000 mg/kg-
day
Female fertility and mating
success was not affected. At
1,000 mg/kg-day group, 5/23
females died, gestation BW
gain was sig. -1. After
subchronic oral TCE
exposure, TCE was detected
in fat, adrenals, and ovaries;
TCA levels in uterine tissue
were high.
At 1,000 mg/kg-day,
neonatal deaths (female
pups) were T on PNDs 1,10,
and 14. Dose-related T seen
in TCA in blood, liver and
milk in stomach of $ pups,
not c?s.
Wu and
Berger (2007)
Rat, Simonson
(S-D derived),
female,
(no./group not
reported)
0 or 0.45%
(0.66 g/kg-d)°
Preovulation days
1-5,6-10, 11-14,
or 1-14
Drinking
water, 3%
Tween
vehicle
LOAEL:
0.45%
In vitro fertilization and
sperm penetration of oocytes
sig. -1 with sperm harvested
from untreated males.
Wu and
Berger (2008)
Rat, Simonson
(S-D derived),
female,
(no./group not
reported)
0 or 0.45%
(0.66 g/kg-d)°
1 or 5 d
Drinking
water, 3%
Tween
vehicle
NOEL: 0.45%
Ovarian mRNA expression
for ALCAM and Cudzl
protein were not altered.
This document is a draft for review purposes only and does not constitute Agency policy.
542 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-87. Summary of mammalian in vivo reproductive toxicity studies—
oral exposures (continued)
Reference
Species/strain/
sex/number
Dose
level/exposure
duration
Route/vehicle
NOAEL;
LOAEL'
Effects
Studies assessing fertility and reproductive outcome in both sexes
George et al.
(1985)
Mouse, CD-I,
male and
female, 20
pairs/treatment
group; 40
controls/sex
0,0.15,0.30, or
0.60 %d micro-
encapsulated
TCE
(TWA dose
estimates: 0,173,
362, or 737
mg/kg-day)c
Breeders exposed
1 wk premating,
then for 13 wk;
pregnant females
throughout
gestation (i.e.,
18 wk total)
Dietary
Parental
systemic
toxicity:
NOAEL:
0.30%
LOAEL:
0.60%
At 0.60%, in F0: sig.T liver
weights in both sexes; sig. -i
testis and seminal vesicle
weight; histopathology of
liver and kidney in both
sexes.
At 0.60%, in Fl: sig. i BW
on PND 74, and in
postpartum Fl dams; sig. T
liver, testis, and epididymis
weights in males, sig. T
kidney weights in both
sexes; sig. -1 testis and
seminal vesicle weight;
histopathology of liver and
kidney in both sexes.
Parental
reproductive
function:
LOAEL:
0.60% c
At 0.60%, in F0 and Fl
males: sig. 1 sperm motility.
Offspring
toxicity:
NOAEL:
0.30%
LOAEL:
0.60%
At 0.60%, inFl pups: sig. i
live birth weights, sig. 1 PND
4 pup BW; perinatal mortality
t (PND 0-21).
George et al.
(1986)
Rat, F334,
males and
female,
20 pairs/
treatment
group,
40 controls/sex
0,0.15,0.30 or
0.60 %d micro-
encapsulated
TCE
Breeders exposed
1 wk premating,
then for 13 wk;
pregnant females
throughout
gestation (i.e.,
18 wk total)
Dietary
Parental
systemic
toxicity:
LOAEL:
0.15%
At 0.60%, in F0: sig. X
postpartum dam BW; sig. -1
term. BW in both sexes; sig.
T liver, and kidney/adrenal
weights in both sexes; sig. T
testis/epididymis weights; in
Fl: sig. i testis weight.
At all doses in Fl: sig. -1
postpartum dam BW; sig.>l
term. BW in both sexes, sig.
t liver weight in both sexes.
At 0.30 and 0.60%, in Fl:
sig. T liver weight in
females.
This document is a draft for review purposes only and does not constitute Agency policy.
543 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-87. Summary of mammalian in vivo reproductive toxicity studies—
oral exposures (continued)
Reference
Species/strain/
sex/number
Dose
level/exposure
duration
Route/vehicle
NOAEL;
LOAEL3
Effects
George et al.
(1986)
(continued)
Parental
reproductive
function:
LOAEL:
0.60%d
At 0.60%, sig ^ mating in
F0 males and females (in
cross-over mating trials).
Offspring
toxicity:
LOAEL:
0.15%
At 0.60%, sig. i F1 BW on
PND 4 and 14.
At all doses, sig. -1 F1 BW
on PND 21 and 80.
At 0.3 and 0.60%, sig. -1 live
F1 pups/litter.
Sig. trend towards -l live
litters per pair
At 0.15 and 0.60%, trend
toward -l F1 survival from
PND 21-80.
Bolded studies carried forward for consideration in dose-response assessment (see Section 5).
1
2 aNOAEL, LOAEL, NOEL, and LOEL (lowest-observed-effect level) are based upon reported study findings.
3 bConcurrent exposure to several ground water contaminants; values given are for TCE levels in the mixture.
4 °Dose conversion calculations by study author(s).
5 fertility and reproduction assessment of last litter from continuous breeding phase and cross-over mating
6 assessment (rats only) were conducted for 0 or 0.60% dose groups only.
7
8 LH = luteinizing hormone.
9
10
11 Kumar et al. (2000b) exposed male Wistar rats in whole-body inhalation chambers to
12 376-ppm TCE for 4 hours/day, 5 days/week over several duration scenarios. These were
13 2-weeks (to observe the effect on the epididymal sperm maturation phase), 10 weeks (to observe
14 the effect on the entire spermatogenic cycle), 5 weeks with 2 weeks rest (to observe the effect on
15 primary spermatocytes differentiation to sperm), 8 weeks with 5 weeks rest (to observe effects
16 on an intermediate stage of spermatogenesis), and 10 weeks with 8 weeks rest (to observe the
17 effect on spermatogonial differentiation to sperm). Control rats were exposed to ambient air.
18 Weekly mating with untreated females was conducted. At the end of the treatment/rest periods,
19 the animals were sacrificed; testes and cauda epididymes tissues were collected. Alterations in
20 testes histopathology (smaller, necrotic spermatogenic tubules), increased sperm abnormalities,
21 and significantly increased pre- and/or postimplantation loss in litters were observed in the
This document is a draft for review purposes only and does not constitute Agency policy.
544 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
groups with 2 or 10 weeks of exposure, or 5 weeks of exposure with 2 weeks rest. It was
hypothesized that postmeiotic cells of spermatogenesis and epididymal sperm were affected by
TCE exposure, leading to reproductive impairment.
To test the hypothesis that TCE exposure adversely affects sperm function and
fertilization, Xu et al. (2004) conducted a study in which male CD-I mice were exposed by
inhalation to atmospheres containing 1,000 ppm (5.37 mg/L) TCE for 1-6 weeks (6 hours/day,
5 days/week). After each TCE exposure, body weights were recorded. Following termination,
the right testis and epididymis of each treated male were weighed, and sperm was collected from
the left epididymis and vas deferens for assessment of the number of total sperm and motile
sperm. Sperm function was evaluated in the following experiments: (1) suspensions of
capacitated vas deferens/cauda epididymal sperm were examined for spontaneous acrosome
reaction, (2) in vitro binding of capacitated sperm to mature eggs from female CF-1 mice
(expressed as the number of sperm bound per egg) was assessed, and (3) in vivo fertilization was
evaluated via mating of male mice to superovulated female CF-1 mice immediately following
inhalation exposure; cumulus masses containing mature eggs were collected from the oviducts of
the females, and the percentage of eggs fertilized was examined. Inhalation exposure to TCE did
not result in altered body weight, testis and epididymis weights, sperm count, or sperm
morphology or motility. Percentages of acrosome-intact sperm populations were similar
between treated and control animals. Nevertheless, for males treated with TCE for 2 or more
weeks decreases were observed in the number of sperm bound to the oocytes in vitro (significant
at 2 and 6 weeks,/? < 0.001). In a follow-up assessment, control sperm were incubated for
30-minutes in buffered solutions of TCE or metabolites (chloral hydrate or trichloroethanol);
while TCE-incubation had no effect on sperm-oocyte binding, decreased binding capacity was
noted for the metabolite-incubated sperm. The ability for sperm from TCE-exposed males to
bind to and fertilize oocytes in vivo was also found to be significantly impaired (p < 0.05).
A study designed to investigate the role of testosterone, and of cholesterol and ascorbic
acid (which are primary precursors of testosterone) in TCE-exposed rats with compromised
reproductive function was conducted by Kumar et al. (2000a). Male Wistar rats (12-13/group)
were exposed (whole body) to 376 ppm TCE by inhalation for 4 hours/day, 5 days/week, for
either 12 or 24 weeks and then terminated. Separate ambient-air control groups were conducted
for the 12- and 24-week exposure studies. Epididymal sperm count and motility were evaluated,
and measures of 17-P-hydroxy steroid dehydrogenase (17-P-HSD), testicular total cholesterol
and ascorbic acid, serum testosterone, and glucose 6-p dehydrogenase (G6PDH) in testicular
homogenate were assayed. In rats exposed to TCE for either 12 or 24 weeks, total epididymal
sperm count and motility, serum testosterone concentration, and specific activities of both
This document is a draft for review purposes only and does not constitute Agency policy.
545 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
17-P-HSD and G6PDH were significantly decreased (p < 0.05), while total cholesterol content
was significantly (p < 0.05) increased. Ascorbic acid levels were not affected.
In another study, Kumar et al. (2001b) utilized the same exposure paradigm to examine
cauda epididymal sperm count and motility, testicular histopathology, and testicular marker
enzymes: sorbitol dehydrogenase (SDH), G6PDH, glutamyl transferase (GT), and glucuronidase,
in Wistar rats (6/group). After 24 weeks of exposure, testes weights and epididymal sperm count
and motility were significantly decreased (p < 0.05). After 12 weeks of TCE exposure,
histopathological examination of the testes revealed a reduced number of spermatogenic cells in
the seminiferous tubules, fewer spermatids as compared to controls, and the presence of necrotic
spermatogenic cells. Testicular atrophy, smaller tubules, hyperplastic Leydig cells, and a lack of
spermatocytes and spermatids in the tubules were observed after 24 weeks of TCE exposure.
After both 12 and 24 weeks of exposure, SDH and G6PDH were significantly (p < 0.05) reduced
while GT and P-glucuronidase were significantly (p < 0.05) increased.
In a study by Land et al. (1981), 8-10 week old male mice (C57BlxC3H)Fl (5 or
10/group) were exposed (whole body) by inhalation to a number of anesthetic agents for
5 consecutive days at 4 hours/day and sacrificed 28 days after the first day of exposure.
Chamber concentration levels for the TCE groups were 0.02 and 0.2%. The control group
received ambient air. Epididymal sperm were evaluated for morphological abnormalities. At
0.2% TCE, the percentage of abnormal sperm in a sample of 1,000 was significantly (p < 0.01)
increased as compared to control mice; no treatment-related effect on sperm morphology was
observed at 0.02% TCE.
Forkert et al. (2002) exposed male CD-I mice by inhalation to 1,000-ppm TCE
(6 hours/day, 5 day/week) for 4 consecutive weeks and observed sloughing of portions of the
epithelium upon histopathological evaluation of testicular and epididymal tissues.
Kan et al. (2007) also demonstrated that damage to the epididymal epithelium and sperm
of CD-I mice (4/group) resulted from exposure to 0 or 1,000-ppm TCE by inhalation for
6 hours/day, 5 days/week, for 1-4 weeks. Segments of the epididymis (caput, corpus, and
cauda) were examined by light and electron microscope. As early as 1 week after TCE exposure,
degeneration and sloughing of epithelial cells from all three epididymal areas were observed by
light microscopy; these findings became more pronounced by 4 weeks of exposure. Vesiculation
in the cytoplasm, disintegration of basolateral cell membranes, and epithelial cell sloughing were
observed with electron microscopy. Sperm were found in situ in the cytoplasm of degenerated
epididymal cells. A large number of sperm in the lumen of the epididymis were abnormal,
including head and tail abnormalities.
This document is a draft for review purposes only and does not constitute Agency policy.
546 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
4.8.1.1.8. Oral exposures
A variety of studies were conducted to assess various aspects of male and/or female
reproductive capacity in laboratory animal species following oral exposures to TCE. These are
described below and summarized in Table 4-87. They include studies that focused on male
reproductive outcomes in rats or rabbits following gavage or drinking water exposures (DuTeaux
et al., 2004a; DuTeaux et al., 2003; Veeramachaneni et al., 2001; Zenick et al., 1984), studies
that focused on female reproductive outcomes in rats following gavage or drinking water
exposures (Berger and Horner, 2003; Cosby and Dukelow, 1992; Manson et al., 1984; Wu and
Berger, 2007, 2008), and studies assessed fertility and reproductive outcome in both sexes
following dietary exposures to CD-I mice or F344 rats (George et al., 1985, 1986).
4.8.1.1.8.1.Studies assessing male reproductive outcomes
Zenick et al. (1984) conducted a study in which sexually experienced Long-Evans
hooded male rats were administered 0, 10, 100, or 1,000 mg/kg-day TCE by gavage in corn oil
for 6 weeks. A 4-week recovery phase was also incorporated into the study design. Endpoints
assessed on Weeks 1 and 5 of treatment included copulatory behavior, ejaculatory plug weights,
and ejaculated or epididymal sperm measures (count, motility, and morphology). Sperm
measures and plug weights were not affected by treatment, nor were Week 6 plasma testosterone
levels found to be altered. TCE effects on copulatory behavior (ejaculation latency, number of
mounts, and number of intromissions) were observed at 1,000 mg/kg-day; these effects were
recovered by 1-4 weeks posttreatment. Although the effects on male sexual behavior in this
study were believed to be unrelated to narcotic effects of TCE, a later study by Nelson and
Zenick (1986) showed that naltrexone (an opioid receptor antagonist, 2.0 mg/kg, i.p.,
administered 15 minutes prior to testing) could block the effect. Thus, it was hypothesized that
the adverse effects of TCE on male copulatory behavior in the rat at 1,000 ppm may in fact be
mediated by the endogenous opioid system at the CNS level.
In a series of experiments by DuTeaux et al. (2004a; 2003), adult male rats were
administered 0, 0.2, or 0.4% TCE (v/v) (equivalent to 0, 2.73 mg/L, or 5.46 mg/L) in a solution
of 3% ethoxylated castor oil in drinking water for 14 days. These concentrations were within the
range of measurements obtained in formerly contaminated drinking water wells, as reported by
ATSDR (1997c). The average ingested doses of TCE (based upon animal body weight and
average daily water consumption of 28 mL) were calculated to be 143 or 270 mg/kg-day for the
low and high-dose groups, respectively (DuTeaux et al., (2003). Cauda epididymal and vas
deferens sperm from treated males were incubated in culture medium with oviductal cumulus
This document is a draft for review purposes only and does not constitute Agency policy.
547 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
masses from untreated females to assess in vitro fertilization capability. Treatment with TCE
resulted in a dose-dependent decrease in the ability of sperm to fertilize oocytes. Terminal body
weights and testis/epididymal weights were similar between control and treated groups.
Evaluation of sperm concentration or motility parameters did not reveal any treatment-related
alterations; acrosomal stability and mitochondrial membrane potential were not affected by
treatment. Although no histopathological changes were observed in the testis or in the caput,
corpus, or cauda epididymis, exposure to 0.2 and 0.4% TCE resulted in slight cellular alterations
in the efferent ductule epithelium.
Veeramachaneni et al. (2001) evaluated the effects of drinking water containing
chemicals typical of ground water near hazardous waste sites (including 9.5- or 28.5-ppm TCE)
on male reproduction. In this study, pregnant Dutch-belted rabbits were administered treated
drinking water from GD 20; treatment continued through the lactation period and to weaned
offspring (7-9/group) through postnatal Week 15. Deionized water was administered from
postnatal weeks 16-61, at which time the animals were terminated. At 57-61 weeks of age,
ejaculatory capability, and seminal, testicular, epididymal, and endocrine characteristics were
evaluated. In both treated groups, long-term effects consisted of decreased copulatory behavior
(interest, erection, and/or ejaculation), significant increases in acrosomal dysgenesis and nuclear
malformations (p < 0.03), and significant decreases in serum concentration of luteinizing
hormone (p < 0.05) and testosterone secretion after human chorionic gonadotropin
administration (p < 0.04). There were no effects on total spermatozoa per ejaculate or on daily
sperm production. The contribution of individual drinking water contaminants to adverse male
reproductive outcome could not be discerned in this study. Additionally, it was not designed to
distinguish between adverse effects that may have resulted from exposures in late gestation (i.e.,
during critical period of male reproductive system development) versus postnatal life.
4.8.1.1.8.2.Studies assessing female reproductive outcomes
In a study that evaluated postnatal growth following gestational exposures, female
B6D2F1 mice (7-12/group) were administered TCE at doses of 0, 1% LD50 (24 mg/kg-day), and
10% LD50 (240 mg/kg-day) by gavage in corn oil from GDs 1-5, 6-10, or 11-15 (day of mating
was defined as GD 1) (Cosby and Dukelow, 1992). Litters were examined for pup count, sex,
weight, and crown-rump measurement until GD 21. Some offspring were retained to 6 weeks of
age, at which time they were killed and the gonads were removed, weighed, and preserved. No
treatment-related effects were observed in the dams or offspring. In a second series of studies
conducted by (Cosby and Dukelow) and reported in the same paper, TCE and its metabolites
This document is a draft for review purposes only and does not constitute Agency policy.
548 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
DC A, TCA, and TCOH were added to culture media with capacitated sperm and cumulus masses
from B6D2F1 mice to assess effects on in vitro fertilization. Dose-related decreases in
fertilization were observed for DCA, TCA, and TCOH at 100 and 1,000 ppm, but not with TCE.
Synergystic effects were not observed with TCA and TCOH.
A study was conducted by Manson et al. (1984) to determine if subchronic oral exposure
to TCE affected female reproductive performance, and if TCE or its metabolites trichloroacetic
acid or trichloroethanol accumulated in female reproductive organs or neonatal tissues. Female
Long-Evans hooded rats (22-23/group) were administered 0 (corn oil vehicle), 10, 100, or
1,000 mg/kg-day of TCE by gavage for 2 weeks prior to mating, throughout mating, and to GD
21. Delivered pups were examined for gross anomalies, and body weight and survival were
monitored for 31 days. Three maternal animals per group and 8-10 neonates per group (killed
on GDs 3 and 31) were analyzed for TCE and metabolite levels in tissues. TCE exposure
resulted in 5 deaths and decreased maternal body weight gain at 1,000 mg/kg-day, but did not
affect estrous cycle length or female fertility at any dose level. There were no evident
developmental anomalies observed at any treatment level; however, at 1,000 mg/kg-day there
was a significant increase in the number of pups (mostly female) born dead, and the cumulative
neonatal survival count through PND 18 was significantly decreased as compared to control.
TCE levels were uniformly high in fat, adrenal glands, and ovaries across treatment groups, and
TCA levels were high in uterine tissue. TCE levels in the blood, liver, and milk contents of the
stomach increased in female PND-3 neonates across treatment groups. These findings suggest
that increased metabolite levels did not influence fertility, mating success, or pregnancy
outcome.
In another study that examined the potential effect of TCE on female reproductive
function, Berger and Horner (2003) conducted 2-week exposures of Sprague-Dawley derived
female Simonson rats to tetrachloroethylene, trichloroethylene, several ethers, and
4-vinylcyclohexene diepoxide in separate groups. The TCE-treated group received 0.45% TCE
in drinking water containing 3% Tween vehicle; control groups were administered either
untreated water, or water containing the 3% Tween vehicle. There were 5-6 females/group, and
three replicates were conducted for each group. At the end of exposure, ovulation was induced,
the rats were killed, and the ovaries were removed. The zona pellucida was removed from
dissected oocytes, which were then placed into culture medium and inseminated with sperm from
untreated males. TCE treatment did not affect female body weight gain, the percentage of
females ovulating, or the number of oocytes per ovulating female. Fertilizability of the oocytes
from treated females was reduced significantly (46% for TCE-treated females versus 56% for
vehicle controls). Oocytes from TCE-treated females had reduced ability to bind sperm plasma
membrane proteins compared with vehicle controls.
This document is a draft for review purposes only and does not constitute Agency policy.
549 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
In subsequent studies, Wu and Berger (2007, 2008) examined the effect of TCE on
oocyte fertilizibility and ovarian gene expression. TCE was administered to female Simonson
rats (number of subjects not reported) in the drinking water at 0 or 0.45% (in 3% Tween vehicle);
daily doses were estimated to be 0.66 g TCE/kg body weight/day. In the oocyte fertilizibility
study (Wu and Berger, 2007), the female rats were treated on Days 1-5, 6-10, 11-14, or 1-14 of
the 2-week period preceding ovulation (on Day 15). Oocytes were extracted and fertilized in
vitro with sperm from a single male donor rat. With any duration of TCE exposure, fertilization
(as assessed by the presence of decondensed sperm heads) was significantly (p < 0.05) decreased
as compared to controls. After exposure on Days 6-10, 11-14, or 1-14, the oocytes from
TCE-treated females had a significantly decreased ability to bind sperm (p < 0.05) in comparison
to oocytes from vehicle controls. Increased protein carbonyls (an indicator of oxidatively
modified proteins) were detected in the granulosa cells of ovaries from females exposed to TCE
for 2 weeks. The presence of oxidized protein was confirmed by Western blot analysis.
Microsomal preparations demonstrated the localization of cytochrome P450 2E1 and glutathione
S-transferase (TCE-metabolizing enzymes) in the ovary. Ovarian mRNA transcription for
ALCAM and Cuzdl protein was not found to be altered after 1 or 5 days of exposure (Wu and
Berger, 2008), suggesting that the posttranslational modification of proteins within the ovary
may partially explain the observed reductions in oocyte fertilization.
4.8.1.1.8.3 .Studies assessing fertility and reproductive outcomes in both sexes
Assessments of reproduction and fertility with continuous breeding were conducted in
NTP studies in CD-I mice (George et al., 1985) and Fischer 344 rats (George et al., 1986). TCE
was administered to the mice and rats at dietary levels of 0, 0.15, 0.30, or 0.60%, based upon the
results of preliminary 14-day dose-range finding toxicity studies. Actual daily intake levels for
the study in mice were calculated from the results of dietary formulation analyses and body
weight/food consumption data at several time points during study conduct; the most conservative
were from the second week of the continuous breeding study: 0, 52.5, 266.3, and
615.0 mg/kg-day. No intake calculations were presented for the rat study. In these studies,
which were designed as described by Chapin and Sloane (1997), the continuous breeding phase
in F0 adults consisted of a 7-day premating exposure, 98-day cohabitation period, and 28-day
segregation period. In rats, a crossover mating trial (i.e., control males x control females; 0.60%
TCE males x control females; control males x 0.60% TCE females) was conducted to further
elucidate treatment-related adverse reproductive trends observed in the continuous breeding
phase. The last litter of the continuous breeding phase was raised to sexual maturity for an
This document is a draft for review purposes only and does not constitute Agency policy.
550 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
assessment of fertility and reproduction in control and high-dose groups; for the rats, this
included an open field behavioral assessment of F1 pups. The study protocol included terminal
studies in both generations, including sperm evaluation (count morphology, and motility) in 10
selected males per dose level, macroscopic pathology, organ weights, and histopathology of
selected organs.
In the continuous breeding phase of the CD-I mouse study (George et al., 1985), no
clinical signs of toxicity were observed in the parental (F0) animals, and there were no treatment-
related effects on the proportion of breeding pairs able to produce a litter, the number of live
pups per litter, the percentage born live, the proportion of pups born live, the sex of pups born
live, absolute live pup weights, or adjusted female pup weights. At the high dose level of 0.60%,
a number of adverse outcomes were observed. In the parental animals, absolute and body-
weight-adjusted male and female liver weight values were significantly increased (p < 0.01), and
right testis and seminal vesicle weights were decreased (p < 0.05), but kidney/adrenal weights
were not affected. Sperm motility was significantly (p < 0.01) decreased by 45% in treated
males as compared to controls. Histopathology examination revealed lesions in the liver
(hypertrophy of the centrilobular liver cells) and kidneys (tubular degeneration and karyomegaly
of the corticomedullary renal tubular epithelium) of F0 males and females. In the pups at 0.60%,
adjusted live birth weights for males and both sexes combined were significantly decreased
(p < 0.01) as compared to control. The last control and high-dose litters of the continuous
breeding assessment were raised to the age of sexual maturity for a further assessment of
reproductive performance. In these F1 pups, body weights (both sexes) were significantly
decreased at PND 4, and male offspring body weights were significantly (p < 0.05) less than
controls at PND 74 (±10). It was reported that perinatal mortality (PND 0-21) was increased,
with a 61.3%) mortality rate for TCE-treated pups versus a 28.3% mortality rate for control pups.
Reproductive performance was not affected by treatment, and postmortem evaluations of the F1
adult mice revealed significant findings at 0.60% TCE that were consistent with those seen in the
F0 adults and additionally demonstrated renal toxicity, i.e., elevated liver and kidney/adrenal
weights and hepatic and renal histopathological lesions in both sexes, elevated testis and
epididymis weights in males, and decreased sperm motility (18% less than control).
The F344 rat study continuous breeding phase demonstrated no evidence of treatment-
related effects on the proportion of breeding pairs able to produce a litter, percentage of pups
born alive, the sex of pups born alive, or absolute or adjusted pup weights (George et al., 1986).
However, the number of live pups per litter was significantly (p < 0.05) decreased at 0.30 and
0.60%) TCE, and a significant (p < 0.01) trend toward a dose-related decrease in the number of
live litters per pair was observed; individual data were reported to indicate a progressive decrease
in the number of breeding pairs in each treatment group producing third, fourth, and fifth litters.
This document is a draft for review purposes only and does not constitute Agency policy.
5 51 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
The crossover mating trial conducted in order to pursue this outcome demonstrated that the
proportion of detected matings was significantly depressed (p < 0.05) in the mating pairs with
TCE-treated partners compared to the control pairs. In the F0 adults at 0.60% TCE, postpartum
dam body weights were significantly decreased (p < 0.01 or 0.05) in the continuous breeding
phase and the crossover mating trials, and terminal body weights were significantly decreased
(p < 0.01) for both male and female rats. Postmortem findings for F0 adults in the high-dose
group included significantly increased absolute and body-weight-adjusted liver and
kidney/adrenal weights in males, increased adjusted liver and kidney/adrenal weights in females,
and significantly increased adjusted left testis/epididymal weights. Sperm assessment did not
identify any effects on motility, concentration or morphology, and histopathological examination
was negative. The last control and high-dose litters of the continuous breeding assessment were
raised to the age of sexual maturity for assessment of open field behavior and reproductive
performance. In these F1 pups at 0.60% TCE, body weights of male and females were
significantly (p < 0.05 or 0.01, respectively) decreased at PND 4 and 14. By PND 21, pup
weights in both sexes were significantly reduced in all treated groups, and this continued until
termination (approximately PND 80). A tendency toward decreased postweaning survival (i.e.,
from PND 21 to PND 81 ± 10) was reported for F1 pups at the 0.15 and 0.60% levels. Open
field testing revealed a significant (p < 0.05) dose-related trend toward an increase in the time
required for male and female F1 weanling pups to cross the first grid in the testing device,
suggesting an effect on the ability to react to a novel environment. Reproductive performance
assessments conducted in this study phase were not affected by treatment. Postpartum F1 dam
body weights were significantly decreased (p < 0.05 or 0.01) in all of the TCE-treated groups as
compared to controls, as were terminal body weights for both adult F1 males and females.
Postmortem evaluations of the F1 adult rats revealed significantly (p < 0.01) decreased left
testis/epididymis weight at 0.60% TCE, and significantly (p < 0.05 or 0.01) increased adjusted
mean liver weight in all treated groups for males and at 0.30 and 0.60% for females. Sperm
assessments for F1 males revealed a significant increase (p < 0.05) in the percentage of abnormal
sperm in the 0.30% TCE group, but no other adverse effects on sperm motility, concentration, or
morphology were observed. As with the F0 adults, there were no adverse treatment-related
findings revealed at histopathological assessment. The study authors concluded that the
observed effects to TCE exposure in this study were primarily due to generalized toxicity and not
to a specific effect on the reproductive system; however, based upon the overall toxicological
profile for TCE, which demonstrates that the male reproductive system is a target for TCE
exposures, this conclusion is not supported.
This document is a draft for review purposes only and does not constitute Agency policy.
552 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
4.8.1.1.9. Discussion/Synthesis of Noncancer Reproductive Toxicity Findings
The human epidemiological findings and animal study evidence consistently indicate that
TCE exposures can result in adverse reproductive outcomes. Although the epidemiological data
may not always be robust or unequivocal, they demonstrate the potential for a wide range of
exposure-related adverse outcomes on female and male reproduction. In animal studies, there is
some evidence for female-specific reproductive toxicity; but there is strong and compelling
evidence for adverse effects of TCE exposure on male reproductive system and function.
4.8.1.1.10. Female reproductive toxicity
Although few epidemiological studies have examined TCE exposure in relation to female
reproductive function (see Table 4-88), the available studies provide evidence of decreased
fertility, as measured by time to pregnancy (Sallmen et al., 1995), and effects on menstrual cycle
patterns, including abnormal cycle length (ATSDR, 2001), amenorrhea (Sagawa et al., 1973;
Zielinski, 1973), and menstrual "disturbance" (Bardodej and Vyskocil, 1956). In experimental
animals, the effects on female reproduction include evidence of reduced in vitro oocyte
fertilizability in rats (Berger and Horner, 2003; Wu and Berger, 2007). However, in other
studies that assessed reproductive outcome in female rodents (Cosby and Dukelow, 1992;
George et al., 1985, 1986; Manson et al., 1984), there was no evidence of adverse effects of TCE
exposure on female reproductive function. Overall, although the data are suggestive, there are
inadequate data to make conclusions as to whether adverse effects on human female
reproduction are caused by TCE.
Table 4-88. Summary of adverse female reproductive outcomes associated
with TCE exposures
Finding
Species
Citation
Menstrual cycle disturbance
Human
ATSDR (2001)a
Bardodej and Vyskocil (1956)
Sagawa et al. (1973)
Zielinski (1973)
Reduced fertility
Humana
Sallmen et al. (1995)
Ratb
Berger and Horner (2003)
Wu and Berger (2007)
This document is a draft for review purposes only and does not constitute Agency policy.
553 DRAFT—DO NOT CITE OR QUOTE
-------
1 aNot significant.
2 bIn vitro oocyte fertilizability.
3
4
4.8.1.1.11. Male reproductive toxicity
5 Notably, the results of a number of studies in both humans and experimental animals
6 have suggested that exposure to TCE can result in targeted male reproductive toxicity (see
7 Table 4-89). The adverse effects that have been observed in both male humans and male animal
8
This document is a draft for review purposes only and does not constitute Agency policy.
554 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-89. Summary of adverse male reproductive outcomes associated
2 with TCE exposures
3
Finding
Species
Citation
Testicular toxicity/pathology
Rat
George et al. (1986)
Kumar et al. (2000b)
Kumar et al. (2001b)
Mouse
Kan et al. (2007)
Epididymal toxicity/pathology
Mouse
Forkert et al. (2002)
Decreased sperm quantity/quality
Human
Chia et al. (1996)
Rasmussen et al. (1988)a
Rat
Kumar et al. (2001b; 2000a;
2000b)
Mouse
George et al. (1985)
Land et al. (1981)
Rabbit
Veeramachaneni et al. (2001)
Altered in vitro sperm-oocyte binding or in vivo
fertilization
Rat
DuTeaux et al. (2004a)
Mouse
Cosby and Dukelow (1992)b
Xu et al. (2004)b
Altered sexual drive or function
Human
El Ghawabi et al. (1973)
Saihan et al. (1978)°
Bardodej and Vyskocil (1956)
Rat
George et al. (1986)
Zenick et al. (1984)
Rabbit
Veeramachaneni et al. (2001)
Altered serum testosterone levels
Human
Chia et al. (1997)d
Goh et al. (1998)e
Rat
Kumar et al. (2000a)
Rabbit
Veeramachaneni et al. (2001)
Reduced fertility
Rat
George et al. (1986)
Gynaecomastia
Human
Saihan et al. (1978)°
4 "Nonsignificant increase in percentage of two YFF in spermatozoa; no effect on sperm count or morphology.
5 bObserved with metabolite(s) of TCE only.
6 °Case study of one individual.
7 dAlso observed altered levels of DHEAS, FSH, and SHBG.
This document is a draft for review purposes only and does not constitute Agency policy.
555 DRAFT—DO NOT CITE OR QUOTE
-------
1 eAlso observed altered levels of SHBG.
2
This document is a draft for review purposes only and does not constitute Agency policy.
556 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
models include altered sperm count, morphology, or motility (Chia et al., 1996; George et al.,
1985; Kumar et al., 2001b; Kumar et al., 2000a; Kumar et al., 2000b; Land et al., 1981;
Rasmussen et al., 1988; Veeramachaneni et al., 2001); decreased libido or copulatory behavior
(Bardodej and Vyskocil, 1956; El Ghawabi et al., 1973; George et al., 1986; Saihan et al., 1978;
Veeramachaneni et al., 2001; Zenick et al., 1984); alterations in serum hormone levels (Chia et
al., 1997; Goh et al., 1998; Kumar et al., 2000a; Veeramachaneni et al., 2001); and reduced
fertility (George et al., 1986). However, other studies in humans did not see evidence of altered
sperm count or morphology (Rasmussen et al., 1988) or reduced fertility (Forkert et al., 2003;
Sallmen et al., 1998), and some animal studies also did not identify altered sperm measures
(Cosby and Dukelow, 1992; George et al., 1986; Xu et al., 2004; Zenick et al., 1984). Additional
adverse effects observed in animals include histopathological lesions of the testes (George et al.,
1986; Kumar et al., 2001b; Kumar et al., 2000b) or epididymides (Forkert et al., 2002; Kan et al.,
2007) and altered in vitro sperm-oocyte binding and/or in vivo fertilization for TCE and/or its
metabolites (DuTeaux et al., 2004a; Xu et al., 2004).
In spite of the preponderance of studies demonstrating effects on sperm parameters, there
is an absence of overwhelming evidence in the database of adverse effects of TCE on overall
fertility in the rodent studies. That is not surprising, however, given the redundancy and
efficiency of rodent reproductive capabilities. Nevertheless, the continuous breeding
reproductive toxicity study in rats (George et al., 1986) did demonstrate a trend towards
reproductive compromise (i.e., a progressive decrease in the number of breeding pairs producing
third, fourth, and fifth litters).
It is noted that in the studies by George et al. (George et al., 1985, 1986), adverse
reproductive outcomes in male rats and mice were observed at the highest dose level tested
(0.060% TCE in diet) which was also systemically toxic (i.e., demonstrating kidney toxicity and
liver enzyme induction and toxicity, sometimes in conjunction with body weight deficits).
Because of this, the study authors concluded that the observed reproductive toxicity was a
secondary effect of generalized systemic toxicity; however, this conclusion is not supported by
the overall toxicological profile of TCE which provides significant evidence indicating that TCE
is a reproductive toxicant.
4.8.1.1.11.1. The role of metabolism in male reproductive toxicity
There has been particular focus on evidence of exposure to male reproductive organs by
TCE and/or its metabolites, as well as the role of TCE metabolites in the observed toxic effects.
This document is a draft for review purposes only and does not constitute Agency policy.
557 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
In humans, a few studies demonstrating male reproductive toxicity have measured levels
of TCE in the body. U-TCA was measured in men employed in an electronics factory, and
adverse effects observed included abnormal sperm morphology and hyperzoospermia and altered
serum hormone levels (Chia et al., 1996, 1997; Goh et al., 1998). U-TCA was also measured as
a marker of exposure to TCE in men occupationally exposed to solvents, although this study did
not report any adverse effects on fertility (Sallmen et al., 1998).
In the study in Long-Evans male rats by Zenick et al. (1984), blood and tissue levels of
TCE, TCA, and TCOH were measured in three rats/group following 6 weeks of gavage treatment
at 0, 10, 100, and 1,000 mg/kg-day. Additionally the levels of TCE and metabolites were
measured in seminal plugs recovered following copulation at Week 5. Marked increases in TCE
levels were observed only at 1,000 mg/kg-day, in blood, muscle, adrenals, and seminal plugs. It
was reported that dose-related increases in TCA and TCOH concentrations were observed in the
organs evaluated, notably including the reproductive organs (epididymis, vas deferens, testis,
prostate, and seminal vesicle), thus, creating a potential for interference with reproductive
function.
This potential was explored further in a study by Forkert et al. (2002), in which male
CD-I mice were exposed by inhalation to 1,000-ppm TCE (6 hours/day, 5 day/week) for
4 consecutive weeks. Urine was obtained on Days 4, 9, 14, and 19 of exposure and analyzed for
concentrations of TCE and TCOH. Microsomal preparations from the liver, testis and
epididymis were used for immunoblotting, determining /;-nitrophenol hydroxylase and CYP2E1
activities, and evaluating the microsomal metabolism of TCE.
Subsequent studies conducted by the same laboratory (Forkert et al., 2003) evaluated the
potential of the male reproductive tract to accumulate TCE and its metabolites including chloral,
TCOH, TCA, and DCA. Human seminal fluid and urine samples from eight mechanics
diagnosed with clinical infertility and exposed to TCE occupationally were analyzed. Urine
samples from two of the eight subjects contained TCA and/or TCOH, suggesting that TCE
exposure and/or metabolism was low during the time just prior to sample collection. TCE,
chloral, and TCOH were detected in seminal fluid samples from all eight subjects, while TCA
was found in one subject, and DCA was found in two subjects. Additionally, TCE and its
metabolites were assessed in the epididymis and testis of CD-I mice (4/group) exposed by
inhalation (6 hours/day, 5 days/week) to 1,000 ppm TCE for 1, 2, and 4 weeks. TCE, chloral,
and TCOH were found in the epididymis at all timepoints, although TCOH levels were increased
significantly (tripled) at 4 weeks of exposure. This study showed that the metabolic disposition
of TCE in humans is similar to that in mice, indicating that the murine model is appropriate for
investigating the effects of TCE-induced toxicity in the male reproductive system. These studies
provide support for the premise that TCE is metabolized in the human reproductive tract, mainly
This document is a draft for review purposes only and does not constitute Agency policy.
558 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
in the epididymis, resulting in the production of metabolites that cause damage to the epididymal
epithelium and affect the normal development of sperm.
Immunohistochemical experiments (Forkert et al., 2002) confirmed the presence of
CYP2E1 in the epididymis and testis of mice; it was found to be localized in the testicular
Leydig cells and the epididymal epithelium. Similar results were obtained with the
immunohistochemical evaluation of human and primate tissue samples. CYP2E1 has been
previously shown by Lipscomb et al. (1998a) to be the predominant CYP enzyme catalyzing the
hepatic metabolism of TCE in both animals and rodents. These findings support the role of
CYP2E1 in TCE metabolism in the male reproductive tract of humans, primates, and mice.
4.8.1.1.11.2. Mode of action for male reproductive toxicity
A number of studies have been conducted to attempt to characterize various aspects of
the mode of action for observed male reproductive outcomes.
Studies by Kumar et al. (2001b; 2000a) suggest that perturbation of testosterone
biosynthesis may have some role in testicular toxicity and altered sperm measures. Significant
decreases in the activity of G6PDH and accumulation of cholesterol are suggestive of an
alteration in testicular steroid biosynthesis. Increased testicular lipids, including cholesterol,
have been noted for other testicular toxicants such as lead (Saxena et al., 1987),
triethylenemelamine (Johnson et al., 1967), and quinalphos (Ray et al., 1987), in association with
testicular degeneration and impaired spermatogenesis. Since testosterone has been shown to be
essential for the progression of spermatogenesis (O'Donnell et al., 1994), alterations in
testosterone production could be a key event in male reproductive dysfunction following TCE
exposure. Additionally, the observed TCE-related reduction of 17-P-HSD, which is involved in
the conversion of androstenedione to testosterone, has also been associated with male
reproductive insufficiency following exposure to phthalate esters (Srivastava and Srivastava,
1991), quinalphos (Ray et al., 1987), and lead (Saxena et al., 1987). Reductions in SDH, which
are primarily associated with the pachytene spermatocyte maturation of germinal epithelium,
have been shown to be associated with depletion of germ cells (Chapin et al., 1982; Mills and
Means, 1972), and the activity of G6PDH is greatest in premeiotic germ cells and Leydig cells of
the interstitium (Blackshaw, 1970). The increased GT and glucuronidase observed following
TCE exposures appear to be indicative of impaired Sertoli cell function (Hodgen and Sherins,
1973; Sherins and Hodgen, 1976). Based upon the conclusions of these studies, Kumar et al.
(2001b) hypothesized that the reduced activity of G6PDH and SDH in testes of TCE-exposed
This document is a draft for review purposes only and does not constitute Agency policy.
559 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
male rats is indicative of the depletion of germ cells, spermatogenic arrest, and impaired function
of the Sertoli cells and Leydig cells of the interstitium.
In the series of experiments by DuTeaux et al. (2004a; 2003), protein dichloroacetyl
adducts were found in the corpus epididymis and in the efferent ducts of rats administered TCE;
this effect was also demonstrated following in vitro exposure of reproductive tissues to TCE.
Oxidized proteins were detected on the surface of spermatozoa from TCE-treated rats in a
dose-response pattern; this was confirmed using a Western blotting technique. Soluble (but not
mitochondrial) cysteine-conjugate P-lyase was detected in the epididymis and efferent ducts of
treated rats. Following a single intraperitoneal injection of DCVC, no dichloroacetylated protein
adducts were detected in the epididymis and efferent ducts. The presence of CYP2E1 was found
in epididymis and efferent ducts, suggesting a role of cytochrome P450-dependent metabolism
in adduct formation. An in vitro assay was used to demonstrate that epididymal and efferent
duct microsomes are capable of metabolizing TCE; TCE metabolism in the efferent ducts was
found to be inhibited by anti-CYP2El antibody. Lipid peroxidation in sperm, presumably
initiated by free radicals, was increased in a significant (p < 0.005) dose-dependent manner after
TCE-exposure.
Overall, it has been suggested (DuTeaux et al., 2004a) that reproductive organ toxicities
observed following TCE exposure are initiated by metabolic bioactivation, leading to subsequent
protein adduct formation. It has been hypothesized that epoxide hydrolases in the rat epididymis
may play a role in the biological activation of metabolites (DuTeaux et al., 2004b). Disruption
of colony stimulating factor and of macrophage development may also play a role in sperm
production (Cohen et al., 1999), and thus may be another route through which immune-related
effects of TCE may operate. In addition, the potential for epigenetic changes, through which
heritable changes in gene mutations occur without changes in DNA sequencing, should also be
considered in the evaluation of transgenerational effects (Guerrero-Bosagna and Skinner, 2009).
4.8.1.1.12. Summary of noncancer reproductive toxicity
The toxicological database for TCE includes a number of studies that demonstrate
adverse effects on the integrity and function of the reproductive system in females and males.
Both the epidemiological and animal toxicology databases provide suggestive, but limited,
evidence of adverse outcomes to female reproductive outcomes. However, much more extensive
evidence exists in support of an association between TCE exposures and male reproductive
toxicity. The available epidemiological data and case reports that associate TCE with adverse
effects on male reproductive function are limited in size and provide little quantitative dose data
This document is a draft for review purposes only and does not constitute Agency policy.
560 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
(Lamb and Hentz, 2006). However, the animal data provide extensive evidence of TCE-related
male reproductive toxicity. Strengths of the database include the presence of both functional and
structural outcomes, similarities in adverse treatment-related effects observed in multiple species,
and evidence that metabolism of TCE in male reproductive tract tissues is associated with
adverse effects on sperm measures in both humans and animals (suggesting that the murine
model is appropriate for extrapolation to human health risk assessment). Additionally some
aspects of a putative MOA (e.g., perturbations in testosterone biosynthesis) appear to have some
commonalities between humans and animals.
4.8.2. Cancers of the Reproductive System
The effects of TCE on cancers of the reproductive system have been examined for males
and females in both epidemiological and experimental animal studies. The epidemiological
literature includes data on prostate in males and cancers of the breast and cervix in females. The
experimental animal literature includes data on prostate and testes in male rodents; and uterus,
ovary, mammary gland, vulva, and genital tract in female rodents. The evidence for these
cancers is generally not robust.
4.8.2.1.1. Human Data
The epidemiologic evidence on TCE and cancer of the prostate, breast, and cervix is from
cohort and geographic based studies. Two additional case-control studies of prostate cancer in
males are nested within cohorts (Greenland et al., 1994; Krishnadasan et al., 2007). The nested
case-control studies are identified in Tables 4-90 through 4-92 with cohort studies given their
source population for case and control identification. One population-based case-control study
examined on TCE exposure and prostate (Siemiatycki, 1991); however, no population case-
control studies on breast or cervical cancers and TCE exposure were found in the peer-reviewed
literature.
4.8.2.1.2. Prostate cancer
Sixteen cohort or PMR studies, two nested case-control, one population case-control, and
two geographic-based studies present relative risk estimates for prostate cancer (Anttila et al.,
1995; ATSDR, 2004a, 2006a; Axelson et al., 1994; Blair et al., 1989; Blair et al., 1998; Boice et
al., 1999; Boice et al., 2006a; Chang et al., 2003a; Chang et al., 2005; Garabrant et al., 1988;
This document is a draft for review purposes only and does not constitute Agency policy.
561 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
Greenland et al., 1994; Hansen et al., 2001; Krishnadasan et al., 2007; Morgan and Cassady,
2002; Morgan et al., 1998; Raaschou-Nielsen et al., 2003; Radican et al., 2008; Ritz, 1999a;
Shannon et al., 1988; Siemiatycki, 1991; Wilcosky et al., 1984). Three small cohort studies
(Costa et al., 1989; Henschler et al., 1995; Sinks et al., 1992), one multiple-site population case-
control (Siemiatycki, 1991) and one geographic-based study (Vartiainen et al., 1993) do not
report estimates for prostate cancer in their published papers. Twelve of the 19 studies with
prostate cancer relative risk estimates had high likelihood of TCE exposure in individual study
subjects and were judged to have met, to a sufficient degree, the standards of epidemiologic
design and analysis (Siemiatycki, 1991);
Table 4-90. Summary of human studies on TCE exposure and prostate
cancer
Studies
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Cohort studies—incidence
Aerospace workers (Rocketdyne)
Krishnadasan et al. (2007)
Low/moderate TCE score
1.3 (0.81, 2.1)ab
90
High TCE score
2.1 (1.2, 3.9fb
45
p for trend
0.02
Low/moderate TCE score
1.3 (0.81, 2.1)a'°
High TCE score
2.4 (1.3, 4.4)a'c
p for trend
0.01
All employees at electronics factory (Taiwan)
0.14(0.00, 0.76)d
1
Chang et al. (2005)
Danish blue-collar worker with TCE exposure
Raaschou-Nielsen et al.
(2003)
Any exposure
0.9 (0.79, 1.08)
163
Biologically-monitored Danish workers
Hansen etal. (2001)
Any TCE exposure, females
0.6 (0,2, 1.3)
6
Aircraft maintenance workers (Hill Air Force Base, UT)
TCE subcohort
Not reported
158
Blair etal. (1998)
Cumulative exposure
0
1.0e
<5 ppm-yr
1.1 (0.7, 1.6)
64
5-25 ppm-yr
1.0 (0.6, 1.6)
38
>25 ppm-yr
1.2 (0.8, 1.8)
56
TCE subcohort
1.2 (0.92, 1.76)
116
Radican et al. (2008)
Cumulative exposure
This document is a draft for review purposes only and does not constitute Agency policy.
562 DRAFT—DO NOT CITE OR QUOTE
-------
0
1.0e
<5 ppm-yr
1.03 (0.65, 1.62)
41
5-25 ppm-yr
1.33 (0.82,2.15)
42
>25 ppm-yr
1.31 (0.84,2.06)
43
Biologically-monitored Finnish workers
1.38 (0.73,2.35)
13
Anttila et al. (1995)
Mean air-TCE (Ikeda extrapolation
<6 ppm
1.43 (0.62, 2.82)
8
6+ppm
0.68 (0.08, 2.44)
2
This document is a draft for review purposes only and does not constitute Agency policy.
563 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-90. Summary of human studies on TCE exposure and prostate
cancer (continued)
Studies
Exposure group
Relative risk
(95% CI)
No. obs.
Events
Reference
Cardboard manufacturing workers in Arnsburg, Germany
Henschler et al. (1995)
Exposed workers
Not reported
Biologically-monitored Swedish workers
1.25 (0.84, 1.84)
26
Axelsonetal. (1994)
Cardboard manufacturing workers, Atlanta area, GA
Not reported
Sinks et al. (1992)
Cohort and PMR-mortality
Aerospace workers (Rocketdyne)
Boice et al. (2006a)
Any TCE (utility/eng flush)
0.82 (0.36, 1.62)
8
View-Master employees
1.69 (0.68, 3.48)f
8
ATSDR (2004a)
All employees at electronics factory (Taiwan)
Not reported
0
Chang et al. (2003a)
Fernald workers
Ritz (1999a)
Any TCE exposure
Not reported
Light TCE exposure, >2 yr duration
0.91 (0.38, 2.18)eg
10
Moderate TCE exposure, >2 yr duration
1.44(0.19, 114)es
1
Aerospace workers (Lockheed)
Boice et al. (1999)
Routine exposure to TCE
1.31 (0.52,2.69)
7
Routine -intermittent
Not reported
Aerospace workers (Hughes)
Morgan et al. (1998, 2000b)
TCE subcohort
1.18 (0.73, 1.80)
21
Low intensity (<50 ppm)
1.03 (0.51, 1.84)
7
High intensity (>50 ppm)
0.47(0.15, 1.11)
14
TCE subcohort (Cox Analysis)
Never exposed
1.00e
Ever exposed
1.58 (0.96, 2.62)h
Peak
No/Low
1.00e
Medium/high
1.39 (0.80, 2.41)h
Cumulative
Referent
1.00e
Low
1.72 (0.78, 3.80)h
High
1.53 (0.85, 2.75)h
This document is a draft for review purposes only and does not constitute Agency policy.
564 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-90. Summary of human studies on TCE exposure and prostate
cancer (continued)
Studies
Exposure group
Relative risk
(95% CI)
No. obs.
Events
Reference
Aircraft maintenance workers (Hill Air Force Base, UT)
Blair et al. (1998)
TCE subcohort
1.1 (0.6, 1.8)
54
Cumulative exposure
0
1.0e
<5 ppm-yr
0.9 (0.5, 1.8)
19
5-25 ppm-yr
1.0(0.5,2.1)
13
>25 ppm-yr
1.3 (0.7, 2.4)
22
Cardboard manufacturing workers in Arnsburg, Germany
Henschler et al. (1995)
TCE exposed workers
Not reported
Deaths reported to GE pension fund (Pittsfield, MA)
0.82 (0.46, 1.46)a
58
Greenland et al. (1994)
Cardboard manufacturing workers, Atlanta area, GA
Not reported
0
Sinks et al. (1992)
U. S. Coast Guard employee
Blair et al. (1989)
Marine inspectors
1.06 (0.51, 1.95)
10
Noninspectors
0.57 (0.15, 1.45)
7
Aircraft manufacturing plant employees (Italy)
Costa et al. (1989)
Aircraft manufacturing plant employees (San Diego,
CA)
0.93 (0.60, 1.37)
25
Garabrant et al. (1988)
Lamp manufacturing workers (GE)
1.56 (0.63, 3.22)
7
Shannon et al. (1988)
Rubber workers
Wilcosky et al. (1984)
Any TCE exposure
0.62 (not reported)
3
Case-control studies
Population of Montreal, Canada
Siemiatycki (1991)
Any TCE exposure
1.1 (0.6,2.1)'
11
Substantial TCE exposure
1.8 (0.8, 4.0)1
7
Geographic based studies
Residents in two study areas in Endicott, NY
1.05 (0.75, 1.43)
40
ATSDR (2006b)
Residents of 13 census tracts in Redlands, CA
1.11 (0.98, 1.25y
483
Morgan and Cassady (2002)
Finnish residents
Vartiainen et al. (1993)
Residents of Hausjarvi
Not reported
Residents of Huttula
Not reported
1
2 aOdds ratio from nested case-control study.
3 bOdds ratio, zero lag.
This document is a draft for review purposes only and does not constitute Agency policy.
565 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Table 4-90. Summary of human studies on TCE exposure and prostate
cancer (continued)
°Odds ratio, 20 yr lag.
dChang et al. (2005) presents SIRs for a category site of all cancers of male genital organs.
"Internal referents, workers without TCE exposure.
Proportional mortality ratio.
8Analysis for >2 yrs exposure duration and a lagged TCE exposure period of 15 yrs.
hRisk ratio from Cox Proportional Hazard Analysis, stratified by age and sex, from Environmental Health Strategies
(1997) Final Report to Hughes Corporation (Communication from Paul A. Cammer, President, Trichloroethylene
Issues Group to Cheryl Siegel Scott, U.S. EPA, December 22, 1997).
'90% CI.
J99% CI.
GE = General Electric, No. obs. events = number of observed events.
This document is a draft for review purposes only and does not constitute Agency policy.
566 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-91. Summary of human studies on TCE exposure and breast cancer
2
Studies
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Cohort studies—incidence
Aerospace workers (Rocketdyne)
Zhao et al. (2005)
Any TCE exposure
Not reported
Low cumulative TCE score
Medium cumulative TCE score
High TCE score
p for trend
All employees at electronics factory (Taiwan)
Females
1.09 (0.96, 1.22)a
286
Sung et al. (2007)
Females
1.19 (1.03, 1.36)
215
Chang et al. (2005)
Danish blue-collar worker with TCE exposure
Raaschou-Nielsen et al.
(2003)
Any exposure, males
0.5 (0.06, 1.90)
2
Any exposure, females
1.1 (0.89, 1.24)
145
Biologically-monitored Danish workers
Hansen etal. (2001)
Any TCE exposure, males
0
(0.2 exp)
Any TCE exposure, females
0.9 (0.2, 2.3)
4
Aircraft maintenance workers (Hill Air Force Base, UT)
Blair et al. (1998)
TCE subcohort
Not reported
34
Females, cumulative exposure
0
1.0b
<5 ppm-yr
0.3 (0.1,1.4)
20
5-25 ppm-yr
0.4 (0.1,2.9)
11
>25 ppm-yr
0.4 (0.4, 1.2)
3
Biologically-monitored Finnish workers
Not reported
Anttila et al. (1995)
Cardboard manufacturing workers in Arnsburg, Germany
Henschler et al. (1995)
Exposed workers
Not reported
Biologically-monitored Swedish workers
Not reported
Axelsonetal. (1994)
Cardboard manufacturing workers, Atlanta area, GA
Not reported
Sinks et al. (1992)
3
This document is a draft for review purposes only and does not constitute Agency policy.
567 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-91. Summary of human studies on TCE exposure and breast cancer
(continued)
Studies
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Cohort and PMR-mortality
Aerospace workers (Rocketdyne)
Any TCE (utility/eng flush)
Not reported
Boice et al. (2006a)
Any exposure to TCE
Not reported
Zhao et al. (2005)
Low cumulative TCE score
Not reported
Medium cumulative TCE score
Not reported
High TCE score
Not reported
p for trend
View-Master employees
ATSDR (2004a)
Males
0
(0.05 exp)
Females
1.02 (0.67, 1.49)°
27
Fernald workers
Ritz (1999a)
Any TCE exposure
Not reported
Light TCE exposure, >2 yr duration
Not reported
Moderate TCE exposure, >2 yr duration
Not reported
Aerospace workers (Lockheed)
Boice et al. (1999)
Routine exposure to TCE
1.31 (0.52, 2.69)d
7
Routine -intermittent3
Not reported
Aerospace workers (Hughes)
Morgan et al. (1998)
TCE subcohort
0.75 (0.43, 1.22)d
16
Low intensity (<50 ppm)
1.03 (0.51, 1.84)d
11
High intensity (>50 ppm)
0.47 (0.15, l.ll)d
5
TCE subcohort (Cox Analysis)
Never exposed
1.00d
NR
Ever exposed
0.94 (0.51, lJS)46
NR
Peak
No/Low
1.00d
Medium/high
1.14(0.48, 2.70)^
NR
Cumulative
Referent
1.00b
Low
1.20 (0.60, 2.40)^
NR
High
0.65 (0.25, l^)46
NR
This document is a draft for review purposes only and does not constitute Agency policy.
568 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-91. Summary of human studies on TCE exposure and breast cancer
(continued)
Studies
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Aircraft maintenance workers (Hill Air Force Base, UT)
TCE subcohort (females)
2.0 (0.9, 4.6)
20
Blair et al. (1998)
Females, cumulative exposure
0
1.0b
<5 ppm-yr
2.4 (1.1, 5.2)
10
5-25 ppm-yr
1.2 (0.3, 5.4)
21
>25 ppm-yr
1.4 (0.6, 3.2)
8
Low level intermittent exposure
3.1 (1.5,6.2)
15
Low level continuous exposure
3.4 (1.4, 8.0)
8
Frequent peaks
1.4 (0.7, 3.2)
10
TCE subcohort (females)
1.23 (0.73, 2.06)
26
Radican et al. (2008)
Females, cumulative exposure
0
1.0b
<5 ppm-yr
1.57 (0.81,3.04)
12
5-25 ppm-yr
1.01 (0.31, 3.30)
3
>25 ppm-yr
1.05 (0.53, 2.07)
11
Low level intermittent exposure
1.92 (1.08,3.43)
18
Low level continuous exposure
1.71 (0.79, 3.71)
8
Frequent peaks
1.08 (0.57, 2.02)
14
Cardboard manufacturing workers in Arnsburg, Germany
Henschler et al. (1995)
TCE exposed workers
Not examined
Deaths reported to GE pension fund (Pittsfield, MA)
Not reported
Greenland et al. (1994)
Cardboard manufacturing workers, Atlanta area, GA
Not reported
0
Sinks et al. (1992)
U. S. Coast Guard employees
Blair et al. (1989)
Marine inspectors
Not reported
Noninspectors
Not reported
Aircraft manufacturing plant employees (Italy)
Not reportedf
Costa et al. (1989)
Aircraft manufacturing plant employees (San Diego, CA)
Garabrant et al. (1988)
All subjects, females
0.81 (0.52, 1.48)d
16
Lamp manufacturing workers (GE)
Shannon et al. (1988)
Coil/wire drawing
2.04 (0.88, 4.02)
8
Other areas
0.97 (0.57, 1.66)
13
This document is a draft for review purposes only and does not constitute Agency policy.
569 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
Table 4-91. Summary of human studies on TCE exposure and breast cancer
(continued)
Studies
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Case-control Studies
Population of Montreal, Canada
Siemiatycki (1991)
Any TCE exposure
Not reported
Substantial TCE exposure
Not reported
Geographic Based Studies
Residents in two study areas in Endicott, NY
0.88 (0.65, 1.18)
46
ATSDR (2006b)
Residents of 13 census tracts in Redlands, CA
1.09(0.97, 1.21)
536
Morgan and Cassady (2002)
Finnish residents
Vartiainen et al. (1993)
Residents of Hausjarvi
Not reported
Residents of Huttula
Not reported
a15yrlag.
internal referents, workers not exposed to TCE.
°Proportional mortality ratio.
dIn Garabrant et al. (1988), Morgan et al. (1998) and Boice et al. (1999), breast cancer risk is for males and females
combined (ICD-9, 174, 175).
eRisk ratio from Cox Proportional Hazard Analysis, stratified by age and sex, from Environmental Health Strategies
(1997) Final Report to Hughes Corporation (Communication from Paul A. Cammer, President, Trichloroethylene
Issues Group to Cheryl Siegel Scott, U.S. EPA, December 22, 1997).
fThe cohort of Blair et al. (1989) and Costa et al. (1989) are composed of males only.
GE = General Electric, No. obs. events = number of observed events; NR = not reported.
This document is a draft for review purposes only and does not constitute Agency policy.
570 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-92. Summary of human studies on TCE exposure and cervical
2 cancer
3
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Cohort studies—incidence
Aerospace workers (Rocketdyne)
Zhao et al. (2005)
Any exposure to TCE
Not reported
Low cumulative TCE score
Not reported
Medium cumulative TCE score
High TCE score
p for trend
All employees at electronics factory (Taiwan)
0.96 (0.86, 1.22)a
337
Sung et al. (2007)
Danish blue-collar worker w/TCE exposure
Raaschou-Nielsen et al.
(2003)
Any exposure
1.9 (1.42,2.37)
62
Exposure lag time
20 yr
1.5 (0.7, 2.9)
9
Employment duration
<1 yr
2.5 (1.7, 3.5)
30
1-4.9 yr
1.6 (1.0,2.4)
22
>5 yr
1.3 (0.6, 2.4)
10
Biologically-monitored Danish workers
Hansen etal. (2001)
Any TCE exposure
3.8 (1.0, 9.8)
4
Cumulative exposure (Ikeda)
<17 ppm-yr
2.9 (0.04, 16)
1
>17 ppm-yr
2.6 (0.03, 14)
1
Mean concentration (Ikeda)
<4 ppm
3.4 (0.4, 12)
2
4+ ppm
4.3 (0.5, 16)
2
Employment duration
<6.25 yr
3.8(0.1,21)
1
>6.25 yr
2.1 (0.03, 12)
1
Aircraft maintenance workers from Hill Air Force Base, UT
Blair et al. (1998)
TCE subcohort
Not reported
Cumulative exposure
Not reported
4
This document is a draft for review purposes only and does not constitute Agency policy.
571 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-92. Summary of human studies on TCE exposure and cervical
cancer (continued)
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Biologically-monitored Finnish workers
Anttila et al. (1995)
All subjects
2.42 (1.05, 4.77)
8
Mean air-TCE (Ikeda extrapolation)
<6 ppm
1.86 (0.38, 5.45)
3
6+ ppm
4.35 (1.41, 10.1)
5
Cardboard manufacturing workers in Arnsburg, Germany
Henschler et al. (1995)
Exposed workers
Not reported
Biologically-monitored Swedish workers
Axelsonetal. (1994)
Any TCE exposure
Not reported
Cardboard manufacturing workers, Atlanta area, GA
Sinks et al. (1992)
All subjects
Not reported
Cohort studies-mortality
Aerospace workers (Rocketdyne)
Any TCE (utility/eng flush)
Not reported
Boice et al. (2006a)
Any exposure to TCE
Not reported
Zhao et al. (2005)
View-Master employees
ATSDR (2004a)
Females
1.77 (0.57, 4.12)b
5
United States uranium-processing workers (Fernald, OH)
Ritz (1999a)
Any TCE exposure
Not reported
Light TCE exposure, >2 yr duration
Not reported
Moderate TCE exposure, >2 yr duration
Not reported
Aerospace workers (Lockheed)
Boice et al. (1999)
Routine exposure
- (0.00, 5.47)
0
Routine -intermittent
Not reported
Aerospace workers (Hughes)
Morgan et al. (1998)
TCE subcohort
(0.00, 1.07)
0
(3.5 exp)
Low intensity (<50 ppm)
0
(1.91 exp)
High intensity (>50 ppm)
0
(1.54 exp)
This document is a draft for review purposes only and does not constitute Agency policy.
572 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-92. Summary of human studies on TCE exposure and cervical
cancer (continued)
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Aircraft maintenance workers (Hill AFB, UT)
TCE subcohort
1.8 (0.5, 6.5)°
5
Blair et al. (1998)
Cumulative exposure
0
1.0C
<5 ppm-yr
0.9 (0.1, 8.3)
1
5-25 ppm-yr
0
>25 ppm-yr
3.0(0.8, 11.7)
4
TCE subcohort
1.67 (0.54, 5.22)
6
Radican et al. (2008)
Cumulative exposure
0
1.0°
<5 ppm-yr
0.76 (0.09, 6.35)
1
5-25 ppm-yr
0
>25 ppm-yr
2.83 (0.86, 9.33)
5
Cardboard manufacturing workers in Arnsburg, Germany
Henschler et al. (1995)
TCE exposed workers
Not reported
Unexposed workers
Not reported
Deaths reported to GE pension fund (Pittsfield, MA)
Not examinedd
Greenland et al. (1994)
Cardboard manufacturing workers, Atlanta area, GA
Not reported
Sinks et al. (1992)
U. S. Coast Guard employees
Not reported6
Blair et al. (1989)
Aircraft manufacturing plant employees (Italy)
Not reported6
Costa et al. (1989)
Aircraft manufacturing plant employees (San Diego, CA)
Garabrant et al. (1988)
All subjects
0.61 (0.25, 1.26)f
7
Lamp manufacturing workers (GE)
Shannon et al. (1988)
Coil/wire drawing
1.05 (0.03, 5.86)
1
Other areas
1.16 (0.32,2.97)
4
Case-control studies
Geographic based studies
Residents in two study areas in Endicott, NY
1.06 (0.29, 2.71)
<6
ATSDR (2006b)
Residents in Texas
Coyle et al. (2005)
Counties reporting any air TCE release
66.4s
Countries not reporting any air TCE release
60.8s
Residents of 13 census tracts in Redlands, CA
0.65 (0.38, 1.02)
29
Morgan and Cassady (2002)
This document is a draft for review purposes only and does not constitute Agency policy.
573 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-92. Summary of human studies on TCE exposure and cervical
cancer (continued)
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Finnish residents
Vartiainen et al. (1993)
Residents of Hausjarvi
Not reported
Residents of Huttula
Not reported
1
2 Standardized incidence ratio for females in Sung et al. (2007) reflects a 15-yr lag period.
3 bProportional mortality ratio.
4 Internal referents, workers not exposed to TCE.
5 dNested case-control analysis.
6 "Males only in cohort.
7 fSMR is for cancer of the genital organs (cervix, uterus, endometrium, etc.).
8 8 Median annual age-adjusted breast cancer rate (1995-2000).
9
10 GE = General Electric, No. obs. events = number of observed events.
11
12
13 (Anttila et al., 1995; Axelson et al., 1994; Blair et al., 1998; Boice et al., 1999; Boice et al.,
14 2006a; Greenland et al., 1994; Hansen et al., 2001; Krishnadasan et al., 2007; Morgan et al.,
15 1998, 2000b; Raaschou-Nielsen et al., 2003; Radican et al., 2008). Krishnadasan et al. (2007) in
16 their nested case-control study of prostate cancer observed a twofold odds ratio estimate with
17 high cumulative TCE exposure score (2.4, 95% CI: 1.3, 4.4, 20 year lagged exposure) and an
18 increasing positive relationship between prostate cancer incidence and TCE cumulative exposure
19 score (p = 0.02). TCE exposure was positively correlated with several other occupational
20 exposures, and Krishnadasan et al. (2007) adjusted for possible confounding from all other
21 chemical exposures as well as age at diagnosis, occupational physical activity, and socio-
22 economic status in statistical analyses. Relative risk estimates in studies other than Krishnadasan
23 et al. (2007) were above 1.0 for overall TCE exposure (1.8, 95% CI: 0.8, 4.0 (Siemiatycki,
24 1991); 1.1, 95% CI: 0.6, 1.8 (Blair et al., 1998) and 1.20, 95% CI: 0.92, 1.76, with an additional
25 10-year follow-up (Radican et al., 2008); 1.58, 95% CI: 0.96, 2.62 (EHS, 1997; Morgan et al.,
26 1998, 2000b); 1.3, 95% CI: 0.52, 2.69 (Boice et al., 1999); 1.38, 95% CI: 0.73, 2.35 (Anttila et
27 al., 1995)) and prostate cancer risks did not appear to increase with increasing exposure. Four
28 studies observed relative risk estimates below 1.0 for overall TCE exposure (0.93, 95% CI: 0.60,
29 1.37 (Garabrant et al., 1988); 0.6, 95% CI: 0.2, 1.30 (Hansen et al., 2001); 0.9, 95% CI: 0.79,
30 1.08 (Raaschou-Nielsen et al., 2003); 0.82, 95% CI: 0.36, 1.62 (Boice et al., 2006a)), and are not
31 considered inconsistent because alternative explanations are possible and included observations
32 are based on few subjects, lowering statistical power, or to poorer exposure assessment
33 approaches that may result in a higher likelihood of exposure misclassification.
This document is a draft for review purposes only and does not constitute Agency policy.
574 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Seven other cohort, PMR, and geographic based studies were given less weight in the
analysis because of their lesser likelihood of TCE exposure and other study design limitations
that would decrease statistical power and study sensitivity (ATSDR, 2004a, 2006a; Blair et al.,
1989; Chang et al., 2005; Morgan and Cassady, 2002; Shannon et al., 1988; Wilcosky et al.,
1984). Chang et al. (2005) observed a statistically significant deficit in prostate cancer risk,
based on one case, and an insensitive exposure assessment (0.14, 95% CI: 0.00, 0.76). Relative
risks in the other five studies ranged from 0.62 (CI not presented in paper) (Wilcosky et al.,
1984) to 1.11 (95% CI: 0.98, 1.25) (Morgan and Cassady, 2002).
Risk factors for prostate cancer include age, family history of prostate cancer, and
ethnicity as causal with inadequate evidence for a relationship with smoking or alcohol (Wigle et
al., 2008). All studies except Krishnadasan et al. (2007) were not able to adjust for possible
confounding from other chemical exposures in the work environment. None of the studies
including Krishnadasan et al. (2007) accounted for other well-established nonoccupational risk
factors for prostate cancer such as race, prostate cancer screening and family history. There is
limited evidence that physical activity may provide a protective effect for prostate cancer (Wigle
et al., 2008). Krishnadasan et al. (2008) examined the effect of physical activity in the
Rocketdyne aerospace cohort (Krishnadasan et al., 2007; Zhao et al., 2005). Their finding of a
protective effect with high physical activity (0.55, 95% CI: 0.32, 0.95, p trend = 0.04) after
control for TCE exposure provides additional evidence (Krishnadasan et al., 2008) and suggests
underlying risk may be obscured in studies lacking adjustment for physical activity.
4.8.2.1.3. Breast cancer
Fifteen studies of TCE exposure reported findings on breast cancer in males and females
combined (Boice et al., 1999; Garabrant et al., 1988; Greenland et al., 1994), in males and
females, separately (ATSDR, 2004a; Clapp and Hoffman, 2008; Hansen et al., 2001; Raaschou-
Nielsen et al., 2003), or in females only (ATSDR, 2006a; Blair et al., 1998; Chang et al., 2005;
Coyle et al., 2005; Morgan et al., 1998; Radican et al., 2008; Shannon et al., 1988; Sung et al.,
2007). Six studies have high likelihood of TCE exposure in individual study subjects and met, to
a sufficient degree, the standards of epidemiologic design and analysis (Blair et al., 1998; Boice
et al., 1999; Hansen et al., 2001; Morgan et al., 1998; Raaschou-Nielsen et al., 2003; Radican et
al., 2008). Four studies with risk estimates for other cancer sites do not report risk estimates for
breast cancer (Anttila et al., 1995; Axelson et al., 1994; Boice et al., 2006a; Siemiatycki, 1991).
No case-control studies were found on TCE exposure, although several studies examine
occupational title or organic solvent as a class (Band et al., 2000; Ji et al., 2008; Rennix et al.,
2005; Weiderpass et al., 1999). While association is seen with occupational title or industry and
This document is a draft for review purposes only and does not constitute Agency policy.
575 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
breast cancer (employment in aircraft and aircraft part industry, 2.48, 95% CI: 1.14, 5.39 (Band
et al., 2000); solvent user: 1.48, 95% CI: 1.03, 2.12 (Rennix et al., 2005)), TCE exposure is not
uniquely identified. The two studies suggest association between organic solvents and female
breast cancer needs further investigation of possible risk factors.
Relative risk estimates in the five studies in which there is a high likelihood of TCE
exposure in individual study subjects and which met, to a sufficient degree, the standards of
epidemiologic design and analysis in a systematic review ranged from 0.75 (0.43, 1.22) (females
and males; (Morgan et al., 1998)) to 2.0 (0.9, 4.6) (mortality in females; (Blair et al., 1998)).
Blair et al. (1998), additionally, observed stronger risk estimates for breast cancer mortality
among females with low level intermittent (3.1, 95% CI: 1.5, 6.2) and low level continuous (3.4,
95% CI: 1.4, 8.0) TCE exposures, but not with frequent peaks (1.4, 95% CI: 0.7, 3.2). A similar
pattern of risks was also observed by Radican et al. (2008) who studied mortality in this cohort
and adding 10 years of follow-up, although the magnitude of breast cancer risk in females was
lower than that observed in Blair et al. (1998). Risk estimates did not appear to increase with
increasing cumulative exposure in the two studies that included exposure-response analyses
(Blair et al., 1998; Morgan et al., 1998). None of these five studies reported a statistically
significant deficit in breast cancer and confidence intervals on relative risks estimates included
1.0 (no risk). Few female subjects in these studies appear to have high TCE exposure. For
example, Blair et al. (1998) identified 8 of the 28 breast cancer deaths and 3 of the 34 breast
cancer cases with high cumulative exposure.
Relative risk estimates in six studies of lower likelihood TCE exposure and other design
deficiencies ranged from 0.81 (95% CI: 0.52, 1.48) (Garabrant et al., 1988) to 1.19 (1.03, 1.36)
(Chang et al., 2005). These studies lack a quantitative surrogate for TCE exposure to individual
subjects and instead classify all subjects as "potentially exposed," with resulting large dilution of
actual risk and decreased sensitivity (ATSDR, 2006a; Chang et al., 2005; Garabrant et al., 1988;
Morgan and Cassady, 2002; NRC, 2006; Shannon et al., 1988; Sung et al., 2007).
Four studies reported on male breast cancer separately (ATSDR, 2004a; Clapp and
Hoffman, 2008; Hansen et al., 2001; Raaschou-Nielsen et al., 2003) and a total of three cases
were observed. Breast cancer in men is a rare disease and is best studied using a case-control
approach (Weiss et al., 2005). Reports exist of male breast cancer among former residents of
Camp Lejuene (ATSDR, 2007, 2010). Further assessment of TCE exposure and male breast
cancer is warranted.
Overall, the epidemiologic studies on TCE exposure and breast cancer are quite limited in
statistical power; observations are based on few breast cancer cases or on inferior TCE exposure
assessment in studies with large numbers of observed cases. Additionally, adjustment for
nonoccupational breast cancer risk factors is less likely in cohort and geographic based studies
This document is a draft for review purposes only and does not constitute Agency policy.
576 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
given their use of employment and public records. Breast cancer mortality observations in Blair
et al. (1998) and further follow-up of this cohort by Radican et al. (2008) of an elevated risk with
overall TCE exposure, particularly low level intermittent and continuous TCE exposure, provide
evidence of an association with TCE. No other study with high likelihood of TCE exposure in
individual study subjects reported a statistically significant association with breast cancer,
although few observed cases leading to lower statistical power or examination of risk for males
and females combined are alternative explanations for the null observations in these studies.
Both Chang et al. (2005) and Sung et al. (2007), two overlapping studies of female electronics
workers exposed to TCE, perchloroethylene, and mixed solvents, reported association with
breast cancer incidence, with breast cancer risk in Chang et al. (2005) appearing to increase with
employment duration. Both studies, in addition to association provided by studies of exposure to
broader category of organic solvents (Band et al., 2000; Rennix et al., 2005), support Blair et al.
(1998) and Radican et al. (2008), although the lack of exposure assessment is an uncertainty.
The epidemiologic evidence is limited for examining TCE and breast cancer, and while these
studies do not provide any strong evidence for association with TCE exposure they in turn do not
provide evidence of an absence of association.
4.8.2.1.4. Cervical cancer
Eleven cohort or PMR studies and 2 geographic based studies present relative risk
estimates (Anttila et al., 1995; ATSDR, 2004a, 2006a; Blair et al., 1998; Boice et al., 1999;
Garabrant et al., 1988; Hansen et al., 2001; Morgan and Cassady, 2002; Morgan et al., 1998;
Raaschou-Nielsen et al., 2003; Radican et al., 2008; Shannon et al., 1988; Sung et al., 2007).
Seven of these studies had high likelihood of TCE exposure in individual study subjects and
were judged to have met, to a sufficient degree, the standards of epidemiologic design and
analysis (Anttila et al., 1995; Blair et al., 1998; Boice et al., 1999; Hansen et al., 2001; Morgan et
al., 1998; Raaschou-Nielsen et al., 2003; Radican et al., 2008). Three small cohort studies (Costa
et al., 1989; Henschler et al., 1995; Sinks et al., 1992) as well as three studies with high
likelihood of TCE exposure in individual study subjects (Axelson et al., 1994; Boice et al.,
2006a; Zhao et al., 2005) did not present relative risk estimates for cervical cancer. Additionally,
one population case-control and one geographic study of several site-specific cancers do not
present information on cervical cancer (Siemiatycki, 1991; Vartiainen et al., 1993).
Five studies with high likelihood of TCE exposure in individual study subjects and which
met, to a sufficient degree, the standards of epidemiologic design and analysis in a systematic
review observed elevated risk for cervical cancer and overall TCE exposure (2.42, 95% CI: 1.05,
This document is a draft for review purposes only and does not constitute Agency policy.
577 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
4.77 (Anttila et al., 1995); 1.8, 95% CI: 0.5, 6.5 (Blair et al., 1998) that changed little with an
additional 10 years follow-up, 1.67, 95% CI: 0.54, 5.22 (Radican et al., 2008); 3.8, 95% CI: 1.42,
2.37 (Hansen et al., 2001); 1.9, 95% CI: 1.42, 2.37 (Raaschou-Nielsen et al., 2003). The
observations of a three- to fourfold elevated cervical cancer risk with high mean TCE exposure
compared to subjects in the low exposure category (6+ ppm: 4.35, 95% CI: 1.41, 10.1 (Anttila et
al., 1995); 4+ ppm: 4.3, 95% CI: 0.5, 16 (Hansen et al., 2001)) or with high cumulative TCE
exposure (0.25-ppm year: 3.0, 95% CI: 0.8, 11.7 (Blair et al., 1998), 2.83, 95% CI: 0.86, 9.33
(Radican et al., 2008)) provides additional support for association with TCE. Cervical cancer
risk was lowest for subjects in the high exposure duration category (Hansen et al., 2001;
Raaschou-Nielsen et al., 2003); however, duration of employment is a poor exposure metric
given subjects may have differing exposure intensity with similar exposure duration (NRC,
2006). No deaths due to cervical cancer were observed in two other studies (Boice et al., 1999;
Morgan et al., 1998), less than 4 deaths were expected, suggesting these cohorts contained few
female subjects with TCE exposure.
Human papilloma virus and low socioeconomic status are known risk factors for cervical
cancer (ACS, 2008). Subjects in Raaschou-Nielsen et al. (2003) are blue-collar workers and low
socioeconomic status likely explains observed associations in this and the other studies. The use
of internal controls in Blair et al. (1998) who are similar in socioeconomic status as TCE subjects
is believed to partly account for possible confounder related to socio-economic status; however,
direct information on individual subjects is lacking.
Six other cohort, PMR, and geographic based studies were given less weight in the
analysis because of their lesser likelihood of TCE exposure and other study design limitations
that would decrease statistical power and study sensitivity (ATSDR, 2004a, 2006a; Garabrant et
al., 1988; Morgan and Cassady, 2002; Shannon et al., 1988; Sung et al., 2007). Cervical cancer
risk estimates in these studies ranged between 0.65 (95% CI: 0.38, 1.02) (Morgan and Cassady,
2002) to 1.77 (proportional mortality ratio; 95% CI: 0.57, 4.12 (ATSDR, 2004a)). No study
reported a statistically significant deficit in cervical cancer risk.
4.8.2.1.5. Animal Studies
Histopathology findings have been noted in reproductive organs in various cancer
bioassay studies conducted with TCE. A number of these findings (summarized in Table 4-93)
do not demonstrate a treatment-related profile.
This document is a draft for review purposes only and does not constitute Agency policy.
578 DRAFT—DO NOT CITE OR QUOTE
-------
1 Cancers of the reproductive system that are associated with TCE exposure and observed
2 in animal studies are comprised of testicular tumors (interstitial cell and Leydig cell). A
3 summary of the incidences of testicular tumors observed in male rats is presented in Table 4-94.
4
This document is a draft for review purposes only and does not constitute Agency policy.
579 DRAFT—DO NOT CITE OR QUOTE
-------
1
Table 4-93. Histopathology findings in reproductive organs
Tumor incidence in mice after 18 mo inhalation exposure"
Tissue
Finding
Control
100 ppm
500 ppm
Males
Number examined:
30
29
30
Prostate
Myoma
1
0
0
Testis
Carcinoma
0
0
1
Cyst
0
0
1
Females
No. examined:
29
30
28
Uterus
Adenocarcinoma
1
0
0
Ovary
Adenocarcinoma
1
0
0
Adenoma
3
1
3
Carcinoma
0
2
2
Granulosa cell tumor
4
0
2
Tumor incidence in rats after 18 mo inhalation exposure"
Tissue
Finding
Control
100 ppm
500 ppm
Males
Number examined:
29
30
30
Testis
Interstitial cell tumors
4
0
3
Females
No. examined:
28
30
30
Mammary
Fibroadenoma
2
0
0
Adenocarcinoma
3
2
2
Uterus
Adenocarcinoma
3
1
4
Ovary
Carcinoma
4
0
1
Granulosa cell tumor
1
0
0
Genital tract
Squamous cell carcinoma
0
2
0
Tumor incidence in hamsters after 18 mo inhalation exposure3
Tissue
Finding
Control
100 ppm
500 ppm
Females
Number examined:
30
29
30
Ovary
Cystadenoma
1
0
0
Tumor incidence in mice after 18 mo gavage administration1"
Tissue
Finding
Con-
trol
TCE
Pure
TCE
Industrial
TCE +
EPC
TCE
+ BO
TCE + EPC
+ BO
Females
Number examined:
50
50
50
50
48
50
Mammary
Carcinoma
1
2
0
0
0
0
Ovary
Granulosa cell tumor
0
1
0
0
0
0
Vulva
Squamous cell carcinoma
0
0
0
0
1
1
2 aHenschler et al. (1980).
3 bHenschler et al. (1984); EPC = epichlorohydrin; BO = 1,2-epoxybutane.
This document is a draft for review purposes only and does not constitute Agency policy.
580 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-94. Testicular tumors in male rats exposed to TCE, adjusted for
2 reduced survival11
3
Interstitial cell tumors after 103 wk gavage exposure, beginning at 6.5-8 wk of age (NTP,
1988,1990)
Administered dose (mg/kg-
day)
Untreated
control
Vehicle
control
500
1,000
Male ACI rats
38/45 (84%)
36/44 (82%)
23/26 (88%)
17/19 (89%)
Male August rats
36/46 (78%)
34/46 (74%)
30/34 (88%)
26/30 (87%)
Male Marshall ratsb
16/46 (35%)
17/46 (37%)
21/33 (64%)
32/39 (82%)
Male Osborne-Mendel rats
1/30 (3%)
0/28 (0%)
0/25 (0%)
1/19 (5%)
Male F344/N rats
44/47 (94%)
47/48 (98%)
47/48 (98%)
32/44 (73%)
Leydig cell tumors after 104 wk inhalation exposure, beginning at 12 wk of age (Maltoni et
al., 1986)
Administered daily
concentration (mg/m3)°
Control
112.5
337.5
675
Male Sprague-Dawley ratsb
6/114(5%)
16/105 (15%)
30/107 (28%)
31/113 (27%)
4
5 aACI rats alive at Week 70, August rats at Week 65, Marshall rats at Week 32, Osborne-Mendel rats at Week 97,
6 F344/N rats at Week 32, Sprague-Dawley rats at Week 81 (except BT304) or Week 62 (except BT304 bis).
7 Equivalent to 100, 300, or 600 ppm (100 ppm = 540 mg/m3), adjusted for 7 h/d, 5 d/wk exposure.
8 Statistically significant by Cochran-Armitage trend test (p < 0.05).
9
10 Sources: NTP (1988) Tables A2, C2, E2, G2; NTP (1990) Table A3; Maltoni et al. (1986) IV/IV Table 21, IV/V
11 Table 21.
12
13
4.8.2.1.6. Mode of Action for Testicular Tumors
14 The database for TCE does not include an extensive characterization of the mode of
15 action for Leydig cell tumorigenesis in the rat, although data exist that are suggestive of
16 hormonal disruption in male rats. A study by Kumar et al. (2000a) found significant decreases in
17 serum testosterone concentration and in 17-P-HSD, G6PDH, and total cholesterol and ascorbic
18 acid levels in testicular homogenate from male rats that had been exposed via inhalation to
19 376 ppm TCE for 12 or 24 weeks. In a follow-up study, Kumar et al. (2001b) also identified
20 decreases in SDH in the testes of TCE-treated rats. These changes are markers of disruption to
21 testosterone biosynthesis. Evidence of testicular atrophy, observed in the 2001 study by
22 Kumar et al., as well as the multiple in vivo and in vitro studies that observed alterations in
23 spermatogenesis and/or sperm function, could also be consistent with alterations in testosterone
This document is a draft for review purposes only and does not constitute Agency policy.
5 81 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
levels. Therefore, while the available data are suggestive of a MOA involving hormonal
disruption for TCE-induced testicular tumors, the evidence is inadequate to specify and test a
hypothesized sequence of key events.
Leydig cell tumors can be chemically induced by alterations of steroid hormone levels,
through mechanisms such as agonism of estrogen, gonadotropin releasing hormone, or dopamine
receptors; antagonism of androgen receptors; and inhibition of 5a-reductase, testosterone
biosynthesis, or aromatase (Cook et al., 1999). For those plausible mechanisms that involve
disruption of the hypothalamic-pituitary-testis (HPT) axis, decreased testosterone or estradiol
levels or recognition is involved, and increased luteinizing hormone (LH) levels are commonly
observed. Although there is evidence to suggest that humans are quantitatively less sensitive
than rats in their proliferative response to LH, evidence of treatment-related Leydig cell tumors
in rats that are induced via HPT disruption is considered to represent a potential risk to humans
(with the possible exception of GnRh or dopamine agonists), since the pathways for regulation of
the HPT axis are similar in rats and humans (Clegg et al., 1997).
4.8.3. Developmental Toxicity
An evaluation of the human and experimental animal data for developmental toxicity,
considering the overall weight and strength of the evidence, suggests a potential for adverse
outcomes associated with pre- and/or postnatal TCE exposures.
4.8.3.1.1. Human Developmental Data
Epidemiological developmental studies (summarized in Table 4-95) examined the
relationship between TCE exposure and prenatal developmental outcomes including spontaneous
abortion and perinatal death; decreased birth weight, small for gestational age, and postnatal
growth; congenital malformations; and other adverse birth outcomes. Postnatal developmental
outcomes examined include developmental neurotoxicity, developmental immunotoxicity, other
developmental outcomes, and childhood cancer.
This document is a draft for review purposes only and does not constitute Agency policy.
582 DRAFT—DO NOT CITE OR QUOTE
-------
4.8.3.1.2. Adverse fetal/birth outcomes
4.8.3.1.2.1.Spontaneous abortion and perinatal death
1 Spontaneous abortion or miscarriage is defined as nonmedically induced premature
2 delivery of a fetus prior to 20 weeks gestation. Perinatal death is defined as stillbirths and deaths
3 before 7 days after birth. Available data comes from several studies of occupational exposures in
4 Finland and Santa Clara, California, and by geographic-based studies in areas with known
5 contamination of water supplies in Woburn, MA; Tucson Valley, AZ; Rocky Mountain Arsenal,
6 CO; Endicott, NY; and New Jersey.
7
This document is a draft for review purposes only and does not constitute Agency policy.
583 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-95. Developmental studies in humans
Subjects
Exposure
Effect
Reference
Adverse fetal/birth outcomes
Spontaneous abortion and perinatal death
371 men occupationally exposed to
solvents in Finland 1973-1983
Questionnaire
Low/rare used <1 d/wk;
Intermediate used 1-4 d/wk or
intermediate/low TCA urine
levels;
High/frequent used daily or high
TCA urine levels
No risk of spontaneous abortion after paternal exposure,
based on 17 cases and 35 controls exposed to TCE
OR: 1.0, 95% CI: 0.6-2.0
Taskinen et al.
(1989)
535 women occupationally exposed to
solvents in Finland 1973-1986
Questionnaire
Rare used 1-2 d/wk;
Frequent used >3 d/wk
Increased risk of spontaneous abortion among frequently-
exposed women, based on 7 cases and 9 controls exposed to
TCE
OR: 1.6, 95% CI: 0.5-4.8
Taskinen et al.
(1994)
3,265 women occupationally exposed
to organic solvents in Finland
1973-1983
Questionnaire
U-TCA: median: 48.1 |imol/L: mean
96.2 ± 19.2 ^mol/L
No increased risk of spontaneous abortion based on 3 cases
and 13 controls exposed to TCE
OR: 0.6, 95% CI: 0.2-2.3
Lindbohm et al.
(1990)
361 women occupationally and
residentially exposed to solvents in
Santa Clara County, CA
June 1986-February 1987 (735
controls)
Questionnaire
Increased risk of spontaneous abortion based on 6 cases and 4
controls exposed to TCEa
OR: 3.1, 95% CI: 0.92-10.4
Windham et al.
(1991)
4,396 pregnancies among residents of
Woburn, MA 1960-1982
TCE: 267 ng/L
Tetrachloroethylene: 21 (ig/L
Chloroform: 12 |ig/L
Increased risk of perinatal death (n = 67) after 1970
{p = 0.55) but not before 1970 (OR: 10, p = 0.003)
No increased risk of spontaneous abortion (n = 520;
p = 0.66)
Lagakos et al.
(1986)
707 parents of children with congenital
heart disease in Tucson Valley, AZ
1969-1987
6-239 ppb TCE, along with DCA and
chromium
No increased risk of fetal death (not quantified) based on 246
exposed and 461 unexposed cases
Goldberg et al.
(1990)
75 men and 71 women living near
Rocky Mountain Arsenal, CO
1981-1986
Low: <5.0 ppb
Medium: >5.0 to <10.0 ppb
High: <10.0 ppb
Increased risk of miscarriage
ORadj: 4.44, 95% CI: 0.76-26.12
Increased risk of no live birth
ORadj: 2.46, 95% CI: 0.24-24.95
ATSDR (2001)
3^
rs
sj
3
a
>!
&
"^S
fs*
, ». si
00 ^3
K O
^ >!
>!
o
a
a
HH
0 &
M ^
'/O >!
fel
T" §
1 ^
o £.
TO
2!
O^
Hl
1—1 s
H S.
W|
O •
-------
Table 4-95. Developmental studies in humans (continued)
Subjects
Exposure
Effect
Reference
1,440 pregnancies among residents of
Endicott, NY
1978-2002
indoor air from soil vapor: 0.18-140
mg/m3
No increase in spontaneous fetal death
SIR: 0.66, 95% CI: 0.22-1.55
ATSDR (ATSDR,
2006b, 2008)
81,532 pregnancies among residents of
75 New Jersey towns
1985-1988 (3 control groups)
55 ppb TCE, along with many other
compounds
No increased risk of fetal death for >10 ppb
OR: 1.12
Bove (1996); Bove
etal. (1995)
Decreased birth weight, small for gestational age, and postnatal growth
361 women occupationally and
residentially exposed to solvents in
Santa Clara County, CA
June 1986-February 1987 (735
controls)
Questionnaire
Increased risk of IUGR based on one case exposed to both
TCE and tetrachloroethylene
OR: 12.5
Windham et al.
(1991)
3,462 births in Woburn, MA
1960-1982
267 |ig/L TCE in drinking water,
along with tetrachloroethylene and
chloroform
No increase in low birth weight (p = 0.77)
Lagakos et al.
(1986)
1,099 singleton birthsb to residents of 3
census tracts near Tucson International
Airport
1979-1981 (877 controls)
<5-107 (ig/L
No increase in full-term low birth weight (OR: 0.81)
No increase in low birth weight (OR: 0.9)
Increase in very low birth weight
OR: 3.3, 95% CI: 0.53-20.6
Rodenbeck et al.
(2000)
1,440 births0 to residents of Endicott,
NY
1978-2002
Indoor air from soil vapor:
0.18-140 mg/m3
Small increase in low birth weight
OR: 1.26, 95% CI: 1.00-1.59
Small increase in small for gestational age
OR: 1.22, 95% CI: 1.02-1.45
Increase in full-term low birth weight
OR: 1.41, 95% CI: 1.01-1.95
ATSDR (ATSDR,
2006b, 2008)
-------
Table 4-95. Developmental studies in humans (continued)
Subjects
Exposure
Effect
Reference
6,289 pregnancies among women
residing at Camp Lejeune, NC
1968-1985 (141 short-term and 31
long-term TCE-exposed, 5,681
unexposed controls)d
Tarrawa Terrace:
TCE: 8 ppb
1,2-DCE: 12 ppb
PCE: 215 ppb
Hadnot Point:
TCE: 1,400 ppb
1,2-DCE: 407 ppb
Change in mean birth weight
Long-term total: -139 g, 90% CI: -277, -1
Long-term males: -312 g, 90% CI: -540, -85
Short term total: +70g, 90% CI: -6, 146
Increase in SGA
Long-term total: OR: 1.5, 90% CI: 0.5, 3.8
Long-term males: OR: 3.9, 90% CI: 1.1-11.9
Short term total: OR: 1.1, 90% CI: 0.2-1.1
ATSDR (1998b)
81,532 pregnancies6 among residents of
75 New Jersey towns 1985-1988
55 ppb TCE, along with many other
compounds
Decreased birth weight at >5 ppb by 17.9g
No increase in prematurity at >10 ppb: OR: 1.02
Increase in low birth weight, term
>10 ppb: OR: 1.23, 50% CI: 1.09-1.39
No risk for very low birth weight
Bove (1996); Bove
etal. (1995)
Congenital malformations
1,148 men and 969 women
occupationally exposed to TCE in
Finland 1963-1976
U-TCA:
<10 to >500 mg/L
No congenital malformations reported
Tola et al. (1980)
371 men occupationally exposed to
solvents in Finland 1973-1983
Low/rare used <1 d/wk;
Intermediate used 1-4 d/wk or if
biological measures indicated high
exposure;
High/frequent used daily or if
biological measures indicated high
exposure
No increase in congenital malformations based on 17 cases
and 35 controls exposed to TCE
OR: 0.6, 95% CI: 0.2-2.0
Taskinen et al.
(1989)
100 babies with oral cleft defects born
to women occupationally exposed in
Europe 1989-1992
Questionnaire
Increase in cleft lip based on 2 of 4 TCE-exposed women
ORadj: 3.21, 95% CI: 0.49-20.9
Increase in cleft palate based on 2 of 4 TCE-exposed women
ORadj: 4.47, 95% CI: 1.02-40.9
Lorente et al.
(2000)
-------
Table 4-95. Developmental studies in humans (continued)
Subjects
Exposure
Effect
Reference
4,396 pregnancies among residents of
Woburn, MA
1960-1982
TCE: 267 ng/L
Tetrachloroethylene: 21 |ig/L
Chloroform: 12 |ig/L
Increase in eye/ear birth anomalies: OR: 14.9, p < 0.0001
Increase in CNS/chromosomal/oral cleft anomalies:
OR: 4.5,/? = 0.01
Increase in kidney/urinary tract disorders:
OR: 1.35,/? = 0.02
Small increase in lung/respiratory tract disorders:
OR: 1.16,/? = 0.05
No increase in cardiovascular anomalies (n = 5):p = 0.91
Lagakos et al.
(1986)
707 children with congenital heart
disease in Tucson Valley, AZ
1969-1987 (246 exposed, 461
unexposed)
Wells contaminated with TCE (range:
6-239 ppb), along with DCA and
chromium
Increase in congenital heart disease
<1981: OR: ~3(p< 0.005)
>1981: OR: ~1
Increased prevalence after maternal exposure during first
trimester (p < 0.001, 95% CI: 1.14-4.14)
Goldberg et al.
(1990)
75 men, 71 women living near Rocky
Mountain Arsenal, CO
1981-1986
Low: <5.0 ppb
Medium: >5.0 to <10.0 ppb
High: <10.0 ppb
Increase in total birth defects (n = 9)
OR: 5.87, 95% CI: 0.59-58.81
ATSDR (2001)
Births to residents of Endicott, NY
1983-2000f
Indoor air from soil vapor: 0.18-140
mg/m3
No increase in total birth defects:
RR: 1.08, 95% CI: 0.82-1.42
Increase in total cardiac defects:
RR: 1.94, 95% CI: 1.21-3.12
Increase in major cardiac defects:
RR: 2.52, 95% CI: 1.2-5.29
Increase in conotruncal heart defects:
RR: 4.83, 95% CI: 1.81-12.89
ATSDR (ATSDR,
2006b, 2008)
-------
Table 4-95. Developmental studies in humans (continued)
Subjects
Exposure
Effect
Reference
81,532 pregnancies among residents of
75 New Jersey towns
1985-1988
55 ppb TCE, along with many other
compounds
No increase in total birth defects: >10 ppb: OR: 1.12
Increase in total CNS defects at high dose
>1-5 ppb: OR: 0.93, 90% CI: 0.47-1.77
>10 ppb: OR: 1.68, 90% CI: 0.76-3.52
Increase in neural tube defects
>1-5 ppb: OR: 1.58, 90% CI: 0.69-3.40
>10 ppb: OR: 2.53, 90% CI: 0.91-6.37
Increase in oral clefts:
>5 ppb: OR: 2.24, 95% CI: 1.16-4.20
Increase in major cardiac defects:
>10 ppb: OR: 1.24, 50% CI: 0.75-1.94
Increase in ventrical septal defects
>5ppb: OR: 1.30, 95% CI: 0.88-1.87
Bove (1996); Bove
etal. (1995)
1,623 children <20 yr old dying from
congenital anomalies in Maricopa
County, AZ
1966-1986
8.9 and 29 ppb TCE in drinking water
Increase in deaths due to congenital anomalies in East Central
Phoenix
1966-1969: RR: 1.4, 95% CI: 1.1-1.7
1970-1981: RR: 1.5, 95% CI: 1.3-1.7
1982-1986: RR: 2.0, 95% CI: 1.5-2.5
AZ DHS (Flood,
1988)
4,025 infants born with congenital heart
defects in Milwaukee, WI
1997-1999
Maternal residence within 1.32 miles
from at least one TCE emissions
source
Increase in congenital heart defects for mothers >38 yr old
Exposed: OR: 6.2, 95% CI: 2.6-14.5
Unexposed: OR: 1.9, 95% CI: 1.1-3.5
No increase in congenital heart defects for exposed mothers
<38 yr old: OR: 0.9, 95% CI: 0.6-1.2
Yauck et al. (2004)
12 children exposed to TCE in well
water in Michigan
5-10 yr to 8-14 ppm
1 born with multiple birth defects
Bernad et al.
(1987), abstract
Other adverse birth outcomes
34 live births for which inhalation of
TCE for anesthesia was used in Japan
1962-1967
2-8 mL (mean 4.3 mL) for 2-98 min
(mean: 34.7 min)
1 case of asphyxia; 3 "sleepy babies" with Apgar scores of
5-9. Delayed appearance of newborn reflexes
Beppu (1968)
-------
L/l
K>
Table 4-95. Developmental studies in humans (continued)
Subjects
Exposure
Effect
Reference
51 UK women whose fetus was
considered to be at risk for hypoxia
during labor administered TCE as an
analgesic (50 controls)
Amount and route of exposure not
reported
TCE caused fetal pH to fall more, base deficit increased
more, and P02 fell more than the control group by fourfold or
more compared to other analgesics used
Phillips and
Macdonald (1971)
Postnatal developmental outcomes
Developmental neurotoxicity
54 individuals from 3 residential
cohorts in the United States exposed to
TCE in drinking water
Woburn, MA
63-400 ppb for <1-12 yr
Alpha, OH
3.3-330 ppb for 5-17 yr
Twin Cities, MN
261-2,440 ppb for 0.25-25 yr
Woburn, MA
Verbal naming/language impairment in 6/13 children (46%)
Alpha, OH
Verbal naming/language impairment in 1/2 children (50%)
Twin Cities, MN
Verbal naming/language impairment in 4/4 children (100%)
Memory impairment in 4/4 children (100%)
Academic impairment in 4/4 children (100%)
Moderate encephalopathy in 4/4 children (100%)
Poor performance on reading/spelling test in 3/4 children
(75%)
Poor performance on information test in 3/4 children (75%)
White etal. (1997)
284 cases of ASD diagnosed <9 yr old
and 657 controls born in the San
Francisco Bay Area
1994
Births geocoded to census tracts, and
linked to HAPs data
Increase in ASD
upper 3rd quartile: OR: 1.37, 95% CI: 0.96-1.95
upper 4th quartile: OR: 1.47, 95% CI: 1.03-2.08
Windham et al.
(2006)
948 children (<18 yr) in the
trichloroethylene Subregistry
0.4 to >5,000 ppb TCE
Increase in speech impairment:
0-9 yr old: RR: 2.45, 99% CI: 1.31-4.58
10-17 yr old: RR: 1.14, 99% CI: 0.46-2.85
Increase in hearing impairment:
0-9 yr old: RR: 2.13, 99% CI: 1.12-4.07
10-17 yr old: RR: 1.12, 99% CI: 0.52-2.24
ATSDR (2003b);
Burg et al. (1995);
Burg and Gist
(1999)
12 children exposed to TCE in well
water in Michigan
5-10 yr to 8-14 ppm
9 of 12 children (75%) had poor learning ability, aggressive
behavior, and low attention span
Bernad et al.
(1987), abstract
3^
rs
sj
3
a
>!
&
fs*
, si
00 ^3
^ g
>!
o
a
a
HH
0 &
M ^
'/O >!
fel
T" §
1 ^
o £.
TO
2; ^
H J
1—1 s
H £.
W|
O •
-------
Table 4-95. Developmental studies in humans (continued)
Subjects
Exposure
Effect
Reference
Developmental immunotoxicity
200 children aged 36 mo old born
prematurely8 and at risk of atopyh in
Lepzig, Germany
1995-1996
Median air level in child's bedroom:
0.42 |ig/m3
No association with allergic sensitization to egg white and
milk, or to cytokine producing peripheral T-cells
Lehmann et al.
(2001)
85 healthy1 full-term neonates born in
Lepzig, Germany
1997-1999
Median air level in child's bedroom
3-4 wk afterbirth: 0.6 |ig/m'
Significant reduction of Thl IL-2 producing T-cells
Lehmann et al.
(2002)
Other developmental outcomes
55 children (6 mo to 10 yr old) were
anesthetized for operations to repair
developmental defects of the jaw and
face in Poland
1964
>10 mL TCE
Reports of bradycardia, accelerated heart rate, and respiratory
acceleration observed; no arrhythmia was observed
Jasinka (1965),
translation
Childhood cancer
98 children (<10 yr old) diagnosed with
brain tumors in Los Angeles County
1972-1977
Questionnaire of parental
occupational exposures
Two cases were reported for TCE exposure, one with methyl
ethyl ketone
Peters and Preston-
Martin (1981)
22 children (<19 yr old) diagnosed with
neuroblastoma in United States and
Canada
1992-1994 (12 controls)
Questionnaire of parental
occupational exposures
Increase in neuroblastoma after paternal exposure
OR: 1.4, 95% CI: 0.7-2.9
Maternal exposure not reported
De Roos et al.
(2001)
61 boys and 62 girls (<10 yr old)
diagnosed with leukemia and 123
controls in Los Angeles County
1980-1984
Questionnaire of parents for
occupational exposure
Increase in leukemia after paternal exposure
Preconception (1 yr): OR: 2.0, p = 0.16
Prenatal: OR: 2.0, p = 0.16
Postnatal: OR: 2.7, p = 0.7
Maternal exposure not reported
Lowengart et al.
(1987)
-------
Table 4-95. Developmental studies in humans (continued)
Subjects
Exposure
Effect
Reference
1,842 children (<15 yr old) diagnosed
with ALL in United States and Canada
1989-1993 (1986 controls)
Questionnaire of parents for
occupational exposure
Increase in ALL after maternal exposure
Preconception: OR: 1.8, 95% CI: 0.6-5.2
Pregnancy: OR: 1.8, 95% CI: 0.5-6.4
Postnatal: OR: 1.4, 95% CI: 0.5-4.1
Anytime: OR: 1.8, 95% CI: 0.8-4.1
No increase in ALL after paternal exposure
Anytime: OR: 1.1, 95% CI: 0.8-1.5
Shuetal. (1999)
109 children (<15 yr old) born in UK
1974-1988 (218 controls)
Questionnaire of parents for
occupational exposure
Increase in leukemia and NHL after paternal exposure
Preconception: OR: 2.27, 95% CI: 0.84-6.16
Prenatal: OR: 4.40, 95% CI: 1.15-21.01
Postnatal: OR: 2.66, 95% CI: 0.82-9.19
No increase in leukemia and NHL after maternal exposure
Preconception: OR: 1.16, 95% CI: 0.13-7.91
McKinney et al.
(1991)
22 children (<15 yr old) diagnosed with
childhood cancer in California
1988-1998
0.09-97 ppb TCE in drinking water
No increase in total cancer: SIR: 0.83, 99% CI: 0.44-1.40
No increase in CNS cancer: SIR: 1.05, 99% CI: 0.24-2.70
No increase in leukemia: SIR: 1.09, 99% CI: 0.38-2.31
Morgan and
Cassady (2002)
1,190 children (<20 yr old) diagnosed
with leukemia in 4 counties in New
Jersey
1979-1987
0-67 ppb TCE in drinking water
Increase in ALL in girls with >5 ppb exposure
<20 yr old: RR: 3.36, 95% CI: 1.29-8.28
<5 yr old: RR: 4.54, 95% CI: 1.47-10.6
Cohnetal. (1994b)
24 children (<15 yr old) diagnosed with
leukemia in Woburn, MA
1969-1997
267 |ig/L TCE in drinking water,
along with tetrachloroethylene,
arsenic, and chloroform
Increase in childhood leukemia
Preconception: ORadj: 2.61, 95% CI: 0.47-14.97
Pregnancy: ORadj: 8.33, 95% CI: 0.73-94.67
Postnatal: ORadj: 1.18, 95% CI: 0.28-5.05
Ever: ORad|: 2.39, 95% CI: 0.54-10.59
Costas et al.
(2002); Cutler et
al. (1986); Lagakos
et al. (1986);
MA DPH (1997a)1
347 children (<20 yr old) diagnosed
with cancer in Endicott, NY
1980-2001
indoor air from soil vapor: 0.18-140
mg/m3
No increase in cancer (<6 cases, similar to expected)
ATSDR (ATSDR,
2006b, 2008)
-------
L/l
K>
Table 4-95. Developmental studies in humans (continued)
Subjects
Exposure
Effect
Reference
189 children (<20 yr old) diagnosed
with cancer in Maricopa County, AZ
1965-1990
8.9 and 29 ppb TCE in drinking water
Increase in leukemia:
1965-1986: SIR: 1.67, 95% CI: 1.20-2.27
1982-1986: SIR: 1.91, 95% CI: 1.11-3.12
No increase in total childhood cancers, lymphoma,
brain/CNS, or other cancers
AZ DHS (Flood,
1988; Flood, 1997)
(1990)k
16 children (<20 yr old) diagnosed with
cancer in East Phoenix, AZ
1965-1986
TCE, TCA, and other contaminants in
drinking water
No increase in leukemia: SIR: 0.85, 95% CI: 0.50-1.35
AZ DHS (Kioski et
al., 1990b)
37 children (<20 yr old) diagnosed with
cancer in Pima County, AZ
1970-1986
1.1-239 ppb TCE, along with
1,1-DCE, chloroform and chromium
in drinking water
Increase in leukemia (n = 11):
SIR: 1.50, 95% CI: 0.76-2.70
No increase in testicular cancer (n = 6):
SIR: 0.78, 95% CI: 0.32-1.59
No increase in lymphoma (n = 2):
SIR: 0.63, 95% CI: 0.13-1.80
No increase in CNS/brain cancer (n = 3):
SIR: 0.84, 95% CI: 0.23-2.16
Increase in other cancer (n = 15):
SIR: 1.40, 95% CI: 0.79-2.30
AZ DHS (Kioski et
al., 1990a)
3^
rs
sj
3
a
>!
&
fs*
, si
\Q ^3
O
Grj
>!
o
a
a
§•
>!
o
rs
o
a
>!
to
o
H
o
o
2!
a
"T3
H S.
W|
O •
p
o
c
o
H
W
H
o;
s
37 wk gestation and <25,000 g).
dUnexposed residents resided at locations not classified for long-term or short-term TCE exposure. Long-term TCE exposed mothers resided at Hospital Point
during 1968-1985 for at least 1 wk prior to birth. Short-term TCE exposed mothers resided at Berkeley Manor, Midway Park, Paradise Point, and Wakins
Village at the time of birth and at least 1 wk during January 27 to February 7, 1985. In addition, the mother's last menstrual period occurred on or before
January 31, 1985 and the birth occurred after February 2, 1985.
eLow birth weight defined as <2,500 g, very low birth weight as <1,500 g.
fl,440 births reported for years 1978-2002, but number not reported for years 1983-2000.
gPremature defined as 1,500-2,500 g at birth.
hRisk of atopy defined as cord blood IgE >0.9 kU/L; double positive family atopy history.
'Healthy birth defined as >2,500 g and >37 wk gestation.
JOnly results from Costas et al. (2002) are reported in the table.
kOnly results from AZ DHS (1990) are reported in the table.
ALL = acute lymphoblastic leukemia, ASD = autism spectrum disorder, DCE = dichloroethylene, HAP = hazardous air pollutant, IUGR = intrauterine growth
restriction, PCE = perchloroethylene, UK = United Kingdom.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
4.8.3.1.2.1.1. Occupational studies
The risks of spontaneous abortion and congenital malformations among offspring of men
occupationally exposed to TCE and other organic solvents were examined by Taskinen et al.
(1989). This nested case-control study was conducted in Finland from 1973-1983. Exposure
was determined by biological measurements of the father and questionnaires answered by both
the mother and father. The level of exposure was classified as "low/rare" if the chemical was
used <1 days/week, "intermediate" if used 1-4 days/week or if TCA urine measurements
indicated intermediate/low exposure, and "high/frequent" if used daily or if TCA urine
measurements indicated clear occupational exposure (defined as above the RfV for the general
population). There was no risk of spontaneous abortion from paternal TCE exposure (OR: 1.0,
95% CI: 0.6-2.0), although there was a significant increase for paternal organic solvent exposure
(OR: 2.7, 95% CI: 1.3-5.6) and a nonsignificant increase for maternal organic solvent exposure
(OR: 1.4, 95% CI: 0.6-3.0). (Also see section below for results from this study for congenital
malformations).
Another case-control study in Finland examined pregnancy outcomes in 1973-1986
among female laboratory technicians aged 20-34 years (Taskinen et al., 1994). Exposure was
reported via questionnaire, and was classified as "rare" if the chemical was used 1-2 days/week,
and "frequent" if used at least 3 days/week. Cases of spontaneous abortion (n = 206) were
compared with controls who had delivered a baby and did not report prior spontaneous abortions
(n = 329). A nonstatistically significant increased risk was seen between spontaneous abortion
and TCE use at least 3-days-a-week (OR: 1.6, 95% CI: 0.5-4.8).
The association between maternal exposure to organic solvents and spontaneous abortion
was examined in Finland for births 1973-1983 (Lindbohm et al., 1990). Exposure was assessed
by questionnaire and confirmed with employment records, and the level of exposure was either
high, low or none based on the frequency of use and known information about typical levels of
exposure for job type. Biological measurements of trichloroacetic acid in urine were also taken
on 64 women, with a median value of 48.1 |imol/L (mean: 96.2 ± 19.2 |imol/L). Three cases and
13 controls were exposed to TCE, with no increased risk seen for spontaneous abortion (OR: 0.6,
95% CI: 0.2-2.3, p. = 0.45).
A case-control study in Santa Clara County, California, examined the association
between solvents and adverse pregnancy outcomes in women >18 years old (Windham et al.,
1991). For pregnancies occurring between June 1986 and February 1987, 361 cases of
spontaneous abortion were compared to 735 women who had a live birth during this time period.
Telephone interviews included detailed questions on occupational solvent exposure, as well as
additional questions on residential solvent use. For TCE exposure, six cases of spontaneous
This document is a draft for review purposes only and does not constitute Agency policy.
593 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
abortion were compared to four controls of live births; of these ten TCE-exposed individuals,
four reported exposure to tetrachloroethylene, and one reported exposure to paint strippers and
thinners. An increased risk of spontaneous abortions was seen with TCE exposure (OR: 3.1,
95% CI: 0.92-10.4), with a statistically significant increased risk for those exposed
>0.5 hours/week (OR: 7.7, 95% CI: 1.3-47.4). An increased risk for spontaneous abortion was
also seen for those reporting a more "intense" exposure based primarily on odor, as well as skin
contact or other symptoms (OR: 3.9,p = 0.04). (Also see section below from this study on low
birth weight.)
4.8.3.1.2.1.2. Geographic-based studies
A community in Woburn, MA with contaminated well water experienced an increased
incidence of adverse birth outcomes and childhood leukemia (Lagakos et al., 1986). In 1979, the
wells supplying drinking water were found to be contaminated with 267 ppb TCE, 21 ppb
tetrachloroethylene, 11.8 ppb, and 12 ppb chloroform, and were subsequently closed. Pregnancy
and childhood outcomes were examined from 4,396 pregnancies among residents (Lagakos et
al., 1986). No association between water access and incidence of spontaneous abortion (n = 520)
was observed {p = 0.66). The town's water distribution system was divided into five zones,
which was reorganized in 1970. Prior to 1970, no association was observed between water
access and incidence of perinatal deaths (n = 46 still births and 21 deaths before 7 days)
(p = 0.55). However, after 1970, a statistically significant positive association between access to
contaminated water and perinatal deaths was observed (OR: 10.0, p = 0.003). The authors could
not explain why this discrepancy was observed, but speculated that contaminants were either not
present prior to 1970, or were increased after 1970. (Also see sections below on decreased birth
weight, congenital malformations, and childhood cancer for additional results from this cohort.)
A community in Tucson Valley, AZ with contaminated well water had a number of
reported cases of congenital heart disease. The wells were found to be contaminated with TCE
(range = 6-239 ppb), along with dichloroethylene and chromium (Goldberg et al., 1990). This
study identified 707 children born with congenital heart disease during the years 1969-1987. Of
the study participants, 246 families had parental residential and occupational exposure during
one month prior to conception and during the first trimester of pregnancy, and 461 families had
no exposure before the end of the first trimester. In addition to this control group, two others
were used: (1) those that had contact with the contaminated water area, and (2) those that had
contact with the contaminated water area and matched with cases for education, ethnicity, and
occupation. Among these cases of congenital heart disease, no significant difference was seen
This document is a draft for review purposes only and does not constitute Agency policy.
594 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
for fetal death (not quantified) for exposed cases compared to unexposed cases. (Also see
section below on congenital malformations for additional results from this cohort.)
A residential study of individuals living near the Rocky Mountain Arsenal in Colorado
examined the outcomes in offspring of 75 men and 71 women exposed to TCE in drinking water
(ATSDR, 2001). TCE exposure was stratified by high (>10.0 ppb), medium (>5.0 ppm to
<10.0 ppb), and low (<5.0 ppb). Among women with >5 ppb exposure experiencing miscarriage
(n = 22/57) compared to unexposed women experiencing miscarriage (n = 2/13) an elevated
nonsignificant association was observed (ORadj: 4.44, 95% CI: 0.76-26.12). For lifetime number
of miscarriages reported by men and women, results were increased but without dose-response
for women (medium: ORadj: 8.56, 95% CI: 0.69-105.99; high: ORadj: 4.16, 95% CI: 0.61-25.99),
but less for men (medium: ORadj: 1.68, 95% CI: 0.26-10.77; high: ORadj: 0.65,
95% CI: 0.12-3.48). Among women with >5 ppb exposure experiencing no live birth (n = 9/57)
compared to unexposed women experiencing no live birth (n = 1/13) an elevated nonsignificant
association was observed (ORadj: 2.46, 95% CI: 0.24-24.95). (Also see below for results from
this study on birth defects.)
NYS DOH and ATSDR conducted a study in Endicott, NY to examine childhood cancer
and birth outcomes in an area contaminated by a number of volatile organic compounds (VOCs),
including "thousands of gallons" of TCE (ATSDR, 2006a). Soil vapor levels tested ranged from
0.18-140 mg/m3 in indoor air. A follow-up study by ATSDR (2008) reported that during the
years 1978-1993 only five spontaneous fetal deaths occurring >20 weeks gestation were
reported when 7.5 were expected (SIR: 0.66, 95% CI: 0.22-1.55). (See sections on low birth
weight, congenital malformations, and childhood cancer for additional results from this cohort.)
Women were exposed to contaminated drinking water while pregnant and living in 75
New Jersey towns during the years 1985-1988 (Bove, 1996; Bove et al., 1995). The water
contained multiple trihalomethanes, including an average of 55 ppb TCE, along with
tetrachloroethylene, 1,1,1-trichloroethane, carbon tetrachloride, 1,2-dichloroethane, and benzene.
A number of birth outcomes were examined for 81,532 pregnancies, which resulted in
80,938 live births and 594 fetal deaths. No association was seen for exposure to >10 ppb TCE
and fetal death (ORadj: 1.12). (See below for results from this study on decreased birth weight
and congenital malformations.)
4.8.3.1.2.2.Decreased birth weight, small for gestational age, and postnatal growth
Available data pertaining to birth weight and other growth-related outcomes come from
the case-control study in Santa Clara, CA (discussed above), and by geographic-based studies as
This document is a draft for review purposes only and does not constitute Agency policy.
595 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
well as geographic areas with known contamination of water supplies areas in Woburn, MA;
Tucson, AZ, Endicott, NY; Camp Lejeune, NC; and New Jersey.
4.8.3.1.2.2.1. Occupational studies
The case-control study of the relationship between solvents and adverse pregnancy
outcomes discussed above (Windham et al., 1991) also examined intrauterine growth restriction
(IUGR). Telephone interviews included detailed questions on occupational solvent exposure, as
well as additional questions on residential solvent use. An increased risk of IUGR was observed
(OR: 12.5), although this was based only on one case that was exposed to both TCE and
tetrachloroethylene (also see section above on spontaneous abortion).
4.8.3.1.2.2.2. Geographic-based studies
The study of Woburn, MA with contaminated well water discussed above (Lagakos et al.,
1986) examined birth weight. Of 3,462 live births surviving to 7 days, 220 were less than
6 pounds at birth (6.4%). No association was observed between water access and low birth
weight (p = 0.77). (See section on spontaneous abortion for study details, and see sections on
spontaneous abortion, congenital malformations, and childhood cancer for additional results
from this cohort.)
An ecological analysis of well water contaminated with TCE in Tucson and birth-weight
was conducted by Rodenbeck et al. (2000). The source of the exposure was a U.S. Air Force
plant and the Tucson International Airport. The wells were taken out of service in 1981 after
concentrations of TCE were measured in the range of <5 |ig/L to 107 |ig/L. The study
population consisted of 1,099 babies born within census tracts between 1979 and 1981, and the
comparison population consisted of 877 babies from nearby unexposed census tracts. There was
a nonsignificant increased risk for maternal exposure to TCE in drinking water and very-low-
birth-weight (<1,501 g) (OR: 3.3, 95% CI: 0.53-20.6). No increases were observed in the low-
birth-weight (<2,501 g) (OR: 0.9) or full-term (>35-week and <46-week gestation) low-birth-
weight (OR: 0.81).
The study of VOC exposure in Endicott, NY reported data on low birth weight and small
for gestational age (ATSDR, 2006a), see section on spontaneous abortion for study details). For
births occurring during the years 1978-2002, low birth weight was slightly but statistically
elevated (OR: 1.26, 95% CI: 1.00-1.59), as was small for gestational age (SGA; OR: 1.22,
95% CI: 1.02-1.45), and full-term low birth weight (OR: 1.41, 95% CI: 1.01-1.95). (Also see
This document is a draft for review purposes only and does not constitute Agency policy.
596 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
sections on spontaneous abortion, congenital malformations, and childhood cancer for additional
results from this cohort.)
Well water at the U.S. Marine Corps Base in Camp Lejeune, NC was identified to be
contaminated with TCE, tetrachloroethylene, and 1,2-dichloroethane in April, 1982 and the wells
were closed in December, 1984. ATSDR examined pregnancy outcomes among women living
on the base during the years 1968-1985 (ATSDR, 1998b). Compared to unexposed residentsl2
(n = 5,681), babies exposed to TCE long-terml3 (// = 31) had a lower mean birth weight after
adjustment for gestational age (-139 g, 90% CL = -277, -1), and babies exposed short-terml4
(n = 141) had a slightly higher mean birth weight (+70 g, 90% CL = -6, 146). For the long-term
group, no effect was seen for very low birth weight (<1,500 g) or prematurity (>5 ppb,
OR: 1.05). No preterm births were reported in the long-term group and those (n = 8) in the
short-term group did not have an increased risk (OR: 0.7, 90% CI: 0.3-1.2). A higher
prevalence of SGA15 was seen in the long-term exposed group (n = 3; OR 1.5, 90% CL: 0.5, 3.8)
compared to the short-term exposed group (OR: 1.1, 90% CI: 0.2-1.1). When the long-term
group was stratified by gender, male offspring were at more risk for both reduced birth weight
(-312 g, 90% CL = -632, -102) and SGA (OR: 3.9, 90% CL: 1.1-11.8). This study is limited
due the mixture of chemicals in the water, as well as it small sample size. ATSDR is currently
reanalyzing the findings because of an error in the exposure assessment related to the start-up
date of a water treatment plant (ATSDR, 2007, 2009; U.S. GAO, 2007)
Pregnancy outcomes among women were exposed to contaminated drinking water while
pregnant and living in 75 New Jersey towns during the years 1985-1988 was examined by
Bove et al. (Bove, 1996; Bove et al., 1995). The water contained multiple trihalomethanes,
including an average of 55 ppb TCE, along with tetrachloroethylene, 1,1,1-trichloroethane,
carbon tetrachloride, 1,2-dichloroethane, and benzene. A number of birth outcomes were
examined for 81,532 pregnancies, which resulted in 80,938 live births and 594 fetal deaths. A
slight decrease of 17.9 g in birth weight was seen for exposure >5 ppb, with a slight increase in
risk for exposure >10 ppb (OR: 1.23), but no effect was seen for very low birth weight or
SGA/prematurity (>5 ppb, OR: 1.05). However, due to the multiple contaminants in the water, it
is difficult to attribute the results solely to TCE exposure. (See below for results from this study
on congenital malformations.)
12Unexposed residents resided at locations not classified for long-term or short-term TCE exposure.
13Long-term TCE exposed mothers resided at Hospital Point during 1968-1985 for at least 1 week prior to birth.
14Short-term TCE exposed mothers resided at Berkeley Manor, Midway Park, Paradise Point, and Wakins Village
at the time of birth and at least 1 week during January 27 to February 7, 1985. In addition, the mother's last
menstrual period occurred on or before January 31, 1985 and the birth occurred after February 2, 1985.
15The criteria for SGA being singleton births less than the 10th percentile of published sex-specific growth curves.
This document is a draft for review purposes only and does not constitute Agency policy.
597 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
4.8.3 .1.2.3 .Congenital malformations
Three studies focusing on occupational solvent exposure and congenital malformations
from Europe provide data pertaining to TCE. Analyses of risk of congenital malformations were
also included in the studies in the four geographic areas described above (Woburn, MA; Tucson,
AZ, Rocky Mountain Arsenal, CO; Endicott, NY; and New Jersey), as well as additional sites in
Phoenix, AZ; and Milwaukee, WI. Specific categories of malformations examined include
cardiac defects, as well as cleft lip or cleft palate.
4.8.3.1.2.3.1. Occupational studies
A study of 1,148 men and 969 women occupationally exposed to TCE in Finland from
1963-1976 to examined congenital malformations of offspring (Tola et al., 1980). Urinary
trichloroacetic acid measurements available for 2,004 employees ranged from <10 to >500 mg/L,
although 91% of the samples were below 100 mg/L. No congenital malformations were seen in
the offspring of women between the ages of 15-49 years, although 3 were expected based on the
national incidence. Expected number of cases for the cohort could not be estimated because the
number of pregnancies was unknown.
Men from Finland occupationally exposed to organic solvents including TCE did not
observe a risk of congenital malformations from paternal organic solvent exposure based on
17 cases and 35 controls exposed to TCE (OR: 0.6, 95% CI: 0.2-2.0) (Taskinen et al., 1989).
(Also see section above on spontaneous abortion for study details and additional results from this
cohort.)
An occupational study of 100 women who gave birth to babies born with oral cleft
defects and 751 control women with normal births were examined for exposure to a number of
agents including TCE during the first trimester of pregnancy (Lorente et al., 2000). All women
were participants in a multicenter European case-referent study whose children were born
between 1989 and 1992. Four women were exposed to TCE, resulting in two cases of cleft lip
(ORadj: 3.21, 95% CI: 0.49-20.9), and two cases of cleft palate (ORadj: 4.47, 95% CI: 1.02-40.9).
Using logistic regression, the increased risk of cleft palate remained high (OR: 6.7, 95%
CI: 0.9-49.7), even when controlling for tobacco and alcohol consumption (OR: 7.8, 95%
CI: 0.8-71.8). However, the number of cases was small, and exposure levels were not known.
4.8.3.1.2.3.2. Geographic-based studies
A community in Woburn, MA with contaminated well water experienced an increased
incidence of adverse birth outcomes and childhood leukemia (Lagakos et al., 1986, see section
This document is a draft for review purposes only and does not constitute Agency policy.
598 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
on spontaneous abortion for study details). Statistically significant positive association between
access to contaminated water and eye/ear birth anomalies (OR: 14.9, p < 0.0001),
CNS/chromosomal/oral cleft anomalies (OR: 4,5, p = 0.01), kidney/urinary tract disorders
(OR: 1.35,/? = 0.02) and lung/respiratory tract disorders (OR: 1.16,p = 0.05) were observed.
There were also five cases of cardiovascular anomalies, but there was not a significant
association with TCE (p = 0.91). However, since organogenesis occurs during gestational weeks
3-5 in humans, some of these effects could have been missed if fetal loss occurred. (Also see
sections on spontaneous abortion, perinatal death, decreased birth weight, and childhood cancer
for additional results from this cohort.)
A high prevalence of congenital heart disease was found within an area of Tucson Valley,
AZ (Goldberg et al., 1990, see section on spontaneous abortion for study details and additional
results). Of the total 707 case families included, 246 (35%) were exposed to wells providing
drinking water found to be contaminated with TCE (range = 6-239 ppb), along with
dichloroethylene and chromium. Before the wells were closed after the contamination was
discovered in 1981, the OR of congenital heart disease was 3 times higher for those exposed to
contaminated drinking water compared to those not exposed; after the wells were closed, there
was no difference seen. This study observed 18 exposed cases of congenital heart disease when
16.4 would be expected (RR: 1.1). Prevalence of congenital heart disease in offspring after
maternal exposure during the first trimester (6.8 in 1,000 live births) was significantly increased
compared to nonexposed families (2.64 in 1,000 live births) (p < 0.001, 95% CI: 1.14-4.14). No
difference in prevalence was seen if paternal data was included, and there was no difference in
prevalence by ethnicity. In addition, no significant difference was seen for cardiac lesions.
A residential study of individuals living near the Rocky Mountain Arsenal in Colorado
examined the outcomes in offspring of 75 men and 71 women exposed to TCE in drinking water
(ATSDR, 2001). The risk was elevated for the nine birth defects observed (OR: 5.87,
95% CI: 0.59-58.81), including one nervous system defect, one heart defect, and one incidence
of cerebral palsy. The remaining cases were classified as "other," and the authors speculate
these may be based on inaccurate reports. (See above for study details and results on
spontaneous abortion.)
The study of VOC exposure in Endicott, NY examined a number of birth defects during
the years 1983-2000 (ATSDR, 2006a), see section on spontaneous for study details). These
include total reportable birth defects, structural birth defects, surveillance birth defects, total
cardiac defects, major cardiac defects, cleft lip/cleft palate, neural tube defects, and choanal
atresia (blocked nasal cavities). There were 56 expected cases of all birth defects and 61 were
observed resulting in no elevation of risk (rate ratio, RR: 1.08, 95% CI: 0.82-1.42). There were
no cases of cleft lip/cleft palate, neural tube defects, or choanal atresia. Both total cardiac
This document is a draft for review purposes only and does not constitute Agency policy.
599 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
defects (n = 15; RR: 1.94, 95% CI: 1.21-3.12) and major cardiac defects (n = 6; RR: 2.52,
95% CI: 1.2-5.29) were statistically increased. A follow-up study by ATSDR (2008) reported
that conotruncal heart malformations were particularly elevated (n = 4; RR: 4.83, 95% CI:
1.81-12.89). The results remained significantly elevated (aRR: 3.74; 95% CI: 1.21-11.62)
when infants with Down syndrome were excluded from the analysis. (Also see sections on
spontaneous abortion, decreased birth weight, and childhood cancer for additional results from
this cohort.)
In the New Jersey study described previously, the prevalence of birth defects reported by
surveillance systems was examined among the women exposed to TCE and other contaminants
in water while pregnant between 1985-1988 (Bove, 1996; Bove et al., 1995). For exposure
>10 ppb (n = 1,372), an increased risk, with relatively wide confidence intervals, was seen for all
birth defects (OR: 2.53, 95% CI: 0.77-7.34). An increased risk was also seen for CNS defects
(>10 ppb: OR: 1.68), specifically 56 cases of neural tube defects (<1-5 ppb: 1.58,
95%) CI: 0.61-3.85; >10 ppb: OR: 2.53, 95% CI: 0.77-7.34). A slight increase was seen in
major cardiac defects (>10 ppb: OR: 1.24, 50% CI: 0.75-1.94), including ventrical septal defects
(>5 ppb: OR: 1.30, 95% CI: 0.88-1.87). An elevated risk was seen for 9 cases of oral clefts
(<5 ppb: OR: 2.24, 95% CI: 1.04-4.66), although no dose-response was seen (>10 ppb,
OR: 1.30). However, due to the multiple contaminants in the water, it is difficult to attribute the
results solely to TCE exposure. (See above for results from this study on fetal death and
decreased birth weight.)
Arizona Department of Health Services (AZ DHS) conducted studies of contaminated
drinking water and congenital malformations (<20 years old) in Maricopa County, which
encompasses Phoenix and the surrounding area (Flood, 1988). TCE contamination was
associated with elevated levels of deaths in children less than 20 years old due to total congenital
anomalies in East Central Phoenix from 1966-1969 (RR: 1.4, 95% CI: 1.1-1.7), from
1970-1981 (RR: 1.5, 95% CI: 1.3-1.7), and from 1982-1986 (RR: 2.0, 95% CI: 1.5-2.5), as
well as in other areas of the county. (See below for results from this study on childhood
leukemia.)
A study was conducted of children born 1997-1999 with congenital heart defects in
Milwaukee, WI (Yauck et al., 2004). TCE emissions data were ascertained from state and EPA
databases, and distance between maternal residence and the emission source was determined
using a GIS. Exposure was defined as those within 1.32 miles from at least one site. Results
showed that an increased risk of congenital heart defects was seen for the offspring of exposed
mothers 38 years old or older (OR: 6.2, 95% CI: 2.6-14.5), although an increased risk was also
seen for offspring of unexposed mothers 38 years old or older (OR: 1.9, 95% CI: 1.1-3.5), and
no risk was seen for offspring of exposed mothers younger than 38 years (OR: 0.9, 95%
This document is a draft for review purposes only and does not constitute Agency policy.
600 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
CI: 0.6-1.2). The authors speculate that studies that did not find a risk only examined younger
mothers. The authors also note that statistically-significant increased risk was seen for mothers
with preexisting diabetes, chronic hypertension, or alcohol use during pregnancy.
An abstract reported that twenty-eight people living in a Michigan town were exposed for
5-10 years to 8-14 ppm TCE in well water (Bernad et al., 1987, abstract). One child was born
with multiple birth defects, with no further details.
4.8.3 .1.2.4.Other adverse birth outcomes
TCE was previously used as a general anesthetic during pregnancy. One study measured
the levels of TCE in maternal and newborn blood after use during 34 vaginal childbirths (Beppu,
1968). TCE was administered through a vaporizer from two to 98 minutes (mean 34.7 minutes)
at volumes from 2-8 mL (mean 4.3 mL). Mean blood TCE concentrations were 2.80 ±1.14
mg/dL in maternal femoral arteries; 2.36 ±1.17 mg/dL in maternal cubital veins; 1.83 ± 1.08
mg/dL in umbilical vein; and 1.91 ± 0.95 mg/dL in the umbilical arteries. A significant
correlation was seen for maternal arterial blood and infants' venous blood, and the concentration
of the fetal blood was lower than that of the mother. Of these newborns, one had asphyxia and
three "sleepy babies" had Apgar scores of 5-9; however, these results could not be correlated to
length of inhalation and there was no difference in the TCE levels in the mother or newborn
blood compared to those without adverse effects. Discussion included delayed newborn reflexes
(raising the head and buttocks, bending the spine, and sound reflex), blood pressure, jaundice,
and body weight gain; however, the results were compared to newborns exposed to other
compounds, not to an unexposed population. This study also examined the concentration of TCE
in one mother at 22-weeks gestation exposed for four minutes, after which the fetus was
"artificially delivered." Maternal blood concentration was 3.0 mg/dL, and 0.9 mg/dL of TCE
was found in the fetal heart, but not in other organs.
Another study of TCE administered during childbirth to the mother as an analgesic
examined perinatal measures, including fetal pH, fetal partial pressure carbon dioxide (PC02.)
fetal base deficit, fetal partial pressure oxygen (P02), Apgar scores, and neonatal capillary blood
(Phillips and Macdonald, 1971). The study consisted of 152 women whose fetus was considered
to be at risk for hypoxia during labor. Out of this group, 51 received TCE (amount and route of
exposure not reported). TCE caused fetal pH to fall more, base deficit increased more, and P02
fell more than the control group by fourfold or more compared to other analgesics used.
This document is a draft for review purposes only and does not constitute Agency policy.
601 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
4.8.3.1.3. Postnatal developmental outcomes
4.8.3.1.3.1 Developmental neurotoxicity
The studies examining neurotoxic effects from TCE exposure are discussed in
Section 4.3, and the human developmental neurotoxic effects are reiterated here.
4.8.3.1.3.1.1. Occupational studies
An occupational study examined the neurodevelopment of the offspring of 32 women
exposed to various organic solvents during pregnancy (Laslo-Baker et al., 2004) Till et al., 2001.
Three of these women were exposed to TCE; however, no levels were measured and the results
for examined outcomes are for total organic solvent exposure, and are not specific to TCE.
4.8.3.1.3.1.2. Geographic-based studies
A study of three residential cohorts (Woburn, MA; Alpha, OH; and Twin Cities, MN)
examined the neurological effects of TCE exposure in drinking water (White et al., 1997). For
Woburn, MA, 28 individuals ranging from 9-55 years old were assessed, with exposure from a
tanning factor and chemical plant at levels 63-400 ppb for <1 to 12 years; the time between
exposure and neurological examination was about 5 years. In this cohort, six of thirteen children
(46%) had impairments in the verbal naming/language domain. For Alpha, OH, 12 individuals
ranging from 12-68 years old were assessed, with exposure from degreasing used at a
manufacturing operation at levels 3.3-330 ppb for 5-17 years; the time between exposure and
neurological examination was 5-17 years. In this cohort, one of two children (50%) had
impairments in the verbal naming/language domain. For Twin Cities, MN, 14 individuals
ranging from 8-62 years old were assessed, with exposure from an army ammunition plant at
levels 261-2,440 ppb for 0.25-25 years; the time between exposure and neurological
examination was 4-22 years. In this cohort, four of four children (100%) had impairments in the
verbal naming/language, memory, and academic domains and were diagnosed with moderate
encephalopathy; and three of four children (15%) performed poorly on the WRAT-R Reading
and Spelling and WAIS-R Information tests.
A case-control study was conducted to examine the relationship between multiple
environmental agents and autism spectrum disorder (ASD) (Windham et al., 2006). Cases
(n = 284) and controls (n = 657) were born in 1994 in the San Francisco Bay Area. Cases were
diagnosed before age nine. Exposure was determined by geocoding births to census tracts, and
rd
linking to hazardous air pollutants data. An elevated risk was seen for TCE in the upper 3
This document is a draft for review purposes only and does not constitute Agency policy.
602 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
quartile (OR: 1.37, 95% CI: 0.96-1.95), and a statistically significant elevated risk was seen for
the upper 4th quartile (OR: 1.47, 95% CI: 1.03-2.08).
The Trichloroethylene Subregistry (Burg and Gist, 1999; Burg et al., 1995), including
948 children <18 years old from 13 sites located in 3 states, was examined for any association of
ingestion of drinking water contaminated with TCE and various health effects (ATSDR, 2003a;
Burg and Gist, 1999; Burg et al., 1995). Exposure groups included (1) maximum TCE exposure,
(2) cumulative TCE exposure, (3) cumulative chemical exposure, and (4) duration of exposure.
Exposed children 0-9 years old had statistically increased hearing impairment compared to
controls (RR: 2.13, 99% CI: 1.12-4.07), with children <5 having a 5.2-fold increase over
controls. Exposed children 0-9 years old also had statistically increased speech impairment
(RR: 2.45, 99% CI: 1.31-4.58). In addition, anemia and other blood disorders were statistically
higher for males 0-9 years old. The authors noted that exposure could have occurred prenatally
or postnatally. There was further analysis on the 116 exposed children and 182 controls who
were under 10 years old at the time that the baseline study was conducted by ATSDR. This
analysis did not find a continued association with speech and hearing impairment in these
children; however, the absence of acoustic reflexes (contraction of the middle ear muscles in
response to sound) remained significant (ATSDR, 2003a). No differences were seen when
stratified by prenatal and postnatal exposure.
Twenty-eight people living in a Michigan town were exposed for 5-10 years to
8-14 ppm TCE in well water (Bernad et al., 1987, abstract). Ten adults and 12 children
completed a questionnaire on neurotoxic endpoints. Nine of the 12 children had poor learning
ability, aggressive behavior, and low attention span.
4.8.3.1.3.2.Developmental immunotoxicity
The studies examining human immunotoxic effects from TCE exposure are discussed in
Section 4.6.1. The studies reporting developmental effects are reiterated briefly here.
Two studies focused on immunological development in children after maternal exposure
to VOCs (Lehmann et al., 2001; Lehmann et al., 2002). The first examined premature neonates
(1,500-2,500 g) and neonates at risk of atopy (cord blood IgE >0.9 kU/L; double positive family
atopy history) at 36 months of age (Lehmann et al., 2001). Median air level in child's bedroom
"3
measured 0.42 |ig/m . There was no association with allergic sensitization to egg white and
milk, or to cytokine producing peripheral T-cells. The second examined healthy, full-term
neonates (>2,500 g; >37 weeks gestation) born in Lepzig, Germany (Lehmann et al., 2002).
This document is a draft for review purposes only and does not constitute Agency policy.
603 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Median air level in the child's bedroom 3-4 weeks after birth measured 0.6 |ig/m . A significant
reduction of Thl IL-2 producing T-cells was observed.
Byers et al. (1988) observed altered immune response in family members of children
diagnosed with leukemia in Woburn, MA (Lagakos et al., 1986, see below for results of this
study). The family members included 13 siblings under 19 years old at the time of exposure;
however, an analysis looking at only these children was not done. This study is discussed in
further detail in Section 4.6.1.
4.8.3 .1.3 .3 .Other developmental outcomes
A study demonstrated the adverse effects of TCE used as an anesthetic in children during
operations during 1964 in Poland to repair developmental defects of the jaw and face (Jasinska,
1965, translation). Fifty-five children ranging from 6 months to 10 years old were anesthetized
with at least 10 mL TCE placed into an evaporator. Bradycardia occurred in 2 children, an
accelerated heart rate of 20-25 beats per minute occurred in 7 children, no arrhythmia was
observed, and arterial blood pressure remained steady or dropped by 10 mmHg only.
Respiratory acceleration was observed in 25 of the children, and was seen more in infants and
younger children.
4.8.3 .1.3 .4.Childhood cancer
Several studies of parental occupational exposure were conducted in North America and
the United Kingdom to determine an association with childhood cancer. A number of
geographic-based studies were conducted in California; New Jersey; Woburn, MA; Endicott,
NY; Phoenix, AZ; and Tucson, AZ. Specific categories of childhood cancers examined include
leukemia, non-Hodgkin lymphoma, and CNS tumors.
4.8.3.1.3.4.1. Occupational studies
Brain tumors in 98 children less than 10 years old at diagnosis from 1972-1977 in Los
Angeles County have been observed in the offspring of fathers (Peters et al., 1985; Peters et al.,
1981). Exposure was determined by questionnaire. Two cases with TCE exposure were
reported: one case of oligodendroglioma in an 8-year-old whose father was a machinist, and
astrocytoma in a 7-year-old whose father was an inspector for production scheduling and parts
also exposed to methyl ethyl ketone (Peters et al., 1981). Peters et al. (1985) also briefly
This document is a draft for review purposes only and does not constitute Agency policy.
604 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
mentioned 5 cases and no controls of paternal exposure to TCE and brain tumors in the offspring
(resulting in an inability to calculate an odds ratio), but without providing any additional data.
A case-control study was conducted to assess an association between parental
occupational exposure and neuroblastoma diagnosed in offspring <19 years old in the United
States and Canada from May 1992 to April 1994 (De Roos et al., 2001). Paternal self-reported
exposure to TCE was reported in 22 cases and 12 controls, resulting in an elevated risk of
neuroblastoma in the offspring (OR: 1.4, 95% CI: 0.7-2.9). Maternal exposure to TCE was not
reported.
A case-control study of parental occupational exposure and childhood leukemia was
conducted in Los Angeles County (Lowengart et al., 1987). Children (61 boys and 62 girls)
diagnosed less than 10 years old (mean age 4 years) from 1980-1984 were included in the
analysis. Paternal occupation exposure to TCE was elevated for 1 year preconception (OR: 2.0,
p = 0.16), prenatal (OR: 2.0, p = 0.16), and postnatal (OR: 2.7, p = 0.7). Maternal exposure to
TCE was not reported.
A case-control study children diagnosed with acute lymphoblastic leukemia (ALL)
examined parental occupational exposure to hydrocarbons in the United States and Canada (Shu
et al., 1999). Children were under the age of 15 years at diagnosis during the years 1989-1993.
Cases were confirmed with a bone marrow sample. 1,842 case-control pairs were given
questionnaires on maternal and paternal exposures, resulting in 15 cases and 9 controls
maternally exposed and 136 cases and 104 controls paternally exposed to TCE. There was an
increased but nonsignificant risk for maternal exposure to TCE during preconception (OR: 1.8,
95% CI: 0.6-5.2), pregnancy (OR: 1.8, 95% CI: 0.5-6.4), postnatally (OR: 1.4,
95% CI: 0.5-4.1), or any of these periods (OR: 1.8, 95% CI: 0.8-4.1). However, there was no
increased risk for paternal exposure to TCE.
Occupational exposure in communities in the United Kingdom was examined to
determine an association with leukemia and non-Hodgkin lymphoma diagnosed in the offspring
(McKinney et al., 1991). Paternal occupational exposure was elevated for exposure occurring
during preconception (OR: 2.27, 95% CI: 0.84-6.16), prenatal (OR: 4.40, 95% CI: 1.15-21.01),
and postnatal (OR: 2.66, 95% CI: 0.82-9.19). Risk from maternal preconception exposure was
not elevated (OR: 1.16, 95% CI: 0.13-7.91). However, the number of cases examined in this
study was low, particularly for maternal exposure.
4.8.3.1.3.4.2. Geographic-based studies
A California community exposed to TCE (0.09-97 ppb) in drinking water from
contaminated wells was examined for cancer (Morgan and Cassady, 2002). A specific emphasis
This document is a draft for review purposes only and does not constitute Agency policy.
605 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
was placed on the examination of 22 cases of childhood cancer diagnosed before 15 years old.
However, the incidence did not exceed those expected for the community for total cancer
(SIR: 0.83, 99% CI: 0.44-1.40), CNS cancer (SIR: 1.05, 99% CI: 0.24-2.70), and leukemia
(SIR: 1.09, 99% CI: 0.38-2.31).
An examination of drinking water was conducted in four New Jersey counties to
determine an association with leukemia and non-Hodgkin lymphoma (Cohn et al., 1994b). A
number of contaminants were reported, including VOCs and trihalomethanes. TCE was found as
high as 67 ppb, and exposure categories were assigned to be >0.1, 0.1-5, and >5 ppb. A
significantly elevated dose-response risk for ALL was observed for girls diagnosed before
20 years old (RR: 3.36, 95% CI: 1.29-8.28), which was increased among girls diagnosed before
5 years old (RR:4.54, 95% CI: 1.47-10.6). A significantly elevated dose-response risk for girls
was also observed for total leukemia (RR: 1.43, 95% CI: 1.07-1.98).
The Woburn, MA community with contaminated well water experienced an increase in
the incidence of childhood leukemia (Costas et al., 2002; Cutler et al., 1986; Lagakos et al.,
1986; MDPH, 1997a). An initial study examined twelve cases of childhood leukemia diagnosed
in children less than 15 years old between 1969-1979, when 5.2 cases were expected, and a
higher risk was observed in boys compared to girls; however, no factors were observed to
account for this increase (Cutler et al., 1986). Another study observed statistically significant
positive association between access to contaminated water and 20 cases of childhood cancer
were observed for both cumulative exposure metric (OR: 1.39,p = 0.03), and none versus some
exposure metric (OR: 3.03,p = 0.02) (Lagakos et al., 1986). Massachusetts Department of
Public Health (MDPH, 1997a) conducted a case-control study of children less than 20 years old
living in Woburn and diagnosed with leukemia between 1969 and 1989 (n = 21) and observed
that consumption of drinking water increased the risk of leukemia (OR: 3.03, 95%
CI: 0.82-11.28), with the highest risk from exposure during fetal development (OR: 8.33,
95% CI: 0.73-94.67). This study found that paternal occupational exposure to TCE was not
related to leukemia in the offspring (MDPH, 1997a). In the most recent update, Costas et al.
(2002) reported that between the years 1969 and 1997, 24 cases of childhood leukemia were
observed when 11 were expected. Risk was calculated for cumulative exposure to contaminated
drinking water two years prior to conception (ORadj: 2.61, 95% CI: 0.47-14.97), during
pregnancy (ORadj: 8.33, 95% CI: 0.73-94.67), postnatal (ORadj: 1.18, 95% CI: 0.28-5.05), and
any of these time periods (ORadj: 2.39, 95% CI: 0.54-10.59). A dose response was observed
during pregnancy only. Cases were more likely to be male (76%), <9 years old at diagnosis
(62%>), breast-fed (OR: 10.17, 95% CI: 1.22-84.50), and exposed during pregnancy (adjusted
OR: 8.33, 95% CI: 0.73-94.67). A dose-response was seen during the pregnancy exposure
period, with the most exposed having an adjusted OR of 14.30 (95% CI: 0.92-224.52). Other
This document is a draft for review purposes only and does not constitute Agency policy.
606 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
elevated risks observed included maternal alcohol intake during pregnancy (OR: 1.50,
95% CI: 0.54-4.20), having a paternal grandfather diagnosed with cancer (OR: 2.01,
95% CI: 0.73-5.58), father employed in a high risk industry (OR: 2.55, 95% CI: 0.78-8.30), and
public water being the subject's primary beverage (OR: 3.03, 95% CI: 0.82-11.28). (Also see
sections on spontaneous abortion, perinatal death, decreased birth weight, and congenital
malformations for additional results from this cohort.)
The study of VOC exposure in Endicott, NY discussed above observed fewer than six
cases of cancer that were diagnosed between 1980 and 2001 in children less than 20 years old,
and did not exceed expected cases or types (ATSDR, 2006a). (See section on spontaneous
abortion for study details, and sections on spontaneous abortion, decreased birth weight, and
congenital malformations for additional results from this cohort.)
The AZ DHS conducted a number of studies of contaminated drinking water and 189
cases of childhood cancer (<20 years old) (ADHS, 1990; Flood, 1988; Flood, 1997; Kioski et al.,
1990a; Kioski et al., 1990b). In Maricopa County, which encompasses Phoenix and the
surrounding area, TCE contamination (8.9 and 29 ppb in two wells) was associated with elevated
levels of childhood leukemia (n = 67) in west central Phoenix during 1965-1986 (SIR: 1.67,
95% CI: 1.20-2.27) and 1982-1986 (SIR: 1.91, 95% CI: 1.11-3.12), but did not observe a
significant increase in total childhood cancers, lymphoma, brain/CNS, or other cancers during
these time periods (ADHS, 1990). (See above for results from this study on congenital
anomalies.) A follow-up study retrospectively asked parents about exposures and found that
residence within 2 miles of wells contaminated with TCE was not a risk factor for childhood
leukemia, but identified a number of other risk factors (Flood, 1997). A further study of East
Phoenix, reported on TCE contamination found along with 1,1,1-trichloroethane and 25 other
contaminants in well water (levels not reported) and found no increase in incidence of childhood
leukemia (SIR: 0.85, 95% CI: 0.50-1.35) based on 16 cases (Kioski et al., 1990b). There were
also 16 cases of other types of childhood cancer, but were too few to be analyzed separately. In
Pima County, which encompasses Tucson and the surrounding area, TCE was found in drinking
wells (1.1-239 ppb), along with 1,1-dichloroethylene (DCE), chloroform and chromium and
found a nonstatistically elevated risk of leukemia was observed (SIR: 1.50, 95% CI: 0.76-2.70),
but no risk was observed for testicular cancer, lymphoma, or CNS/brain cancer (Kioski et al.,
1990a).
This document is a draft for review purposes only and does not constitute Agency policy.
607 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
4.8.3.1.4. Summary of human developmental toxicity
Epidemiological developmental studies examined the association between TCE exposure
and a number of prenatal and postnatal developmental outcomes. Prenatal developmental
outcomes examined include spontaneous abortion and perinatal death; decreased birth weight,
small for gestational age, and postnatal growth; congenital malformations; and other adverse
birth outcomes. Postnatal developmental outcomes examined include developmental
neurotoxicity, developmental immunotoxicity, other developmental outcomes, and childhood
cancer related to TCE exposure.
More information on developmental outcomes is expected. A follow-up study of the
Camp Lejeune cohort (ATSDR, 1998b) for birth defects and childhood cancers was initiated in
1999 (ATSDR, 2003c) and expected to be completed soon (ATSDR, 2009; U.S. GAO, 2007).
Out of a total of 106 potential cases of either birth defects or childhood cancer, 57 have been
confirmed and will constitute the cases. These will be compared 548 control offspring of
mothers who also lived at Camp Lejeune during their pregnancy from 1968-1985. As part of
this study, a drinking water model was developed to determine a more accurate level and
duration of exposure to these pregnant women (ATSDR, 2007). Additional health studies have
been suggested, including adverse neurological or behavioral effects or pregnancy loss.
4.8.3.1.5. Animal Developmental Toxicology Studies
A number of animal studies have been conducted to assess the potential for
developmental toxicity of TCE. These include studies conducted in rodents by prenatal
inhalation or oral exposures (summarized in Tables 4-96 and 4-97), as well as assessments in
nonmammalian species (e.g., avian, amphibian, and invertebrate species) exposed to TCE during
development. Studies have been conducted that provide information on the potential for effects
Table 4-96. Summary of mammalian in vivo developmental toxicity
studies—inhalation exposures
Reference
Species/strain/
sex/number
Exposure level/
duration
NOAEL; LOAEL"
Effects
Carney et
al. (2006)
Rat, Sprague-
Dawley, females,
27 dams/group
0,50,150, or
600 ppm
(600 ppm =
3.2 mg/L)b
6 h/d;
GD 6-20
Mat. NOAEL: 150
ppm
Mat. LOAEL: 600
ppm
•I BW gain (22% less than control)
on GD 6-9 at 600 ppm.
Dev. NOAEL: 600 ppm
No evidence of developmental toxicity,
including heart defects.
Dorfmueller
et al. (1979)
Rat, Long-Evans,
females, 30
0 or 1,800+200
ppm
Mat. NOAEL: 1,800 +
200 ppm
No maternal abnormalities.
This document is a draft for review purposes only and does not constitute Agency policy.
608 DRAFT—DO NOT CITE OR QUOTE
-------
dams/group
(9,674 + 1,075
mg/m3)b
2 wk, 6 h/d,
5 d/wk; prior to
mating and/or on
GD 0-20
Dev. LOAEL: 1,800 +
200 ppm
Sig. T skeletal and soft tissue
anomalies in fetuses from dams
exposed during pregnancy only. No
sig. treatment effects on behavior of
offspring 10, 20, or 100 d postpartum.
B W gains sig. 1 in pups from dams
with pregestational exposure.
Hardin et al.
(1981)
Rat, Sprague-
Dawley, female,
nominal 30/group
0 or 500 ppm
6-7 h/d;
GD 1-19
Mat. NOAEL: 500 ppm
No maternal toxicity
Dev. NOAEL: 500 ppm
No embryonic or fetal toxicity.
Rabbit, New
Zealand white,
female, nominal
20/group
0 or 500 ppm
6-7 h/d;
GD 1-24
Mat. NOAEL: 500 ppm
No maternal toxicity.
Dev. LOAEL: 500 ppm
Hydrocephaly observed in 2 fetuses of
2 litters, considered equivocal evidence
of teratogenic potential.
Healy et al.
(1982)
Rat, Wistar,
females, 31-32
dams/group
0 or 100 ppm
4 h/d;
GD 8-21
Mat. NOAEL: 100 ppm
No maternal abnormalities.
Dev. LOAEL: 100
ppm
Litters with total resorptions sig. T.
Sig. -i fetal weight, and T bipartite or
absent skeletal ossification centers.
Schwetz et
al. (1975)
Rat, Sprague-
Dawley, female,
20-35/group
Mouse, Swiss-
Webster, females,
30-40 dams/group
0 or 300 ppm
7 h/d;
GD 6-15
Mat. LOAEL: 300
ppm
4-5% 4^ maternal BW
Dev. NOAEL: 300 ppm
No embryonic or fetal toxicity; not
teratogenic.
Westergren
et al. (1984)
Mouse, NMRI, male
and female, 6-12
offspring/group
0 or 150 ppm
24 h/d;
30 d (during 7 d
of mating and
until GD 22)
Dev. LOAEL:
150 ppm°
Specific gravity of brains sig. 1 at PND
0, 10, and 20-22. Similar effects at
PND 20-22 in occipital cortex and
cerebellum. No effects at 1 mo of age.
Bolded studies carried forward for consideration in dose-response assessment (see Section 5).
1 'NOAEL and LOAEL are based upon reported study findings.
2 bDose conversions provided by study author(s).
3 °Parental observations not reported.
4 Dev. = developmental; Mat. = maternal; sig. = statistically significant.
5
This document is a draft for review purposes only and does not constitute Agency policy.
609 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-97. Ocular defects observed (Narotsky et al., 1995)
2
Dose TCE (mg/kg-day)
Incidence
(number affected
pups/total number pups)a
Percentage of
pups with eye
defects
0
1/197
0.51
10.1
0/71
0.00
32
0/85
0.00
101
3/68
4.41
320
3/82
3.66
475
6/100
6.00
633
6/100
6.00
844
7/58
12.07
1,125
12/44
27.27
3
4 aReported in Barton and Das (1996).
5
6
7 on specific organ systems, including the developing nervous, immune, and pulmonary systems.
8 Additionally, a number of research efforts have focused on further characterization of the mode
9 of action for cardiac malformations that have been reported to be associated with TCE exposure.
10
4.8.3.1.6. Mammalian studies
11 Studies that have examined the effects of TCE on mammalian development following
12 either inhalation or oral exposures are described below and summarized in Tables 4-96 and 4-98,
13 respectively.
14
4.8.3 .1.6.1.Inhalation exposures
15 Dorfmueller et al. (1979) conducted a study in which TCE was administered by
16 inhalation exposure to groups of approximately 30 female Long-Evans hooded rats at a
17 concentration of 1,800 ± 200 ppm before mating only, during gestation only, or throughout the
18 premating and gestation periods. Half of the dams were killed at the end of gestation and half
19 were allowed to deliver. There were no effects on body weight change or relative liver weight in
20 the dams. The number of corpora lutea, implantation sites, live fetuses, fetal body weight,
This document is a draft for review purposes only and does not constitute Agency policy.
610 DRAFT—DO NOT CITE OR QUOTE
-------
1 resorptions, and sex ratio were not affected by treatment. In the group exposed only during
2 gestation, a significant increase in four specific sternebral, vertebral, and rib findings, and a
3
This document is a draft for review purposes only and does not constitute Agency policy.
611 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-98. Summary of mammalian in vivo developmental toxicity
2 studies—oral exposures
3
Reference
Species/strain/
sex/number
Dose level/
exposure
duration
Route/
vehicle
NOAEL;
LOAEL'
Effects
Blossom
and Doss
(2007)
Mouse, MRL +/+,
dams and both sexes
offspring, 3
litters/group, 8-12
offspring/group
0,0.5, or 2.5
mg/mL
Parental mice
and/or offspring
exposed from GD
0 to 7-8 mo of age
Drinking
water
Dev. LOAEL
= 0.5 mg/mLb
At 0.5 mg/mL: sig j
postweaning weight; sig. t
IFNy produced by splenic
CD4+ cells at 5-6 wk; sig |
splenic CD8+and B220+
lymphocytes; sig. t lgG2a and
histone; sig. altered
CD4-/CD8- and
CD4+/CD8+ thymocyte
profile.
At 2.5 mg/mL: Sig [
postweaning weight; sig. t
IFNy produced by splenic
CD4+ and CD8+ cells at 4-5
and 5-6 wk; sig | splenic
CD4+, CD8+, and B220+
lymphocytes; sig. altered
CD4+/CD8+ thymocyte
profile.
Blossom et
al. (2008)
Mouse, MRL +/+,
dams and both sexes
offspring, 8
litters/group, 3-8
offspring/group
0 or 0.1 mg/mL
(maternal dose =
25.7 mg/kg-day;
offspring PND
24-42 dose—31.0
mg/kg-day
Parental mice
and/or offspring
exposed from GD
0 to PND 42
Drinking
water
Dev. LOAEL
= 1,400 ppbb
At 0.1 mg/mL: at PND 20,
sig. t thymocyte cellularity
and distribution, associated
with sig. t in thymocyte
subset distribution; sig. t
reactive oxygen species
generation in total
thymocytes; sig. t in splenic
CD4+ T-cell production of
IFN-y and IL-2 in females
and TNF-a in males at PND
42.
Significantly impaired nest-
building behaviors at PND
35. Increased aggressive
activities, and increased
oxidative stress and impaired
thiol status in the cerebellar
tissue of male offspring at
PND 40.
4
This document is a draft for review purposes only and does not constitute Agency policy.
612 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-98. Summary of mammalian in vivo developmental toxicity
studies—oral exposures (continued)
Reference
Species/strain/
sex/number
Dose level/
exposure
duration
Route/
vehicle
NOAEL;
LOAEL'
Effects
Collier et al.
(2003)
Rat, Sprague-
Dawley, female,
number dams/group
not reported
0,0.11, or 1.1
mg/mL
(0, 830, or 8,300
ligM)c
GD 0-11
Drinking
water
Dev. LOEL:
0.11 mg/mL
Embryos collected between
GD 10.5 and 11. Gene
expression at 1.1 mg/mL
TCE: 8 housekeeping genes
T, and one gene j; 3 stress
response genes T, IL-10 I; 2
cyto-skeletal/cell
adhesion/blood related genes
T, 3 genes 1; 2 heart-specific
genes t- Effects at 0.11
mg/mL reduced considerably.
Two possible markers for
fetal TCE exposure identified
as Serca-2 Ca+2 ATPase and
GPI-pl37.
Cosby and
Dukelow
(1992)
Mouse, B6D2F1,
female, 28-62
dams/group
0, 24, or 240
mg/kg-day
GD 1-5, 6-10, or
11-15
Gavage in
corn oil
Mat. NOAEL:
240 mg/kg-
day
No maternal toxicity.
Dev. NOAEL:
240 mg/kg-
day
No effects on embryonic or
fetal development.
Dawson et al.
(1993)
Rat, Sprague-
Dawley, 116
females allocated to
11 groups
0, 1.5, or 1,100
ppm
2 mo before
mating and/or
during gestation
Drinking
water
Mat. NOAEL:
1,100 ppm
No maternal toxicity.
Dev. LOAEL:
1.5 ppm
Sig. T in heart defects,
primarily atrial septal defects,
found at both dose levels in
groups exposed prior to
pregnancy and during
pregnancy, as well as in group
exposed to 1,100 ppm dose
during pregnancy only. No
sig. T in congenital heart
defects in groups exposed
prior to pregnancy only.
Fisher et al.
(2001);
Warren et al.
(2006)
Rat, Sprague-
Dawley, female,
20-25 dams/group
0 or 500 mg/kg-
day
GD 6-15
Gavage in
soybean oil
Mat. NOAEL:
500 mg/kg-
day
No maternal toxicity.
Dev. NOAEL:
500 mg/kg-
day
No developmental toxicity.
The incidence of heart
malformations for fetuses
from TCE-treated dams
(3-5%) did not differ from
negative controls. No eye
defects observed.
This document is a draft for review purposes only and does not constitute Agency policy.
613 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-98. Summary of mammalian in vivo developmental toxicity
studies—oral exposures (continued)
Reference
Species/strain/
sex/number
Dose level/
exposure
duration
Route/
vehicle
NOAEL;
LOAEL'
Effects
Fredriksson
et al. (1993)
Mouse, NMRI,
male pups, 12 pups
from 3-4 different
litters/group
0, 50, or 290
mg/kg-day
PND 10-16
Gavage in
a 20% fat
emulsion
prepared
from egg
lecithin
and
peanut oil
Dev.
LOAEL: 50
mg/kg-day
Rearing activity sig. i at
both dose levels on PND 60.
George et al.
(1986)
Rat, F334, male
and female, 20
pairs/treatment
group,
40 controls/sex
0,0.15,0.30 or
0.60% micro-
encapsulated
TCE
Breeders exposed
1 wk premating,
then for 13 wk;
pregnant *- s
throughout
pregnancy (i.e.,
18 wk total)
Dietary
LOAEL:
0.15%
Open field testing in pups: a
sig. dose-related trend
toward T time required for
male and female pups to
cross the first grid in the
test devise.
Isaacson and
Taylor
(1989)
Rat, Sprague-
Dawley, females, 6
dams/group
0,312, or 625
mg/L.
(0,4.0, or 8.1
mg/d)c
Dams (and pups)
exposed from 14
d prior to mating
until end of
lactation.
Drinking
water
Dev.
LOAEL: 312
mg/Lb
Sig. 4^ myelinated fibers in
the stratum lacunosum-
moleculare of pups.
Reduction in myelin in the
hippocampus.
Johnson et
al. (2003)
Rat, Sprague-
Dawley, female,
9-13/group, 55 in
control group
0,2.5 ppb, 250
ppb, 1.5 ppm, or
1,100 ppm
(0, 0.00045, 0.048,
0.218, or 129
mg/kg-day)c
GD 0-22
Drinking
water
Dev.
NOAEL: 2.5
ppb
Dev.
LOAEL: 250
ppbb
Sig. T in percentage of
abnormal hearts and the
percentage of litters with
abnormal hearts at >250
ppb.
This document is a draft for review purposes only and does not constitute Agency policy.
614 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-98. Summary of mammalian in vivo developmental toxicity
studies—oral exposures (continued)
Reference
Species/strain/
sex/number
Dose level/
exposure
duration
Route/
vehicle
NOAEL;
LOAEL'
Effects
Narotsky et
al. (1995)
Rat, Fischer 344,
females, 8-12
dams/group
0,10.1,32,101,
320, 475, 633,
844, or 1,125
mg/kg-day
GD 6-15
Gavage in
corn oil
Mat.
LOAEL: 475
mg/kg-day
Sig. dose-related -i dam BW
gain at all dose levels on GD
6-8 and 6-20. Delayed
parturition at >475
mg/kg-day; ataxia at >633
mg/kg-day; mortality at
1,125 mg/kg-day.
Narotsky et
al. (1995)
(continued)
Dev.
NOAEL: 32
mg/kg-day
Dev.
LOAEL: 101
mg/kg-day
T full litter resorption and
postnatal mortality at >425
mg/kg-day. Sig. prenatal
loss at 1,125 mg/kg-day.
Pup BW -l (not sig.) on PND
1 and 6. Sig. T in pups with
eye defects at 1,125
mg/kg-day. Dose-related
(not sig.) T in pups with eye
defects at >101 mg/kg-day.
Narotsky and
Kavlock
(1995)
Rat, Fischer 344,
females, 16-21
dams/group
0, 1,125, or
1,500 mg/kg-day
GD 6-19
Gavage in
corn oil
Mat. LOAEL:
1,125 mg/kg-
day
Ataxia, 1 activity,
piloerection; dose-related 1
BW gain.
Dev. LOAEL:
1,125 mg/kg-
day
Sig. T full litter resorptions, 1
live pups/litter; sig. 1 pup BW
on PND 1; sig. T incidences of
microophthalmia and
anophthalmia.
Noland-
Gerbec et al.
(1986)
Rat, Sprague-
Dawley, females,
9-11 dams/group
0 or 312 mg/L
(Average total
intake of dams:
825 mg TCE over
61 d)°
Dams (and pups)
exposed from 14 d
prior to mating
until end of
lactation
Drinking
water
Dev. LOEL:
312 mg/Lb
Sig. 1 uptake of 3H-2-DG in
whole brains and cerebella
(no effect in hippocampus)
of exposed pups at 7, 11,
and 16 d, but returned to
control levels by 21 d.
This document is a draft for review purposes only and does not constitute Agency policy.
615 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-98. Summary of mammalian in vivo developmental toxicity
studies—oral exposures (continued)
Reference
Species/strain/
sex/number
Dose level/
exposure
duration
Route/
vehicle
NOAEL;
LOAEL3
Effects
Peden-
Adams et al.
(2006)
Mouse, B6C3F1,
dams and
both sexes
offspring, 5
dams/group; 5-7
pups/group at 3
wk; 4-5
pups/sex/group at 8
wk
0,1,400, or
14,000 ppb
Parental mice
and/or offspring
exposed during
mating, and from
GD 0 thru 3 or 8
wk of age
Drinking
water
Dev.
LOAEL:
1,400 ppbb
At 1,400 ppb: Suppressed
plaque-forming cell (PFC)
responses in males at 3 and
8 wk of age and in females
at 8 wk of age. Delayed
hypersensitivity response
increased at 8 wk of age in
females.
At 14,000 ppb: Suppressed
PFC responses in males and
females at 3 and 8 wk of
age. Splenic cell population
decreased in 3 wk old pups.
Increased thymic T-cells at
8 wk of age. Delayed
hypersensitivity response
increased at 8 wk of age in
males and females.
Peden-
Adams et al.
(2008)
Mouse, MRL +/+,
dams and both sexes
offspring, unknown
number
litters/group, 6-10
offspring/sex/group
0, 1,400, or 14,000
ppb (vehicle = 1%
emulphore)
Parental mice
and/or offspring
exposed from GD
0 to 12 mo of age
Drinking
water
Dev. LOAEL
= 1,400 ppbb
At 1,400 ppb: splenic
CD4-/CD8- cells sig. t in
females; thymic CD4+/CD8+
cells sig. 1 in males; 18% f in
male kidney weight.
At 14,000 ppb: thymic T-cell
subpopulations (CD8+,
CD4/CD8-, CD4+) sig. ^ in
males.
Taylor et al.
(1985)
Rat, Sprague-
Dawley, females,
no. dams/group not
reported
0, 312, 625, or
1,250 mg/L
Dams (and pups)
exposed from 14
d prior to mating
until end of
lactation
Drinking
water
Dev.
LOAEL:
312 mg/Lb
Exploratory behavior sig. f
in 60- and 90-d old male
rats at all treatment levels.
Locomotor activity was
higher in rats from dams
exposed to 1,250 ppm TCE.
Bolded studies carried forward for consideration in dose-response assessment (see Section 5).
1
2 "NO A EL. LOAEL, and LOEL (lowest-observed-effect level) are based upon reported study findings.
3 bDose conversions provided by study author(s).
4 °Maternal observations not reported.
5
6 Dev. = developmental; Mat. = maternal; sig. = statistically significant.
7
8
This document is a draft for review purposes only and does not constitute Agency policy.
616 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
significant increase in displaced right ovary were observed upon fetal skeletal and soft tissue
evaluation. Mixed function oxidase enzymes (ethoxycoumarin and ethoxyresorbin) which are
indicative of cytochrome P450 and P448 activities, respectively, were measured in the livers of
dams and fetuses, but no treatment-related findings were identified. Postnatal growth was
significantly (p < 0.05) decreased in the group with gestation-only exposures. Postnatal
behavioral studies, consisting of an automated assessment of ambulatory response in a novel
environment on GDs 10, 20, and 100, did not identify any effect on general motor activity of
offspring following in utero exposure to TCE.
In a study by Schwetz et al. (1975), pregnant Sprague-Dawley rats and Swiss Webster
mice (30-40 dams/group) were exposed to TCE via inhalation at a concentration of 300 ppm for
7 hours/day on GDs 6-15. The only adverse finding reported was a statistically significant
4-5% decrease in maternal rat body weight. There were no treatment related effects on pre- and
postimplantation loss, litter size, fetal body weight, crown-rump length, or external, soft tissue,
or skeletal findings.
Hardin et al. (1981) summarized the results of inhalation developmental toxicology
studies conducted in pregnant Sprague-Dawley rats and New Zealand white rabbits for a number
of industrial chemicals, including TCE. Exposure concentrations of 0 or 500 ppm TCE were
administered for 6-7 hours/day, on gestations days 1-19 (rats) or 1-24 (rabbits), and cesarean
sections were conducted on GDs 21 or 30, respectively. There were no adverse findings in
maternal animals. No statistically significant increase in the incidence of malformations was
reported for either species; however, the presence of hydrocephaly in two fetuses of two TCE-
treated rabbit litters was interpreted as a possible indicator of teratogenic potential.
Healy et al. (1982) did not identify any treatment-related fetal malformations following
"3
inhalation exposure of pregnant inbred Wistar rats to 0 or 100 ppm (535 mg/m ) on GD 8-21. In
this study, significant differences between control and treated litters were observed as an
increased incidence of total litter loss (p < 0.05), decreased mean fetal weight (p < 0.05), and
increased incidence of minor ossification variations (p = 0.003) (absent or bipartite centers of
ossification).
Carney et al. (2006) investigated the effects of whole-body inhalation exposures to
pregnant Sprague-Dawley rats at nominal (and actual) chamber concentrations of 0, 50, 150, or
600 ppm TCE for 6 hours/day, 7 days/week, on GDs 6-20. This study was conducted under
Good Laboratory Practice regulations according to current EPA and Organisation for Economic
Co-operation and Development (OECD) regulatory testing guidelines (i.e., OPPTS 870.3700 and
OECD GD 414). Maternal toxicity consisted of a statistically significant decrease (22%) in body
weight gain during the first 3 days of exposure to 600-ppm TCE, establishing a no-observed-
effect concentration (NOEC) of 150 ppm for dams. No significant difference between control
This document is a draft for review purposes only and does not constitute Agency policy.
617 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
and TCE-treated groups was noted for pregnancy rates, number of corpora lutea, implantations,
viable fetuses per litter, percentage pre- and postimplantation loss, resorption rates, fetal sex
ratios, or gravid uterine weights. External, soft tissue, and skeletal evaluation of fetal specimens
did not identify any treatment-related effects. No cardiac malformations were identified in
treated fetuses. The fetal NOEC for this study was established at 600 ppm.
Westergren et al. (1984) examined brain specific gravity of NMRI mice pups following
developmental exposures to TCE. Male and female mice were separately exposed 24 hours/day
(except for limited periods of animal husbandry activities) to 0- or 150-ppm TCE for 30 days and
mated during exposure for 7 days. Exposure of the females was continued throughout gestation,
until the first litter was born. Offspring (6-12/group; litter origin not provided in report) were
terminated by decapitation on PND 1, 10, 21-22, or 30. The specific gravity of the brain frontal
cortex, cortex, occipital cortex, and cerebellum were measured. The cortex specific gravity was
significantly decreased at PND 1 (p < 0.001) and 10 (p < 0.01) in pups from exposed mice.
There were also significant differences (p < 0.05) in the occipital cortex and cerebellum at
PND 20-22. This was considered suggestive of delayed maturation. No significant differences
between control and treated pups were observed at 1 month of age.
4.8.3.1.6.2.Oral exposures
A screening study conducted by Narotsky and Kavlock (1995) assessed the
developmental toxicity potential of a number of pesticides and solvents, including TCE. In this
study, Fischer 344 rats were administered TCE by gavage at 0, 1,125, and 1,500 mg/kg-day on
GDs 6-19, and litters were examined on GDs 1, 3, and 6. TCE-related increased incidences of
full-litter resorptions, decreased litter sizes, and decreased mean pup birth weights were observed
at both treatment levels. Additionally, TCE treatment was reported to be associated with
increased incidences of eye abnormalities (microphthalmia or anophthalmia). Increased
incidences of fetal loss and percentage of pups with eye abnormalities were confirmed by
Narotsky et al. (1995) in a preliminary dose-setting study that treated Fischer 344 rats with TCE
by gavage doses of 475, 633, 844, or 1,125 mg/kg-day on GDs 6-15, and then in a 5 x 5 x 5
mixtures study that used TCE doses of 0, 10.1, 32, 101, and 320 mg/kg-day on GD 6-15. In
both studies, dams were allowed to deliver, and pups were examined postnatally. The incidence
of ocular defects observed across all TCE treatment levels tested is presented in Table 4-97.
Other developmental findings in this study included increased full litter resorption at 475,
844, and 1,125 mg/kg-day; increased postnatal mortality at 425 mg/kg-day. Pup body weights
were decreased (not significantly) on PND 1 and 6 at 1,125 mg/kg-day. In both the Narotsky
and Kavlock (1995) and Narotsky et al. (1995) studies, significantly decreased maternal body
This document is a draft for review purposes only and does not constitute Agency policy.
618 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
weight gain was observed at the same treatment levels at which full litter resorption was noted.
Additionally, in Narotsky et al. (1995) maternal observations included delayed parturition at 475,
844, and 1,125 mg/kg-day, ataxia at 633 mg/kg-day, and mortality at 1,125 mg/kg-day.
Cosby and Dukelow (1992) administered TCE in corn oil by gavage to female B6D2F1
mice (28-62/group) on GDs 1-5, 6-10, or 11-15 (where mating = GD 1). Dose levels were 0,
1/100 and 1/10 of the oral LD50 (i.e., 0, 24.02, and 240.2 mg/kg body weight). Dams were
allowed to deliver; litters were evaluated for pup count sex, weight, and crown-rump length until
weaning (PND 21). Some litters were retained until 6 weeks of age at which time gonads (from
a minimum of 2 litters/group) were removed, weighed, and examined. No treatment-related
reproductive or developmental abnormalities were observed.
A single dose of TCE was administered by gavage to pregnant CD-I mice (9-19/group)
at doses of 0, 0.1, or 1.0 (J,g/kg in distilled water, or 0, 48.3, or 483 mg/kg in olive oil, 24 hours
after premating human chorionic gonadotropin (hCG) injection (Coberly et al., 1992). At
53 hours after the hCG-injection, the dams were terminated, and the embryos were flushed from
excised oviducts. Chimera embryos were constructed, cultured, and examined. Calculated
proliferation ratios did not identify any differences between control and treated blastomeres. A
lack of treatment-related adverse outcome was also noted when the TCE was administered by i.p.
injection to pregnant mice (16-39/group) at 24 and 48 hours post-hCG at doses of 0, 0.01, 0.02,
or 10 (J,g/kg body weight.
In a study intended to confirm or refute the cardiac teratogenicity of TCE that had been
previously observed in chick embryos, Dawson et al. (1990b) continuously infused the gravid
uterine horns of Sprague-Dawley rats with solutions of 0-, 15-, or 1,500-ppm TCE (or 1.5 or
150-ppm dichloroethylene) on GDs 7-22. At terminal cesarean section on GD 22, the uterine
contents were examined, and fetal hearts were removed and prepared for further dissection and
examination under a light microscope. Cardiac malformations were observed in 3% of control
fetuses, 9% of the 15-ppm TCE fetuses (p = 0.18), and 14% of the 1,500-ppm TCE fetuses, (p =
0.03). There was a >60% increase in the percentage of defects with a 100-fold increase in dose.
No individual malformation or combination of abnormalities was found to be selectively induced
by treatment.
To further examine these TCE-induced cardiac malformations in rats, Dawson et al.
(1993) administered 0, 1.5 or 1,100-ppm TCE in drinking water to female Sprague-Dawley rats.
Experimental treatment regimens were (1) a period of approximately 2 months prior to
pregnancy plus the full duration of pregnancy, (2) the full duration of pregnancy only, or (3) an
average of 3 months before pregnancy only. The average total daily doses of TCE consumed for
each exposure group at both dose levels were
This document is a draft for review purposes only and does not constitute Agency policy.
619 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
1.5 ppm 1,100 ppm
Group 1
Group 2
Group 3
23.5 |j,L 1,206 |j,L
0.78 |j,L 261 |j,L
3.97 nL 1,185 nL
The study also evaluated 0, 0.15, or 110-ppm dichloroethylene in drinking water, with treatment
administered (1) two months prior to pregnancy plus the full duration of pregnancy, or (2) an
average of 2 months before pregnancy only. At terminal cesarean section, uterine contents were
examined, fetuses were evaluated for external defects, and the heart of each fetus was removed
for gross histologic examination under a dissecting microscope, conducted without knowledge of
treatment group. There were no differences between TCE-treated and control group relative to
percentage of live births, implants, and resorptions. The percentage of cardiac defects in
TCE-treated groups ranged from 8.2-13.0%, and was statistically significant as compared to the
control incidence of 3%. The dose-response was relatively flat, even in spite of the extensive
difference between the treatment levels. There was a broad representation of various types of
cardiac abnormalities identified, notably including multiple transposition, great artery, septal,
and valve defects (see Table 4-99). No particular combination of defects or syndrome
predominated. Exposure before pregnancy did not appear to be a significant factor in the
incidence of cardiac defects.
In an attempt to determine a threshold for cardiac anomalies following TCE exposures,
Johnson et al. (2003) 2005 compiled and reanalyzed data from five studies conducted from
1989-1995. In these studies, TCE was administered in drinking water to Sprague-Dawley rats
throughout gestation (i.e., a total of 22 days) at levels of 2.5 ppb (0.0025 ppm), 250 ppb
(0.25-ppm), 1.5, or 1,100 ppm. The dams were terminated on the last day of pregnancy and
fetuses were evaluated for abnormalities of the heart and great vessels. The control data from the
five studies were combined prior to statistical comparison to the individual treated groups, which
were conducted separately. The study author reported that significant increases in the percentage
of abnormal hearts and the percentage of litters with abnormal hearts were observed in a
generally dose-responsive manner at 250 ppb and greater (see Table 4-100).
In a study by Fisher et al. (2001), pregnant Sprague-Dawley rats were administered daily
gavage doses on GDs 6-15 of TCE (500 mg/kg-day), TCA (300 mg/kg-day), or DCA
(300 mg/kg-day). Cesarean delivery of fetuses was conducted on GD 21. Water and soybean oil
negative control groups, and a retinoic acid positive control group were also conducted
simultaneously. Maternal body weight gain was not significantly different from control for any
of the treated groups. No significant differences were observed for number of implantations,
This document is a draft for review purposes only and does not constitute Agency policy.
620 DRAFT—DO NOT CITE OR QUOTE
-------
1
2 Table 4-99. Types of congenital cardiac defects observed in TCE-exposed
3 fetuses (Dawson et al., 1993, Table 3)
4
TCE concentrations
Premating
Premating/gestation
Gestation only
Cardiac abnormalities
Control
1,100 ppm
1.5 ppm
1,100 ppm
1.5 ppm
1,100 ppm
1.5 ppm
d-transposition (right chest)
2
1-transposition (left chest)
2
1
Great artery defects
1
2
1
Atrial septal defects
1
7
3
19
5
7
4
Mitral valve defects
5
8
Tricuspid valve defects
1
1
2
Ventricular septal defects
Subaortic
1
4
1
1
2
Membranous
2
Muscular
2
1
1
4
4
1
Endocardial cushion defect
1
1
Pulmonary valve defects
3
2
1
1
Aortic valve defects
1
2
2
2
Situs inversus
1
Total abnormalities
7
9
8
41
23
15
10
Total abnormal hearts
7
9
8
40
23
11
9
5
6
This document is a draft for review purposes only and does not constitute Agency policy.
621 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-100. Types of heart malformations per 100 fetuses (Johnson et al.,
2 2003, Table 2, p. 290)
3
Type of defect/100 fetuses
Control
TCE dose group
1,100 ppm
1.5 ppm
250 ppb
2.5 ppb
Abnormal looping
0.33
1
Coronary artery/sinus
1.82
Aortic hypoplasia
0.55
Pulmonary artery hypoplasia
0.55
Atrial septal defect
1.16
6.67
2.21
0.91
Mitral valve defect
0.17
0.91
Tricuspid valve defect
0.91
Ventricular septal defect
Perimembranous (subaortic)
0.33
2.86
1.66
Muscular
0.33
0.95
0.55
Atriventricular septal defect
0.17
0.95
Pulmonary valve defect
Aortic valve defects
1.9
0.91
Fetuses with abnormal hearts (n)
13
11
9
5
0
Total fetuses (n)
606
105
181
110
144
Litters with fetuses with abnormal hearts/litter («)
9/55
6/9
5/13
4/9
0/12
Litter with fetuses with abnormal hearts/number litters (%)
16.4
66.7
38.5
44.4
0.0
4
5
6 resorptions, or litter size. Mean fetal body weight was reduced by treatment with TCA and
7 DC A. The incidence of heart malformations was not significantly increased in treated groups as
8 compared to controls. The fetal rate of cardiac malformations ranged from 3-5% across the
9 TCE, TCA, and DC A dose groups and from 6.5-2.9% for the soybean and water control dose
10 groups, respectively. It was suggested that the apparent differences between the results of this
11 study and the Dawson et al. (1993) study may be related to factors such as differences in purity
12 of test substances or in the rat strains, or differences in experimental design (e.g., oral gavage
13 versus drinking water, exposure only during the period of organogenesis versus during the entire
14 gestation period, or the use of a staining procedure). The rats from this study were also
15 examined for eye malformations to follow-up on the findings of Narotsky (1995). As reported in
16 Warren et al. (2006), gross evaluation of the fetuses as well as computerized morphometry
This document is a draft for review purposes only and does not constitute Agency policy.
622 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
conducted on preserved and sectioned heads revealed no ocular anomalies in the groups treated
with TCE. This technique allowed for quantification of the lens area, global area, medial
canthus, distance, and interlocular distance. DCA treatment was associated with statistically
significant reductions in the lens area, globe area, and interlocular distance. All four measures
were reduced in the TCA-treated group, but not significantly. The sensitivity of the assay was
demonstrated successfully with the use of a positive control group that was dosed on GD 6-15
with a known ocular teratogen, retinoic acid (15 mg/kg-day).
Johnson et al. (1998a; 1998b) conducted a series of studies to determine whether specific
metabolites of TCE or dichloroethylene were responsible for the cardiac malformations observed
in rats following administration during the period of organogenesis. Several metabolites of the
two chemicals were administered in drinking water to Sprague-Dawley rats from GD 1-22.
These included carboxy methylcystine, dichloroacetaldehyde, dichlorovinyl cystine,
monochloroacetic acid, trichloroacetic acid, trichloroacetaldehyde, and trichloroethanol.
Dichloroacetic acid, a primary common metabolite of TCE and dichloroethylene, was not
included in these studies. The level of each metabolite administered in the water was based upon
the dosage equivalent expected if 1,100 ppm (the limit of solubility) TCE broke down
completely into that metabolite. Cesarean sections were performed on GD 22, uterine contents
were examined, and fetuses were processed and evaluated for heart defects according to the
procedures used by Dawson et al. (1993). No treatment-related maternal toxicity was observed
for any metabolite group. Adverse fetal outcomes were limited to significantly increased
incidences of fetuses with abnormal hearts (see Table 4-101). Significant increases in fetuses
with cardiac defects (on a per-fetus and per-litter basis) were observed for only one of the
metabolites evaluated, i.e., trichloroacetic acid (2,730 ppm, equivalent to a dose of 291
mg/kg-day). Notably, significant increases in fetuses with cardiac malformations were also
observed with 1.5 or 1,100-ppm TCE (0.218 or 129 mg/kg-day), or with 0.15 or 110-ppm DCE
(0.015 or 10.64 mg/kg-day), but in each case only with prepregnancy-plus-pregnancy treatment
regimens. The cardiac abnormalities observed were diverse and did not segregate to any
particular anomaly or grouping. Dose related increases in response were observed for the overall
number of fetuses with any cardiac malformation for both TCE and DCE; however, no dose-
related increase occurred for any specific cardiac anomaly (Johnson et al., 1998b).
The TCE metabolites TCA and DCA were also studied by Smith et al. (1992; 1989).
Doses of 0, 330, 800, 1,200, or 1,800 mg/kg TCA were administered daily by oral gavage to
Long-Evan hooded rats on GDs 6-15. Similarly, DCA was administered daily by gavage to
Long-Evans rats on GD 6-15 in two separate studies, at 0, 900, 1,400, 1,900, or
2,400 mg/kg-day and 0, 14, 140, or 400 mg/kg-day. Embryo lethality and statistically or
This document is a draft for review purposes only and does not constitute Agency policy.
623 DRAFT—DO NOT CITE OR QUOTE
-------
1 biologically significant incidences of orbital anomalies (combined soft tissue and skeletal
2 findings) were observed for TCA at >800 mg/kg-day, and for DCA at >900 mg/kg-day. Fetal
This document is a draft for review purposes only and does not constitute Agency policy.
624 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-101. Congenital cardiac malformations (Johnson et al., 1998a, Table 2, p. 997)
Heart abnormalities
Treatment group
Normal
water
TCE
p+p
1,100
ppm
TCE
p+p
1.5
ppm
TCE
P
1,100
ppm
DCE
p+p
110
ppm
DCE
p+p
0.15
ppm
TCAA
P
2,730
ppm
MCAA
P
1,570
ppm
TCEth
P
1,249
ppm
TCAld
P
1,232
ppm
DCAld
P
174
ppm
CMC
P
473
ppm
DCVC
P
50
ppm
Abnormal looping
2
-
2
-
-
-
-
-
-
-
-
-
-
Aortic hypoplasia
-
1
1
-
1
-
1
-
1
-
1
-
1
Pulmonary artery hypoplasia
-
-
1
-
-
-
2
1
-
-
2
-
-
Atrial septal defects
7
19
5
7
11
7
3
3
-
2
-
-
1
Mitral valve defects, hypoplasia or ectasia
1
5
8
-
4
3
1
-
1
2
-
-
1
Tricuspid valve defects, hypoplasia or ectasia
-
1
1
-
1
-
-
-
1
-
-
-
-
Ventricular septal defects
Perimcmbranous"
2
6
2
1
4
1
4
-
-
3
-
1
-
Muscular
2
4
-
4
2
1
1
-
1
-
-
2
2
Atrioventricual septal defects
1
-
-
1
1
-
-
-
-
-
-
-
-
Pulmonary valve defects
-
2
1
-
1
-
1
3
1
1
-
-
-
Aortic valve defects
-
2
2
2
2
3
-
-
1
-
-
1
-
Situs inversus
-
1
-
-
-
-
-
-
-
-
-
-
-
Total
Abnormal hearts
15
41
23
15
25
15
13
7
6
8
3
4
5
Fetuses with abnormal hearts
13
40a
22a
lla
24a
14a
12a
6
5
8
3
4
5
Fetuses
605
434
255
105
184
121
114
132
121
248
101
85
140
Subaortic.
bPer-fetus statistical significance (Fisher's exact test),
p = pregnancy ; p+p = pregnancy; and prepregnancy.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
growth (body weight and crown-rump length) was affected at >330 mg/kg-day for TCE and at
>400 mg/kg-day for DCA. For TCA, the most common cardiac malformations observed were
levocardia at >330 mg/kg-day and interventricular septal defect at >800 mg/kg-day. For DCA,
levocardia was observed at >900 mg/kg-day, interventricular septal defect was observed at
>1,400 mg/kg-day, and a defect between the ascending aorta and right ventricle was observed in
all treated groups (i.e., >14 mg/kg-day, although the authors appeared to discount the single fetal
finding at the lowest dose tested). Thus, NOAELs were not definitively established for either
metabolite, although it appears that TCA was generally more potent than DCA in inducing
cardiac abnormalities.
These findings were followed up by a series of studies on DCA reported by Epstein et al.
(1992), which were designed to determine the most sensitive period of development and further
characterize the heart defects. In these studies, Long-Evans hooded rats were dosed by oral
gavage with a single dose of 2,400 mg/kg-day on selected days of gestation (6-8, 9-11, or
12-15); with a single dose of 2,400 mg/kg on Days 10, 11, 12, or 13; or with a single dose of
3,500 mg/kg on Days 9, 10, 11, 12, or 13. The heart defects observed in these studies were
diagnosed as high interventricular septal defects rather than membranous type interventricular
septal defects. The authors hypothesized that high intraventricular septal defects are a specific
type of defect produced by a failure of proliferating interventricular septal tissue to fuse with the
right tubercle of the atrioventricular cushion tissue. This study identified GDs 9 through 12 as a
particularly sensitive period for eliciting high interventricular septal defects. It was postulated
that DCA interferes with the closure of the tertiary interventricular foramen, allowing the aorta to
retain its embryonic connection with the right ventricle. Further, it was suggested that the
selectivity of DCA in inducing cardiac malformations may be due to the disruption of a discrete
cell population.
TCE and its metabolites DCE and TCAA were administered in drinking water to
pregnant Sprague-Dawley rats from GDs 0-11 (Collier et al., 2003). Treatment levels were 0,
110, or 1,100 ppm (i.e., 0, 830 or 8,300 (J,gM) TCE; 0, 11, or 110 ppm (i.e., 0, 110, or 1,100
(j,gM) DCE; 0, 2.75, or 27.3 mg/mL (i.e., 0, 10, or 100 mM) TCAA. Embryos (including hearts)
were harvested between embryonic days 10.5-11, since this is the stage at which the
developmental processes of myoblast differentiation, cardiac looping, atrioventricular valve
formation, and trabeculation would typically be occurring. A PCR based subtraction scheme
was used to identify genes that were differentially regulated with TCE or metabolite exposure.
Numerous differentially regulated gene sequences were identified. Up-regulated transcripts
included genes associated with stress response (Hsp 70) and homeostasis (several ribosomal
proteins). Down-regulated transcripts included extracellular matrix components (GPI-pl37 and
vimentin) and Ca2+ responsive proteins (Serca-2 Ca2+-ATPase and P-catenin). Serca-2 Ca2+ and
This document is a draft for review purposes only and does not constitute Agency policy.
626 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
GPI-pl37 were identified as two possible markers for fetal TCE exposure. Differential
regulation of expression of these markers by TCE was confirmed by dot blot analysis and
semiquantitative real time PCR with decreased expression seen at levels of TCE exposure
between 100 and 250 ppb (0.76 and 1.9 (xM).
4.8.3.1.6.2.1. Developmental neurotoxicity and developmental immunotoxicity
Several studies were conducted that included assessments of the effects of TCE oral
exposure on the developing nervous system (Blossom et al., 2008; Dorfmueller et al., 1979;
Fredriksson et al., 1993; George et al., 1986; Isaacson and Taylor, 1989; Noland-Gerbec et al.,
1986) or immune system (Blossom and Doss, 2007; Blossom et al., 2008; Peden-Adams et al.,
2006; Peden-Adams et al., 2008). These studies, summarized below, are addressed in additional
detail in Section 4.3 (nervous system) and Section 4.6.2.1.2 (immune system).
4.8.3.1.6.2.2. Developmental neurotoxicity
Fredriksson et al. (1993) conducted a study in male NMRI weanling mice (12/group,
selected from 3-4 litters), which were exposed to trichloroethylene by oral gavage at doses of 0
(vehicle), 50, or 290 mg/kg-day TCE in a fat emulsion vehicle, on PNDs 10-16. Locomotor
behavior (horizontal movement, rearing and total activity) were assessed over three 20-minute
time periods at GDs 17 and 60. There were no effects of treatment in locomotor activity at PND
17. At PND 60, the mice treated with 50 and 290 mg/kg-day TCE showed a significant
(p < 0.01) decrease in rearing behavior at the 0-20 and 20-40 minute time points, but not at the
40-60 minute time point. Mean rearing counts were decreased by over 50% in treated groups as
compared to control. Horizontal activity and total activity were not affected by treatment.
Open field testing was conducted in control and high-dose F1 weanling Fischer 344 rat
pups in an NTP reproduction and fertility study with continuous breeding (George et al., 1986).
In this study, TCE was administered at dietary levels of 0, 0.15, 0.30, or 0.60%. The open field
testing revealed a significant (p < 0.05) dose-related trend toward an increase in the time required
for male and female pups to cross the first grid in the testing device, suggesting an effect on the
ability to react to a novel environment.
Taylor et al. (1985) administered TCE in drinking water (0, 312, 625, or 1,250 ppm) to
female Sprague-Dawley rats for 14 days prior to breeding, and from GD 0 through offspring GD
21. The number of litters/group was not reported, nor did the study state how many pups per
litter were evaluated for behavioral parameters. Exploratory behavior was measured in the pups
in an automated apparatus during a 15-minute sampling period on PND 28, 60, and 90.
This document is a draft for review purposes only and does not constitute Agency policy.
627 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Additionally, wheel-running, feeding, and drinking behavior was monitored 24 hours/day on
PND 55-60. The number of exploratory events was significantly increased by approximately
25-50% in 60- and 90-day old male TCE-treated rats at all dose levels, with the largest effect
observed at the highest dose level tested, although there were no effects of treatment on the
number of infrared beam-breaks. No difference between control and treated rats was noted for
pups tested on PND 28. Wheel-running activity was increased approximately 40% in 60-day old
males exposed to 1,25-ppm TCE as compared to controls. It is notable that adverse outcomes
reported in the developmentally-exposed offspring on this study were observed long after
treatment ceased.
Using a similar treatment protocol, the effects of TCE on development of myelinated
axons in the hippocampus was evaluated by Isaacson and Taylor (1989) in Sprague-Dawley rats.
Female rats (6/group) were exposed in the drinking water from 14 days prior to breeding and
through the mating period; then the dams and their pups were exposed throughout the prenatal
period and until PND 21, when they were sacrificed. The dams received 0, 312 or 625 ppm (0,
4, or 8.1 mg/day TCE in the drinking water. Myelinated fibers were counted in the hippocampus
of 2-3 pups per treatment group at PND 21, revealing a decrease of approximately 40% in
myelinated fibers in the CA1 area of the hippocampus of pups from dams at both treatment
levels, with no dose-response relationship. There was no effect of TCE treatment on myelination
in several other brain regions including the internal capsule, optic tract or fornix.
A study by Noland-Gerbec et al. (1986) examined the effect of pre- and perinatal
exposure to TCE on 2-deoxyglucose (2-DG) uptake in the cerebellum, hippocampus and whole
brain of neonatal rats. Sprague-Dawley female rats (9-11/group) were exposed via drinking
water to 0 or 312 mg TCE/liter distilled water from 14 days prior to mating until their pups were
euthanized at GD 21. The total TCE dose received by the dams was 825 mg over the 61-day
exposure period. Pairs of male neonates were euthanized on PND 7, 11, 16, and 21. There was
no significant impairment in neonatal weight or brain weight attributable to treatment, nor were
other overt effects observed. 2-DG uptake was significantly reduced from control values in
neonatal whole brain (9—11%) and cerebellum (8—16%) from treated rats at all ages studied, and
hippocampal 2-DG uptake was significantly reduced (7—21 % from control) in treated rats at all
ages except at PND 21.
In a study by Blossom et al. (2008), MRL +/+ mice were treated in the drinking water
with 0 or 0.1 mg/mL TCE from maternal GD 0 through offspring PND 42. Based on drinking
water consumption data, average maternal doses of TCE were 25.7 mg/kg-day, and average
offspring (PND 24-42) doses of TCE were 31.0 mg/kg-day. In this study, a subset of offspring
(3 randomly selected neonates from each litter) was evaluated for righting reflex on PNDs 6, 8,
and 10; bar-holding ability on PNDs 15 and 17; and negative geotaxis on PNDs 15 and 17; none
This document is a draft for review purposes only and does not constitute Agency policy.
628 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
of these were impaired by treatment. In an assessment of offspring nest building on PND 35,
there was a significant association between impaired nest quality and TCE exposure; however,
TCE exposure did not have an effect on the ability of the mice to detect social and nonsocial
odors on PND 29 using olfactory habituation and dishabituation methods. Resident intruder
testing conducted on PND 40 to evaluate social behaviors identified significantly more
aggressive activities (i.e., wrestling and biting) in TCE-exposed juvenile male mice as compared
to controls. Cerebellar tissue homogenates from the male TCE-treated mice had significantly
lower GSH levels and GSH:GSSG ratios, indicating increased oxidative stress and impaired thiol
status; these have been previously reported to be associated with aggressive behaviors (Franco et
al., 2006). Qualitative histopathological examination of the brain did not identify alterations
indicative of neuronal damage or inflammation. Although the study author attempted to link the
treatment-related alterations in social behaviors to the potential for developmental exposures to
TCE to result in autism in humans, this association is not supported by data and is considered
speculative at this time.
As previously noted, postnatal behavioral studies conducted by Dorfmueller et al. (1979)
did not identify any changes in general motor activity measurements of rat offspring on PND 10,
20, and 100 following maternal gestational inhalation exposure to TCE at 1,800 ± 200 ppm.
4.8.3.1.6.2.3. Developmental immunotoxicity
Peden-Adams et al. (2006) assessed the potential for developmental immunotoxicity
following TCE exposures. In this study, B6C3F1 mice (5/sex/group) were administered TCE via
drinking water at dose levels of 0, 1,400 or 14,000 ppb from maternal GD 0 to either postnatal 3
or 8, when offspring lymphocyte proliferation, NK cell activity, SRBC-specific IgM production
(PFC response), splenic B220+ cells, and thymus and spleen T-cell immunophenotypes were
assessed. (A total of 5-7 pups per group were evaluated at Week 3, and the remainder were
evaluated at Week 8.) Observed positive responses consisted of suppressed PFC responses in
males at both ages and both TCE treatment levels, and in females at both ages at 14,000 ppb and
at 8 weeks of age at 1,400 ppb. Spleen numbers of B220+ cells were decreased in 3-week old
pups at 14,000 ppb. Pronounced increases in all thymus T-cell subpopulations (CD4+, CD8+,
CD4+/CD8+, and CD4-/CD8-) were observed at 8 weeks of age. Delayed hypersensitivity
response, assessed in offspring at 8 weeks of age, was increased in females at both treatment
levels and in males at 14,000 ppb only. No treatment-related increase in serum anti-dsDNA
antibody levels was found in the offspring at 8 weeks of age.
In a study by Blossom and Doss (2007), TCE was administered to groups of pregnant
MRL +/+ mice in drinking water at levels of 0, 0.5, or 2.5 mg/mL. TCE was continuously
This document is a draft for review purposes only and does not constitute Agency policy.
629 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
administered to the offspring until young adulthood (i.e., 7-8 weeks of age). Offspring
postweaning body weights were significantly decreased in both treated groups. Decreased
spleen cellularity and reduced numbers of CD4+, CD8+, and B220+ lymphocyte subpopulations
were observed in the postweaning offspring. Thymocyte development was altered by TCE
exposures (significant alterations in the proportions of double-negative subpopulations and
inhibition of in vitro apoptosis in immature thymocytes). A dose-dependent increase in CD4+
and CD8+ T-lymphocyte IFNy was observed in peripheral blood by 4-5 weeks of age, although
these effects were no longer observed at 7-8 weeks of age. Serum antihistone autoantibodies
and total IgG2a were significantly increased in treated offspring; however, no histopathological
signs of autoimmunity were observed in the liver and kidneys at sacrifice.
Blossom et al. (2008) administered TCE to MRL +/+ mice (8 dams/group) in the drinking
water at levels of 0 or 0.1 mg/mL from GD 0 through offspring GD 42. Average maternal doses
of TCE were 25.7 mg/kg-day, and average offspring (PND 24-42) doses of TCE were 31.0
mg/kg-day. Subsets of offspring were sacrificed at PND 10 and 20, and thymus endpoints (i.e.,
total cellularity, CD4+/CD8+ ratios, CD24 differentiation markers, and double-negative
subpopulation counts) were evaluated. Evaluation of the thymus identified a significant
treatment-related increase in cellularity, accompanied by alterations in thymocyte subset
distribution, at PND 20 (sexes combined). TCE treatment also appeared to promote T-cell
differentiation and maturation at PND 42. Indicators of oxidative stress were measured in the
thymus at PND 10 and 20, and in the brain at PND 42,.and ex vivo evaluation of cultured
thymocytes indicated increased ROS generation. Mitogen-induced intracellular cytokine
production by splenic CD4+ and CD8+ T-cells was evaluated in juvenile mice and brain tissue
was examined at PND 42 for evidence of inflammation. Evaluation of peripheral blood
indicated that splenic CD4+ T-cells from TCE-exposed PND 42 mice produced significantly
greater levels of IFN-y and IL-2 in males and TNF-a in both sexes. There was no effect on
cytokine production on PND 10 or 20.
Peden-Adams et al. (2008) administered TCE to MRL+/+ mice (unspecified number of
dams/group) in drinking water at levels of 0, 1,400, or 14,000 ppb from GD 0 and continuing
until the offspring were 12 months of age. At 12 months of age, final body weight; spleen,
thymus, and kidney weights; spleen and thymus lymphocyte immunophenotyping (CD4 or
CD8); splenic B-cell counts; mitogen-induced splenic lymphocyte proliferation; serum levels of
autoantibodies to dsDNA and GA, periodically measured from 4-12 months of age; and urinary
protein measures were recorded. Reported sample sizes for the offspring measurements varied
from 6-10 per sex per group; the number of source litters represented within each sample was
not specified. The only organ weight alteration was an 18% increase in kidney weight in the
1,400 ppb males. Splenic CD4-/CD8- cells were altered in female mice (but not males) at 1,400
This document is a draft for review purposes only and does not constitute Agency policy.
630 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
ppm only. Splenic T-cell populations, numbers of B220+ cells, and lymphocyte proliferation
were not affected by treatment. Populations of thymic T-cell subpopulations (CD8+,
CD4-/CD8-, and CD4+) were significantly decreased in male but not female mice following
exposure to 14,000 ppb TCE, and CD4+/CD8+ cells were significantly reduced in males by
treatment with both TCE concentrations. Autoantibody levels (anti-dsDNA and anti-GA) were
not increased in the offspring over the course of the study.
Although all of the developmental immunotoxicity studies with TCE (Peden-Adams et
al., 2006, 2008; Blossom and Doss, 2007; Blossom et al., 2008) exposed the offspring during
critical periods of pre- and postnatal immune system development, they were not designed to
assess issues such as posttreatment recovery, latent outcomes, or differences in severity of
response that might be attributed to the early life exposures.
4.8.3 .1.6.3 .Intraperitoneal exposures
The effect of TCE on pulmonary development was evaluated in a study by Das and Scott
(1994). Pregnant Swiss-Webster mice (5/group) were administered a single intraperitoneal
injection of TCE in peanut oil at doses of 0 or 3,000 mg/kg on GD 17 (where mating = Day 1).
Lungs from GD 18 and 19 fetuses and from neonates on PND 1, 5, and 10 were evaluated for
phospholipid content, DNA, and microscopic pathology. Fetal and neonatal (PND 1) mortality
was significantly increased (p < 0.01) in the treated group. Pup body weight and absolute lung
weight were significantly decreased (p < 0.05) on PND 1, and mean absolute and relative (to
body weight) lung weights were significantly decreased on GDs 18 and 19. Total DNA content
([j,g/mg lung) was similar between control and treated mice, but lung phospholipid was
significantly (p < 0.05) reduced on GD 19 and significantly increased (p < 0.05) on PND 10 in
the TCE-treated group. Microscopic examination revealed delays in progressive lung
morphological development in treated offspring, first observed at GD 19 and continuing at least
through PND 5.
4.8.3.1.7. Studies in nonmammalian species
4.8.3.1.7.l.Avian
Injection of White Leghorn chick embryos with 1, 5, 10, or 25 [j,mol TCE per egg on
Days 1 and 2 of embryogenesis demonstrated mortality, growth defects, and morphological
anomalies at evaluation on Day 14 (Bross et al., 1983). These findings were consistent with a
This document is a draft for review purposes only and does not constitute Agency policy.
631 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
previous study that had been conducted by Elovaara et al. (1979). Up to 67% mortality was
observed in the treated groups, and most of the surviving embryos were malformed (as compared
to a complete absence of malformed chicks in the untreated and mineral-oil-treated control
groups). Reported anomalies included subcutaneous edema, evisceration (gastroschisis), light
dermal pigmentation, beak malformations, club foot, and patchy feathering. Retarded growth
was observed as significantly (p < 0.05) reduced crown-rump, leg, wing, toe, and beak lengths as
compared to untreated controls. This study did not identify any liver damage or cardiac
anomalies.
In a study by Loeber et al. (1988), 5, 10, 15, 20, or 25 [j.mol TCE was injected into the air
space of White Longhorn eggs at embryonic stages 6, 12, 18, or 23. Embryo cardiac
development was examined in surviving chicks in a double-blinded manner at stages 29, 34, or
44. Cardiac malformations were found in 7.3% of TCE-treated hearts, compared to 2.3% of
saline controls and 1.5% of mineral oil controls. The observed defects included septal defects,
cor biloculare, conotruncal abnormalities, atrioventricular canal defects, and abnormal cardiac
muscle.
Drake et al. (2006a) injected embryonated White Leghorn chicken eggs (Babcock or
Bovan strains) with 0, 0.4, 8, or 400 ppb TCE per egg during the period of cardiac valvuloseptal
morphogenesis (i.e., 2-3.3 days incubation). The injections were administered in four aliquots at
Hamberger and Hamilton (HH) stages 13, 15, 17, and 20, which spanned the major events of
cardiac cushion formation, from induction through mesenchyme transformation and migration.
Embryos were harvested 22 hours after the last injection (i.e., HH 24 or HH 30) and evaluated
for embryonic survival, apoptosis, cellularity and proliferation, or cardiac function. Survival was
significantly reduced for embryos at 8 and 400 ppb TCE at HH 30. Cellular morphology of
cushion mesenchyme, cardiomyocytes, and endocardiocytes was not affected by TCE treatment;
however, the proliferative index was significantly increased in the atrioventricular canal (AVC)
cushions at both treatment levels and in the outflow tract (OFT) cushions at 8 ppb. This resulted
in significant cushion hypercellularity for both the OFT and AVC of TCE-treated embryos.
Similar outcomes were observed in embryos when TCA or TCOH was administered, and the
effects of TCA were more severe than for TCE. Doppler ultrasound assessment of cardiac
hemodynamics revealed no effects of TCE exposure on cardiac cycle length or heart rate;
however, there was a reduction in dorsal aortic blood flow, which was attributed to a 30.5%
reduction in the active component of atrioventricular blood flow. Additionally the passive-to-
active atrioventricular blood flow was significantly increased in treated embryos, and there was a
trend toward lower stroke volume. The overall conclusion was that exposure to 8 ppb TCE
during cushion morphogenesis reduced the cardiac output of the embryos in this study. The
findings of cardiac malformations and/or mortality following in ovo exposure to chick embryos
This document is a draft for review purposes only and does not constitute Agency policy.
632 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
with 8 ppb TCE during the period of valvuloseptal morphogenesis has also been confirmed by
Ruferetal. (2008; 2010).
In a follow-up study, Drake et al. (2006b) injected embryonated White Leghorn chicken
eggs with TCE or TCA during the critical window of avian heart development, beginning at HH
stage 3+ when the primary heart field is specified in the primitive streak and ending
approximately 50 hours later at HH stage 17, at the onset of chambering. Total dosages of 0, 0.2,
2, 4, 20, or 200 nmol (equivalent to 0, 0.4, 4, 8, 40, or 400 ppb) were injected in four aliquots
into each egg yolk during this window (i.e., at stages 3+, 6, 13, and 17: hours 16, 24, 46, and 68).
Embryos were harvested at 72 hours, 3.5 days, 4 days or 4.25 days (HH stages 18, 21, 23, or 24,
respectively) and evaluated for embryonic survival, cardiac function, or cellular parameters.
Doppler ultrasound technology was utilized to assess cardiovascular effects at HH 18, 21, and
23. In contrast with the results of Drake et al. (2006a), all of the functional parameters assessed
(i.e., cardiac cycle length, heart rate, stroke volume, and dorsal aortic and atrioventricular blood
flow) were similar between control and TCE- or TCA-treated embryos. The authors attributed
this difference in response between studies to dependence upon developmental stage at the time
of exposure. In this case, the chick embryo was relatively resistant to TCE when exposure
occurred during early cardiogenic stages, but was extremely vulnerable when TCE exposure
occurred during valvuloseptal morphogenesis. It was opined that this could explain why some
researchers have observed no developmental cardiac effects after TCE exposure to mammalian
models, while others have reported positive associations.
4.8.3.1.7.2. Amphibian
The developmental toxicity of TCE was evaluated in the Frog Embryo Teratogenesis
Assay: Xenopus by Fort et al. (1991; 1993). Late Xenopus laevis blastulae were exposed to TCE,
with and without exogenous metabolic activation systems, or to TCE metabolites (dichloroacetic
acid, trichloroacetic acid, trichloroethanol, or oxalic acid), and developmental toxicity ensued.
Findings included alterations in embryo growth, and increased types and severity of induced
malformations. Findings included cardiac malformations that were reportedly similar to those
that had been observed in avian studies. It was suggested that a mixed function oxidase-
mediated reactive epoxide intermediate (i.e., TCE-oxide) may play a significant role in observed
developmental toxicity in in vitro tests.
Likewise, McDaniel et al. (2004) observed dose-dependent increases in developmental
abnormalities in embryos of four North American amphibian species (wood frogs, green frogs,
American toads, and spotted salamanders) following 96-hour exposures to TCE. Median
effective concentrations (EC50) for malformations was 40 mg/L for TCE in green frogs, while
This document is a draft for review purposes only and does not constitute Agency policy.
633 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
American toads were less sensitive (with no EC50 at the highest concentration tested—85 mg/L).
Although significant mortality was not observed, the types of malformations noted would be
expected to compromise survival in an environmental context.
4.8.3.1.7.3.Invertebrate
The response of the daphnid Ceriodaphnia dubia to six industrial chemicals, including
TCE, was evaluated by Niederlehner et al. (1998). Exposures were conducted for 6-7 days,
according to standard EPA testing guidelines. Lethality, impairment of reproduction, and
behavioral changes, such as narcosis and abnormal movement, were observed with TCE
exposures. The reproductive sublethal effect concentration value for TCE was found to be
82 |xM.
4.8.3.1.8. In vitro studies
Rat whole embryo cultures were used by Saillenfait et al. (1995) to evaluate the
embryotoxicity of TCE, tetrachloroethylene, and four metabolites (trichloroacetic acid,
dichloroacetic acid, chloral hydrate, and trichloroacetyl chloride). In this study, explanted
embryos of Sprague-Dawley rats were cultured in the presence of the test chemicals for 46 hours
and subsequently evaluated. Concentration-dependant decreases in growth and differentiation,
and increases in the incidence of morphologically abnormal embryos were observed for TCE at
>5 mM.
Whole embryo cultures were also utilized by Hunter et al. (1996) in evaluating the
embryotoxic potential of a number of disinfection by-products, including the TCE metabolites
DCA and TCA. CD-I mouse conceptuses (GD 9; 3-6 somites) were cultured for 24-26 hours in
treated medium. DCA levels assessed were 0, 734, 1,468, 4,403, 5,871, 7,339, 11,010, or
14,680 [xM; TCA levels assessed were 0, 500, 1,000, 2,000, 3,000, 4,000, 5,000 [xM. For DCA,
neural tube defects were observed at levels of >5,871 [xM, heart defects were observed at
>7,339 [xM, and eye defects were observed at levels of >11,010 [xM. For TCA, neural tube
defects were observed at levels of >2,000 [xM, heart and eye defects were observed at
>3,000 [xM. The heart defects for TCA were reported to include incomplete looping, a reduction
in the length of the heart beyond the bulboventricular fold, and a marked reduction in the caliber
of the heart tube lumen. Overall benchmark concentrations (i.e., the lower limit of the 95%
confidence interval required to produce a 5% increase in the number of embryos with neural tube
defects) were 2,451.9 [xM for DCA and 1,335.8 [xM for TCA (Richard and Hunter, 1996).
This document is a draft for review purposes only and does not constitute Agency policy.
634 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Boyer et al. (2000) used an in vitro chick-AVC culture to test the hypothesis that TCE
might cause cardiac valve and septal defects by specifically perturbing epithelial-mesenchymal
cell transformation of endothelial cells in the AVC and outflow tract areas of the heart. AV
explants from Stage 16 White Leghorn chick embryos were placed in hydrated collagen gels,
with medium and TCE concentrations of 0, 50, 100, 150, 200, or 250 ppm. TCE was found to
block the endothelial cell-cell separation process that is associated with endothelial activation as
well as to inhibit mesenchymal cell formation across all TCE concentrations tested. TCE did
not, however, have an effect on the cell migration rate of fully formed mesenchymal cells. TCE-
treatment was also found to inhibit the expression of transformation factor Mox-1 and
extracellular matrix protein fibrillin 2, two protein markers of epithelial-mesenchyme cell
transformation.
4.8.3.1.9. Discussion/Synthesis of Developmental Data
In summary, an overall review of the weight of evidence in humans and experimental
animals is suggestive of the potential for developmental toxicity with TCE exposure. A number
of developmental outcomes have been observed in the animal toxicity and the epidemiological
data, as discussed below. These include adverse fetal/birth outcomes including death
(spontaneous abortion, perinatal death, pre- or postimplantation loss, resorptions), decreased
growth (low birth weight, small for gestational age, intrauterine growth restriction, decreased
postnatal growth), and congenital malformations, in particular cardiac defects. Postnatal
developmental outcomes include developmental neurotoxicity, developmental immunotoxicity,
and childhood cancer.
4.8.3.1.10. Adverse fetal and early neonatal outcomes
Studies that demonstrate adverse fetal or early neonatal outcomes are summarized in
Table 4-102. In human studies of prenatal TCE exposure, increased risk of spontaneous abortion
was observed in some studies (ATSDR, 2001; Taskinen et al., 1994; Windham et al., 1991), but
not in others (ATSDR, 2001, 2008; Goldberg et al., 1990; Lagakos et al., 1986; Lindbohm et al.,
1990; Taskinen et al., 1989). In addition, perinatal deaths were observed after 1970, but not
before 1970 (Lagakos et al., 1986). In rodent studies that examined offspring viability and
survival, there was an indication that TCE exposure may have resulted in increased pre-and/or
postimplantation loss (Healy et al., 1982; Kumar et al., 2000b; Narotsky and Kavlock, 1995), and
in reductions in live pups born as well as in postnatal and postweaning survival (George et al.,
1985, 1986).
This document is a draft for review purposes only and does not constitute Agency policy.
63 5 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Decreased birth weight and small for gestational age was observed (ATSDR, 1998b,
2006b; Rodenbeck et al., 2000; Windham et al., 1991), however, no association was observed in
other studies (Bove, 1996; Bove et al., 1995; Lagakos et al., 1986). While comprising both
occupational and environmental exposures, these human studies are overall not highly
informative due to their small numbers of cases and limited exposure characterization or to the
fact that exposures to mixed solvents were involved. However, decreased fetal weight, live birth
weights and postnatal growth were also observed in rodents, (George et al., 1985, 1986; Healy et
al., 1982; Narotsky and Kavlock, 1995), adding to the weight of evidence for this endpoint. It is
noted that the rat studies reporting effects on fetal or neonatal viability and growth used Fischer
344 or Wistar rats, while several other studies, which used Sprague-Dawley rats, reported no
increased risk in these developmental measures (Carney et al., 2006; Hardin et al., 1981;
Schwetz et al., 1975).
Table 4-102. Summary of adverse fetal and early neonatal outcomes
associated with TCE exposures
Positive finding
Species
Citation
Spontaneous abortion, miscarriage, pre-
and/or postimplantation loss
Human
ATSDR (2001)a
Taskinen et al. (1994)a
Windham et al. (1991)
Rat
Kumar et al. (2000b)
Healy etal. (1982)
Narotsky and Kavlock (1995)
Narotsky et al. (1995)
Perinatal death, reduction in live births
Human
Lagakos et al. (1986)b
Mouse
George et al. (1985)
Rat
George et al. (1986)
Postnatal and postweaning survival
Mouse
George et al. (1985)
Rat
George et al. (1986)
Decreased birth weight, small for
gestational age, postnatal growth
Human
ATSDR (1998b)
ATSDR (2006a)
Rodenbeck et al. (2000)°
Windham et al. (1991)
Mouse
George et al. (1985)
Rat
George et al. (1986)
Healy et al. (1982)
Narotsky and Kavlock (1995)
Narotsky et al. (1995)
This document is a draft for review purposes only and does not constitute Agency policy.
636 DRAFT—DO NOT CITE OR QUOTE
-------
1 aNot significant.
2 bObserved for exposures after 1970, but not before.
3 Increased risk for very low birth weight but not low birth weight or Ml-term low birth weight.
4
5
6 Overall, based on weakly suggestive epidemiologic data and fairly consistent laboratory
7 animal data, it can be concluded that TCE exposure poses a potential hazard for prenatal losses
8 and decreased growth or birth weight of offspring.
9
4.8.3.1.11. Cardiac malformations
10 A discrete number of epidemiological studies and studies in laboratory animal models
11 have identified an association between TCE exposures and cardiac defects in developing
12 embryos and/or fetuses. These are listed in Table 4-103. Additionally, a number of avian and
13 rodent in vivo studies and in vitro assays have examined various aspects of the induction of
14 cardiac malformations.
15
16 Table 4-103. Summary of studies that identified cardiac malformations
17 associated with TCE exposures
18
Finding
Species
Citations
Cardiac defects
Human
AT SDR (2006a, 2008)
Yauck et al. (2004)
Rat
Dawson et al. (1993; 1990b)
Johnson et al. (2003) (2005)
Johnson et al. (1998a;
1998b)a
Smith etal. (1989),a (1992)a
Epstein et al. (1992)a
Chicken
Bross et al. (1983)
Boyer et al. (2000)
Loeber et al. (1988)
Drake et al. (2006a; 2006b)
Mishima et al. (2006)
Rufer et al. (2008; 2010)
Altered heart rate
Human
Jasinka (1965, translation)
19
20 aMetabolites of TCE.
21
22
This document is a draft for review purposes only and does not constitute Agency policy.
637 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
In humans, an increased risk of cardiac defects has been observed after exposure to TCE
in studies reported by ATSDR (ATSDR, 2006a, 2008) and Yauck et al. (2004), although others
saw no significant effect (Bove, 1996; Bove et al., 1995; Goldberg et al., 1990; Lagakos et al.,
1986), possibly due to a small number of cases. In addition, altered heart rate was seen in one
study (Jasinska, 1965, translation). A cohort of water contamination in Santa Clara County, CA
is often cited as a study of TCE exposure and cardiac defects; however, the chemical of exposure
is in fact trichloroethane, not TCE (Deane et al., 1989; Swan et al., 1989).
In laboratory animal models, avian studies were the first to identify adverse effects of
TCE exposure on cardiac development. As described in Section 4.8.2.2.1, cardiac malformations
have been reported in chick embryos exposed to TCE (Boyer et al., 2000; Bross et al., 1983;
Drake et al., 2006a; Drake et al., 2006b; Loeber et al., 1988; Mishima et al., 2006; Rufer et al.,
2008). Additionally, a number of studies were conducted in rodents in which cardiac
malformations were observed in fetuses following the oral administration of TCE to maternal
animals during gestation (Dawson et al., 1993; Dawson et al., 1990b; Johnson et al., 2003) 2005;
see Section 4.8.2.2.1.2). Cardiac defects were also observed in rats following oral gestational
treatment with metabolites of TCE (Epstein et al., 1992; Johnson et al., 1998a; Johnson et al.,
1998b; Smith et al., 1992; Smith et al., 1989).
However, cardiac malformations were not observed in a number of other studies in
laboratory animals in which TCE was administered during the period of cardiac organogenesis
and fetal visceral findings were assessed. These included inhalation studies in rats (Carney et al.,
2006; Dorfmueller et al., 1979; Hardin et al., 1981; Healy et al., 1982; Schwetz et al., 1975) and
rabbits (Hardin et al., 1981), and oral gavage studies in rats (Fisher et al., 2001; Narotsky and
Kavlock, 1995; Narotsky et al., 1995) and mice (Cosby and Dukelow, 1992).
It is generally recognized that response variability among developmental bioassays
conducted with the same chemical agent may be related to factors such as the study design (e.g.,
the species and strain of laboratory animal model used, the day(s) or time of day of dose
administration in relation to critical developmental windows, the route of exposure, the vehicle
used, the day of study termination), or the study methodologies (e.g., how fetuses were
processed, fixed, and examined; what standard procedures were used in the evaluation of
morphological landmarks or anomalies, and whether there was consistency in the fetal
evaluations that were conducted). In the case of studies that addressed cardiac malformations,
there is additional concern as to whether detailed visceral observations were conducted, whether
or not cardiac evaluation was conducted using standardized dissection procedures (e.g., with the
use of a dissection microscope or including confirmation by histopathological evaluation, and
whether the examinations were conducted by technicians who were trained and familiar with
fetal cardiac anatomy). Furthermore, interpretation of the findings can be influenced by the
This document is a draft for review purposes only and does not constitute Agency policy.
63 8 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
analytical approaches applied to the data as well as by biological considerations such as the
historical incidence data for the species and strain of interest. These issues have been critically
examined in the case of the TCE developmental toxicity studies (Hardin et al., 2005; Watson et
al., 2006).
In the available animal developmental studies with TCE, differences were noted in the
procedures used to evaluate fetal cardiac morphology following TCE gestational exposures
across studies, and some of these differences may have resulted in inconsistent fetal outcomes
and/or the inability to detect cardiac malformations. Most of the studies that did not identify
cardiac anomalies used a traditional free-hand sectioning technique (as described in Wilson,
1965) on fixed fetal specimens (Dorfmueller et al., 1979; Hardin et al., 1981; Healy et al., 1982;
Schwetz et al., 1975). Detection of cardiac anomalies can be enhanced through the use of a fresh
dissection technique as described by Staples (1974) and Stuckhardt and Poppe (1984); a
significant increase in treatment-related cardiac heart defects was observed by Dawson et al.
(1990b) when this technique was used. Further refinement of this fresh dissection technique was
employed by Dawson and colleagues at the University of Arizona (UA), resulting in several
additional studies that reported cardiac malformations (Dawson et al., 1993; Johnson et al., 2003)
2005. However, two studies conducted in an attempt to verify the teratogenic outcomes of the
UA laboratory studies used the same or similar enhanced fresh dissection techniques and were
unable to detect cardiac anomalies (Carney et al., 2006; Fisher et al., 2001). Although the
Carney et al. study was administered via inhalation (a route which has not previously been
shown to produce positive outcomes), the Fisher et al. study was administered orally and
included collaboration between industry and UA scientists. It was suggested that the apparent
differences between the results of the Fisher et al. study and the Dawson et al. (1993) and
Johnson et al. studies may be related to factors such as differences in purity of test substances or
in the rat strains, or differences in experimental design (e.g., oral gavage versus drinking water,
exposure only during the period of organogenesis versus during the entire gestation period, or the
use of a staining procedure).
It is notable that all studies that identified cardiac anomalies following gestational
exposure to TCE or its metabolites were (1) conducted in rats and (2) dosed by an oral route of
exposure (gavage or drinking water). Cross-species and route-specific differences in fetal
response may be due in part to toxicokinetic factors. Although a strong accumulation and
retention of TCA was found in the amniotic fluid of pregnant mice following inhalation
exposures to TCE (Ghantous et al., 1986), other toxicokinetic factors may be critical. The
consideration of toxicokinetics in determining the relevance of murine developmental data for
human risk assessment is briefly discussed by Watson et al. (2006). There are differences in the
metabolism of TCE between rodent and humans in that TCE is metabolized more efficiently in
This document is a draft for review purposes only and does not constitute Agency policy.
639 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
rats and mice than humans, and a greater proportion of TCE is metabolized to DCA in rodents
versus to TCA in humans. Studies that examined the induction of cardiac malformations with
gestational exposures of rodents to various metabolites of TCE identified TCA and DCA as
putative cardiac teratogens. Johnson et al. (1998a; 1998b) and Smith et al. (1989) reported
increased incidences of cardiac defects with gestational TCA exposures, while Smith et al.
(1992) and Epstein et al. (1992) reported increased incidences following DCA exposures.
In all studies that observed increased cardiac defects, either TCE or its metabolites were
administered during critical windows of in utero cardiac development, primarily during the entire
duration of gestation, or during the period of major organogenesis (e.g., GD 6-15 in the rat).
The study by Epstein et al. (1992) used dosing with DCA on discrete days of gestation and had
identified GDs 9 through 12 as a particularly sensitive period for eliciting high interventricular
septal defects associated with exposures to TCE or its metabolites.
In the oral studies that identified increased incidences of cardiac malformations following
gestational exposure to TCE, there was a broad range of administered doses at which effects
were observed. In drinking water studies, Dawson et al. (1993) observed cardiac anomalies at
1.5 and 1,100 ppm (with no NOAEL) and Johnson et al. (2003) 2005 reported effects at 250 ppb
(with a NOAEL of 2.5 ppb). One concern is the lack of a clear dose-response for the incidence
of any specific cardiac anomaly or combination of anomalies was not identified, a disparity for
which no reasonable explanation for this disparity has been put forth.
The analysis of the incidence data for cardiac defects observed in the Johnson et al.
(2003) 2005 studies has been critiqued (Watson et al., 2006). Issues of concern that have been
raised include the statistical analyses of findings on a per-fetus (rather than the more appropriate
per-litter) basis (Benson, 2004), and the use of nonconcurrent control data in the analysis (Hardin
et al., 2004). In response, the study author has further explained procedures used (Johnson et al.,
2004) and has provided individual litter incidence data to the EPA for independent statistical
analysis (P. Johnson, personal communication, 2008) (see Section 5.1.2.8, dose-response). In
sum, while the studies by Dawson et al. (1993) and Johnson et al. (2003)(2005) have significant
limitations, there is insufficient reason to dismiss their findings.
4.8.3 .1.11.1. Mode of action for cardiac malformations
A number of in vitro studies have been conducted to further characterize the potential for
alterations in cardiac development that have been attributed to exposures with TCE and/or its
metabolites. It was noted that many of the cardiac defects observed in humans and laboratory
species (primarily rats and chickens) involved septal and valvular structures.
This document is a draft for review purposes only and does not constitute Agency policy.
640 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
During early cardiac morphogenesis, outflow tract and AV endothelial cells differentiate
into mesenchymal cells. These mesenchymal cells have characteristics of smooth muscle-like
myofibroblasts and form endocardial cushion tissue, which is the primordia of septa and valves
in the adult heart. Events that take place in cardiac valve formation in mammals and birds are
summarized by NRC (2006) and reproduced in Table 4-104.
Methods have been developed to extract the chick stage 16 atrioventricular canal from the
embryo and culture it on a hydrated collagen gel for 24-48 hours, allowing evaluation of the
described stages of cardiac development and their response to chemical treatment. Factors that
have been shown to influence the induction of endocardial cushion tissue include molecular
components such as fibronectin, laminin, and galactosyltransferase (Loeber and Runyan, 1990;
Mjaatvedt et al., 1987), components of the extracellular matrix (Mjaatvedt et al., 1991), and
smooth muscle a-actin and transforming growth factor P3 (Nakajima et al., 1997; Ramsdell and
Markwald, 1997).
Boyer et al. (2000) utilized the in vitro chick AVC culture system to examine the
molecular mechanism of TCE effects on cardiac morphogenesis. AVC explants from stage 16
chick embryos (15/treatment level) were placed onto collagen gels and treated with 0, 50, 100,
150, 200, or 250-ppm TCE and incubated for a total of 54 hours. Epithelial-mesenchymal
transformation, endothelial cell density, cell migration, and immunohistochemistry were
evaluated. TCE treatment was found to inhibit endothelial cell activation and normal
Table 4-104. Events in cardiac valve formation in mammals and birds"
Stage and event
Structural descriptionb
Early cardiac development
The heart is a hollow, linear, tube-like structure with two cell layers. The outer surface
is a myocardial cell layer, and the inner luminal surface is an endothial layer.
Extracellular matrix is between the two cell layers.
Epithelial-mesenchymal
cell transformation
A subpopulation of endothelial cells lining the atrioventricular canal detaches from
adjacent cells and invades the underlying extracellular matrix.
Three events occur
> Endothelial cell activation (avian stage 14)
> Mesenchymal cell formation (avian stage 16)
> Mesenchymal cell migration into the extracellular matrix (avian stages 17 and 18).
Mesenchymal cell
migration and proliferation
Endothelial-derived mesenchymal cells migrate toward the surrounding myocardium
and proliferate to populate the AVC extracellular matrix.
Development of septa and
valvular structures
Cardiac mesenchyme provides cellular constituents for
> Septum intermedium
> Valvular leaflets of the mitral and tricuspid AV valves.
The septum intermedium subsequently contributes to
> Lower portion of the interatrial septum
> Membranous portion of the interventricular septum.
aAs summarized in NRC (2006).
bMarkwald et al. (1996; 1984), Boyer et al. (2000).
This document is a draft for review purposes only and does not constitute Agency policy.
641 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
mesenchymal cell transformation, endothelial cell-cell separation, and protein marker expression
(i.e., transcription factor Mox-1 and extracellular matrix protein fibrillin 2). Mesenchymal cell
migration was not affected, nor was the expression of smooth muscle a-actin. The study authors
proposed that TCE may cause cardiac valvular and septal malformations by inhibiting
endothelial separation and early events of mesenchymal cell formation. Hoffman et al. (2004)
has proposed alternatively that TCE may be affecting the adhesive properties of the endocardial
cells. No experimental data are currently available that address the levels of TCE in cardiac
tissue in vivo, resulting in some questions (Dugard, 2000) regarding the relevance of these
mechanistic findings to human health risk assessment.
In a study by Mishima et al. (2006), White Leghorn chick whole embryo cultures (stage
13 and 14) were used to assess the susceptibility of endocardial epithelial-mesenchymal
transformation in the early chick heart to TCE at analytically determined concentrations of 0, 10,
20, 40, or 80 ppm. This methodology maintained the anatomical relationships of developing
tissues and organs, while exposing precisely staged embryos to quantifiable levels of TCE and
facilitating direct monitoring of developmental morphology. Following 24 hours of incubation
the numbers of mesenchymal cells in the inferior and superior AV cushions were counted. TCE
treatment significantly reduced the number of mesenchymal cells in both the superior and
inferior AV cushions at 80 ppm.
Ou et al. (2003) examined the possible role of endothelial nitric oxide synthase (which
generates nitric oxide that has an important role in normal endothelial cell proliferation and
hence normal blood vessel growth and development) in TCE-mediated toxicity. Cultured
proliferating bovine coronary endothelial cells were treated with TCE at 0-100 [xM and
stimulated with a calcium ionophore to determined changes in endothelial cells and the
generation of endothelial nitric oxide synthase, nitric oxide, and superoxide anion. TCE was
shown to alter heat shock protein interactions with endothelial nitric oxide synthase and induce
endothelial nitric oxide synthase to shift nitric oxide to superoxide-anion generation. These
findings provide insight into how TCE impairs endothelial proliferation.
Several studies have also identified a TCE-related perturbation of several proteins
2+
involved in regulation of intracellular Ca . After 12 days of maternal exposure to TCE in
2+
drinking water, Serca2a (sarcoendoplasmic reticulum Ca ATPase) mRNA expression was
reduced in rat embryo cardiac tissues (Collier et al., 2003). Selmin et al. (2008) conducted a
microarray analysis of a P19 mouse stem cell line exposed to 1-ppm TCE in vitro, identifying
2+
altered expression of Ryr2 (ryanodine receptor isoform 2), a Ca release channel that is
2+
important in normal rhythmic heart activity (Gyorke and Terentyev, 2008). Alterations in Ca
This document is a draft for review purposes only and does not constitute Agency policy.
642 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
cycling and resulting contractile dysfunction is a recognized pathogenic mechanism of cardiac
arrhythmias and sudden cardiac death (Leandri et al., 1995; Lehnart et al., 2008; Yano et al.,
2008). Caldwell et al. (2008b) used real-time PCR and digital imaging microscopy to
characterize the effects of various doses of TCE on gene expression and Ca2+ response to
vasopressin in rat cardiac myocytes (H9c2) Serca2a and Ryr2 expression were reduced at 12
2+
and 48 hours following exposure to TCE. Additionally, Ca response to vasopressin was altered
following TCE treatment. Makwana et al. (2010) dosed chick embryos in ovo with 8 or 800 ppb
TCE; RT-PCR analysis of RNA isolated during specific windows of cardiac development
demonstrated effects on the expression of genes associated with reduced blood flow. Although it
has been hypothesized that TCE might interfere with the folic acid/methylation pathway in liver
and kidney and alter gene regulation by epigenetic mechanisms, Caldwell et al. (2010) found that
the effects of TCE exposure on normal gene expression in rat embryonic hearts was not altered
by the administration of exogenous folate. Overall, these data suggest that TCE may disrupt the
2+
ability to regulate cellular Ca fluxes, altering blood flow and leading to morphogenic
consequences in the developing heart. This remains an open area of research.
Thus, in summary, a number of studies have been conducted in an attempt to characterize
the MOA for TCE-induced cardiac defects. A major research focus has been on disruptions in
cardiac valve formation, using avian in ovo and in vitro studies. These studies demonstrated
treatment-related alterations in endothelial cushion development that could plausibly be
associated with defects involving septal and valvular morphogenesis in rodents and chickens.
However, a broad array of cardiac malformations has been observed in animal models following
TCE exposures (Dawson et al., 1993; Johnson et al., 2003) 2005, and other evidence of
2+
molecular disruption of Ca during cardiac development has been examined (Caldwell et al.,
2008b; Collier et al., 2003; Selmin et al., 2008) suggesting the possible existence of multiple
MO As. The observation of defective myocardial development in a mouse model deficient for
gpl30, a signal transducer receptor for IL-6 (Yoshida et al., 1996), suggests the potential
involvement of immune-mediated effects.
4.8.3 .1.11.2. Association of peroxisome proliferator activated receptor alpha (PPARa)
with developmental outcomes
The PPARs are ligand activated receptors that belong to the nuclear hormone receptor
family. Three isotypes have been identified (PPARa, PPARS [also known as PPARP], and
PPARy). These receptors, upon binding to an activator, stimulate the expression of target genes
implicated in important metabolic pathways. In rodents, all three isotypes show specific time
and tissue-dependent patterns of expression during fetal development and in adult animals. In
This document is a draft for review purposes only and does not constitute Agency policy.
643 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
development, they have been especially implicated in several aspects of tissue differentiation,
e.g., of the adipose tissue, brain, placenta and skin. Epidermal differentiation has been linked
strongly with PPARa and PPARS (Michalik et al., 2002). PPARa starts late in development,
with increasing levels in organs such as liver, kidney, intestine, and pancreas; it is also
transiently expressed in fetal epidermis and CNS (Braissant and Wahli, 1998) and has been
linked to phthalate-induced developmental and testicular toxicity (Corton and Lapinskas, 2005).
Liver, kidney, and heart are the sites of highest PPARa expression (Toth et al., 2007). PPARS
and PPARy have been linked to placental development and function, with PPARy found to be
crucial for vascularization of the chorioallantoic placenta in rodents (Wendling et al., 1999), and
placental anomalies mediated by PPARy have been linked to rodent cardiac defects (Barak et al.,
2008). While it might be hypothesized that there is some correlation between PPAR signaling,
fetal deaths, and/or cardiac defects observed following TCE exposures in rodents, no definitive
data have been generated that elucidate a possible PPAR-mediated MOA for these outcomes.
4.8.3.1.11.3. Summary of the weight of evidence on cardiac malformations
The evidence for an association between TCE exposures in the human population and the
occurrence of congenital cardiac defects is not particularly strong. Many of the epidemiological
study designs were not sufficiently robust to detect exposure-related birth defects with a high
degree of confidence. However, two well-conducted studies by ATSDR (2006a, 2008) clearly
demonstrated an elevation in cardiac defects. It could be surmised that the identified cardiac
defects were detected because they were severe, and that additional cases with less severe
cardiac anomalies may have gone undetected.
The animal data provide strong, but not unequivocal, evidence of the potential for
TCE-induced cardiac malformations following oral exposures during gestation. Strengths of the
evidence are the duplication of the adverse response in several studies from the same laboratory
group, detection of treatment-related cardiac defects in both mammalian and avian species (i.e.,
rat and chicken), general cross-study consistency in the positive association of increased cardiac
malformations with test species (i.e., rat), route of administration (i.e., oral), and the
methodologies used in cardiac morphological evaluation (i.e., fresh dissection of fetal hearts).
Furthermore, when differences in response are observed across studies they can generally be
attributed to obvious methodological differences, and a number of in ovo and in vitro studies
demonstrate a consistent and biologically plausible MOA for one type of malformation observed.
Weaknesses in the evidence include lack of a clear dose-related response in the incidence of
This document is a draft for review purposes only and does not constitute Agency policy.
644 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
cardiac defects, and the broad variety of cardiac defects observed, such that they cannot all be
grouped easily by type or etiology.
Taken together, the epidemiological and animal study evidence raise sufficient concern
regarding the potential for developmental toxicity (increased incidence of cardiac defects) with
in utero TCE exposures.
4.8.3.1.12. Other structural developmental outcomes
A summary of other structural developmental outcomes that have been associated with
TCE exposures is presented in Table 4-105.
In humans, a variety of birth defects other than cardiac have been observed. These
include total birth defects (ATSDR, 2001; Bove, 1996; Bove et al., 1995; Flood, 1988) CNS
birth defects (ATSDR, 2001; Bove, 1996; Bove et al., 1995; Lagakos et al., 1986), eye/ear birth
anomalies (Lagakos et al., 1986); oral cleft defects (Bove, 1996; Bove et al., 1995; Lagakos et
al., 1986; Lorente et al., 2000); kidney/urinary tract disorders (Lagakos et al., 1986);
musculoskeletal birth anomalies (Lagakos et al., 1986); anemia/blood disorders (Burg and Gist,
1999); and lung/respiratory tract disorders (Lagakos et al., 1986). While some of these results
were statistically significant, they have not been reported elsewhere. Occupational cohort
studies, while not reporting positive results, are generally limited by the small number of
observed or expected cases of birth defects (Lorente et al., 2000; Taskinen et al., 1989; Tola et
al., 1980).
Table 4-105. Summary of other structural developmental outcomes
associated with TCE exposures
Finding
Species
Citations
Eye/ear birth anomalies
Human
Lagakos et al. (1986)
Rat
Narotsky (1995)
Narotsky and Kavlock (1995)
Oral cleft defects
Human
Bove (1996)
Bove et al. (1995)
Lagakos et al. (1986)
Lorente et al. (2000)
Kidney/urinary tract
disorders
Human
Lagakos et al. (1986)
Musculoskeletal birth
anomalies
Human
Lagakos et al. (1986)
This document is a draft for review purposes only and does not constitute Agency policy.
645 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Anemia/blood disorders
Human
Burg and Gist (1999)
Lung/respiratory tract
disorders
Human
Lagakos et al. (1986)
Mouse
Das and Scott (1994)
Skeletal
Rat
Healy et al. (1982)
Othera
Human
AT SDR (2001)
aAs reported by the authors.
In experimental animals, a statistically significant increase in the incidence of fetal eye
defects, primarily micropththalmia and anopththalmia, manifested as reduced or absent eye
bulge, was observed in rats following gavage administration of 1,125 mg/kg-day TCE during the
period of organogenesis (Narotsky and Kavlock, 1995; Narotsky et al., 1995). Dose-related
nonsignificant increases in the incidence of Fischer 344 rat pups with eye defects were also
observed at lower dose levels (101, 320, 475, 633, and 844 mg/kg-day) in the Narotsky et al.
(1995) study (also reported in Barton and Das, 1996). However, no other developmental or
reproductive toxicity studies identified abnormalities of eye development following TCE
exposures. For example, in a study reported by Warren et al. (2006), extensive computerized
morphometric ocular evaluation was conducted in Sprague-Dawley rat fetuses that had been
examined for cardiac defects by Fisher et al. (2001); the dams had been administered TCE
(500 mg/kg-day), DCA (300 mg/kg-day), or TCA (300 mg/kg-day) during GDs 6-15. No ocular
defects were found with TCE exposures; however, significant reductions in the lens area, globe
area, and interocular distance were observed with DCA exposures, and nonsignificant decreases
in these measures as well as the medial canthus distance were noted with TCA exposures.
Developmental toxicity studies conducted by Smith et al. (1992; 1989) also identified orbital
defects (combined soft tissue and skeletal abnormalities) in Long Evans rat fetuses following GD
6-15 exposures with TCA and DCA (statistically or biologically significant at >800 and >900
mg/kg-day, respectively). Overall, the study evidence indicates that TCE and its oxidative
metabolites can disrupt ocular development in rats. In addition to the evidence of alteration to
the normal development of ocular structure, these findings may also be an indicator of
disruptions to nervous system development. It has been suggested by Warren et al. (2006) and
Williams and DeSesso (2008) that the effects of concern (defined as statistically significant
outcomes) are observed only at high dose levels and are not relevant to risk assessment for
environmental exposures. On the other hand, Barton and Das (1996) point out that benchmark
dose modeling of the quantal eye defect incidence data provides a reasonable approach to the
development of oral toxicity values for TCE human health risk assessment. It is also noted that
This document is a draft for review purposes only and does not constitute Agency policy.
646 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
concerns may exist not only for risks related to low level environmental exposures, but also for
risks resulting from acute or short-term occupational or accidental exposures, which may be
associated with much higher inadvertent doses.
It was also notable that a study using a single intraperitoneal dose of 3,000 mg/kg TCE to
mice during late gestation (GD 17) identified apparent delays in lung development and increased
neonatal mortality (Das and Scott, 1994). No further evaluation of this outcome has been
identified in the literature.
Healy et al. (1982) did not identify any treatment-related fetal malformations following
"3
inhalation exposure of pregnant inbred Wistar rats to 0 or 100 ppm (535 mg/m ) on GD 8-21. In
this study, significant differences between control and treated litters were observed as an
increased incidence of minor ossification variations (p = 0.003) (absent or bipartite centers of
ossification).
4.8.3.1.13. Developmental neurotoxicity
Studies that address effects of TCE on the developing nervous system are discussed in
detail in Section 4.3, addressed above in the sections on human developmental toxicity (see
Section 4.8.3) and on mammalian studies (see Section 4.8.3.2.1) by route of exposure, and
summarized in Table 4-106. The available data collectively suggest that the developing brain is
susceptible to TCE exposures.
In humans, CNS birth defects were observed in a few studies (ATSDR, 2001; Bove,
1996; Bove et al., 1995; Lagakos et al., 1986). Postnatally, observed adverse effects in humans
include delayed newborn reflexes following use of TCE during childbirth (Beppu, 1968),
impaired learning or memory (Bernad et al., 1987, abstract; White et al., 1997); aggressive
Table 4-106. Summary of developmental neurotoxicity associated with TCE
exposures
Positive findings
Species
Citations
CNS defects, neural tube defects
Human
ATSDR (2001)
Bove (1996); Bove et al. (1995)
Lagakos et al. (1986)
Eye defects
Rat
Narotsky (1995);
Narotsky and Kavlock (1995)
Delayed newborn reflexes
Human
Beppu (1968)
Impaired learning or memory
Human
Bernad et al. (1987), abstract
This document is a draft for review purposes only and does not constitute Agency policy.
647 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
White et al. (1997)
Aggressive behavior
Human
Bernad et al. (1987), abstract
Rat
Blossom et al. (2008)
Hearing impairment
Human
ATSDR (2003a);
Burg et al. (1995);
Burg and Gist (1999)
Beppu (1968)
Speech impairment
Human
ATSDR (2003a);
Burg et al. (1995);
Burg and Gist (1999)
White et al. (1997)
Encephalopathy
Human
White et al. (1997)
Impaired executive function
Human
White et al. (1997)
Impaired motor function
Human
White et al. (1997)
Attention deficit
Human
Bernad et al. (1987), abstract
ASD
Human
Windham et al. (2006)
Delayed or altered biomarkers of
CNS development
Rat
Isaacson and Taylor (1989);
Noland-Gerbec et al. (1986);
Westergren et al. (1984)
Behavioral alterations
Mice
Blossom et al. (2008); Fredriksson
et al. (1993)
Rat
George et al. (1986);
Taylor et al. (1985)
behavior (Bernad et al., 1987, abstract); hearing impairment (ATSDR, 2003a; Beppu, 1968; Burg
and Gist, 1999; Burg et al., 1995); speech impairment (Burg and Gist, 1999; Burg et al., 1995;
White et al., 1997); encephalopathy (White et al., 1997); impaired executive and motor function
(White et al., 1997); attention deficit (Bernad et al., 1987, abstract; White et al., 1997), and
autism spectrum disorder (Windham et al., 2006). While there are broad developmental
neurotoxic effects that have been associated with TCE exposure, there are many limitations in
the studies.
More compelling evidence for the adverse effect of TCE exposure on the developing
nervous system is found in the animal study data, although a rigorous evaluation of potential
outcomes has not been conducted. For example, there has not been an assessment of cognitive
function (i.e., learning and memory) following developmental exposures to TCE, nor have most
This document is a draft for review purposes only and does not constitute Agency policy.
648 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
of the available studies characterized the pre- or postnatal exposure of the offspring to TCE or its
metabolites. Nevertheless, there is evidence of treatment-related alterations in brain
development and in behavioral parameters (e.g., spontaneous motor activity and social
behaviors) associated with exposures during neurological development. The animal study
database includes the following information: Following inhalation exposures of 150 ppm to mice
during mating and gestation, the specific gravity of offspring brains were significantly decreased
at postnatal time points through the age of weaning; however, this effect did not persist to
1 month of age (Westergren et al., 1984). In studies reported by Taylor et al. (1985), Isaacson
and Taylor (1989), and Noland-Gerbec et al. (1986), 312 mg/L exposures in drinking water that
were initiated 2 weeks prior to mating and continued to the end of lactation resulted,
respectively, in (a) significant increases in exploratory behavior at GDs 60 and 90, (b) reductions
in myelination in the brains of offspring at weaning, and (c) significantly decreased uptake of
2-deoxyglucose in the neonatal rat brain (suggesting decreased neuronal activity). Ocular
malformations in rats observed by Narotsky (1995) and Narotsky and Kavlock (1995) following
maternal gavage doses of 1,125 mg/kg-day during gestation may also be indicative of alterations
of nervous system development. Gestational exposures to mice (Fredriksson et al., 1993)
resulted in significantly decreased rearing activity on GD 60, and dietary exposures during the
course of a continuous breeding study in rats (George et al., 1986) found a significant trend
toward increased time to cross the first grid in open field testing. In a study by Blossom et al.
(2008), alterations in social behaviors (deficits in nest-building quality and increased aggression
in males) were observed in pubertal-age MRL +/+ mice that had been exposed to 0.1 mg/mL
TCE via drinking water during prenatal and postnatal development (until PND 42). Dorfmueller
et al. (1979) was the only study that assessed neurobehavioral endpoints following in utero
exposure (maternal inhalation exposures of 1,800 ± 200 ppm during gestation) and found no
adverse effects that could be attributed to TCE exposure. Specifically, an automated assessment
of ambulatory response in a novel environment on GDs 10, 20 and 100, did not identify any
effect on general motor activity of offspring.
4.8.3.1.14. Developmental immunotoxicity
Studies that address the developmental immunotoxic effects of TCE are discussed in
detail in Section 4.6, addressed above in the sections on human developmental toxicity (see
Section 4.8.3) and on mammalian studies (see Section 4.8.3.2.1) by route of exposure, and
summarized in Table 4-107.
This document is a draft for review purposes only and does not constitute Agency policy.
649 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Table 4-107. Summary of developmental immunotoxicity associated with
TCE exposures
Finding
Species (strain)
Citations
Significant reduction in Thl IL-2 producing cells
Human
Lehmann et al.
(2002)
Altered immune response
Human
Byers et al. (1988)
Suppression of PFC responses, increased T-cell
subpopulations, decreased spleen cellularity, and
increased hypersensitivity response
Mouse (B6C3F1)
Peden-Adams et al.
(2006)
Altered splenic and thymic T-cell subpopulations
Mouse (MRL +/+)
Peden-Adams et al.
(2008)
Altered thymic T-cell subpopulations; transient
increased proinflammatory cytokine production by
T-cells; increased autoantibody levels and IgG
Mouse (MRL +/+)
Blossom and Doss
(2007)
Increased proinflammatory cytokine production by
T-cells
Mouse (MRL +/+)
Blossom et al. (2008)
Two epidemiological studies that addressed potential immunological perturbations in
children that were exposed to TCE were reported by Lehmann et al. (2001; 2002). In the 2001
study, no association was observed between TCE and allergic sensitization to egg white and
milk, or to cytokine producing peripheral T-cells, in premature neonates and 36-month-old
neonates that were at risk of atopy. In the 2002 study, there was a significant reduction in Thl
IL-2 producing cells. Another study observed altered immune response in family members of
those diagnosed with childhood leukemia, including 13 siblings under age 19 at the time of
exposure, but an analysis looking at only these children was not done (Byers et al., 1988).
Several studies were identified (Blossom and Doss, 2007; Blossom et al., 2008; Peden-
Adams et al., 2006; Peden-Adams et al., 2008) which assessed the potential for developmental
immunotoxicity in mice following oral (drinking water) TCE exposures during critical pre- and
postnatal stages of immune system development. Peden-Adams et al. (2006) noted evidence of
immune system perturbation (suppression of PFC responses, increased T-cell subpopulations,
decreased spleen cellularity, and increased hypersensitivity response) in B6C3F1 offspring
following in utero and 8 weeks of postnatal exposures to TCE. Evidence of autoimmune
response was not observed in the offspring of this nonautoimmune-prone strain of mice.
However, in a study by Peden-Adams et al. (2008) MRL +/+ mice, which are autoimmune-
prone, were exposed from conception until 12 months of age. Consistent with the Peden-Adams
This document is a draft for review purposes only and does not constitute Agency policy.
650 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
et al. (2006) study, no evidence of increased autoantibody levels was observed in the offspring.
In two other studies focused on autoimmune responses following drinking water exposures of
MRL +/+ mice to TCE during in utero development and continuing until the time of sexual
maturation, Blossom and Doss (2007) and Blossom et al. (2008) reported some peripheral blood
changes that were indicative of treatment-related autoimmune responses in offspring. Positive
response levels were 0.5 and 2.5 mg/mL for Blossom and Doss (2007) and 0.1 mg/mL for
Blossom et al. (2008). None of these studies were designed to extensively evaluate recovery,
latent outcomes, or differences in severity of response that might be attributed to the early life
exposures. Consistency in response in these animal studies was difficult to ascertain due to the
variations in study design (e.g., animal strain used, duration of exposure, treatment levels
evaluated, timing of assessments, and endpoints evaluated). Likewise, the endpoints assessed in
the few epidemiological studies that evaluated immunological outcomes following
developmental exposures to TCE were dissimilar from those evaluated in the animal models, and
so provided no clear cross-species correlation. The most sensitive immune system response
noted in the studies that exposed developing animals were the decreased PFC and increased
hypersensitivity observed by Peden-Adams et al. (2006); treatment-related outcomes were noted
in mice exposed in the drinking water at a concentration of 1,400 ppb. None of the other studies
that treated mice during immune system development assessed these same endpoints; therefore,
direct confirmation of these findings across studies was not possible. It is noted, however, that
similar responses were not observed in studies in which adult animals were administered TCE
(e.g., Woolhiser et al., 2006), suggesting increased susceptibility in the young. Differential
lifestage-related responses have been observed with other diverse chemicals (e.g.,
diethylstilbestrol; diazepam; lead; 2,3,7,8-tetrachlorobenzo-p dioxin; and tributyltin oxide) in
which immune system perturbations were observed at lower doses and/or with greater
persistence when tested in developing animals as compared to adults (Luebke et al., 2006).
Thus, such an adverse response with TCE exposure is considered biologically plausible and an
issue of concern for human health risk assessment.
4.8.3.1.15. Childhood cancers
A summary of childhood cancers that have been associated with TCE exposures
discussed above is presented in Table 4-108. A summary of studies that observed childhood
leukemia is also discussed in detail in Section 4.6.1.3 and Section 4.8.3.1.2.4 contains details of
epidemiologic studies on childhood brain cancer.
This document is a draft for review purposes only and does not constitute Agency policy.
651 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Table 4-108. Summary of childhood cancers associated with TCE exposures
Finding
Species
Citations
Leukemia
Human
AZ DHS (1990; Flood, 1988)
AZ DHS (Kioski et al., 1990a)
Cohn et al. (1994b)
Cutler et al. (1986); Costas et al. (2002); Lagakos et al. (1986);
MA DPH( 1997a)
Lowengart et al. (1987)
McKinney et al. (1991)
Shu et al. (1999)
Neuroblastoma
Human
De Roos et al. (2001)
Peters et al. (1985; 1981)
A nonsignificant increased risk of leukemia diagnosed during childhood has been
observed in a number of studies examining TCE exposure (1990; Flood, 1988; Kioski et al.,
1990a), (Cohn et al., 1994b; Costas et al., 2002; Lagakos et al., 1986; Lowengart et al., 1987;
McKinney et al., 1991; MDPH, 1997a; Shu et al., 1999). However, other studies did not
observed an increased risk for childhood leukemia after TCE exposure (Flood, 1997; Kioski et
al., 1990b; Morgan and Cassady, 2002), possibly due to the limited number of cases or the
analysis based on multiple solvents.
CNS cancers during childhood have been reported on in a few studies. Neuroblastomas were not
statistically elevated in one study observing parental exposure to multiple chemicals, including
TCE (De Roos et al., 2001). Brain tumors were observed in another study, but the odds ratio
could not be determined (Peters et al., 1985; Peters et al., 1981). CNS cancers were not elevated
in other studies (Kioski et al., 1990a; Morgan and Cassady, 2002). Other studies did not see an
excess risk of total childhood cancers (ATSDR, 2006a; Morgan and Cassady, 2002).
A follow-up study of the Camp Lejeune cohort that will examine childhood cancers
(along with birth defects) was initiated in 1999 (ATSDR, 2003c), is expected to be completed
soon (ATSDR, 2009; U.S. GAO, 2007), and may provide additional insight.
No studies of cancers in experimental animals in early lifestages have been identified.
This document is a draft for review purposes only and does not constitute Agency policy.
652 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
4.9. OTHER SITE-SPECIFIC CANCERS
4.9.1. Esophageal Cancer
Increasing esophageal cancer incidence has been observed in males, but not females in
the United States between 1975 and 2002, a result of increasing incidence of esophageal
adenocarcinoma (Ward et al., 2006). Males also have higher age-adjusted incidence and
mortality rates (incidence, 7.8 per 100,000; mortality, 7.8 per 100,000) than females (incidence,
2.0 per 100,000; mortality, 1.7 per 100,000) (Ries et al., 2008). Survival for esophageal cancer
remains poor and age-adjusted mortality rates are just slightly lower than incidence rates. Major
risk factors associated with esophageal cancer are smoking and alcohol for squamous cell
carcinoma, typically found in the upper third of the esophagus, and obesity, gastroesophageal
reflux, and Barrett's esophagus for adenocarcinoma that generally occurs in the lower esophagus
(Ward et al., 2006).
Seventeen epidemiologic studies on TCE exposure reported relative risks for esophageal
cancer (ATSDR, 2004a, 2006b; Blair et al., 1989; Blair et al., 1998; Boice et al., 1999; Boice et
al., 2006b; Clapp and Hoffman, 2008; Costa et al., 1989; Garabrant et al., 1988; Greenland et al.,
1994; Hansen et al., 2001; Raaschou-Nielsen et al., 2003; Radican et al., 2008; Ritz, 1999a;
Siemiatycki, 1991; Sung et al., 2007; Zhao et al., 2005). Ten studies had high likelihood of TCE
exposure in individual study subjects and were judged to have met, to a sufficient degree, the
standards of epidemiologic design and analysis (Blair et al., 1998; Boice et al., 1999; Boice et al.,
2006b; Greenland et al., 1994; Hansen et al., 2001; Raaschou-Nielsen et al., 2003; Radican et al.,
2008; Ritz, 1999a; Siemiatycki, 1991; Zhao et al., 2005). Four studies with TCE exposure
potential assigned to individual subjects (Anttila et al., 1995; Axelson et al., 1994; Blair et al.,
1998 [Incidence]; Morgan et al., 1998) do not present relative risk estimates for esophageal
cancer and TCE exposure nor do two other studies which carry less weight in the analysis
because of design limitations (Henschler et al., 1995; Sinks et al., 1992). Only Raaschou-
Nielsen et al. (2003) examines esophageal cancer histologic type, an important consideration
given differences between suspected risk factors for adenocarcinoma and those for squamous cell
carcinoma. Appendix B identifies these studies' design and exposure assessment characteristics.
Several population case-control studies (Engel et al., 2002; Gustavsson et al., 1998;
Parent et al., 2000b; Ramanakumar et al., 2008; Santibanez et al., 2008; Weiderpass et al., 2003;
Yu et al., 1988) examine esophageal cancer and organic solvents or occupational job titles with
past TCE use documented (Bakke et al., 2006). Relative risk estimates in case-control studies
that examine metal occupations or job titles, or solvent exposures are found in Table 4-109. The
This document is a draft for review purposes only and does not constitute Agency policy.
653 DRAFT—DO NOT CITE OR QUOTE
-------
1 lack of exposure assessment to TCE, low prevalence of exposure to chlorinated hydrocarbon
2 solvents, or few exposed cases and controls in those studies lowers their sensitivity for informing
3 evaluations of TCE and esophageal cancer.
This document is a draft for review purposes only and does not constitute Agency policy.
654 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-109. Selected observations from case-control studies of TCE exposure and esophageal cancer
All esophageal cancers
Squamous cell cancer
Adenocarcinoma
Study
population
Exposure group
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Reference
Population of regions in Eastern Spain
Santibanez et al.
Metal molders, welders, etc.
0.94 (0.14, 6.16)
3
0.40 (0.05,3.18)
2
3.55 (0.28, 44.70)
1
(2008)
Metal-processing plant operators
1.14(0.29, 4.44)
5
1.23 (0.23,6.51)
4
0.86 (0.08, 8.63)
1
Chlorinated hydrocarbon solvents
Low exposure
1.05 (0.15,7.17)
2
0
4.92 (0.69, 34.66)
2
High exposure
1.76 (0.40, 7.74)
6
2.18(0.41, 11.57)
5
3.03 (0.28, 32.15)
1
Population of Montreal, Canada
Ramanakumar
Painter, Metal coatings
et al. (2008);
Parent et al.
(2000b)
Any exposure
1.3 (0.4,4.2)
6
Substantial exposure
4.2 (1.1, 17.0)
4
Solvents
Any exposure
1.1 (0.7, 1.7)
39
1.4(0.8,2.5)
30
Nonsubstantial exposure
1.0(0.5, 1.9)
16
1.3 (0.6,2.6)
12
Substantial exposure
1.1 (0.6, 1.9)
39
1.4 (0.8, 2.5)
30
Population of Sweden
Jansson et al.,
Organic solvents
(2005; 2006)
No exposure
1.0
145
1.0
128
Moderate exposure
0.7 (0.4, 1.5)
15
1.2(0.6,2.3)
14
High exposure
1.3 (0.7,2.3)
21
1.4(0.7,2.5)
18
Test for trend
p = 0.47
p = 0.59
No exposure
1.0
1.0
Moderate exposure
0.5 (0.1, 3.9)a
1
0.4(0.1, 1.5)a
2
High exposure
0.4(0.1, 1.8)a
2
0.9 (0.5, 1.6)a
12
Test for trend
p = 0.44
p = 0.36
-------
Table 4-109. Selected observations from case-control studies of TCE exposure and esophageal cancer (continued)
Study
population
Exposure group
All esophageal cancers
Squamous cell cancer
Adenocarcinoma
Reference
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Relative risk
(95% CI)
No. obs.
events
Population of Finland (Females)
Weiderpass et
al., (2003)
Chlorinated hydrocarbon solvents
Low level exposure
0.95 (0.54, 1.66)
Not
reported
High level exposure
0.62 (0.34, 1.13)
Not
reported
Population of NJ, CT, WA State
Engel et al.
(2002)
Precision metal workers
Not reported
0.7(0.3, 1.5)
12
1.4(0.8,2.3)
25
Metal product manufacturing
Not reported
0.8(0.3, 1.8)
15
1.3 (0.8,2.3)
26
a Jansson et al. (2006) is a registry-based study of the Swedish Construction Worker Cohort. Relative risks are incidence rate ratios from Cox regression analysis
using calendar time and adjustment for attained age, calendar period at entry into the cohort, tobacco smoking status at entry into the cohort and BMI at entry
into the cohort.
No. obs. events = number of observable events.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Table 4-110 presents risk estimates for TCE exposure and esophageal cancer observed in
cohort, PMR, case-control, and geographic based studies. Ten studies in which there is a high
likelihood of TCE exposure in individual study subjects (e.g., based on job-exposure matrices or
biomarker monitoring) reported risk estimates for esophageal cancer (Blair et al., 1998; Boice et
al., 1999; Boice et al., 2006b; Greenland et al., 1994; Hansen et al., 2001; Raaschou-Nielsen et
al., 2003; Radican et al., 2008; Ritz, 1999a; Siemiatycki, 1991; Zhao et al., 2005). Some
evidence for association with esophageal cancer and overall TCE exposure comes from studies
with high likelihood of TCE exposure (5.6, 95% CI: 0.7, 44.5 (Blair et al., 1998) and 1.88, 95%
CI: 0.61, 5.79 [Radican et al. (2008), which was an update of Blair et al. (1998) with an
additional 10 years of follow-up]; 4.2, 95% CI: 1.5, 9.2, (Hansen et al., 2001); 1.2, 95% CI: 0.84,
1.57 (Raaschou-Nielsen et al., 2003)). Two studies support an association with adenocarcinoma
histologic type of esophageal cancer and TCE exposure (five of the six observed esophageal
cancers were adenocarcinomas [less than 1 expected; Hansen et al. (2001)]; 1.8, 95% CI: 1.2, 2.7
(Raaschou-Nielsen et al., 2003)). Risk estimates in other studies are based on few deaths, low
statistical power to detect a doubling of esophageal cancer risk, and confidence intervals which
include a risk estimate of 1.0 (no increased risk).
Seven other studies (ATSDR, 2004a, 2006a; Blair et al., 1989; Clapp and Hoffman, 2008;
Costa et al., 1989; Garabrant et al., 1988; Sung et al., 2007) with lower likelihood for TCE
exposure, in addition to limited statistical power and other design limitations, observed relative
risk estimates between 0.21 (95% CI: 0.0.01, 1.17) (Costa et al., 1989) to 1.14 (95% CI: 0.62,
1.92) (Garabrant et al., 1988). For these reasons, esophageal cancer observations in these studies
are not inconsistent with Blair et al. (1998) and its update Radican et al. (2008), Hansen et al.
(2001), or Raaschou-Nielsen et al. (2003). No study reported a statistically significant deficit in
the esophageal cancer risk estimate and overall of TCE exposure. Of those studies with
exposure-response analyses, a pattern of increasing esophageal cancer relative risk with
increasing exposure metric is not generally noted (Blair et al., 1998; Boice et al., 1999; Radican
et al., 2008; Siemiatycki, 1991; Zhao et al., 2005) except for Hansen et al. (2001) and
Raaschou-Nielsen et al. (2003). In these last two studies, esophageal cancer relative risk
estimates associated with long employment duration were slightly higher (SIR: 6.6, 95% CI: 1.8,
7.0.8, 3.7 (Hansen et al., 2001); SIR: 1.9, 95% CO: 0.8, 3.7 (Raaschou-Nielsen et al., 2003)) than
those for short employment duration (SIR: 4.4, 95% CI: 0.5, 19 (Hansen et al., 2001); SIR: 1.7,
95% CI: 0.6, 3.6 (Raaschou-Nielsen et al., 2003)). Hansen et al. (2001) also reports risk for two
other TCE exposure surrogates, average intensity and cumulative exposure, and in both cases
observed lower risk estimates with the higher exposure surrogate.
This document is a draft for review purposes only and does not constitute Agency policy.
657 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-110. Summary of human studies on TCE exposure and esophageal
2 cancer
3
Exposure group
Relative risk
(95% CI)
No. obs. events
Reference
Cohort studies—incidence
Aerospace workers (Rocketdyne)
Zhao et al. (2005)
Any exposure to TCE
Not reported
Low cumulative TCE score
1.00a
9
Med cumulative TCE score
1.66 (0.62, 4.41)b
8
High TCE score
0.82 (0.17, 3.95)b
2
p for trend
p = 0.974
All employees at electronics factory (Taiwan)
Sung et al. (2007)
Males
Not reported
Females
1.16(0.0.14, 4.20)c
2
Danish blue-collar worker with TCE exposure
Raaschou-Nielsen et al.
Any exposure, all subjects
1.2 (0.84, 1.57)
44
(2003)
Any exposure, males
1.1 (0.81, 1.53)
40
Any exposure, females
2.0 (0.54, 5.16)
4
Any exposure, males
1.8 (1.15, 2.73)d
23
Any exposure, females
0 (0.4 exp)d
Exposure lag time
20 yr
1.7 (0.8, 3.0)d
10
Employment duration
<1 yr
1.7 (0.6, 3.6)d
6
1-4.9 yr
1.9 (0.9, 3.6)d
9
>5 yr
1.9 (0.8, 3.7)d
8
Subcohort with higher exposure
Any TCE exposure
1.7 (0.9, 2.9)d
13
Employment duration
1-4.9 yr
1.6 (0.6, 3.4)d
6
>5 yr
1.9 (0.8, 3.8)d
7
4
This document is a draft for review purposes only and does not constitute Agency policy.
658 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-110. Summary of human studies on TCE exposure and esophageal
cancer (continued)
Exposure group
Relative risk
(95% CI)
No. obs. events
Reference
Biologically-monitored Danish workers
4.0 1.5, 8.72)
6
Hansen etal. (2001)
Any TCE exposure, males
4.2(1.5,9.2)
6
Adenocarcinoma histologic type
3.6(1.2, 8.3)e
5
Any TCE exposure, females
0 (0.1 exp)
Cumulative exposure (Ikeda)
<17 ppm-yr
6.5 (1.3, 19)
3
>17 DDm-vr
4.2(1.5,9.2)
3
Mean concentration (Ikeda)
<4 ppm
8.0 (2.6, 19)
5
4+ ppm
1.3 (0.02, 7.0)
1
Employment duration
<6.25 yr
4.4 (0.5, 16)
2
>6.25 yr
6.6 (1.8, 17)
4
Aircraft maintenance workers from Hill Air Force Base
Blair etal. (1998)
TCE subcohort
Not reported
Males, cumulative exposure
0
1.0a
<5 ppm-yr
Not reported
5-25 ppm-yr
Not reported
>25 ppm-yr
Not reported
Females, cumulative exposure
0
1.0a
<5 ppm-yr
Not reported
5-25 ppm-yr
Not reported
>25 ppm-yr
Not reported
Biologically-monitored Finnish workers
Anttila et al. (1995)
All subjects
Not reported
Mean air-TCE (Ikeda extrapolation)
<6 ppm
Not reported
6+ ppm
Not reported
This document is a draft for review purposes only and does not constitute Agency policy.
659 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-110. Summary of human studies on TCE exposure and esophageal
cancer (continued)
Exposure group
Relative risk
(95% CI)
No. obs. events
Reference
Cardboard manufacturing workers in Arnsburg, Germany
Henschler et al. (1995)
Exposed workers
Not reported
Biologically-monitored Swedish workers
Axelsonetal. (1994)
Any TCE exposure, males
Not reported
Any TCE exposure, females
Not reported
Cardboard manufacturing workers, Atlanta area, GA
Sinks et al. (1992)
All subjects
Not reported
Cohort and PMR studies-mortality
Computer manufacturing workers (IBM), NY
Clapp and Hoffman (2008)
Males
1.12 (0.30, 2.86)f
5.24 (0.13, 29.2)f
Aerospace workers (Rocketdyne)
Any TCE (utility/eng flush)
0.88 (0.18,2.58)
3
Boice et al. (2006b)
Any exposure to TCE
Not reported
Zhao et al. (2005)
Low cumulative TCE score
1.00a
18
Medium cumulative TCE score
1.40 (0.70, 2.82)b
15
High TCE score
1.27 (0.52, 3.13)b
7
p for trend
p = 0.535
View-Master employees
ATSDR (2004a)
Males
0.62 (0.02, 3.45)f
1
Females
0 (1.45 exp)f
All employees at electronics factory (Taiwan)
Chang et al. (2003b)
Males
0 (3.34 exp)
Females
0 (0.83 exp)
United States uranium-processing workers (Fernald)
Ritz (1999a)
Any TCE exposure
Not reported
Light TCE exposure, >2 yr duration
2.61 (0.99, 6.88)8
12
Moderate TCE exposure, >2 yr duration
0
This document is a draft for review purposes only and does not constitute Agency policy.
660 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-110. Summary of human studies on TCE exposure and esophageal
cancer (continued)
Exposure group
Relative risk
(95% CI)
No. obs. events
Reference
Aerospace workers (Lockheed)
Boice et al. (1999)
Routine exposure
0.83 (0.34, 1.72)
7
Routine -intermittent3
Not presented
11
Duration of exposure
0 yr
1.0a
28
<1 yr
0.23 (0.05, 0.99)
2
1-4 yr
0.57 (0.20, 1.67)
4
>5 vr
0.91 (0.38,2.22)
7
p for trend
p > 0.20
Aerospace workers (Hughes)
Morgan et al. (1998)
TCE subcohort
Not reported
Low intensity (<50 ppm)
High intensity (>50 ppm)
TCE subcohort (Cox Analysis)
Not reported
Never exposed
Ever exposed
Peak
Not reported
No/Low
Medium/high
Cumulative
Not reported
Referent
Low
High
Aircraft maintenance workers (Hill AFB, UT)
Blair etal. (1998)
TCE subcohort
5.6(0.7, 44.5)a
10
Males, cumulative exposure
0
1.0a
<5 ppm-yr
Not reported11
3
5-25 ppm-yr
Not reported11
2
>25 ppm-yr
Not reported11
4
This document is a draft for review purposes only and does not constitute Agency policy.
661 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-110. Summary of human studies on TCE exposure and esophageal
cancer (continued)
Exposure group
Relative risk
(95% CI)
No. obs. events
Reference
Females, cumulative exposure
Blair etal. (1998)
(continued)
0
1.0a
<5 ppm-yr
3.6 (0.2, 58)
1
5-25 ppm-yr
0
>25 ppm-yr
0
TCE subcohort
1.88 (0.61,5.79)
17
Radican et al. (2008)
Males, cumulative exposure
1.66 (0.48, 5.74)
15
0
1.0a
<5 ppm-yr
1.84(0.48,7.14)
7
5-25 ppm-yr
1.33 (0.27, 6.59)
3
>25 ppm-yr
1.67 (0.40, 7.00)
5
Females, cumulative exposure
2.81 (0.25, 31.10)
2
0
1.0a
<5 ppm-yr
3.99 (0.25, 63.94)
1
5-25 ppm-yr
9,.59 (0.60, 154.14)
1
>25 ppm-yr
0
Cardboard manufacturing workers in Arnsburg, Germany
Henschler et al. (1995)
TCE exposed workers
Not reported
Unexposed workers
Not reported
Deaths reported to among GE pension fund
(Pittsfield, MA)
0.95 (0.1, 3.17)1
13
Greenland et al. (1994)
Cardboard manufacturing workers, Atlanta area,
GA
Not reported
Sinks et al. (1992)
U. S. Coast Guard employees
Blair etal. (1989)
Marine inspectors
0.72 (0.09, 2.62)
2
Noninspectors
0.74 (0.09, 2.68)
2
Aircraft manufacturing plant employees (Italy)
Costa et al. (1989)
All subjects
0.21 (0.01, 1.17)
1
Rubber Workers
Not reported1
Wilcosky et al. (1984)
Aircraft manufacturing plant employees (San Diego, CA)
Garabrant et al. (1988)
All subjects
1.14(0.62, 1.92)
14
This document is a draft for review purposes only and does not constitute Agency policy.
662 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Table 4-110. Summary of human studies on TCE exposure and esophageal
cancer (continued)
Exposure group
Relative risk
(95% CI)
No. obs. events
Reference
Case-control studies
Population of Montreal, Canada
Siemiatycki et al. (1991);
Parent et al. (2000b)
Any TCE exposure
0.5 (0.1, 2.5)"
1
Substantial TCE exposure
0.8 (0.1, 4.6)J
1
Geographic based studies
Residents in two study areas in Endicott, NY
0.78 (0.29, 1.70)
6
ATSDR (2006a)
Residents of 13 census tracts in Redlands, CA
Not reported
Morgan and Cassady
(2002)
Finnish residents
Vartiainen et al. (1993)
Residents of Hausjarvi
Not reported
Residents of Huttula
Not reported
"Internal referents, workers not exposed to TCE.
bRitz (1999a) and Zhao et al. (2005) reported relative risks for the combined site of esophagus and stomach.
°Sung et al. (2007) and Chang et al. (2005)—SIR for females and reflects a 10-yr lag period.
dSIR for adenocarcinoma of the esophagus.
eThe SIR for adenocarcinoma histologic type cannot be calculated because Hansen et al. (2001) do not present
expected numbers for adenocarcinoma histologic type of esophageal cancer. An approximation of the SIR for
adenocarcinoma histologic type is presented using the expected number of total number of expected esophageal
cancers for males (n = 1.4). The expected numbers of esophageal adenocarcinomas in males will be lower;
Hansen et al. (2001) noted the proportion of adenocarcinomas among the comparable Danish male population
during the later period of the study (1990-1996) as 38%. A rough approximation of the expected number of
esophageal carcinomas would be 0.5 expected cases and an approximated SIR of 9.4 (3.1, 22).
Proportional mortality ratio.
8Adjusted relative risks for >2 yr exposure duration and 15 yr lag from 1st exposure.
hNo esophageal cancer deaths occurred in the referent population in Blair et al. (1998) and relative risk in could not
be calculated for this reason.
'Odds ratio from nested case-control analysis.
J90% CI.
GE = General Electric, IBM = International Business Machines Corporation, No. obs. events = number of
observable events.
Meta-analysis is not adopted as a tool for examining the body of epidemiologic evidence
on esophageal cancer and TCE exposure given the absence of reported relative risk estimates in
several of the studies in which there is a high likelihood of TCE exposure in individual study
subjects and which met, to a sufficient degree, the standards of epidemiologic design and analysis
in a systematic review (Anttila et al., 1995; Axelson et al., 1994; Morgan et al., 1998).
This document is a draft for review purposes only and does not constitute Agency policy.
663 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Overall, three cohort studies in which there is a high likelihood of TCE exposure in
individual study subjects and which met, to a sufficient degree, the standards of epidemiologic
design and analysis in a systematic review provide some evidence of association for esophageal
cancer and TCE exposure. The finding in two of these studies of esophageal risk estimates
among subjects with long employment duration were higher than those associated with low
employment duration provides additional evidence (Hansen et al., 2001; Raaschou-Nielsen et al.,
2003). The cohort studies are unable to directly examine possible confounding due to suspected
risk factors for esophageal cancer such as smoking, obesity and alcohol. The use of an internal
referent group, similar in socioeconomic status as exposed subjects, is believed to minimize but
may not completely control for possible confounding related to smoking and health status (Blair
et al. (1998); its follow-up Radican et al. (2008); Zhao et al. (2005); Boice et al. (2006b).
Observation of a higher risk for adenocarcinoma histologic type than for a combined category of
esophageal cancer in Raaschou-Nielsen et al. (2003) also suggests minimal confounding from
smoking. Smoking is not identified as a possible risk factor for the adenocarcinoma histologic
type of esophageal cancer but is believed a risk factor for squamous cell histologic type.
Furthermore, the magnitude of lung cancer risk in Raaschou-Nielsen et al. (2003) suggests a high
smoking rate is unlikely. The lack of association with overall TCE exposure and the absence of
exposure-response patterns in the other studies of TCE exposure may reflect limitations in
statistical power, the possibility of exposure misclassification, and differences in measurement
methods. These studies do not provide evidence against an association between TCE exposure
and esophageal cancer.
4.9.2. Bladder Cancer
Twenty-five epidemiologic studies present risk estimates for bladder cancer Pesch et al.,
2000a (Anttila et al., 1995; ATSDR, 2004a, 2006a; Axelson et al., 1994; Blair et al., 1989; Blair
et al., 1998; Boice et al., 1999; Boice et al., 2006b; Chang et al., 2003b; Chang et al., 2005;
Costa et al., 1989; Garabrant et al., 1988; Greenland et al., 1994; Hansen et al., 2001; Mallin,
1990; Morgan and Cassady, 2002; Morgan et al., 1998; Raaschou-Nielsen et al., 2003; Radican
et al., 2008; Shannon et al., 1988; Siemiatycki, 1991; Sinks et al., 1992; Sung et al., 2007; Zhao
et al., 2005). Table 4-111 presents risk estimates for TCE exposure and bladder cancer observed
in cohort, case-control, and geographic based studies. Thirteen studies, all either cohort or case-
control studies, which there is a high likelihood of TCE exposure in individual study subjects
(e.g., based on job-exposure matrices or biomarker monitoring) or which met, to a sufficient
degree, the standards of epidemiologic design and analysis in a systematic review, reported
This document is a draft for review purposes only and does not constitute Agency policy.
664 DRAFT—DO NOT CITE OR QUOTE
-------
1 relative risk estimates for bladder or urothelial cancer between 0.6 (Siemiatycki, 1991) and 1.7
2 (Boice et al., 2006b)and overall TCE exposure. Relative risk estimates were generally based on
3 small numbers of cases or deaths, except for one study (Raaschou-Nielsen et al., 2003), with the
4 result of wide confidence intervals on the estimates. Of thesestudies, two reported statistically
5 significant elevated bladder or urothelial cancer risks with the highest cumulative TCE exposure
6 category (2.71, 95% CI: 1.10, 6.65 (Morgan et al., 1998); 1.8, 95% CI: 1.2, 2.7 [Pesch et al.,
7 2000b]) and five presented risk estimates and categories of increasing cumulative TCE exposure
8 Pesch et al., 2000b (Blair et al., 1998; Morgan et al., 1998; Radican et al., 2008; Zhao et al.,
9 2005). Risk estimates in
10
This document is a draft for review purposes only and does not constitute Agency policy.
665 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-111. Summary of human studies on TCE exposure and bladder
2 cancer
3
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Cohort studies—incidence
Aerospace workers (Rocketdyne)
Zhao et al. (2005)
Any exposure to TCE
Not reported
Low cumulative TCE score
1.00a
20
Medium cumulative TCE score
1.54 (0.81, 2.92)b
19
High TCE score
1.98 (0.93, 4.22)b
11
p for trend
p = 0.069
TCE, 20 yr exposure lag
Low cumulative TCE score
1.00a
20
Medium cumulative TCE score
1.76(0.61,5.10)°
20
High TCE score
3.68 (0.87, 15.5)°
10
p for trend
p = 0.064
All employees at electronics factory (Taiwan)
Males
Not reported
Sung et al. (2007)
Females
0.34 (0.07, 1.00)
10
Males
1.06 (0.45, 2.08)d
8
Chang et al. (2005)
Females
1.09 (0.56, 1.9 l)d
12
Danish blue-collar worker with TCE exposure
Raaschou-Nielsen et al.
Any exposure, all subjects
1.1 (0.92, 1.21)
220
(2003)
Any exposure, males
1.0 (0.89, 1.18)
203
Any exposure, females
1.6 (0.93, 2.57)
17
Biologically-monitored Danish workers
1.0 (0.48, 1.86)
10
Hansen etal. (2001)
Any TCE exposure, males
1.1 (0.50,2.0)
10
Any TCE exposure, females
0.5 expected
0
Aircraft maintenance workers from Hill Air Force Base
Blair et al. (1998)
TCE subcohort
Not reported
Males, cumulative exposure
0
1.0a
<5 ppm-yr
1.7(0.6,4.4)
13
5-25 ppm-yr
1.7(0.6,4.9)
9
>25 ppm-yr
1.4(0.5,4.1)
9
4
This document is a draft for review purposes only and does not constitute Agency policy.
666 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-111. Summary of human studies on TCE exposure and bladder
cancer (continued)
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Females, cumulative exposure
Blair et al. (1998) (continued)
0
1.0a
<5 ppm-yr
1.1 (0.1, 10.8)
1
5-25 ppm-yr
0
>25 ppm-yr
1.0(0.1,9.1)
1
Biologically-monitored Finnish workers
Anttila et al. (1995)
All subjects
0.82 (0.27, 1.90)
5
Biologically-monitored Swedish workers
Axelsonetal. (1994)
Any TCE exposure, males
1.02 (0.44, 2.00)
8
Any TCE exposure, females
Not reported
Cohort and PMR studies-mortality
Aerospace workers (Rocketdyne)
Any TCE (utility/eng flush)
1.66 (0.54, 3.87)
5
Boice et al. (2006b)
Any exposure to TCE
Not reported
Zhao et al. (2005)
Low cumulative TCE score
1.00a
8
Med cumulative TCE score
1.27 (0.43, 3.73)b
6
High TCE score
1.15 (0.29, 4.5 l)b
3
p for trend
p = 0.809
TCE, 20 yr exposure lag
Low cumulative TCE score
1.00a
8
Medium cumulative TCE score
0.95 (0.15, 6.02)c
7
High TCE score
1.85(0.12, 21.If
2
p for trend
p = 0.533
View-Master employees
ATSDR (2004a)
Males
1.22(0.15,4.40)
Females
0.78 (0.09, 2.82)
United States uranium-processing workers (Fernald)
Ritz (1999a)
Any TCE exposure
Not reported
Light TCE exposure, >2 yr duration
Not reported
Moderate TCE exposure, >2 yr duration
Not reported
Aerospace workers (Lockheed)
Boice et al. (1999)
Routine exposure
0.55 (0.18, 1.28)
5
Routine-intermittent3
Not reported
This document is a draft for review purposes only and does not constitute Agency policy.
667 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-111. Summary of human studies on TCE exposure and bladder
cancer (continued)
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Aerospace workers (Hughes)
Morgan et al. (1998)
TCE subcohort
1.36 (0.59, 2.68)
8
Low intensity (<50 ppm)
0.51 (0.01,2.83)
1
High intensity (>50 ppm)
1.79 (0.72, 3.69)
7
TCE subcohort (Cox Analysis)
Never exposed
1.0a
Ever exposed
2.05 (0.86, 4.85)e
8
Peak
No/low
1.0a
Medium/high
1.41 (0.52, 3.81)
5
Cumulative
Referent
1.0a
Low
0.69 (0.09, 5.36)
1
High
2.71 (1.10,6.65)
7
Aircraft maintenance workers (Hill AFB, UT)
Blair et al. (1998)
TCE subcohort
1.2 (0.5, 2.9)a
17
Males, cumulative exposure
0
1.0a
<5 ppm-yr
1.8(0.5,6.2)
7
5-25 ppm-yr
2.1 (0.6,8.0)
5
>25 ppm-yr
1.0(0.2,5.1)
3
Females, cumulative exposure
0
1.0a
<5 ppm-yr
0
5-25 ppm-yr
0
>25 ppm-yr
0.8(0.1,7.5)
1
TCE subcohort
0.80 (0.41, 1.58)
25
Radican et al. (2008)
Males, cumulative exposure
1.05 (0.47, 2.35)
24
0
1.0a
<5 ppm-yr
0.96 (0.37,2.51)
9
5-25 ppm-yr
1.77 (0.70, 4.52)
10
>25 ppm-yr
0.67 (0.15,2.95)
5
This document is a draft for review purposes only and does not constitute Agency policy.
668 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-111. Summary of human studies on TCE exposure and bladder
cancer (continued)
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Females, cumulative exposure
0.22 (0.03, 1.83)
1
Radican et al. (2008)
0
1.0a
(continued)
<5 ppm-yr
0
5-25 ppm-yr
2.86 (0.27, 29.85)
1
>25 ppm-yr
0
Cardboard manufacturing workers in Arnsburg, Germany
Henschler et al. (1995)
TCE exposed workers
Not reported
Unexposed workers
Not reported
Deaths reported to GE pension fund (Pittsfield,
MA)
0.85 (0.32, 2.23)f
20
Greenland et al. (1994)
Cardboard manufacturing workers, Atlanta area, GA
Sinks et al. (1992)
0.3 (0.0, 1.6)
1
U. S. Coast Guard employees
Blair et al. (1989)
Marine inspectors
0.50 (0.06, 1.79)
2
Noninspectors
0.90(0.18,2.62)
3
Aircraft manufacturing plant employees (Italy)
Costa et al. (1989)
All subjects
0.74 (0.30, 1.53)
7
Aircraft manufacturing plant employees (San Diego, CA)
Garabrant et al. (1988)
All subjects
1.26 (0.74, 2.03)
17
Lamp manufacturing workers (GE)
0.93 (0.19, 2.72)
3
Shannon et al. (1988)
Case-control studies
Population of 5 regions in Germany
Pesch et al., 2000a
Any TCE exposure
Not reported
Males
Not reported
Females
Not reported
Males
Medium
0.8 (0.6, 1.2)8
47
High
1.3 (0.8, 1.7)8
74
Substantial
1.8 (1.2, 2.7)g
36
Population of Montreal, Canada
Siemiatycki (1991);
Any TCE exposure
0.6(0.3, 1.2)
8
Siemiatycki et al. (1994)
Substantial TCE exposure
0.7(0.3, 1.6)
5
Geographic based studies
Residents in two study areas in Endicott, NY
ATSDR (2006a)
0.71 (0.38, 1.21)
13
This document is a draft for review purposes only and does not constitute Agency policy.
669 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Table 4-111. Summary of human studies on TCE exposure and bladder
cancer (continued)
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Residents of 13 census tracts in Redlands, CA
Morgan and Cassady (2002)
0.98 (0.71, 1.29)h
82
Finnish residents
Vartiainen et al. (1993)
Residents of Hausjarvi
Not reported
Residents of Huttula
Not reported
Residents of 9 county area in Northwestern Illinois
Mallin (1990)
All zip codes in study area
Males
1.4(1.1, 1.9)
47
Females
1.8(1.2,2.7)
21
Cluster community
Males
1.7(1.1,2.6)
21
Females
2.6(1.2,4.7)
10
Adjacent community
Males
1.2(0.6,2.0)
12
Females
1.6(0.5,3.8)
5
Remainder of zip code areas
Males
1.4(0.8,2.2)
14
Females
1.4(0.5,3.0)
6
internal referents, workers not exposed to TCE.
bRelative risk estimates for TCE exposure after adjustment for 1st employment, socioeconomic status, and age at
event.
°Relative risk estimates for TCE exposure after adjustment for 1st employment, socioeconomic status, age at event,
and all other carcinogen exposures, including hydrazine.
dChang et al. (2005) and Costa et al. (1989) report estimated risks for a combined site of all urinary organ cancers.
eRisk ratio from Cox Proportional Hazard Analysis, stratified by age, sex and decade (EHS, 1997).
fOdds ratio from nested case-control analysis.
8Odds ratio for urothelial cancer, a category of bladder, ureter, and renal pelvis cancers) and cumulative TCE
exposure, as assigned using a JTEM approach (Pesch et al., 2000a).
h99% CI.
GE = General Electric, No. obs. events = number of observable events.
Morgan et al. (1998), Pesch et al. (2000a), and Zhao et al. (2005) appeared to increase with
increasing cumulative TCE exposure with the />value for trend of 0.07 in Zhao et al. (2005), the
only study to present a formal statistical test for linear trend. Risk estimates did not appear to
either increase or decrease with increasing cumulative TCE exposure in Blair et al. (1998) or its
update Radican et al. (2008), which added another 10 years of follow-up. Twelve additional
This document is a draft for review purposes only and does not constitute Agency policy.
670 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
studies were given less weight because of their lesser likelihood of TCE exposure and other
design limitations that would decrease statistical power and study sensitivity (ATSDR, 2004a,
2006a; Blair et al., 1989; Chang et al., 2003b; Chang et al., 2005; Costa et al., 1989; Garabrant et
al., 1988; Mallin, 1990; Morgan and Cassady, 2002; Shannon et al., 1988; Sinks et al., 1992;
Sung et al., 2007).
Meta-analysis is not adopted as a tool for examining the body of epidemiologic evidence
on bladder cancer and TCE.
Overall, three cohort or case-control studies in which there is a high likelihood of TCE
exposure in individual study subjects and which met, to a sufficient degree, the standards of
epidemiologic design and analysis in a systematic review provide some evidence of association
for bladder or urothelial cancer and high cumulative TCE exposure Pesch et al., 2000a (Morgan
et al., 1998; Zhao et al., 2005). The case-control study of Pesch et al. (2000a) adjusted for age,
study center, and cigarette smoking, with a finding of a statistically significant risk estimate
between urothelial cancer and the highest TCE exposure category. Cancer cases in this study are
of several sites, bladder, ureter, and renal pelvis, and grouping different site-specific cancers with
possible etiologic heterogeneity may introduce misclassification bias. The cohort studies are
unable to directly examine possible confounding due to suspected risk factors for esophageal
cancer such as smoking, obesity, and alcohol. The use of an internal referent group, similar in
socioeconomic status as exposed subjects, by Morgan et al. (1998) and Zhao et al. (2005) is
believed to minimize but may not completely control for possible confounding related to
smoking and health status. The lack of association with overall TCE exposure in other studies
and the absence of exposure-response patterns with TCE exposure in Blair et al. (1998) and
Radican et al. (2008) may reflect limitations in statistical power, the possibility of exposure
misclassification, and differences in measurement methods. These studies do not provide
evidence against an association between TCE exposure and bladder cancer.
4.9.3. Central Nervous System and Brain Cancers
Brain cancer is examined in most cohort studies and in one case-control study (Anttila et
al., 1995; Blair et al., 1989; Blair et al., 1998; Boice et al., 1999; Boice et al., 2006b; Chang et
al., 2003b; Chang et al., 2005; Clapp and Hoffman, 2008; Costa et al., 1989; Garabrant et al.,
1988; Greenland et al., 1994; Hansen et al., 2001; Heineman et al., 1994; Henschler et al., 1995;
Morgan et al., 1998; Raaschou-Nielsen et al., 2003; Radican et al., 2008; Ritz, 1999a; Sung et
al., 2007; Zhao et al., 2005). Overall, these epidemiologic studies do not provide strong
evidence for or against association between TCE and brain cancer in adults (see Table 4-112).
This document is a draft for review purposes only and does not constitute Agency policy.
671 DRAFT—DO NOT CITE OR QUOTE
-------
1 Relative risk estimates in well designed and conducted cohort studies, Axelson et al. (1994),
2 Anttila et al. (1995), Blair et al. (1998), its follow-up reported in Radican et al. (2008), Morgan
3 et al. (1998), Boice et al. (1999), Zhao et al. (2005), and Boice et al. (2006b), are near a risk of
4 1.0 and imprecise, confidence intervals all include a risk estimate of 1.0. All studies except
5 Raaschou-Nielsen et al. (2003),
6
This document is a draft for review purposes only and does not constitute Agency policy.
672 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-112. Summary of human studies on TCE exposure and brain cancer
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Cohort studies—incidence
Aerospace workers (Rocketdyne)
Zhao et al. (2005)
Any exposure to TCE
Not reported
Low cumulative TCE score
1.00a
7
Medium cumulative TCE score
0.46 (0.09, 2.25)b
2
High TCE score
0.47 (0.06, 3.95)b
1
p for trend
p= 0.382
All employees at electronics factory (Taiwan)
Males
Not reported
Sung et al. (2007)
Females
1.07 (0.59, 1.80)°
Males
0.40 (0.05, 1.46)
2
Chang et al. (2005)
Females
0.97 (0.54, 1.61)
15
Danish blue-collar worker with TCE exposure
Raaschou-Nielsen et al.
Any exposure, all subjects
1.0 (0.84, 1.24)
104
(2003)
Any exposure, males
1.0 (0.76, 1.18)
85
Any exposure, females
1.1 (0.67, 1.74)
19
Biologically-monitored Danish workers
0.3 (0.01, 1.86)
1
Hansen etal. (2001)
Any TCE exposure, males
0.4(0.01,2.1)
1
Any TCE exposure, females
0.5 expected
0
Aircraft maintenance workers from Hill Air Force Base
Blair etal. (1998)
TCE subcohort
Not reported
Males, cumulative exposure
0
1.0a
<5 ppm-yr
2.0 (0.2, 19.7)
3
5-25 ppm-yr
3.9 (0.4, 34.9)
4
>25 ppm-yr
0.8(0.1, 13.2)
1
Females, cumulative exposure
0
1.0a
<5 ppm-yr
0
5-25 ppm-yr
0
>25 ppm-yr
0
Biologically-monitored Finnish workers
Anttila et al. (1995)
All subjects
1.09 (0.50, 2.07)
9
Mean air-TCE (Ikeda extrapolation)
<6 ppm
1.52(0.61,3.13)
7
6+ ppm
0.76 (0.01, 2.74)
2
Biologically-monitored Swedish workers
Axelsonetal. (1994)
Any TCE exposure, males
Not reported
Any TCE exposure, females
Not reported
This document is a draft for review purposes only and does not constitute Agency policy.
673 DRAFT—DO NOT CITE OR QUOTE
-------
1
This document is a draft for review purposes only and does not constitute Agency policy.
674 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-112. Summary of human studies on TCE exposure and brain cancer
(continued)
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Cohort and PMR studies-mortality
Computer manufacturing workers (IBM), NY
Clapp and Hoffman (2008)
Males
1.90 (0.52, 4.85)
4
Females
0
Aerospace workers (Rocketdyne)
Any TCE (utility/eng flush)
0.81 (0.17, 2.36)
3
Boice et al. (2006b)
Any exposure to TCE
Not reported
Zhao et al. (2005)
Low cumulative TCE score
1.00a
12
Medium cumulative TCE score
0.42(0.12, 1.50)
3
High TCE score
0.83 (0.23, 3.08)
3
p for trend
p = 0.613
View-Master employees
ATSDR (2004a)
Males
Not reported
Females
Not reported
All employees at electronics factory (Taiwan)
Chang et al. (2003b)
Males
0.96 (0.01, 5.36)
1
Females
0.96 (0.01,5.33)
1
United States uranium-processing workers (Fernald)
Ritz (1999a)
Any TCE exposure
Not reported
Light TCE exposure, >2 yr duration, 0 lag
1.81 (0.49, 6.71)d
6
Moderate TCE exposure, >2 yr duration, 0
lag
3.26 (0.37, 28.9)d
1
Light TCE exposure, >5 yr duration, 15 yr
lag
5.41 (0.87, 33.9)d
3
Moderate TCE exposure, >5 yr duration,
15 yr lag
14.4 (1.24, 167)d
1
Aerospace workers (Lockheed)
Boice et al. (1999)
Routine exposure
0.54 (0.15, 1.37)
4
Routine-intermittent3
Not presented
Aerospace workers (Hughes)
Morgan et al. (1998)
TCE subcohort
0.99 (0.64, 1.47)
4
Low intensity (<50 ppm)d
0.73 (0.09, 2.64)
2
High intensity (>50 ppm)d
0.44 (0.05, 1.58)
2
Aircraft maintenance workers (Hill AFB, UT)
Blair etal. (1998)
TCE subcohort
0.8 (0.2, 2.2)a
11
This document is a draft for review purposes only and does not constitute Agency policy.
675 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-112. Summary of human studies on TCE exposure and brain cancer
(continued)
Exposure group
Relative risk
(95% CI)
No. obs.
events
Reference
Males, cumulative exposure
Blair et al. (1998) (continued)
0
1.0a
<5 ppm-yr
0.7 (0.7, 3,3)
3
5-25 ppm-yr
2.0 (0.5, 8.4)
5
>25 ppm-yr
0.9 (0.2, 4.4)
2
Females, cumulative exposure
0
1.0a
<5 ppm-yr
0
5-25 ppm-yr
0
>25 ppm-yr
0
TCE subcohort
1.02 (0.39, 2.67)
17
Radican et al. (2008)
Males, cumulative exposure
1.26 (0.43, 3.75)
17
0
1.0a
<5 ppm-yr
1.46 (0.44, 4.86)
8
5-25 ppm-yr
1.74(0.49,6.16)
6
>25 ppm-yr
0.66 (0.15,2.95)
3
Females, cumulative exposure
0
0
<5 ppm-yr
5-25 ppm-yr
>25 ppm-yr
Cardboard manufacturing workers in Arnsburg, Germany
Henschler et al. (1995)
TCE exposed workers
3.70 (0.09, 20.64)
1
Unexposed workers
9.38 (1.93,27.27)
3
Deaths reported to GE pension fund (Pittsfield,
MA)
0.93 (0.32, 2.69)e
16
Greenland et al. (1994)
Cardboard manufacturing workers, Atlanta area, GA
Sinks et al. (1992)
Not reported
U. S. Coast Guard employees
Blair etal. (1989)
Marine inspectors
1.70 (0.55, 3.95)
5
Noninspectors
1.36 (0.44,3.17)
5
Aircraft manufacturing plant employees (Italy)
Costa et al. (1989)
All subjects
0.79 (0.16,2.31)
3
Aircraft manufacturing plant employees (San Diego, CA)
Garabrant et al. (1988)
All subjects
0.78 (0.42, 1.34)
16
Case-control studies
Children's Cancer Group/Pediatric Oncology Group
De Roos et al. (2001)
Any TCE exposure
1.64 (0.95, 2.84)
37
This document is a draft for review purposes only and does not constitute Agency policy.
676 DRAFT—DO NOT CITE OR QUOTE
-------
Table 4-112. Summary of human studies on TCE exposure and brain cancer
(continued)
Relative risk
No. obs.
Exposure group
(95% CI)
events
Reference
Neuroblastoma, <15 yr age
Paternal TCE exposure
De Roos et al. (2001)
Self-reported exposure
1.4 (0.7, 2.9)
22
(continued)
IH assignment of probable exposure
0.9 (0.3, 2.5)
9
Population of So. LA, NJ, Philadelphia, PA
Heineman et al. (1994)
Any TCE exposure
1.1 (0.8, 1.6)
128
Low exposure
1.1 (0.7, 1.7)
27
Medium exposure
1.1 (0.6, 1.8)
42
High exposure
1.1 (0.5,2.8)
12
p for trend
0.45
Geographic based studies
Residents in two study areas in Endicott, NY
ATSDR (2006a)
Brain/CNS, <19 yr of age
Not reported
<6
Residents of 13 census tracts in Redlands, CA
Morgan and Cassady (2002)
Brain/CNS, <15 yr of age
1.05 (0.24, 2.70)f
6
Resident of Tucson Airport Area, AZ
AZDHS (1990, 1995)
Brain/CNS, <19 yr of age
1970-1986
0.84(0.23,2.16)
3
1987-1991
0.78 (0.26, 2.39)
2
1
2 internal referents, workers not exposed to TCE.
3 bRelative risks for TCE exposure after adjustment for 1st employment, socioeconomic status, and age at event.
4 Standardized incidence ratio from analyses lagging exposure 10 yrs prior to end of follow-up or date of incident
5 cancer.
6 dRelative risks for TCE exposure after adjustment for time since 1st hired, external and internal radiation dose, and
7 same chemical at a different level.
8 eOdds ratio from nested case-control analysis.
9 f99% CI.
10
11 IBM = International Business Machines Corporation, No. obs. events = number of observable events.
12
13
14 observations are based on few events and lowered statistical power. Bias resulting from
15 exposure misclassification is likely in these studies, although of a lower magnitude compared to
16 other cohort studies identified in Table 4-112, and may partly explain observations. Exposure
17 misclassification is also likely in the case-control study of occupational exposure of Heineman et
18 al. (1994) who do not report association with TCE exposure.
19 Three geographic-based studies and one case-control study examined childhood brain
20 cancer (ADHS, 1990, 1995; ATSDR, 2006a; De Roos et al., 2001; Morgan and Cassady, 2002).
This document is a draft for review purposes only and does not constitute Agency policy.
677 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
The strongest study, De Roos et al. (2001), a population case-control study which examined
paternal exposure, used expert judgment to evaluate the probably of TCE exposure from self-
reported information in an attempt to reduce exposure misclassification bias. The odds ratio
estimate in this study was 0.9 (95% CI: 0.3, 2.5). Like many population case-control studies, a
low prevalence of TCE exposure was found, only 9 fathers were identified with probable TCE
exposure by the industrial hygiene review, and greatly impacted statistical power. There is some
concern for childhood brain cancer and organic solvent exposure based on Peters et al. (1981)
whose case-control study of childhood brain cancer reported to the Los Angeles County Cancer
Surveillance Program observed a high odds ratio estimate for paternal employment in the aircraft
industry (OR: co,p< 0.001). This study does not present an odds ratio for TCE exposure only
although it did identify two of the 14 case and control fathers with previous employment in the
aircraft industry reported exposure to TCE.
4.10. SUSCEPTIBLE LIFESTAGES AND POPULATIONS
Variation in response among segments of the population may be due to age, genetics, and
ethnicity, as well as to differences in lifestyle, nutrition, and disease status. These could be
potential risk factors that play an important role in determining an individual's susceptibility and
sensitivity to chemical exposures. Available studies on TCE toxicity in relation to some of these
risk factors including lifestage, gender, genetics, race/ethnicity, preexisting health status, and
lifestyle are discussed below. However, there is a general lack of data demonstrating the
modulation of health effects from TCE exposure based on these factors. Additional data
examining these factors would provide further understanding of the populations that may be
more susceptible to the health effects from TCE exposure. Others have also reviewed factors
related to human variability and their potential for susceptibility to TCE (ATSDR, 1997b, 1998a;
Barton et al., 1996; Clewell et al., 2000; Davidson and Beliles, 1991; NRC, 2006; Pastino et al.,
2000).
4.10.1. Lifestages
Individuals of different lifestages are physiologically, anatomically, and biochemically
different. Early (infants and children) and later (the elderly) lifestages differ greatly from
adulthood in body composition, organ function, and many other physiological parameters that
can influence the toxicokinetics of chemicals and their metabolites in the body (ILSI, 1992). The
limited data on TCE exposure among these segments of the population—particularly individuals
This document is a draft for review purposes only and does not constitute Agency policy.
678 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
in early lifestages—suggest they may have greater susceptibility than does the general
population. This section presents and evaluates the pertinent published literature available to
assess how individuals of differing lifestages may respond differently to TCE.
4.10.1.1.1. Early Lifestages
4.10.1.1.2. Early lifestage-specific exposures
Section 2.4 describes the various exposure pathways of concern for TCE. For all
postnatal lifestages, the primary exposure routes of concern include inhalation and contaminated
drinking water. In addition, there are exposure pathways to TCE are unique to early lifestages.
Fetal and infant exposure to TCE can occur through placental transfer and breast milk
consumption if the mother has been exposed, and could potentially increase overall TCE
exposure. Placental transfer of TCE has been demonstrated in humans (Beppu, 1968; Laham,
1970), rats (Withey and Karpinski, 1985), mice (Ghantous et al., 1986), rabbits (Beppu, 1968),
and sheep and goats (Helliwell and Hutton, 1950). Similarly, TCE has been found in breast milk
in humans (Fisher et al., 1997; Pellizzari et al., 1982), goats (Hamada and Tanaka, 1995), and
rats (Fisher et al., 1990). Pellizzari et al. (1982) conducted a survey of environmental
contaminants in human milk, using samples from cities in the northeastern region of the United
States and one in the southern region and detected TCE in 8 milk samples taken from 42
lactating women. No details of when the samples were taken postpartum, milk lipid content, or
TCE concentration in milk or blood were reported. Fisher et al. (1997) predicted that a nursing
infant would consume 0.496 mg TCE during a 24-hour period. In lactating rats exposed to 600
"3
ppm (3,225 mg/m ) TCE for 4 hours resulted in concentrations of TCE in milk of 110 |ig/mL
immediately following the cessation of exposure (Fisher et al., 1990).
Direct childhood exposures to TCE from oral exposures may also occur. A
contamination of infant formula resulted in levels of 13 ppb (Fan, 1988). Children consume high
levels of dairy products, and TCE has been found in butter and cheese (Wu and Schaum, 2000).
In addition, TCE has been found in food and beverages containing fats such as margarine
(Wallace et al., 1984), grains and peanut butter (Wu and Schaum, 2000), all of which children
consume in high amounts. A number of studies have examined the potential adverse effects of
prenatal or postnatal exposure to drinking water contaminated with TCE (ATSDR, 1998a, 2001;
Bernad et al., 1987, abstract; Bove, 1996; Bove et al., 1995; Burg and Gist, 1999; Goldberg et
al., 1990; Lagakos et al., 1986; Rodenbeck et al., 2000; Sonnenfeld et al., 2001; White et al.,
1997; see Section 4.10.2.1). TCE in residential water may also be a source of dermal or
inhalation exposure during bathing and showering (Fan, 1988; Franco et al., 2007; Giardino and
This document is a draft for review purposes only and does not constitute Agency policy.
679 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Andelman, 1996; Lee et al., 2002; Weisel and Jo, 1996; Wu and Schaum, 2000); it has been
estimated that showering and bathing scenarios in water containing 3-ppm TCE, a child of 22 kg
receives a higher dose (about 1.5 times) on a mg/kg basis than a 70 kg adult (Fan, 1988).
Direct childhood inhalation exposure to TCE have been documented in both urban and
rural settings. A study of VOCs measured personal, indoor and outdoor TCE in 284 homes, with
72 children providing personal measures and time-activity diaries (Adgate et al., 2004a). The
"3
intensive-phase of the study found a mean personal level of 0.8 |ig/m and mean indoor and
outdoor levels of 0.6 |ig/m , with urban homes have significantly higher indoor levels of TCE
than nonurban homes (t = 2.3, p = 0.024) (Adgate et al., 2004a). A similar study of personal,
indoor and outdoor TCE was conducted in two inner-city elementary schools as well as in the
homes of 113 children along with time-activity diaries, and found a median a median personal
3 3
level of 0.3 |ig/m , a median school indoor level of 0.2 |ig/m , a median home indoor level of
3 3
0.3 |ig/m , a median outdoor level of 0.3 |ig/m in the winter, with slightly lower levels in the
spring (Adgate et al., 2004b). Studies from Leipzig, Germany measured the median air level of
3 3
TCE in children's bedrooms to be 0.42 |ig/m (Lehmann et al., 2001) and 0.6 |ig/m (Lehmann et
al., 2002). A study of VOCs in Hong Kong measured air levels in schools, including an 8-hour
"3
average of 1.28 |ig/m , which was associated with the lowest risk of cancer in the study (Guo et
al., 2004). Another found air TCE levels to be highest in school/work settings, followed by
outside, in home, in other, and in transit settings (Sexton et al., 2007). Measured indoor air
"3
levels ranged from 0.18-140 (J,g/m for children exposed through vapor intrusion from soil vapor
(ATSDR, 2006b). Contaminated soil may be a source of either dermal or ingestion exposure of
TCE for children (Wu and Schaum, 2000).
Additional TCE exposure has also been documented to have occurred during medical
procedures. TCE was used in the past as an anesthetic during childbirth (Beppu, 1968; Phillips
and Macdonald, 1971) and surgery during childhood (Jasinska, 1965). These studies are
discussed in more detail in Section 4.8.3.1.1. In addition, the TCE metabolite chloral hydrate has
been used as an anesthetic for children for CAT scans (Steinberg, 1993).
Dose received per body weight for 3-ppm TCE via oral, dermal, dermal plus inhalation,
and bathing scenarios was estimated for a 10-kg infant, a 22-kg child, and a 70-kg adult (Fan,
1988; see Table 4-113). For the oral route (drinking water), an infant would receive a higher
daily dose than a child, and the child more than the adult. For the dermal and dermal plus
inhalation route, the child would receive more than the adult. For the bathing scenario, the infant
and child would receive comparable amounts, more than the adult.
This document is a draft for review purposes only and does not constitute Agency policy.
680 DRAFT—DO NOT CITE OR QUOTE
-------
4.10.1.1.3. Early lifestage-specific toxicokinetics
1 Section 3 describes the toxicokinetics of TCE. However, toxicokinetics in developmental
2 lifestages are distinct from toxicokinetics in adults (Benedetti et al., 2007; Ginsberg et al., 2004a;
3 Ginsberg et al., 2004b; Ginsberg et al., 2002; Hattis et al., 2003) due to, for example, altered
4 ventilation rates, percentage of adipose tissue, and metabolic enzyme expression. Early
5 lifestage-specific information is described below for absorption, distribution, metabolism, and
6 excretion, followed by available early lifestage-specific PBPK models.
7
This document is a draft for review purposes only and does not constitute Agency policy.
681 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 4-113. Estimated lifestage-specific daily doses for TCE in water"
2
Body weight
Infant (10 kg)
Child (22 kg)
Adult (70 kg)
Drinking water
0.3 mg/kg
0.204 mg/kg
0.086 mg/kg
Showering—dermal
-
0.1 mg/kg
0.064 mg/kg
Showering—dermal
and inhalation
-
0.129 mg/kg
0.083 mg/kg
Bathing—15 min
-
0.24 mg/kg
0.154 mg/kg
Bathing—5 min
0.08 mg/kg
0.08 mg/kg
0.051 mg/kg
3
4 a Adapted from Fan (1988).
5
6
4.10.1.1.3.1. Absorption
7 As discussed in Section 3.1, exposure to TCE may occur via inhalation, ingestion, and
8 dermal absorption. In addition, prenatal exposure may result in absorption via the transplacental
9 route. Exposure via inhalation is proportional to the ventilation rate, duration of exposure, and
10 concentration of expired air, and children have increased ventilation rates per kg body weight
11 compared to adults, with an increased alveolar surface area per kg body weight for the first two
12 years (U.S. EPA, 2008b). It is not clear to what extent dermal absorption may be different for
13 children compared to adults; however, infants have a twofold increase in surface area compared
14 to adults, although similar permeability (except for premature babies) compared to adults (U.S.
15 EPA, 2008b).
16
4.10.1.1.3.2. Distribution
17 Both human and animal studies provide clear evidence that TCE distributes widely to all
18 tissues of the body (see Section 3.2). For lipophilic compounds such as TCE, percentage adipose
19 tissue, which varies with age, will affect absorption and retention of the absorbed dose. Infants
20 have a lower percentage of adipose tissue per body weight than adults, resulting in a higher
21 concentration of the lipophilic compound in the fat of the child (NRC, 1993).
22 During pregnancy of humans and experimental animals, TCE is distributed to the
23 placenta (Beppu, 1968; Ghantous et al., 1986; Helliwell and Hutton, 1950; Laham, 1970; Withey
24 and Karpinski, 1985). In humans, TCE has been found in newborn blood after exposure to TCE
25 during childbirth with ratios of concentrations in fetal:maternal blood ranging from
This document is a draft for review purposes only and does not constitute Agency policy.
682 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
approximately 0.5 to approximately 2 (Laham, 1970). In childhood, blood levels concentrations
of TCE were found to range from 0.01-0.02 ng/mL (Sexton et al., 2005). Pregnant rats exposed
to TCE vapors on GD 17 resulted in concentrations of TCE in fetal blood approximately one-
third the concentration in corresponding maternal blood, and was altered based upon the position
along the uterine horn (Withey and Karpinski, 1985). TCE has also been found in the organs of
prenatal rabbits including the brain, liver, kidneys and heart (Beppu, 1968). Rats prenatally
exposed to TCE had increased levels measured in the brain at PND 10, compared to rats exposed
as adults (Rodriguez et al., 2007). TCE can cross the blood:brain barrier during both prenatal
and postnatal development, and may occur to a greater extent in younger children. It is also
important to note that it has been observed in mice that TCE can cycle from the fetus into the
amniotic fluid and back to the fetus (Ghantous et al., 1986).
Studies have examined the differential distribution by age to a mixture of six VOCs
including TCE to children aged 3-10 years and adults aged 20-82 years old (Mahle et al., 2007)
and in rats at PND 10, 2 months (adult), and 2 years (aged) (Mahle et al., 2007; Rodriguez et al.,
2007). In humans, the blood:air partition coefficient for male or female children was
significantly lower compared to adult males (Mahle et al., 2007). In rats, the difference in
tissue:air partition coefficients increased with age (Mahle et al., 2007). Higher peak
concentrations of TCE in the blood were observed in the PND 10 rat compared to the adult rat
after inhalation exposure, likely due to the lower metabolic capacity of the young rats (Rodriguez
et al., 2007).
4.10.1.1.3.3. Metabolism
Section 3.3 describes the enzymes involved in the metabolism of TCE, including CYP
and GST. Expression of these enzymes changes during various stages of fetal development
(Dome et al., 2005; Hakkola et al., 1996a; Hakkola et al., 1998a; Hakkola et al., 1996b; 1998b;
Hines and McCarver, 2002; Shao et al., 2007; van Lieshout et al., 1998) and during postnatal
development (Blake et al., 2005; Dome et al., 2005; Tateishi et al., 1997), and may result in
altered susceptibility.
Expression of CYP enzymes have been shown to play a role in decreasing the
metabolism of TCE during pregnancy in rats, although metabolism increased in young rats
(3-week-old) compared to adult rats (18-week-old) (Nakajima et al., 1992b). For TCE, CYP2E1
is the main metabolic CYP enzyme, and expression of this enzyme has been observed in humans
in prenatal brain tissue at low levels beginning at 8-weeks gestation and increasing throughout
gestation (Brzezinski et al., 1999). Very low levels of CYP2E1 have been detected in some
This document is a draft for review purposes only and does not constitute Agency policy.
683 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
samples fetal liver during the second trimester (37% of samples) and third trimester (80% of
samples) (Carpenter et al., 1996; Johnsrud et al., 2003), although hepatic expression surges
immediately after birth in most cases (Johnsrud et al., 2003; Vieira et al., 1996) and in most
infants reaches adult values by 3 months of age (Johnsrud et al., 2003; Vieira et al., 1996).
Although there is some uncertainty as to which GST isoforms mediate TCE conjugation,
it should be noted that their expression changes with fetal development (McCarver and Hines,
2002; Raijmakers et al., 2001; van Lieshout et al., 1998).
4.10.1.1.3.4. Excretion
The major processes of excretion of TCE and its metabolites are discussed in Section 3.4,
yet little is known about whether there are age-related differences in excretion of TCE. The
major pathway for elimination of TCE is via exhalation, and its metabolites via urine and feces,
and it is known that renal processes are not mature until about 6 months of age (NRC, 1993).
Only one case study was identified that measured TCE or its metabolites in exhaled breath and
urine in a 17-year old who ingested a large quantity of TCE (Briining et al., 1998). TCE has also
been measured in the breast milk in lactating women (Fisher et al., 1997; Pellizzari et al., 1982),
goats (Hamada and Tanaka, 1995), and rats (Fisher et al., 1990).
4.10.1.1.3.5. Physiologically based pharmacokinetic (PBPK) models
Early lifestage-specific information regarding absorption, distribution, metabolism, and
excretion needs to be considered for a child-specific and chemical-specific PBPK model. To
adequately address the risk to infants and children, age-specific parameters for these values
should be used in PBPK models that can approximate the internal dose an infant or child receives
based on a specific exposure level (see Section 3.5).
Fisher et al. developed PBPK models to describe the toxicokinetics of TCE in the
pregnant rat (Fisher et al., 1989), lactating rat and nursing pup (Fisher et al., 1990). The prenatal
study demonstrates that approximately two-thirds of maternal exposure to both TCE and TCA
reached the fetus after maternal inhalation, gavage, or drinking water exposure (Fisher et al.,
1989). After birth, only 2% of maternal exposure to TCE reaches the pup; however, 15% and
30%) of maternal TCA reaches the pup after maternal inhalation and drinking water exposure,
respectively (Fisher et al., 1990). One analysis of PBPK models examined the variability in
response to VOCs including TCE between adults and children, and concluded that the
This document is a draft for review purposes only and does not constitute Agency policy.
684 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
intraspecies uncertainty factor for pharmacokinetics is sufficient to capture variability between
adults and children (Pelekis et al., 2001).
4.10.1.1.4. Early lifestage-specific effects
Although limited data exist on TCE toxicity as it relates to early lifestages, there is
enough information to discuss the qualitative differences. In addition to the evidence described
below, Section 4.8 contains information on reproductive and developmental toxicity. In
addition, Sections 4.3 on neurotoxicity and Section 4.6 on immunotoxicity characterize a wide
array of postnatal developmental effects.
4.10.1.1.4.1. Differential noncancer outcomes in early lifestages
Some adverse health outcomes, in particular birth defects, are observed only after early
lifestage exposure to TCE. A summary of structural developmental outcomes that have been
associated with TCE exposures is presented in Sections 4.8.2.3.
Cardiac birth defects have been observed after exposure to TCE in humans (ATSDR,
2006a; Goldberg et al., 1990; Lagakos et al., 1986; Yauck et al., 2004), rodents (Johnson et al.,
2005)(Dawson et al., 1990a, 1993; Johnson et al., 1998a; Johnson et al., 2003; Johnson et al.,
1998b; Smith et al., 1992; Smith et al., 1989), and chicks (Boyer et al., 2000; Bross et al., 1983;
Drake et al., 2006a; Drake et al., 2006b; Loeber et al., 1988; Mishima et al., 2006; Rufer et al.,
2008). However, it is notable that cardiac malformations were not observed in a number of other
studies in humans (Lagakos et al., 1986; Taskinen et al., 1989; Tola et al., 1980), rodents
(Carney et al., 2006; Coberly et al., 1992; Cosby and Dukelow, 1992; Dorfmueller et al., 1979;
Fisher et al., 2001; Hardin et al., 1981; Healy et al., 1982; Narotsky and Kavlock, 1995;
Narotsky et al., 1995; Schwetz et al., 1975), and rabbits (Hardin et al., 1981). See Section
4.8.2.3.2 for further discussion on cardiac malformations.
Structural CNS birth defects were observed in humans (ATSDR, 2001; Bove, 1996; Bove
et al., 1995; Lagakos et al., 1986). In addition, a number of postnatal nonstructural adverse
effects on the CNS system have been observed in humans and experimental animals following
prenatal exposure to TCE. See Sections 4.3.10 and 4.8.2.3.3 for further discussion on
developmental neurotoxicity.
A variety of other birth defects have been observed—including eye/ear birth anomalies in
humans and rats (Lagakos et al., 1986; Narotsky and Kavlock, 1995; Narotsky et al., 1995);
lung/respiratory tract disorders in humans and mice (Das and Scott, 1994; Lagakos et al., 1986);
This document is a draft for review purposes only and does not constitute Agency policy.
685 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
and oral cleft defects (Bove, 1996; Bove et al., 1995; Lagakos et al., 1986), kidney/urinary tract
disorders, musculoskeletal birth anomalies (Lagakos et al., 1986), and anemia/blood disorders
(Burg and Gist, 1999) in humans. See Section 4.8.2.3.5 for further discussion on other structural
developmental outcomes. A current follow-up study of the Camp Lejeune cohort will examine
birth defects and may provide additional insight (ATSDR, 2003c, 2009; U.S. GAO, 2007).
4.10.1.1.4.2. Susceptibility to noncancer outcomes in early lifestages
There are a number of adverse health outcomes observed after exposure to TCE that are
observed in both children and adults. Below is a discussion of differential exposure, incidence
and/or severity in early lifestages compared to adulthood.
Occupational TCE poisonings via inhalation exposure resulted in an elevated percentage
of cases in the adolescents aged 15-19 years old compared those >20 years old (McCarthy and
Jones, 1983). In addition, there is concern for intentional exposure to TCE during adolescence,
including a series of deaths involving inhaling typewriter correction fluid (King et al., 1985) a
case of glue sniffing likely associated with cerebral infarction in a 12-year-old boy with a 2-year
history of exposure (Parker et al., 1984), and a case of attempted suicide by ingestion of 70 mg
TCE in a 17-year-old boy (Briining et al., 1998).
4.10.1.1.4.2.1. Neurotoxicity
Adverse CNS effects observed after early lifestage exposure to TCE in humans include
delayed newborn reflexes (Beppu, 1968); impaired learning or memory (Bernad et al., 1987,
abstract; White et al., 1997); aggressive behavior (Bernad et al., 1987, abstract; Blossom et al.,
2008); hearing impairment (Burg and Gist, 1999; Burg et al., 1995); speech impairment (Burg
and Gist, 1999; Burg et al., 1995; White et al., 1997); encephalopathy (White et al., 1997)
impaired executive and motor function (White et al., 1997); attention deficit (Bernad et al., 1987,
abstract) (White et al., 1997), and autism spectrum disorder (Windham et al., 2006). One
analysis observed a trend for increased adversity during development, with those exposed during
childhood demonstrating more deficits than those exposed during adulthood (White et al., 1997).
In experimental animals, observations include decreased specific gravity of newborn brains until
weaning (Westergren et al., 1984), reductions in myelination in the brains at weaning,
significantly decreased uptake of 2-deoxyglucose in the neonatal rat brain, significant increase in
exploratory behavior (Isaacson and Taylor, 1989; Noland-Gerbec et al., 1986; Taylor et al.,
This document is a draft for review purposes only and does not constitute Agency policy.
686 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
1985), decreased rearing activity (Fredriksson et al., 1993), and increased time to cross the first
grid in open field testing (George et al., 1986).
Few studies addressed whether or not children are more susceptible to CNS effects
compared to adults (Burg and Gist, 1999; Burg et al., 1995; White et al., 1997). An analysis of
three residential exposures of TCE observed speech impairments in younger children and not at
any other lifestage (White et al., 1997). A national TCE exposure registry also observed
statistically significant speech impairment and hearing impairment in 0-9 year olds and no other
age group (Burg and Gist, 1999; Burg et al., 1995). However, a follow-up study did not find a
continued association with speech and hearing impairment in these children, although the
absence of acoustic reflexes remained significant (ATSDR, 2003b). See Section 4.3 for further
information on central nervous system toxicity, and Section 4.8.3.3.3 for further information on
developmental neurotoxicity.
4.10.1.1.4.2.2. Liver toxicity
No early lifestage-specific effects were observed after TCE exposure. See Section 4.4 for
further information on liver toxicity.
4.10.1.1.4.2.3. Kidney toxicity
Residents of Woburn, Massachusetts including 4,978 children were surveyed on
residential and medical history to examine an association between observed adverse health
outcomes and wells contaminated with TCE and other chemicals; among these children, an
association was observed for higher cumulative exposure measure and history of kidney and
urinary tract disorders (primarily kidney or urinary tract infections) and with lung and respiratory
disorders (asthma, chronic bronchitis, or pneumonia) (Lagakos et al., 1986). Comparisons were
not made for the adults living in this community. See Section 4.5 for further information on
kidney toxicity.
4.10.1.1.4.2.4. Immunotoxicity
Several studies in exposure to TCE in early lifestages of humans (Lehmann et al., 2001;
Lehmann et al., 2002) and experimental animals (Adams et al., 2003; Blossom and Doss, 2007;
Blossom et al., 2008; Peden-Adams et al., 2006; Peden-Adams et al., 2008) were identified that
assessed the potential for developmental immunotoxicity. While some noted evidence of
immune system perturbation (Adams et al., 2003; Blossom and Doss, 2007; Blossom et al.,
This document is a draft for review purposes only and does not constitute Agency policy.
687 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
2008; Lehmann et al., 2002; Peden-Adams et al., 2006), others did not (Lehmann et al., 2001;
Peden-Adams et al., 2008). However, none of these studies assessed whether exposure during
early life resulted in evidence of increased susceptibility as compared to exposure during
adulthood; this is an area for future research. See Section 4.6 for further information on
immunotoxicity, and Section 4.8.2.3.4 for further discussion on developmental immunotoxicity.
4.10.1.1.4.2.5. Respiratory toxicity
Residents of Woburn, Massachusetts including 4,978 children were surveyed on
residential and medical history to examine an association between observed adverse health
outcomes and wells contaminated with TCE and other chemicals; among these children, an
association was observed for lung and respiratory disorders (asthma, chronic bronchitis, or
pneumonia) (Lagakos et al., 1986). Comparisons were not made for the adults living in this
community. See Section 4.7 for further information on respiratory tract toxicity.
4.10.1.1.4.3. Susceptibility to cancer outcomes in early lifestages
The epidemiologic and experimental animal evidence is limited regarding susceptibility
to cancer from exposure to TCE during early lifestages. The human epidemiological evidence is
summarized above for cancer diagnosed during childhood (see Sections 4.8.2.1 and 4.8.2.3.5),
including a discussion of childhood cancers of the nervous system including neuroblastoma and
the immune system including leukemia (see Section 4.6.1.3). A current follow-up study of the
Camp Lejeune cohort will examine childhood cancers and may provide additional insight
(ATSDR, 2003c, 2009; U.S. GAO, 2007). No studies of cancers in experimental animals in
early lifestages have been observed.
4.10.1.1.4.3.1. Total childhood cancer
Total childhood cancers have been examined in relationship to TCE exposure (ATSDR,
2006a; Morgan and Cassady, 2002). Two studies examining total childhood cancer in relation to
TCE in drinking water did not observe an association. A study in Endicott, NY contaminated by
a number of VOCs, including "thousands of gallons" of TCE observed fewer than 6 cases of
cancer diagnosed between 1980 and 2001 in children aged 0-19 years, and did not exceed
expected cases or types (ATSDR, 2006a). A California community exposed to TCE in drinking
water from contaminated wells was examined for cancer, with a specific emphasis on childhood
This document is a draft for review purposes only and does not constitute Agency policy.
688 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
cancer (<15 years old); however, the incidence did not exceed those expected for the community
(Morgan and Cassady, 2002). A third study of childhood cancer in relation to TCE in drinking
water in Camp Lejeune, NC is currently underway (U.S. GAO, 2007).
4.10.1.1.4.3.2. Childhood leukemia
Childhood leukemia has been examined in relationship to TCE exposure (Cohn et al.,
1994b; Costas et al., 2002; Lagakos et al., 1986; Lowengart et al., 1987; McKinney et al., 1991;
Shu et al., 1999). In a study examining drinking water exposure to TCE in 75 New Jersey towns,
childhood leukemia, (including ALL) was significantly increased for girls (n = 6) diagnosed
before age 20 years, but this was not observed for boys (Cohn et al., 1994b). A community in
Woburn, MA with contaminated well water including TCE experienced 20 cases of childhood
leukemia, significantly more than expected (Lagakos et al., 1986); however, the incidence of
leukemia among children was not compared to the incidence rate among adults living in this
community. Further analysis by Costas et al. (2002) also observed a greater than twofold
increase over expected cases of childhood leukemia. Cases were more likely to be male (76%),
<9 years old at diagnosis (62%), breast-fed (OR: 10.17, 95% CI: 1.22-84.50), and exposed
during pregnancy (adjusted OR: 8.33, 95% CI: 0.73-94.67). The highest risk was observed for
exposure during pregnancy compared to preconception or postnatal exposure, and a dose-
response was seen for exposure during pregnancy (Costas et al., 2002). In addition, family
members of those diagnosed with childhood leukemia, including 13 siblings under age 19 at the
time of exposure, had altered immune response, but an analysis looking at only these children
was not done (Byers et al., 1988).
Case-control studies examined children diagnosed with ALL for parental occupational
exposures and found a nonsignificant two- to fourfold increase of childhood leukemia risk for
exposure to TCE during preconception, pregnancy, postnatally, or all developmental periods
combined (Lowengart et al., 1987; McKinney et al., 1991; Shu et al., 1999). Some studies
showed an elevated risk for maternal (Shu et al., 1999) or paternal exposure (Lowengart et al.,
1987; McKinney et al., 1991), while others did not show an elevated risk for maternal
(McKinney et al., 1991) or paternal exposure (Shu et al., 1999), possibly due to the small number
of cases. No variability was observed in the developmental stages in Shu et al. (1999), although
Lowengart et al. (1987) observed the highest risk to be paternal exposure to TCE after birth.
This document is a draft for review purposes only and does not constitute Agency policy.
689 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
4.10.1.1.4.3.3. CNS tumors
In a case-control study of parental occupational exposures, paternal self-reported
exposure to TCE was not significantly associated with neuroblastoma in the offspring (OR = 1.4,
95% CI: 0.7-2.9) (De Roos et al., 2001). Brain tumors have also been observed in the offspring
of fathers exposed to TCE, but the odds ratio could not be determined (Peters et al., 1985; Peters
et al., 1981).
4.10.1.1.4.3.4. Age-dependent adjustment factors (ADAFs)
According to EPA's Supplemental Guidance for Assessing Susceptibility from Early-Life
Exposure to Carcinogens (U.S. EPA, 2005d) there may be increased susceptibility to early-life
exposures for carcinogens with a mutagenic MOA. Therefore, because the weight of evidence
supports a mutagenic MOA for TCE carcinogenicity in the kidney (see Section 4.4.7), the lack of
data suggesting an absence of GSTT1 expression in neonates, and in the absence of chemical-
specific data to evaluate differences in susceptibility, early-life susceptibility should be assumed
and the ADAFs should be applied, in accordance with the Supplemental Guidance.
4.10.1.1.5. Later Lifestages
Few studies examine the differential effects of TCE exposure for elderly adults
(>65 years old). These limited studies suggest that older adults may experience increased
adverse effects than younger adults. However, there is no further evidence for elderly
individuals exposed to TCE beyond these studies.
Toxicokinetics in later lifestages can be distinct from toxicokinetics in younger adults
(Benedetti et al., 2007; Ginsberg et al., 2005), although there is limited evidence showing a
possible age-related difference in CYP expression (Dome et al., 2005; George et al., 1995;
Parkinson et al., 2004). GST expression has been observed to decrease with age in human
lymphocytes, with the lowest expression in those aged 60-80 years old (van Lieshout and Peters,
1998).
Studies have examined the age differences in TK after exposure to a mixture of six VOCs
including TCE for humans (Mahle et al., 2007) and rats (Mahle et al., 2007; Rodriguez et al.,
2007). In humans, the blood:air partition coefficient for adult males (20-82 years) was
significantly (p < 0.05) higher (11.7 ± 1.9) compared to male (11.2 ± 1.8) or female (11.0 ± 1.6)
children (3-10 years) (Mahle et al., 2007); when the data was stratified for adults above and
below 55 years of age, there was no significant difference observed between adults (20-55
This document is a draft for review purposes only and does not constitute Agency policy.
690 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
years) and aged (56-82) (data not reported). In rats, the difference in tissue:air partition
coefficients also increased from PND10 to adult (2 months) to aged (2 years) rat (Mahle et al.,
2007). TCE has also been measured in the brain of rats, with an increased level observed in
older (2 year old) rats compared to adult (2 month old) rats (Rodriguez et al., 2007). It was also
observed that aged rats reached steady state slower with higher concentrations compared to the
adult rat; the authors suggest that the almost twofold greater percentage of body fat in the elderly
is responsible for this response (Rodriguez et al., 2007).
One cohort of TCE exposed metal degreasers found an increase in psychoorganic
syndrome and increased vibration threshold related to increasing age (Rasmussen et al., 1993b;
Rasmussen et al., 1993c; Rasmussen et al., 1993d), although the age groups were <29 years,
30-39 years, and 40+ years, but the age ranged only from 18-68 years and did not examine >65
years as a separate category.
4.10.2. Other Susceptibility Factors
Aside from age, many other factors may affect susceptibility to TCE toxicity. A partial
list of these factors includes gender, genetic polymorphisms, preexisting disease status,
nutritional status, diet, and previous or concurrent exposures to other chemicals. The toxicity
that results due to changes in multiple factors may be quite variable, depending on the exposed
population and the type of exposure. Qualitatively, the presence of multiple susceptibility
factors will increase the variability that is seen in a population response to TCE toxicity.
4.10.2.1.1. Gender
Individuals of different genders are physiologically, anatomically, and biochemically
different. Males and females can differ greatly in many physiological parameters such as body
composition, organ function, and ventilation rate, which can influence the toxicokinetics of
chemicals and their metabolites in the body (Gandhi et al., 2004; Gochfeld, 2007).
4.10.2.1.2. Gender-specific toxicokinetics
Section 3 describes the toxicokinetics of TCE. Gender-specific information is described
below for absorption, distribution, metabolism, and excretion, followed by available gender-
specific PBPK models.
This document is a draft for review purposes only and does not constitute Agency policy.
691 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
4.10.2.1.2.1. Absorption
As discussed in Section 3.1, exposure to TCE may occur via inhalation, ingestion, and
skin absorption. Exposure via inhalation is proportional to the ventilation rate, duration of
exposure, and concentration of expired air, and women have increased ventilation rates during
exercise compared to men (Gochfeld, 2007). Percentage of body fat varies with gender
(Gochfeld, 2007), which for lipophilic compounds such as TCE will affect absorption and
retention of the absorbed dose. After experimental exposure to TCE, women were found to
absorb a lower dose due to lower alveolar intake rates compared to men (Sato, 1993; Sato et al.,
1991b).
4.10.2.1.2.2. Distribution
Both human and animal studies provide clear evidence that TCE distributes widely to all
tissues of the body (see Section 3.2). The distribution of TCE to specific organs will depend on
organ blood flow and the lipid and water content of the organ, which may vary between genders
(Gochfeld, 2007). After experimental exposure to humans, higher distribution of TCE into fat
tissue was observed in women leading to a greater blood concentration 16 hours after exposure
compared to men (Sato, 1993; Sato et al., 1991b). In experimental animals, male rats generally
have higher levels of TCE in tissues compared to female rats, likely due to gender differences in
metabolism (Lash et al., 2006). In addition, TCE has been observed in the male reproductive
organs (epididymis, vas deferens, testis, prostate, and seminal vesicle) (Zenick et al., 1984).
4.10.2.1.2.3. Metabolism
Section 3.3 describes the metabolic processes involved in the metabolism of TCE,
including CYP and GST enzymes. In addition, the role of metabolism in male reproductive
toxicity is discussed in Section 4.8.1.3.2. In general, there is some indication that TCE
metabolism is different between males and females, with females more rapidly metabolizing
TCE after oral exposure to rats (Lash et al., 2006), intraperitoneal injections in rats (Verma and
Rana, 2003), and in mouse, rat and human liver microsomes (Elfarra et al., 1998).
In general, CYP expression may differ between genders (Gandhi et al., 2004; Gochfeld,
2007; Parkinson et al., 2004), although no gender-related difference in CYP2E1 activity is
observed in the human liver microsomes (George et al., 1995; Parkinson et al., 2004). After
exposure to TCE, CYP2E1 was detected in the epididymis and testes of mice (Forkert et al.,
This document is a draft for review purposes only and does not constitute Agency policy.
692 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
2002), and CYP2E1 and GST-alpha has been detected in the ovaries of rats (Wu and Berger,
2008), indicating that metabolism of TCE can occur in both the male and female reproductive
tracts. One study of TCE exposure in mice observed induced CYP2E1 expression in the liver of
males only (Nakajima et al., 2000). Male rats have been shown to have higher levels of TCE
metabolites in the liver (Lash et al., 2006), and lower levels of TCE metabolites in the kidney
(Lash et al., 2006) compared to female rats. However, another study did not observe ant sex-
related differences in the metabolism of TCE in rats (Nakajima et al., 1992b).
Unlike CYP-mediated oxidation, quantitative differences in the polymorphic distribution
or activity levels of GST isoforms in humans are not presently known. However, the available
data (Lash et al., 1999a; Lash et al., 1999b) do suggest that significant variation in GST-
mediated conjugation of TCE exists in humans. One study observed that GSH conjugation is
higher in male rats compared to female rats (Lash et al., 2000b); however, it has also been
speculated that any gender difference may be due to a polymorphism in GSH conjugation of
TCE rather than a true gender difference (Lash et al., 1999a). Also, induction of PPARa
expression in male mice after TCE exposure was greater than that in females (Nakajima et al.,
2000).
4.10.2.1.2.4. Excretion
The major processes of excretion of TCE and its metabolites are discussed in Section 3.4.
Two human voluntary inhalation exposure studies observed the levels of TCE and its metabolites
in exhaled breath and urine (Kimmerle and Eben, 1973b; Nomiyama and Nomiyama, 1971).
Increased levels of TCE in exhaled breath in males were observed in one human voluntary
inhalation exposure study of 250-380 ppm for 160 minutes (Nomiyama and Nomiyama, 1971),
but no difference was observed in another study of 40 ppm for 4 hours or 50 ppm for 4 hours for
5 days (Kimmerle and Eben, 1973b).
After experimental exposure to TCE, women were generally found to excrete higher
levels of TCE and TCA compared to men (Kimmerle and Eben, 1973b; Nomiyama and
Nomiyama, 1971). However, other studies observed an increase in TCE in the urine of males
(Inoue et al., 1989), an increase in TCA in the urine of males (Sato et al., 1991b), or no
statistically significant (p > 0.10) gender difference for TCA in the urine (Inoue et al., 1989).
Others found that the urinary elimination half-life of TCE metabolites is longer in women
compared to men (Ikeda, 1977; Ikeda and Imamura, 1973).
In addition to excretion pathways that occur in both genders, excretion occurs uniquely in
men and women. In both humans and experimental animals, it has been observed that females
This document is a draft for review purposes only and does not constitute Agency policy.
693 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
can excrete TCE and metabolites in breast milk (Fisher et al., 1997; Fisher et al., 1990; Hamada
and Tanaka, 1995; Pellizzari et al., 1982), while males can excrete TCE and metabolites in
seminal fluid (Forkert et al., 2003; Zenick et al., 1984).
4.10.2.1.2.5. Physiologically based pharmacokinetic (PBPK) models
Gender-specific differences in uptake and metabolism of TCE were incorporated into a
PBPK model using human exposure data (Fisher et al., 1998). The chemical-specific parameters
included cardiac output at rest, ventilation rates, tissue volumes, blood flow, and fat volume.
This model found that gender differences for the toxicokinetics of TCE are minor.
4.10.2.1.3. Gender -specific effects
4.10.2.1.3.1. Gender susceptibility to noncancer outcomes
4.10.2.1.3.1.1. Liver toxicity
No gender susceptibility to noncancerous outcomes in the liver was observed. A detailed
discussion of the studies examining the effects of TCE on the liver can be found in Section 4.4.
4.10.2.1.3.1.2. Kidney toxicity
A detailed discussion of the studies examining the noncancer effects of TCE on the
kidney can be found in Section 4.5. A residential study found that females aged 55-64 years old
had an elevated risk of kidney disease (RR = 4.57, 99% CI: 2.10-9.93) compared to males ,
although an elevated risk of urinary tract disorders was reported for both males and females
(Burg et al., 1995). Additionally, a higher rate of diabetes in females compared to males exposed
to TCE was reported in two studies (Burg et al., 1995; Davis et al., 2005). In rodents, however,
and kidney weights were increased more in male mice than in females (Kjellstrand et al., 1983a;
Kjellstrand et al., 1983b), and male rats have exhibited increased renal toxicity to TCE compared
to females (Lash et al., 2001b; Lash et al., 1998a).
This document is a draft for review purposes only and does not constitute Agency policy.
694 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
4.10.2.1.3.1.3. Immunotoxicity
A detailed discussion of the studies examining the immunotoxic effects of TCE can be
found in Section 4.6. Most of the immunotoxicity studies present data stratified by sex. The
prevalence of exposure to TCE is generally lower in women compared with men. In men, the
studies generally reported odds ratios between 2.0 and 8.0, and in women, the odds ratios were
between 1.0 and 2.0 (Cooper et al., 2009). Based on small numbers of cases, an occupational
study of TCE exposure found an increased risk for systemic sclerosis for men (OR: 4.75,
95% CI: 0.99-21.89) compared to women (OR: 2.10; 95% CI: 0.65-6.75) (Diot et al., 2002).
Another study found similar results, with an elevated risk for men with a maximum intensity,
cumulative intensity and maximum probability of exposure to TCE compared to women (Nietert
et al., 1998). These two studies, along with one focused exclusively on the risk of scleroderma
to women (Garabrant et al., 2003), were included in a meta-analysis conducted by the EPA
resulting in a combined estimate for "any" exposure, was OR = 2.5 (95% CI: 1.1, 5.4) for men
and OR =1.2 (95% CI: 0.58, 2.6) in women.
4.10.2.1.3.1.4. Respiratory toxicity
No gender susceptibility to noncancerous outcomes in the respiratory tract after TCE
exposure was observed. A detailed discussion of the studies examining the respiratory effects of
TCE can be found in Section 4.7.
4.10.2.1.3.1.5. Reproductive toxicity
A detailed discussion of the studies examining the gender-specific noncancer
reproductive effects of TCE can be found in Section 4.8.1.
Studies examining males after exposure to TCE observed altered sperm morphology and
hyperzoospermia (Chia et al., 1996), altered endocrine function (Chia et al., 1997; Goh et al.,
1998), decreased sexual drive and function (Bardodej and Vyskocil, 1956; El Ghawabi et al.,
1973; Saihan et al., 1978), and altered fertility to TCE exposure. Infertility was not associated
with TCE exposure in other studies (Forkert et al., 2003; Sallmen et al., 1998), and sperm
abnormalities were not observed in another study (Rasmussen et al., 1988).
There is more limited evidence for reproductive toxicity in females. There are
epidemiological indicators of a possible effect of TCE exposure on female fertility (Sallmen et
al., 1998), increased rate of miscarriage (ATSDR, 2001), and menstrual cycle disturbance
(ATSDR, 2001; Bardodej and Vyskocil, 1956; Zielinski, 1973). In experimental animals, the
effects on female reproduction include evidence of reduced in vitro oocyte fertilizability in rats
This document is a draft for review purposes only and does not constitute Agency policy.
695 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
(Berger and Horner, 2003; Wu and Berger, 2007, 2008). However, in other studies that assessed
reproductive outcome in female rodents (Cosby and Dukelow, 1992; George et al., 1985, 1986;
Manson et al., 1984), there was no evidence of adverse effects of TCE exposure on female
reproductive function.
4.10.2.1.3.1.6. Developmental toxicity
A detailed discussion of the studies examining the gender-specific noncancer
developmental effects of TCE can be found in Section 4.8.3. Only one study of contaminated
drinking water exposure in Camp Lejeune, NC observed a higher risk of SGA in males compared
to females (ATSDR, 1998b; Sonnenfeld et al., 2001).
4.10.2.1.3.2. Gender susceptibility to cancer outcomes
A detailed discussion of the studies examining the carcinogenic effects of TCE can be
found on the liver in Section 4.4, on the kidney in Section 4.5, in the immune system in
Section 4.6.4, in the respiratory system in Sections 4.7.1.2 and 4.7.3, and on the reproductive
system in Section 4.8.2.
4.10.2.1.3.2.1. Liver cancer
An elevated risk of liver cancer was observed for females compared to males in both
human (Raaschou-Nielsen et al., 2003) and rodent (Elfarra et al., 1998) studies. In addition,
gallbladder cancer was significantly elevated for women compared to men (Raaschou-Nielsen et
al., 2003). A detailed discussion of the studies examining the gender-specific liver cancer effects
of TCE can be found in Section 4.4.
4.10.2.1.3.2.2. Kidney cancer
One study of occupational exposure to TCE observed an increase in renal cell carcinoma
for women compared to men (Dosemeci et al., 1999), but no gender difference was observed in
other studies (Pesch et al., 2000b; Raaschou-Nielsen et al., 2003). Blair et al. (1998) and Hansen
et al. (2001) also present some results by sex, but both of these studies have too few cases to be
informative about a sex difference for kidney cancer. Exposure differences between males and
females in Dosemeci et al. (1999) may explain their finding. These studies, however, provide
little information to evaluate susceptibility between sexes because of their lack of quantitative
This document is a draft for review purposes only and does not constitute Agency policy.
696 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
exposure assessment and lower statistical power. A detailed discussion of the studies examining
the gender-specific kidney cancer effects of TCE can be found in Section 4.5.
4.10.2.1.3.2.3. Cancers of the immune system
Two drinking water studies suggest that there may be an increase of leukemia (Cohn et
al., 1994b; Fagliano et al., 1990) and NHL (Cohn et al., 1994b) among females compared to
males. An occupational study also observed an elevated risk of leukemia in females compared to
males (Raaschou-Nielsen et al., 2003), although a study of contaminated drinking water in
Woburn, MA observed an increased risk of childhood leukemia in males compared to females
(Costas et al., 2002). A detailed discussion of the studies examining the gender-specific cancers
of the immune system following TCE exposure can be found in Section 4.6.4.
4.10.2.1.3.2.4. Respiratory cancers
One study observed significantly elevated risk of lung cancer following occupational
TCE exposure for both men and women, although the risk was found to be higher for women
compared to men (Raaschou-Nielsen et al., 2003). This same study observed a nonsignificant
elevated risk in both men and women for laryngeal cancer, again with an increased risk for
women compared to men (Raaschou-Nielsen et al., 2003). Conversely, a study of Iowa residents
with TCE-contaminated drinking water observed a sevenfold increased annual age-adjusted
incidence for males compared to females (Isacson et al., 1985). However, other studies did not
observe a gender-related difference (ATSDR, 2003b; Blair et al., 1998; Hansen et al., 2001). A
detailed discussion of the studies examining the gender-specific respiratory cancers following
TCE exposure can be found in Sections 4.7.1.2 and 4.7.3.
4.10.2.1.3.2.5. Reproductive cancers
Breast cancer in females and prostate cancer in males was reported after exposure to TCE
in drinking water (Isacson et al., 1985). A statistically elevated risk for cervical cancer, but not
breast, ovarian or uterine cancer, was observed in women in another study (Raaschou-Nielsen et
al., 2003). This study also did not observe elevated prostate or testicular cancer (Raaschou-
Nielsen et al., 2003). A detailed discussion of the studies examining the gender-specific
reproductive cancers following TCE exposure can be found in Section 4.8.2.
This document is a draft for review purposes only and does not constitute Agency policy.
697 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
4.10.2.1.3.2.6. Other Cancers
Bladder and rectal cancer was increased in men compared to women after exposure to
TCE in drinking water, but no gender difference was observed for colon cancer (Isacson et al.,
1985). After occupational TCE exposure, bladder, stomach, colon, and esophageal cancer was
nonsignificantly elevated in women compared to men (Raaschou-Nielsen et al., 2003).
4.10.2.1.4. Genetic Variability
Section 3.3 describes the metabolic processes involved in the metabolism of TCE.
Human variation in response to TCE exposure may be associated with genetic variation. TCE is
metabolized by both CYP and GST; therefore, it is likely that polymorphisms will alter the
response to exposure (Garte et al., 2001; Nakajima and Aoyama, 2000), as well as exposure
other chemicals that may alter the metabolism of TCE (Lash et al., 2007) (see Section 4.10.2.6).
It is important to note that even with a given genetic polymorphism, metabolic expression is not
static, and depends on lifestage (see Section 4.10.1.1.2), obesity (see Section 4.10.2.4.1), and
alcohol intake (see Section 4.10.2.5.1).
4.10.2.1.5. CYP genotypes
In general, variability in CYP expression occurs within humans (Dome et al., 2005), and
variability in CYP expression has been observed in experimental animals exposed to TCE
(Nakajima et al., 1993). In particular, increased CYP2E1 activity may lead to increased
susceptibility to TCE (Lipscomb et al., 1997). The CYP2E1*3 allele and the CYP2E1*4 allele
were more common among those who developed scleroderma who were exposed to solvents
including TCE (Povey et al., 2001). A PBPK model of CYP2E1 expression after TCE exposure
has been developed for rats and humans (Yoon et al., 2007).
In experimental animals, toxicokinetics of TCE differed among CYP2E1 knockout and
wild-type mice (Kim and Ghanayem, 2006). This study found that exhalation was more
prevalent among the knockout mice, whereas urinary excretion was more prevalent among the
wild-type mice. In addition, the dose was found to be retained to a greater degree by the
knockout mice compared to the wild-type mice.
4.10.2.1.6. GST genotype
There is a possibility that GST polymorphisms could play a role in variability in toxic
response to TCE (Caldwell and Keshava, 2006), but this has not been sufficiently tested
This document is a draft for review purposes only and does not constitute Agency policy.
698 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
(NRC, 2006). One study of renal cell cancer in workers exposed to TCE demonstrated a
significant increased for those with GSTM1+ and GSTT1+ polymorphisms, compared to a
negative risk for those with GSTM1- and GSTT1-polymorphisms (Briining et al., 1997a).
Another study of occupational TCE exposure found that renal cell carcinoma was significantly
associated with the GSTT+ polymorphism but not with GSTT- (Moore et al., 2010). However,
another study did not confirm this hypothesis, observing no clear relationship between GSTM1
and GSTT1 polymorphisms and renal cell carcinoma among TCE-exposed individuals, although
they did see a possible association with the homozygous wild-type allele GSTP1*A
(Wiesenhiitter et al., 2007). Unrelated to TCE exposure, Sweeney et al. (2000) found GSTT1-
to be associated with an increased risk of renal cell carcinoma, but no difference was seen for
GSTM1 and GSTP1 alleles. The role of GST polymorphisms in the development of renal cell
carcinoma is an area in need of future research.
4.10.2.1.7. Other genotypes
Other genetic polymorphisms could play a role in variability in toxic response, in
particular TCE-related skin disorders. Studies have found that many TCE-exposed patients
diagnosed with skin conditions exhibited the slow-acetylator NAT2 genotype (Huang et al.,
2002; Nakajima et al., 2003); whereas there was no difference in NAT2 status for those
diagnosed with renal cell carcinoma (Wiesenhiitter et al., 2007). Other studies have found that
many TCE-exposed patients diagnosed with skin conditions expressed variant HLA alleles (Li et
al., 2007; Yue et al., 2007), in particular HLA-B*1301 which is more common in Asians
compared to whites (Cao et al., 2001; Williams et al., 2001); or TNF a-308 allele (Dai et al.,
2004). Also, an in vitro study of human lung adenocarcinoma cells exposed to TCE varied in
response based on their p53 status, with p53-wild-type cells resulting in severe cellular damage,
but not the p53-null cells (Chen et al., 2002a).
4.10.2.1.8. Race/Ethnicity
Different racial or ethnic groups may express metabolic enzymes in different ratios and
proportions due to genetic variability (Garte et al., 2001). In particular, ethnic variability in CYP
(Dome et al., 2005; McCarver et al., 1998; Parkinson et al., 2004; Shimada et al., 1994; Stephens
et al., 1994) and GST (Nelson et al., 1995) expression has been reported.
It has been observed that the metabolic rate for TCE may differ between the Japanese and
Chinese (Inoue et al., 1989). Also, body size varies among ethnic groups, and increased body
This document is a draft for review purposes only and does not constitute Agency policy.
699 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
size was related to increased absorption of TCE and urinary excretion of TCE metabolites (Sato
etal., 1991b).
4.10.2.1.9. Preexisting Health Status
It is known that kidney and liver diseases can affect the clearance of chemicals from the
body, and therefore, poor health may lead to increased half-lives for TCE and its metabolites.
There are some data indicating that obesity/metabolic syndrome, diabetes and hypertension may
increase susceptibility to TCE exposure through altered toxicokinetics. In addition, some of
these conditions lead to increased risk for adverse effects that have also been associated with
TCE exposure, though the possible interaction between TCE and known risk factors for these
effects is not understood.
4.10.2.1.10. Obesity
TCE is lipophilic and stored in adipose tissue; therefore, obese individuals may
experience altered toxicokinetics of TCE compared to thin individuals. The absorption of TCE
is increased in obese individuals compared to thin individuals (Clewell et al., 2000), as observed
by lower blood concentrations immediately after exposure in obese men compared to thin men
(Sato, 1993; Sato et al., 1991b). Once absorbed, obese individuals have increased storage of
TCE in the adipose tissue compared to thin men (Clewell et al., 2000) which prolongs internal
exposures (Davidson and Beliles, 1991; Lash et al., 2000b). Obesity also likely alters TCE
metabolism, since increased CYP2E1 expression has been observed in obese individuals
compared to thin individuals (McCarver et al., 1998). Finally, delayed excretion has been
observed in obese individuals compared to thin individuals in both exhaled air (Monster, 1979)
and urine (Sato, 1993; Sato et al., 1991b). In sum, obese individuals have altered toxicokinetics
of TCE compared to thin individuals due to increased storage of TCE, increased CYP2E1
metabolism, and a slower rate of elimination.
In addition, individuals with high BMI are at increased risk of some of the same health
effects associated with TCE exposure. For example, renal cell carcinoma, liver cancer, and
prostate cancer may be positively associated with BMI or obesity (Asal et al., 1988a; Asal et al.,
1988b; Benichou et al., 1998; El-Serag and Rudolph, 2007; Wigle et al., 2008). However,
whether and how TCE interacts with known risk factors for such diseases is unknown, as
existing epidemiologic studies have only examined these factors as possible confounders for
effects associated with TCE, or vice versa (Charbotel et al., 2006; Krishnadasan et al., 2008).
This document is a draft for review purposes only and does not constitute Agency policy.
700 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
4.10.2.1.11. Diabetes
A higher rate of diabetes in females compared to males exposed to TCE was reported in
two studies (Burg et al., 1995; Davis et al., 2005). Whether the TCE may have caused the
diabetes or the diabetes may have increased susceptibility to TCE is not clear. However, it has
been observed that CYP2E1 expression is increased in obese Type II diabetics (Wang et al.,
2003), and in poorly controlled Type I diabetics (Song et al., 1990), which may consequently
alter the metabolism of TCE.
4.10.2.1.12. Hypertension
One study found no difference in risk for renal cell carcinoma among those diagnosed
with hypertension among those living in an area with high TCE exposure; however, a slightly
elevated risk was seen for those being treated for hypertension (OR: 1.57, 95% CI: 0.90-2.72)
(Charbotel et al., 2006). Unrelated to TCE exposure, hypertension has been associated with
increased risk of renal cell carcinoma in women compared to men (Benichou et al., 1998).
4.10.2.1.13. Lifestyle Factors and Nutrition Status
4.10.2.1.14. Alcohol intake
A number of studies have examined the interaction between TCE and ethanol exposure in
both humans (Bardodej and Vyskocil, 1956; Barret et al., 1984; McCarver et al., 1998; Muller et
al., 1975; Sato, 1993; Sato et al., 1991a; Sato et al., 1981; Stewart et al., 1974) and experimental
animals (Kaneko et al., 1994a; Larson and Bull, 1989; Nakajima et al., 1988; Nakajima et al.,
1992a; Nakajima et al., 1990; Okino et al., 1991; Sato and Nakajima, 1985; Sato et al., 1980,
1983; White and Carlson, 1981).
The coexposure causes metabolic inhibition of TCE in humans (Muller et al., 1975;
Windemuller and Ettema, 1978), male rats (Kaneko et al., 1994a; Larson and Bull, 1989;
Nakajima et al., 1988; Nakajima et al., 1990; Nakanishi et al., 1978; Okino et al., 1991; Sato and
Nakajima, 1985; Sato et al., 1981), and rabbits (White and Carlson, 1981). Similarly, individuals
exposed to TCE reported an increase in alcohol intolerance (Bardodej and Vyskocil, 1956;
This document is a draft for review purposes only and does not constitute Agency policy.
701 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
Grandjean et al., 1955; Rasmussen and Sabroe, 1986). Disulfiram, used to treat alcoholism, has
also been found to decrease the elimination of TCE and TCA (Bartonicek and Teisinger, 1962).
A "degreasers flush" has been described, reflecting a reddening of the face of those
working with TCE after drinking alcohol, and measured an elevated level of TCE in exhaled
breath compared to nondrinkers exposed to TCE (Stewart et al., 1974). This may be due to
increased CYP2E1 expression in those that consume alcohol compared to nondrinkers, unrelated
to TCE exposure (Caldwell et al., 2008a; Liangpunsakul et al., 2005; Lieber, 2004; McCarver et
al., 1998; Parkinson et al., 2004; Perrot et al., 1989).
In experimental animals, male rats pretreated with ethanol experienced an induction of
TCE metabolism (Nakajima et al., 1992a), although another study of male rats observed that
pretreatment with ethanol did not decrease CYP activity (Okino et al., 1991). It is important to
note that there a further increased response of TCE and ethanol has been reported when also
combined with low fat diets or low carbohydrate diets in male rats (Sato et al., 1983).
Since the liver is a target organ for both TCE and alcohol, decreased metabolism of TCE
could be related to cirrhosis of the liver as a result of alcohol abuse (McCarver et al., 1998), and
an in increase in clinical liver impairment along with degreasers flush has been observed (Barret
et al., 1984).
The central nervous system may also be impacted by the coexposure. Individuals
exposed to TCE and ethanol reported an increase in altered mood states (Reif et al., 2003),
decreased mental capacity as described as small increases in functional load (Windemuller and
Ettema, 1978), and those exposed to TCE and tetrachloroethylene who consumed alcohol had an
elevated color confusion index (Valic et al., 1997).
4.10.2.1.15. Tobacco smoking
Individuals who smoke tobacco may be at increased risk of the health effects from TCE
exposure. One study examining those living in an area with high TCE exposure found an
increasing trend of risk (p = 0.008) for renal cell carcinoma among smokers, with the highest OR
among those with >40 pack-years (OR = 3.27, 95% CI: 1.48-7.19) (Charbotel et al., 2006). It
has been shown that renal cell carcinoma is independently associated with smoking in a dose-
response manner (Yuan et al., 1998), particularly in men (Benichou et al., 1998). While
Charbotel et al. (2006) adjusted for smoking effects in analyses examining TCE exposure and
renal cell carcinoma, this study provides no information on potential effect modification of TCE
exposure by smoking.
This document is a draft for review purposes only and does not constitute Agency policy.
702 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
A number of factors correlated to smoking (e.g., socioeconomic status, diet, alcohol
consumption) may positively confound results if greater smoking rates were over-represented, as
observed in an occupational cohort exposed to TCE (Raaschou-Nielsen et al., 2003). Absence of
smoking information, on the other hand, could introduce a negative bias. In a drinking water
study with exposures to TCE and perchlorate, Morgan and Cassidy (2002) noted the relatively
high education high income levels and high access to health care of subjects in this study
compared to the averages for the county as a whole likely leads to a lower smoking rate.
4.10.2.1.16. Nutritional status
Malnutrition may also increase susceptibility to TCE. Bioavailability of TCE after oral
and intravenous exposure increased with fasting from approximately 63% in nonfasted rats to
greater than 90% in fasted rats, with blood levels in fasted rats were elevated two- to threefold,
and increased half-life in the blood of fasted rats (D'Souza et al., 1985). Food deprivation (Sato
and Nakajima, 1985) and carbohydrate restriction (Nakajima et al., 1982; Sato and Nakajima,
1985) enhanced metabolism of TCE in male rats, but this was not observed for dietary changes
in protein or fat levels (Nakajima et al., 1982).
Vitamin intake may also alter susceptibility to TCE. An in vitro study of cultured normal
human epidermal keratinocyte demonstrated an increased lipid peroxidation in a dose-dependent
manner after exposure to TCE, which were then attenuated by exposure to Vitamin E (Ding et
al., 2006).
4.10.2.1.17. Physical activity
Increased inhalation during physical activity increases TCE concentrations in the alveoli
when compared to inhalation in a resting state (Astrand, 1975). Studies have examined the time
course of inhaled TCE and metabolites in blood and urine in individuals with different workloads
(Astrand and Ovrum, 1976; Jakubowski and Wieczorek, 1988; Monster et al., 1976; Vesterberg
and Astrand, 1976; Vesterberg et al., 1976). These studies demonstrate that an increase in
pulmonary ventilation increases the amount of TCE taken up during exposure (Astrand and
Ovrum, 1976; Jakubowski and Wieczorek, 1988; Monster et al., 1976; Sato, 1993).
The Rocketdyne aerospace cohort exposed to TCE (and other chemicals) found a
protective effect with high physical activity, but only after controlling for TCE exposure and
socioeconomic status (OR = 0.55, 95% CI: 0.32-0.95, p trend = 0.04) (Krishnadasan et al.,
This document is a draft for review purposes only and does not constitute Agency policy.
703 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
2008). In general, physical activity may provide a protective effect for prostate cancer (Wigle et
al., 2008) (see Section 4.8.3.1.1).
4.10.2.1.18. Socioeconomic status
Socioeconomic status (SES) can be an indicator for a number of coexposures, such as
increased tobacco smoking, poor diet, education, income, and health care access, which may play
a role in the results observed in the health effects of TCE exposure (Morgan and Cassady, 2002).
Children's exposure to TCE was measured in a low SES community, as characterized by
income, educational level, and receipt of free or reduced cost school meals (Sexton et al., 2005);
however, this study did not compare data to a higher SES community, nor examine health
effects.
An elevated risk of NHL and esophagus/adenocarcinoma after exposure to TCE was
observed for blue-collar workers compared to white collar workers and workers with unknown
SES (Raaschou-Nielsen et al., 2003). Authors speculate that these results could be confounded
due to other related factors to SES such as smoking.
4.10.2.1.19. Mixtures
TCE exposure often occurs concurrently with other chemical substances. In general, the
effects of exposures to multiple chemicals is considered by EPA in the Framework for
Cumulative Risk Assessment (U.S. EPA, 2003 a). A summary of the interactive effects of TCE
and other chemical coexposures is addressed in Caldwell et al. (2008a) and in Chapter 10 of the
National Research Council's report Assessing the Human Health Risks of'/richloroethylene: Key
Scientific Issues (NRC, 2006).
Section 2 discusses that other parent compounds produce similar metabolites to TCE (see
Table 2-1) or have similar properties or industrial uses (see Tables 2-3 and 2-14). The metabolic
pathway of TCE in discussed in Section 3.3; due its metabolism into multiple compounds,
exposure to TCE itself can be considered as exposure to a mixture (NRC, 2006). Many of the
studies discussed above in Section 4 demonstrate that exposure to TCE and other chemical
substances often occur together in both occupational and nonoccupational settings.
Coexposures to other solvents may induce or saturate toxicokinetic pathways, altering the
way in which TCE is metabolized and cleared from the body. The limited data summarized by
the ATSDR in its interaction profile on TCE, 1,1,1-trichloroethane, 1,1-dichloroethane, and
tetrachloroethylene suggest that additive joint action is plausible (ATSDR, 2004b; Pohl et al.,
2003). Joint exposure to TCE and the fungicide fenarimol has been shown to alter TCE
This document is a draft for review purposes only and does not constitute Agency policy.
704 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
metabolism and genetic expression in mice (Hrelia et al., 1994). Joint exposure to TCE, benzene
and methyl mercury has been shown to induce genetic expression in the liver and the kidney of
rats (Hendriksen et al., 2007). Metabolic competition was also observed for TCE and various
agents in another study by Jakobson et al. (1986).
PBPK models have been developed demonstrating the interaction between
1,1-dichloroethylene and TCE (Andersen et al., 1987b) and the interaction between TCE,
tetrachloroethylene, and 1,1,1-trichloroethane in rats (Dobrev et al., 2001) and humans (Dobrev
et al., 2002). Other PBPK models also showed metabolic inhibition at higher doses for TCE and
toluene (Thrall and Poet, 2000), and for TCE and chloroform (Isaacs et al., 2004). Another
PBPK model of TCE and multiple VOCs showed metabolic inhibition and induction when
exposure occurs concurrently (Haddad et al., 2000).
4.10.3. Uncertainty of Database and Research Needs for Susceptible Populations
There is some evidence that certain populations may be more susceptible to exposure to
TCE. These populations include early and later lifestages, gender, genetic polymorphisms,
race/ethnicity, preexisting health status, and lifestyle factors and nutrition status. In general, this
database would be improved by future epidemiologic and toxicological studies of TCE exposure
that provide data on effect modification, including the factors discussed here.
Although the toxicokinetic variability has been characterized by population PBPK
modeling (see Section 3.5), the available data are limited due to the relative small numbers of
individuals (n < 100), their all being adults, and the fact that subjects were selected nonrandomly
(healthy human volunteers).
Although there is more information on early life exposure to TCE than on other
potentially susceptible populations, there remain a number of uncertainties and data gaps
regarding children's susceptibility. Improved PBPK modeling for using childhood parameters
early lifestages as recommended by the NRC (2006), and validation of these models, will aid in
determining how variations in metabolic enzymes affect TCE metabolism. In particular, the
NRC states that it is prudent to assume children need greater protection than adults—unless
sufficient data are available to justify otherwise (NRC, 2006).
More studies specifically designed to evaluate effects in early and later lifestages are
needed in order to more fully characterize potential life stage-related TCE toxicity. Because the
neurological effects of TCE constitute the most sensitive endpoints of concern for noncancer
effects, it is quite likely that the early lifestages may be more susceptible to these outcomes than
are adults. Lifestage-specific neurotoxic effects, particularly in the developing fetus, need
This document is a draft for review purposes only and does not constitute Agency policy.
705 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
further evaluation. It is important to consider the use of age-appropriate testing for assessment of
these and other outcomes, both for cancer and noncancer outcomes. Data specific to the
carcinogenic effects of TCE exposure during the critical periods of development of experimental
animals and humans also are sparse.
There is a need to better characterize the implications of TCE exposures to susceptible
populations. There is suggestive evidence that there may be greater susceptibility for exposures
to the elderly. Gender and race/ethnic differences in susceptibility are likely due to variation in
physiology and exposure, and genetic variation likely has an effect on the toxicokinetics of TCE.
In particular, the relationship between genetic variation and generalized hypersensitivity skin
diseases is relevant for future study (see Sections 4.6.1.1.2 and 4.10.2.2). Diminished health
status (e.g., impaired kidney liver or kidney), alcohol consumption, tobacco smoking, and
nutritional status will likely affect an individual's ability to metabolize TCE. In addition, further
evaluation of the effects due to coexposures to other compounds with similar or different MO As
need to be evaluated. Future research should better characterize possible susceptibility for
certain lifestages or populations.
4.11. HAZARD CHARACTERIZATION
4.11.1. Characterization of Noncancer Effects
4.11.1.1.1. Neurotoxicity
Both human and animal studies have associated TCE exposure with effects on several
neurological domains. The strongest neurological evidence of hazard in humans is for changes
in trigeminal nerve function or morphology and impairment of vestibular function. Fewer and
more limited evidence exists in humans on delayed motor function, and changes in auditory,
visual, and cognitive function or performance. Acute and subchronic animal studies show
morphological changes in the trigeminal nerve, disruption of the peripheral auditory system
leading to permanent function impairments and histopathology, changes in visual evoked
responses to patterns or flash stimulus, and neurochemical and molecular changes. Additional
acute studies reported structural or functional changes in hippocampus, such as decreased
myelination or decreased excitability of hippocampal CA1 neurons, although the relationship of
these effects to overall cognitive function is not established. Some evidence exists for motor-
related changes in rats/mice exposed acutely/subchronically to TCE, but these effects have not
been reported consistently across all studies.
This document is a draft for review purposes only and does not constitute Agency policy.
706 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Epidemiologic evidence supports a relationship between TCE exposure and trigeminal
nerve function changes, with multiple studies in different populations reporting abnormalities in
trigeminal nerve function in association with TCE exposure Ruitjen et al., 2001;(Barret et al.,
1982; Barret et al., 1984; Barret et al., 1987; Feldman et al., 1988; Feldman et al., 1992; Kilburn,
2002b; Kilburn and Warshaw, 1993a; Mhiri et al., 2004). Of these, two well conducted
occupational cohort studies, each including more than 100 TCE-exposed workers without
apparent confounding from multiple solvent exposures, additionally reported statistically
significant dose-response trends based on ambient TCE concentrations, duration of exposure,
and/or urinary concentrations of the TCE metabolite TCA (Barret et al., 1984; 1987). Limited
additional support is provided by a positive relationship between prevalence of abnormal
trigeminal nerve or sensory function and cumulative exposure to TCE (most subjects) or CFC-
113 (<25% of subjects) (Rasmussen et al., 1993d). Test for linear trend in this study was not
statistically significant and may reflect exposure misclassification since some subjects included
in this study did not have TCE exposure. The lack of association between TCE exposure and
overall nerve function in three small studies (trigeminal: (El Ghawabi et al., 1973); ulnar and
medial: (Triebig et al., 1983; 1982)) does not provide substantial evidence against a causal
relationship between TCE exposure and trigeminal nerve impairment because of limitations in
statistical power, the possibility of exposure misclassification, and differences in measurement
methods. Laboratory animal studies have also shown TCE-induced changes in the trigeminal
nerve. Although one study reported no significant changes in trigeminal somatosensory evoked
potential in rats exposed to TCE for 13 weeks (Albee et al., 2006), there is evidence of
morphological changes in the trigeminal nerve following short-term exposures in rats (Barret et
al., 1992; 1991).
Human chamber, occupational, geographic based/drinking water, and laboratory animal
studies clearly established TCE exposure causes transient impairment of vestibular function.
Subjective symptoms such as headaches, dizziness, and nausea resulting from occupational
(Grandjean et al., 1955; Liu et al., 1988; Rasmussen and Sabroe, 1986; Smith, 1970),
environmental (Hirsch et al., 1996), or chamber exposures (Smith, 1970; Stewart et al., 1970)
have been reported extensively. A few laboratory animal studies have investigated vestibular
function, either by promoting nystagmus or by evaluating balance (Niklasson et al., 1993; Tham
et al., 1984; 1979; Umezu et al., 1997).
In addition, mood disturbances have been reported in a number of studies, although these
effects also tend to be subjective and difficult to quantify (Gash et al., 2008; Kilburn, 2002a,
2002b; Kilburn and Warshaw, 1993a; McCunney, 1988; Mitchell and Parsons-Smith, 1969;
Rasmussen and Sabroe, 1986; Troster and Ruff, 1990), and a few studies have reported no
This document is a draft for review purposes only and does not constitute Agency policy.
707 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
effects from TCE on mood (Reif et al., 2003; Triebig et al., 1976; Triebig et al., 1977a). Few
comparable mood studies are available in laboratory animals, although both Moser et al. (2003)
and Albee et al. (2006) report increases in handling reactivity among rats exposed to TCE.
Finally, significantly increased number of sleep hours was reported by Arito et al. (1994) in rats
exposed via inhalation to 50-300-ppm TCE for 8 hours/day for 6 weeks.
Four epidemiologic studies of chronic exposure to TCE observed disruption of auditory
function. One large occupational cohort study showed a statistically significant difference in
auditory function with cumulative exposure to TCE or CFC-113 as compared to control groups
after adjustment for possible confounders, as well as a positive relationship between auditory
function and increasing cumulative exposure (Rasmussen et al., 1993b). Of the three studies
based on populations from ATSDR's TCE Subregistry from the National Exposure Registry,
more limited than Rasmussen et al. (1993b) due to inferior exposure assessment, Burg et al.
(1995) and Burg and Gist (1999) reported a higher prevalence of self-reported hearing
impairments. The third study reported that auditory screening revealed abnormal middle ear
function in children less than 10 years of age, although a dose-response relationship could not be
established and other tests did not reveal differences in auditory function (ATSDR, 2003b).
Further evidence for these effects is provided by numerous laboratory animal studies
demonstrating that high dose subacute and subchronic TCE exposure in rats disrupts the auditory
system leading to permanent functional impairments and histopathology.
Studies in humans exposed under a variety of conditions, both acutely and chronically,
report impaired visual functions such as color discrimination, visuospatial learning tasks, and
visual depth perception in subjects with TCE exposure. Abnormalities in visual depth perception
were observed with a high acute exposure to TCE under controlled conditions (Vernon and
Ferguson, 1969). Studies of lower TCE exposure concentrations also observed visuofunction
effects. One occupational study (Rasmussen et al., 1993b) reported a statistically significant
positive relationship between cumulative exposure to TCE or CFC-113 and visual gestalts
learning and retention among Danish degreasers. Two studies of populations living in a
community with drinking water containing TCE and other solvents furthermore suggested
changes in visual function (Kilburn, 2002a)(Reif et al., 2003). These studies used more direct
measures of visual function as compared to Rasmussen et al. (1993b), but their exposure
assessment is more limited because TCE exposure is not assigned to individual subjects
(Kilburn 2002a), or because there are questions regarding control selection (Kilburn 2002a) and
exposure to several solvents (Kilburn, 2002a)(Reif et al., 2003).
Additional evidence of effects of TCE exposure on visual function is provided by a
number of laboratory animal studies demonstrating that acute or subchronic TCE exposure
This document is a draft for review purposes only and does not constitute Agency policy.
708 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
causes changes in visual evoked responses to patterns or flash stimulus (Boyes et al., 2005)(Blain
et al., 1994; Boyes et al., 2003). Animal studies have also reported that the degree of some
effects is correlated with simultaneous brain TCE concentrations (Boyes et al., 2003; Boyes et
al., 2005) and that, after a recovery period, visual effects return to control levels (Blain et al.,
1994; Rebert et al., 1991). Overall, the human and laboratory animal data together suggest that
TCE exposure can cause impairment of visual function, and some animal studies suggest that
some of these effects may be reversible with termination of exposure.
Studies of human subjects exposed to TCE either acutely in chamber studies or
chronically in occupational settings have observed deficits in cognition. Five chamber studies
reported statistically significant deficits in cognitive performance measures or outcome measures
suggestive of cognitive effects (Gamberale et al., 1976; Stewart et al., 1970; Triebig et al., 1976;
Triebig et al., 1977a). Danish degreasers with high cumulative exposure to TCE or CFC-113 had
a high risk (OR = 13.7, 95% CI: 2.0-92.0) for psychoorganic syndrome characterized by
cognitive impairment, personality changes, and reduced motivation, vigilance, and initiative
compared to workers with low cumulative exposure. Studies of populations living in a
community with contaminated groundwater also reported cognitive impairments (Kilburn,
2002b; Kilburn and Warshaw, 1993a), although these studies carry less weight in the analysis
because TCE exposure is not assigned to individual subjects and their methodological design is
weaker.
Laboratory studies provide some additional evidence for the potential for TCE to affect
cognition, although the predominant effect reported has been changes in the time needed to
complete a task, rather than impairment of actual learning and memory function (Kishi et al.,
1993; Kulig, 1987; Umezu et al., 1997). In addition, in laboratory animals, it can be difficult to
distinguish cognitive changes from motor-related changes. However, several studies have
reported structural or functional changes in the hippocampus, such as decreased myelination
(Isaacson and Taylor, 1989; Isaacson et al., 1990) or decreased excitability of hippocampal CA1
neurons (Ohta et al., 2001), although the relationship of these effects to overall cognitive
function is not established.
Two studies of TCE exposure, one chamber study of acute exposure duration and one
occupational study of chronic duration, reported changes in psychomotor responses. The
chamber study of Gamberale et al. (1976) reported a dose-related decrease in performance in a
choice reaction time test in healthy volunteers exposed to 100 and 200-ppm TCE for 70 minutes
as compared to the same subjects without exposure. Rasmussen et al. (1993d) reported a
statistically significant association with cumulative exposure to TCE or CFC-113 and
dyscoordination trend among Danish degreasers. Observations in a third study (Gun et al., 1978)
This document is a draft for review purposes only and does not constitute Agency policy.
709 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
are difficult to judge given the author's lack of statistical treatment of data. In addition, Gash et
al. (2008) reported that 14 out of 30 TCE-exposed workers exhibited significantly slower fine
motor hand movements as measured through a movement analysis panel test. Studies of
populations living in communities with TCE and other solvents detected in groundwater supplies
reported significant delays in simple and choice reaction times in individuals exposed to TCE in
contaminated groundwater as compared to referent groups (Kilburn, 2002b; Kilburn and
Warshaw, 1993a) (Kilburn and Thornton, 1996). Observations in these studies are more
uncertain given questions of the representativeness of the referent population, lack of exposure
assessment to individual study subjects, and inability to control for possible confounders
including alcohol consumption and motivation. Finally, in a presentation of 2 case reports,
decrements in motor skills as measured by the grooved pegboard and finger tapping tests were
observed (Troster and Ruff, 1990).
Laboratory animal studies of acute or subchronic exposure to TCE observed psychomotor
effects, such as loss of righting reflex (Shih et al., 2001J_Umezu et al., 1997) and decrements in
activity, sensory-motor function, and neuromuscular function (Kishi et al., 1993; Moser et al.,
1995; 2003). However, two studies also noted an absence of significant changes in some
measures of psychomotor function (Albee et al., 2006; Kulig, 1987). In addition, less consistent
results have been reported with respect to locomotor activity in rodents. Some studies have
reported increased locomotor activity after an acute i.p. dosage (Wolff and Siegmund, 1978) or
decreased activity after acute or short term oral gavage dosing (Moser et al., 1995; 2003). No
change in activity was observed following exposure through drinking water (Waseem et al.,
2001), inhalation (Kulig, 1987) or orally during the neurodevelopment period (Fredriksson et al.,
1993).
Several neurochemical and molecular changes have been reported in laboratory
investigations of TCE toxicity. Kjellstrand et al. (1987) reported inhibition of sciatic nerve
regeneration in mice and rats exposed continuously to 150-ppm TCE via inhalation for 24 days.
Two studies have reported changes in GABAergic and glutamatergic neurons in terms of GABA
or glutamate uptake (Briving et al., 1986) or response to GABAergic antagonistic drugs (Shih et
al., 2001) as a result of TCE exposure, with the Briving et al. (1986) conducted at 50 ppm for
12 months. Although the functional consequences of these changes is unclear, Tham et al.
(1984; 1979) described central vestibular system impairments as a result of TCE exposure that
may be related to altered GABAergic function. In addition, several in vitro studies have
demonstrated that TCE exposure alters the function of inhibitory ion channels such as receptors
for GABAa glycine, and serotonin (Beckstead et al., 2000; Krasowski and Harrison, 2000;
Lopreato et al., 2003) or of voltage-sensitive calcium channels (Shafer et al., 2005).
This document is a draft for review purposes only and does not constitute Agency policy.
710 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
4.11.1.1.2. Kidney Toxicity
There are few human data pertaining to TCE-related noncancer kidney toxicity.
Observation of elevated excretion of urinary proteins in the available studies (Bolt et al., 2004;
Briining et al., 1999a; Briining et al., 1999b; Green et al., 2004; Rasmussen et al., 1993d)
indicates the occurrence of a toxic insult among TCE-exposed subjects compared to unexposed
controls. Two studies are of subjects with previously diagnosed kidney cancer (Bolt et al., 2004;
Briining et al., 1999a), while subjects in the other studies are disease free. Urinary proteins are
considered nonspecific markers of nephrotoxicity and include al-microglobulin, albumin, and
NAG (Lybarger et al., 1999; Price et al., 1996; 1999). Four studies measure al-microglobulin
with elevated excretion observed in the German studies (Bolt et al., 2004; Briining et al., 1999a;
Briining et al., 1999b) but not Green et al. (2004). However, Rasmussen et al. (1993d) reported a
positive relationship between increasing urinary NAG, another nonspecific marker of tubular
toxicity, and increasing exposure duration; and Green et al. (2004) found statistically significant
group mean differences in NAG. Observations in Green et al. (2004) provide evidence of
tubular damage among workers exposed to trichloroethylene at current occupational levels.
Elevated excretion of NAG has also been observed with acute TCE poisoning (Carried et al.,
2007). Some support for TCE nephrotoxicity in humans is provided by a study of end-stage
renal disease in a cohort of workers at Hill Air Force Base (Radican et al., 2006), although
subjects in this study were exposed to hydrocarbons, JP-4 gasoline, and solvents in addition to
TCE, including 1,1,1-trichloroethane, and a second reporting a twofold elevated risk for
progression of glomerulonephritis to ESRD with TCE exposure (Jacob et al., 2007).
Laboratory animal and in vitro data provide additional support for TCE nephrotoxicity.
Multiple studies with both gavage and inhalation exposure show that TCE causes renal toxicity
in the form of cytomegaly and karyomegaly of the renal tubules in male and female rats and
mice (summarized in Section 4.4.4). Further studies with TCE metabolites have demonstrated a
potential role for DCVC, TCOH, and TCA in TCE-induced nephrotoxicity. Of these, available
data suggest that DCVC induced renal effects most like those of TCE and is formed in sufficient
amounts following TCE exposure to account for these effects. TCE or DCVC have also been
shown to be cytotoxic to primary cultures of rat and human renal tubular cells (Cummings et al.,
2000a; Cummings and Lash, 2000; Cummings et al., 2000b).
Overall, multiple lines of evidence support the conclusion that TCE causes nephrotoxicity
in the form of tubular toxicity, mediated predominantly through the TCE GSH conjugation
product DCVC.
This document is a draft for review purposes only and does not constitute Agency policy.
711 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
4.11.1.1.3. Liver Toxicity
Few studies on liver toxicity and TCE exposure are found in humans. Of these, three
studies reported significant changes in serum liver function tests, widely used in clinical settings
in part to identify patients with liver disease, in metal degreasers whose TCE exposure was
assessed using urinary trichloro-compounds as a biomarker (Nagaya et al., 1993; Rasmussen et
al., 1993b; Xu et al., 2009). Two additional studies reported plasma or serum bile acid changes
(Driscoll et al., 1992; Neghab et al., 1997). One study of subjects from the TCE subregistry of
ATSDR's National Exposure Registry is suggestive of liver disorders but limitations preclude
inferences whether TCE caused these conditions is not possible given the study's limitations
(Davis et al., 2005). Furthermore, a number of case reports exist of liver toxicity including
hepatitis accompanying immune-related generalized skin diseases described as a variation of
erythema multiforme, Stevens-Johnson syndrome, toxic epidermal necrolysis patients, and
hypersensitivity syndrome (Kamijima et al., 2007) in addition to jaundice, hepatomegaly,
hepatosplenomegaly, and liver failure TCE-exposed workers (Huang et al., 2002; Thiele et al.,
1982). Cohort studies have examined cirrhosis mortality and either TCE exposure (ATSDR,
2004a; Blair et al., 1989; 1998; Boice et al., 1999; Boice et al., 2006b; Garabrant et al., 1988;
Morgan et al., 1998; Radican et al., 2008; Ritz, 1999a) or solvent exposure (Leigh and Jiang,
1993), but are greatly limited by their use of death certificates where there is a high degree (up to
50%) of underreporting (Blake et al., 1988), so these null findings do not rule out an effect of
TCE on cirrhosis. Overall, while there some evidence exists of liver toxicity as assessed from
liver function tests, the data are inadequate for making conclusions regarding causality.
In laboratory animals, TCE exposure is associated with a wide array of hepatotoxic
endpoints. Like humans, laboratory animals exposed to TCE have been observed to have
increased serum bile acids (Bai et al., 1992a; Neghab et al., 1997), although the toxicological
importance of this effect is unclear. Most other effects in laboratory animals have not been
studied in humans, but nonetheless provide evidence that TCE exposure leads to hepatotoxicity.
These effects include increased liver weight, small transient increases in DNA synthesis,
cytomegaly in the form of "swollen" or enlarged hepatocytes, increased nuclear size probably
reflecting polyploidization, and proliferation of peroxisomes. Liver weight increases
proportional to TCE dose are consistently reported across numerous studies and appear to be
accompanied by periportal hepatocellular hypertrophy (Berman et al., 1995; Buben and
O'Flaherty, 1985; Dees and Travis, 1993; Elcombe et al., 1985; Goel et al., 1992; Goldsworthy
and Popp, 1987; Kjellstrand et al., 1983a; Kjellstrand et al., 1983b; Kjellstrand et al., 1981b;
This document is a draft for review purposes only and does not constitute Agency policy.
712 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Laughter et al., 2004; Melnick et al., 1987; Merrick et al., 1989; Nakajima et al., 2000; Nunes et
al., 2001; Tao et al., 2000a; Tucker et al., 1982). There is also evidence of increased DNA
synthesis in a small portion of hepatocytes at around 10 days in vivo exposure (Channel et al.,
1998; Dees and Travis, 1993; Elcombe et al., 1985; Mirsalis et al., 1989). The lack of
correlation of hepatocellular mitotic figures with whole liver DNA synthesis or DNA synthesis
observed in individual hepatocytes (Dees and Travis, 1993; Elcombe et al., 1985) supports the
conclusions that cellular proliferation is not the predominant cause of increased DNA synthesis
and that nonparenchymal cells may also contribute to such synthesis. Indeed, nonparenchymal
cell activation or proliferation has been noted in several studies (Goel et al., 1992; Kjellstrand et
al., 1983a). Moreover, the histological descriptions of TCE-exposed livers are consistent with
and, in some cases, specifically note increased polyploidy (Buben and O'Flaherty, 1985).
Interestingly, changes in TCE-induced hepatocellular ploidy, as indicated by histological
changes in nuclei, have been noted to remain after the cessation of exposure (Kjellstrand et al.,
1983a). In regard to apoptosis, TCE has been reported either to have no effect or to cause a
slight increase at high doses (Channel et al., 1998; Dees and Travis, 1993). Some studies have
also noted effects from dosing vehicle alone (such as corn oil, in particular) not only on liver
pathology, but also on DNA synthesis (Channel et al., 1998; Merrick et al., 1989). Available
data also suggest that TCE does not induce substantial cytotoxicity, necrosis, or regenerative
hyperplasia, as only isolated, focal necroses and mild to moderate changes in serum and liver
enzyme toxicity markers having been reported (Channel et al., 1998; Dees and Travis, 1993;
Elcombe et al., 1985). Data on peroxisome proliferation, along with increases in a number of
associated biochemical markers, show effects in both mice and rats (Channel et al., 1998;
Elcombe et al., 1985; Goldsworthy and Popp, 1987). These effects are consistently observed
across rodent species and strains, although the degree of response at a given mg/kg-day dose
appears to be highly variability across strains, with mice on average appearing to be more
sensitive.
While it is likely that oxidative metabolism is necessary for TCE-induced effects in the
liver, the specific metabolite or metabolites responsible is less clear. TCE, TCA, and DCA
exposures have all been associated with induction of changes in liver weight, DNA synthesis,
and peroxisomal enzymes. The available data strongly support TCA not being the sole or
predominant active moiety for TCE-induced liver effects, particularly with respect to
hepatomegaly. In particular, TCE and TCA dose-response relationships are quantitatively
inconsistent, for TCE leads to greater increases in liver/body weight ratios that expected from
predicted rates of TCA production (see analysis in Section 4.5.6.2.1). In fact, above a certain
dose of TCE, liver/body weight ratios are greater than that observed under any conditions studied
This document is a draft for review purposes only and does not constitute Agency policy.
713 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
so far for TCA. Histological changes and effects on DNA synthesis are generally consistent with
contributions from either TCA or DC A, with a degree of polyploidization, rather than cell
proliferation, likely to be significant for TCE, TCA, and DC A.
Overall, TCE, likely through its oxidative metabolites, clearly leads to liver toxicity in
laboratory animals, with mice appearing to be more sensitive than other laboratory animal
species, but there is only limited epidemiologic evidence of hepatotoxicity being associated with
TCE exposure.
4.11.1.1.4. Immunotoxicity
Studies in humans provide evidence of associations between TCE exposure and a number
of immunotoxicological endpoints. The relation between systemic autoimmune diseases, such as
scleroderma, and occupational exposure to TCE has been reported in several recent studies. A
meta-analysis of scleroderma studies (Diot et al., 2002; Garabrant et al., 2003; Nietert et al.,
1998) conducted by the EPA resulted in a statistically significant combined odds ratio for any
exposure in men (OR: 2.5, 95% CI: 1.1, 5.4), with a lower relative risk seen in women (OR: 1.2,
95% CI: 0.58, 2.6). The incidence of systemic sclerosis among men is very low (approximately
1 per 100,000 per year), and is approximately 10 times lower than the rate seen in women
(Cooper and Stroehla, 2003). Thus, the human data at this time do not allow determination of
whether the difference in effect estimates between men and women reflects the relatively low
background risk of scleroderma in men, gender-related differences in exposure prevalence or in
the reliability of exposure assessment (Messing et al., 2003), a gender-related difference in
susceptibility to the effects of TCE, or chance. Changes in levels of inflammatory cytokines
were reported in an occupational study of degreasers exposed to TCE (Iavicoli et al., 2005) and a
study of infants exposed to TCE via indoor air (Lehmann et al., 2001; 2002).
Experimental studies provide additional support for these effects. Numerous studies have
demonstrated accelerated autoimmune responses in autoimmune-prone mice (Blossom et al.,
2007; Blossom et al., 2004; Cai et al., 2008; Griffin et al., 2000a; Griffin et al., 2000b). With
shorter exposure periods, effects include changes in cytokine levels similar to those reported in
human studies. More severe effects, including autoimmune hepatitis, inflammatory skin lesions,
and alopecia, were manifest at longer exposure periods, and interestingly, these effects differ
somewhat from the "normal" expression in these mice. Immunotoxic effects, including increases
in anti-dsDNA antibodies in adult animals, decreased thymus weights, and decreased plaque
forming cell response with prenatal and neonatal exposure, have been also reported in B6C3F1
mice, which do not have a known particular susceptibility to autoimmune disease (Gilkeson et
This document is a draft for review purposes only and does not constitute Agency policy.
714 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
al., 2004; Keil et al., 2009; Peden-Adams et al., 2006). Recent mechanistic studies have focused
on the roles of various measures of oxidative stress in the induction of these effects by TCE
(Wang et al., 2007b; Wang et al., 2008).
There have been a large number of case reports of a severe hypersensitivity skin disorder,
distinct from contact dermatitis and often accompanied by hepatitis, associated with occupational
exposure to TCE, with prevalences as high as 13% of workers in the same location (Kamijima et
al., 2007; 2008). Evidence of a treatment-related increase in delayed hypersensitivity response
accompanied by hepatic damage has been observed in guinea pigs following intradermal
injection (Tang et al., 2002; Tang et al., 2008), and hypersensitivity response was also seen in
mice exposed via drinking water pre- and postnatally (GD 0 through to 8 weeks of age) (Peden-
Adams et al., 2006).
Human data pertaining to TCE-related immunosuppression resulting in an increased risk
of infectious diseases is limited to the report of an association between reported history of
bacteria of viral infections in Woburn, MA (Lagakos et al., 1986). Evidence of localized
immunosuppression, as measured by pulmonary response to bacterial challenge (i.e., risk of
Streptococcal pneumonia-related mortality and clearance of Klebsiella bacteria) was seen in an
acute exposure study in CD-I mice (Aranyi et al., 1986). A 4-week inhalation exposure in
Sprague-Dawley rats reported a decrease in plaque forming cell response at exposures of
1,000 ppm (Woolhiser et al., 2006).
Overall, the human and animal studies of TCE and immune-related effects provide strong
evidence for a role of TCE in autoimmune disease and in a specific type of generalized
hypersensitivity syndrome, while there are less data pertaining to immunosuppressive effects.
4.11.1.1.5. Respiratory Tract Toxicity
There are very limited human data on pulmonary toxicity and TCE exposure. Two recent
reports of a study of gun manufacturing workers reported asthma-related symptoms and lung
function decrements associated with solvent exposure (Cakmak et al., 2004; Saygun et al., 2007),
but these studies are limited by multiple solvent exposures and the significant effect of smoking
on pulmonary function. Laboratory studies in mice and rats have shown toxicity in the bronchial
epithelium, primarily in Clara cells, following acute exposures to TCE by inhalation (see
Section 4.7.2.1.1). A few studies of longer duration have reported more generalized toxicity,
such as pulmonary fibrosis 90 days after a single 2,000 mg/kg i.p. dose in mice and pulmonary
vasculitis after 13-week oral gavage exposures to 2,000 mg/kg-day in rats (Forkert and Forkert,
1994; NTP, 1990). However, respiratory tract effects were not reported in other longer-term
This document is a draft for review purposes only and does not constitute Agency policy.
715 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
studies. Acute pulmonary toxicity appears to be dependent on oxidative metabolism, although
the particular active moiety is not known. While earlier studies implicated chloral produced in
situ by CYP enzymes in respiratory tract tissue was responsible for toxicity (reviewed in Green,
2000), the evidence is inconsistent, and several other possibilities are viable. First, substantial
"accumulation" of chloral is unlikely, as it is likely either to be rapidly converted to TCOH in
respiratory tract tissue or to diffuse rapidly into blood and be converted to TCOH in erythrocytes
or the liver. Conversely, a role for systemically produced oxidative metabolites cannot be
discounted, as CH and TCOH in blood have both been reported following inhalation dosing in
mice. In addition, a recent study reported dichloroacetyl chloride protein adducts in the lungs of
mice to which TCE was administered by i.p. injection, suggesting dichloroacetyl chloride, which
is not believed to be derived from chloral, may also contribute to TCE respiratory toxicity.
Although humans appear to have lower overall capacity for enzymatic oxidation in the lung
relative to mice, CYP enzymes do reside in human respiratory tract tissue, suggesting that,
qualitatively, the respiratory tract toxicity observed in rodents is biologically plausible in
humans. However, quantitative estimates of differential sensitivity across species due to
respiratory metabolism are highly uncertain due to limited data. Therefore, overall, data are
suggestive of TCE causing respiratory tract toxicity, based primarily on short-term studies in
mice and rats, and no data suggest that such hazards would be biologically precluded in humans.
4.11.1.1.6. Reproductive Toxicity
Reproductive toxicity related to TCE exposure has been evaluated in human and
experimental animal studies for effects in males and females. Only a limited number of studies
have examined whether TCE causes female reproductive toxicity. Epidemiologic studies have
identified possible associations of TCE exposure with effects on female fertility (ATSDR, 2001;
Sallmen et al., 1995) and with menstrual cycle disturbances (ATSDR, 2001; Bardodej and
Vyskocil, 1956; Sagawa et al., 1973; Zielinski, 1973). Reduced in vitro oocyte fertilizability has
been reported as a result of TCE exposure in rats (Berger and Horner, 2003; Wu and Berger,
2007), but a number of other laboratory animal studies did not report adverse effects on female
reproductive function (Cosby and Dukelow, 1992; George et al., 1985, 1986; Manson et al.,
1984). Overall, there are inadequate data to conclude whether adverse effects on human female
reproduction are caused by TCE.
By contrast, a number of human and laboratory animal studies suggest that TCE exposure
has the potential for male reproductive toxicity. In particular, human studies have reported TCE
exposure to be associated, in several cases statistically-significantly, with increased sperm
This document is a draft for review purposes only and does not constitute Agency policy.
716 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
density and decreased sperm quality (Chia et al., 1996; Rasmussen et al., 1988), altered sexual
drive or function (Bardodej and Vyskocil, 1956; El Ghawabi et al., 1973; Saihan et al., 1978), or
altered serum endocrine levels (Chia et al., 1997; Goh et al., 1998). In addition, three studies
that reported measures of fertility did not or could not report changes associated with TCE
exposure (ATSDR, 2001; Forkert et al., 2003; Sallmen et al., 1998), although the statistical
power of these studies is quite limited. Further evidence of similar effects is provided by several
laboratory animal studies that reported effects on sperm (George et al., 1985; Kumar et al.,
2001b; Kumar et al., 2000a; Kumar et al., 2000b; Land et al., 1981; Veeramachaneni et al.,
2001), libido/copulatory behavior (George et al., 1986; Veeramachaneni et al., 2001; Zenick et
al., 1984), and serum hormone levels (Kumar et al., 2000a; Veeramachaneni et al., 2001). As
with the human database, some studies that assessed sperm measures did not report treatment-
related alterations (Cosby and Dukelow, 1992; George et al., 1986; Xu et al., 2004; Zenick et al.,
1984). Additional adverse effects on male reproduction have also been reported, including
histopathological lesions in the testes or epididymides (Forkert et al., 2002; George et al., 1986;
Kan et al., 2007; Kumar et al., 2001b; Kumar et al., 2000b) and altered in vitro sperm-oocyte
binding or in vivo fertilization due to TCE or metabolites (DuTeaux et al., 2004b; Xu et al.,
2004). While reduced fertility in rodents was only observed in one study (George et al., 1986),
this is not surprising given the redundancy and efficiency of rodent reproductive capabilities.
Furthermore, while George et al. (1986) proposed that the adverse male reproductive outcomes
observed in rats were due to systemic toxicity, the database as a whole suggests that TCE does
induce reproductive toxicity independent of systemic effects. Therefore, overall, the human and
laboratory animal data together support the conclusion that TCE exposure poses a potential
hazard to the male reproductive system.
4.11.1.1.7. Developmental Toxicity
The relationship between TCE exposure (direct or parental) and adverse developmental
outcomes has been investigated in a number of epidemiologic and laboratory animal studies.
Prenatal effects examined include death (spontaneous abortion, perinatal death, pre- or
postimplantation loss, resorptions), decreased growth (low birth weight, small for gestational
age, intrauterine growth restriction, decreased postnatal growth), and congenital malformations,
in particular eye and cardiac defects. Postnatal developmental outcomes examined include
growth and survival, developmental neurotoxicity, developmental immunotoxicity, and
childhood cancers.
This document is a draft for review purposes only and does not constitute Agency policy.
717 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
A few epidemiological studies have reported associations between parental exposure to
TCE and spontaneous abortion or perinatal death (ATSDR, 2001; Taskinen et al., 1994;
Windham et al., 1991), although other studies reported mixed or null findings (ATSDR, 2006b,
2008; Bove, 1996; Bove et al., 1995; Goldberg et al., 1990; Lagakos et al., 1986; Lindbohm et
al., 1990; Taskinen et al., 1989). Studies examining associations between TCE exposure and
decreased birth weight or small for gestational age have reported small, often nonstatistically
significant, increases in risk for these effects ATSDR, 2006b,(ATSDR, 2008; Windham et al.,
1991). However, other studies observed mixed or no association (Bove, 1996; Bove et al., 1995;
Lagakos et al., 1986; Rodenbeck et al., 2000). While comprising both occupational and
environmental exposures, these studies are overall not highly informative due to their small
numbers of cases and limited exposure characterization or to the fact that exposures to mixed
solvents were involved. However, a number of laboratory animal studies show analogous effects
of TCE exposure in rodents. In particular, pre- or postimplantation losses, increased resorptions,
perinatal death, and decreased birth weight have been reported in multiple well-conducted
studies in rats and mice (George et al., 1985, 1986; Healy et al., 1982; Kumar et al., 2000b;
Narotsky and Kavlock, 1995; Narotsky et al., 1995). Interestingly, the rat studies reporting these
effects used Fischer 344 or Wistar rats, while several other studies, all of which used
Sprague-Dawley rats, reported no increased risk in these developmental measures (Carney et al.,
2006; Hardin et al., 1981; Schwetz et al., 1975). Overall, based on weakly suggestive
epidemiologic data and fairly consistent laboratory animal data, it can be concluded that TCE
exposure poses a potential hazard for prenatal losses and decreased growth or birth weight of
offspring.
Epidemiologic data provide some support for the possible relationship between maternal
TCE exposure and birth defects in offspring, in particular cardiac defects. Other developmental
outcomes observed in epidemiology and experimental animal studies include an increase in total
birth defects (ATSDR, 2001; Flood, 1988), CNS defects (ATSDR, 2001; Bove, 1996; Bove et
al., 1995; Lagakos et al., 1986), oral cleft defects (Bove, 1996; Bove et al., 1995; Lagakos et al.,
1986; Lorente et al., 2000), eye/ear defects (Lagakos et al., 1986; Narotsky and Kavlock, 1995;
Narotsky et al., 1995), kidney/urinary tract disorders (Lagakos et al., 1986), musculoskeletal
birth anomalies (Lagakos et al., 1986), lung/respiratory tract disorders (Das and Scott, 1994;
Lagakos et al., 1986), and skeletal defects (Healy et al., 1982). Occupational cohort studies,
while not consistently reporting positive results, are generally limited by the small number of
observed or expected cases of birth defects (Lorente et al., 2000; Taskinen et al., 1989; Tola et
al., 1980).
This document is a draft for review purposes only and does not constitute Agency policy.
718 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
While only one of the epidemiological studies specifically reported observations of eye
anomalies (Lagakos et al., 1986), studies in rats have identified increases in the incidence of fetal
eye defects following oral exposures during the period of organogenesis with TCE (Narotsky and
Kavlock, 1995; Narotsky et al., 1995) or its oxidative metabolites DCA and TCA (Smith et al.,
1989; Warren et al., 2006)(Smith et al., 1992). No other developmental or reproductive toxicity
studies identified abnormalities of eye development following TCE exposures, which may have
been related to the administered dose or other aspects of study design (e.g., level of detail applied
to fetal ocular evaluation). Overall, the study evidence suggests a potential for the disruption of
ocular development by exposure to TCE and its oxidative metabolites.
The epidemiological studies, while individually limited, as a whole show relatively
consistent elevations, some of which were statistically significant, in the incidence of cardiac
effects in TCE-exposed populations compared to reference groups (ATSDR, 2001, 2006a, 2008;
Bove, 1996; Bove et al., 1995; Goldberg et al., 1990; Yauck et al., 2004). Interestingly,
Goldberg et al. (1990) noted that the odds ratio for congenital heart disease in offspring declined
from threefold to no difference as compared to controls after TCE-contaminated drinking water
wells were closed, suggestive of a causal relationship. However, this study reported no
significant differences in cardiac lesions between exposed and nonexposed groups (Goldberg et
al., 1990). One additional community study reported that, among the 5 cases of cardiovascular
anomalies, there was no significant association with TCE (Lagakos et al., 1986), but due to the
small number of cases this does not support an absence of effect. In laboratory animal models,
avian studies were the first to identify adverse effects of TCE exposure on cardiac development,
and the initial findings have been confirmed multiple times (Boyer et al., 2000; Bross et al.,
1983; Drake et al., 2006a; Drake et al., 2006b; Loeber et al., 1988; Mishima et al., 2006; Rufer et
al., 2008). Additionally, administration of TCE and TCE metabolites TCA and DCA in maternal
drinking water during gestation has been reported to induce cardiac malformations in rat fetuses
(Johnson et al.,2005; Smith et al., 1989)(Dawson et al., 1993; Dawson et al., 1990b; Epstein et
al., 1992; Johnson et al., 1998a; Johnson et al., 2003; Johnson et al., 1998b; Smith et al., 1992).
However, it is notable that a number of other studies, several of which were well conducted, did
not report induction of cardiac defects in rats or rabbits from TCE administered by inhalation
(Carney et al., 2006; Dorfmueller et al., 1979; Hardin et al., 1981; Healy et al., 1982; Schwetz et
al., 1975) or in rats and mice by gavage (Cosby and Dukelow, 1992; Fisher et al., 2001; Narotsky
and Kavlock, 1995; Narotsky et al., 1995).
The potential importance of these effects warrants a more detailed discussion of possible
explanations for the apparent inconsistencies in the laboratory animal studies. Many of the
studies that did not identify cardiac anomalies used a traditional free-hand section technique on
This document is a draft for review purposes only and does not constitute Agency policy.
719 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
fixed fetal specimens (Dorfmueller et al., 1979; Hardin et al., 1981; Healy et al., 1982; Schwetz
et al., 1975). Detection of such anomalies can be enhanced through the use of a fresh dissection
technique as described by Staples (1974) and Stuckhardt and Poppe (1984) and this was the
technique used in the study by Dawson et al. (1990b) with further refinement of the technique
used in the positive studies by Dawson et al. (1993) and Johnson et al. (2005)(2003). However,
two studies that used the same or similar fresh dissection technique did not report cardiac
anomalies (Carney et al., 2006; Fisher et al., 2001), although it has been suggested that
differences in experimental design (e.g., inhalation versus gavage versus drinking water route of
administration, exposure during organogenesis versus the entire gestational period, or varied
dissection or evaluation procedures) may have been contributing factors to the differences in
observed response. A number of other limitations in the studies by Dawson et al. (1993) and
Johnson et al. (2005)(2003) have been suggested (Hardin et al., 2005; Watson et al., 2006). One
concern is the lack of clear dose-response relationship for the incidence of any specific cardiac
anomaly or combination of anomalies, a disparity for which no reasonable explanation has been
put forth. In addition, analyses on a fetal- rather than litter-basis and the pooling of data
collected over an extended period, including nonconcurrent controls, have been criticized. With
respect to the first issue, the study authors provided individual litter incidence data to EPA for
analysis (see Section 5, Dose-Response Assessment), and, in response to the second issue, the
study authors provided further explanation as to their experimental procedures (Johnson et al.,
2004). In sum, while the studies by Dawson et al. (1993) and Johnson et al. (2005)(2003) have
significant limitations, there is insufficient reason to dismiss their findings.
Finally, mechanistic studies, particularly based on the avian studies mentioned above,
provide additional support for TCE-induced fetal cardiac malformation, particularly with respect
to defects involving septal and valvular morphogenesis. As summarized by NRC (2006), there is
substantial concordance in the stages and events of cardiac valve formation between mammals
and birds. While quantitative extrapolation of findings from avian studies to humans is not
possible without appropriate kinetic data for these experimental systems, the treatment-related
alterations in endothelial cushion development observed in avian in ovo and in vitro studies
(Boyer et al., 2000; Mishima et al., 2006; Ou et al., 2003) provide a plausible mechanistic basis
for defects in septal and valvular morphogenesis observed in rodents, and consequently support
the plausibility of cardiac defects induced by TCE in humans.
Postnatal developmental outcomes examined after TCE prenatal and/or postnatal
exposure in both humans and experimental animals include developmental neurotoxicity,
developmental immunotoxicity, and childhood cancer. Effects on the developing nervous
system included a broad array of structural and behavioral alterations in humans (ATSDR,
This document is a draft for review purposes only and does not constitute Agency policy.
720 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
2003b; Beppu, 1968; Bernad et al., 1987; Burg and Gist, 1997; Burg et al., 1995; Laslo-Baker et
al., 2004; Till et al., 2001a; White et al., 1997, abstract; Windham et al., 2006) and animals
(Blossom et al., 2008; Fredriksson et al., 1993; George et al., 1986; Isaacson and Taylor, 1989;
Narotsky and Kavlock, 1995; Noland-Gerbec et al., 1986; Taylor et al., 1985; Westergren et al.,
1984). Adverse immunological findings in humans following developmental exposures to TCE
were reported by Lehmann et al. (2002) and Byers et al. (1988). In mice, alterations in T-cell
subpopulations, spleen and/or thymic cellularity, cytokine production, autoantibody levels (in an
autoimmune-prone mouse strain), and/or hypersensitivity response were observed after
exposures during development (Blossom and Doss, 2007; Blossom et al., 2008; Peden-Adams et
al., 2006; Peden-Adams et al., 2008), Childhood cancers included leukemia and non-Hodgkin
lymphoma (ADHS, 1990; Cohn et al., 1994b; Costas et al., 2002; Cutler et al., 1986; Flood,
1988; Flood, 1997; Kioski et al., 1990a; Kioski et al., 1990b; Lagakos et al., 1986; Lowengart et
al., 1987; McKinney et al., 1991; MDPH, 1997a; Morgan and Cassady, 2002; Shu et al., 1999),
CNS tumors (ADHS, 1990; Flood, 1988; Flood, 1997; Kioski et al., 1990a)Peters and Preston-
Martin, 1981)(De Roos et al., 2001; Morgan and Cassady, 2002; Peters et al., 1985; Peters et al.,
1981), and total cancers ((ADHS, 1990; Flood, 1988; Flood, 1997; Porter, 1993){Morgan, 2002,
707097;ATSDR, 2008, 730402}(ATSDR, 2006a). These outcomes are discussed in the other
relevant sections for neurotoxicity, immunotoxicity, and carcinogenesis.
4.11.2. Characterization of Carcinogenicity
Following EPA (2005c) Guidelines for Carcinogen Risk Assessment, TCE is
characterized as "carcinogenic to humans" by all routes of exposure. This conclusion is based
on convincing evidence of a causal association between TCE exposure in humans and kidney
cancer. The kidney cancer association cannot be reasonably attributed to chance, bias, or
confounding. The human evidence of carcinogenicity from epidemiologic studies of TCE
exposure is compelling for NHL but less convincing than for kidney cancer, and more limited for
liver and biliary tract cancer. In addition to the body of evidence pertaining to kidney cancer,
NHL, and liver cancer, the available epidemiologic studies also provide more limited evidence of
an association between TCE exposure and other types of cancer, including bladder, esophageal,
prostate, cervical, breast, and childhood leukemia. Differences between these sets of data and
the data for kidney cancer, NHL, and liver cancer are observations from fewer numbers of
studies, a mixed pattern of observed risk estimates, and the general absence of exposure-response
data from the studies using a quantitative TCE-specific exposure measure.
There are several lines of supporting evidence for TCE carcinogenicity in humans. First,
TCE induces site-specific tumors in rodents given TCE by oral gavage and inhalation. Second,
This document is a draft for review purposes only and does not constitute Agency policy.
721 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
toxicokinetic data indicate that TCE absorption, distribution, metabolism, and excretion are
qualitatively similar in humans and rodents. Finally, there is sufficient weight of evidence to
conclude that a mutagenic MOA is operative for TCE-induced kidney tumors, and this MOA is
clearly relevant to humans. MO As have not been established for other TCE-induced tumors in
rodents, and no mechanistic data indicate that any hypothesized key events are biologically
precluded in humans.
4.11.2.1.1. Summary Evaluation of Epidemiologic Evidence of Trichloroethylene (TCE)
and Cancer
The available epidemiologic studies provide convincing evidence of a causal association
between TCE exposure and cancer. The strongest epidemiologic evidence consists of reported
increased risks of kidney cancer, with more limited evidence for NHL and liver cancer, in
several well-designed cohort and case-control studies (discussed below). The summary
evaluation below of the evidence for causality is based on guidelines adapted from Hill (1965)
by EPA (2005c), and focuses on evidence related to kidney cancer, NHL, and liver cancer.
4.11.2.1.2. (a) Consistency of observed association
Elevated risks for kidney cancer have been observed across many independent studies.
Eighteen studies in which there is a high likelihood of TCE exposure in individual study subjects
(e.g., based on job-exposure matrices or biomarker monitoring) and which were judged to have
met, to a sufficient degree, the standards of epidemiologic design and analysis, were identified in
a systematic review of the epidemiologic literature. Of the 15 of these studies reporting risks of
kidney cancer (Anttila et al., 1995; Axelson et al., 1994; Boice et al., 1999; Briining et al., 2003;
Charbotel et al., 2006; Dosemeci et al., 1999; Greenland et al., 1994; Hansen et al., 2001; Moore
et al., 2010; Morgan et al., 1998; Pesch et al., 2000b; Raaschou-Nielsen et al., 2003; Radican et
al., 2008; Siemiatycki, 1991; Zhao et al., 2005), most estimated relative risks between 1.1 and
1.9 for overall exposure to TCE. Six of these 15 studies reported statistically significant
increased risks either for overall exposure to TCE (Briining et al., 2003; Dosemeci et al., 1999;
Moore et al., 2010; Raaschou-Nielsen et al., 2003) or for one of the highest TCE exposure group
(Charbotel et al., 2006; Moore et al., 2010; Raaschou-Nielsen et al., 2003; Zhao et al., 2005).
Thirteen other cohort, case-control, and geographic based studies were given less weight because
of their lesser likelihood of TCE exposure and other study design limitations that would decrease
statistical power and study sensitivity (see Sections 4.1 and 4.4.2).
This document is a draft for review purposes only and does not constitute Agency policy.
722 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
The consistency of association between TCE exposure and kidney cancer is further
supported by the results of the meta-analyses of the 15 cohort and case-control studies of
sufficient quality and with high probability TCE exposure potential to individual subjects. These
analyses observed a statistically significant increased summary relative risk estimate (RRm) for
kidney cancer of 1.27 (95% CI: 1.13, 1.43) for overall TCE. The summary relative risk were
robust and did not change appreciably with the removal of any individual study or with the use
of alternate relative risk estimates from individual studies. In addition, there was no evidence for
heterogeneity or publication bias.
The consistency of increased kidney cancer relative risk estimates across a large number
of independent studies of different designs and populations from different countries and
industries argues against chance, bias or confounding as the basis for observed associations.
This consistency, thus, provides substantial support for a causal effect between kidney cancer
and TCE exposure.
Some evidence of consistency is found between TCE exposure and NHL and liver
cancer. In a weight-of-evidence review of the NHL studies, 17 studies in which there is a high
likelihood of TCE exposure in individual study subjects (e.g., based on job-exposure matrices or
biomarker monitoring) and which met, to a sufficient degree, the standards of epidemiologic
design and analysis were identified. These studies generally reported excess relative risk
estimates for NHL between 0.8 and 3.1 for overall TCE exposure. Statistically significant
elevated relative risk estimates for overall exposure were observed in two cohort (Hansen et al.,
2001; Raaschou-Nielsen et al., 2003) and one case-control (Hardell et al., 1994) studies. The
other 14 identified studies reported elevated relative risk estimates with overall TCE exposure
that were not statistically significant (Anttila et al., 1995; Axelson et al., 1994; Boice et al., 1999;
Cocco et al., 2010; Greenland et al., 1994; Miligi et al., 2006; Morgan et al., 1998; Nordstrom et
al., 1998; Persson and Fredrikson, 1999; Purdue et al., 2011; Radican et al., 2008; Siemiatycki,
1991; Wang et al., 2009; Zhao et al., 2005). Fifteen additional studies were given less weight
because of their lesser likelihood of TCE exposure and other design limitations that would
decrease study power and sensitivity (see Sections 4.1 and 4.6.1.2). The observed lack of
association with NHL in these studies likely reflects study design and exposure assessment
limitations and is not considered inconsistent with the overall evidence on TCE and NHL.
Consistency of the association between TCE exposure and NHL is further supported by
the results of meta-analyses. These meta-analyses found a statistically significant increased
summary relative risk estimate for NHL of 1.23 (95% CI: 1.07, 1.42) for overall TCE exposure.
This result and its statistical significance were not overly influenced by most individual studies.
Some heterogeneity was observed across the 17 studies of overall exposure, though it was not
This document is a draft for review purposes only and does not constitute Agency policy.
723 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
statistically significant (p = 0.16). Analyzing the cohort and case-control studies separately
resolved most of the heterogeneity, but the result for the summary case-control studies was only
about a 7% increased relative risk estimate and was not statistically significant. The sources of
heterogeneity are uncertain but may be the result of some bias associated with exposure
assessment and/or disease classification, or from differences between cohort and case-control
studies in average TCE exposure. In addition, there is some evidence of potential publication
bias in this data set; however, it is uncertain that this is actually publication bias rather than an
association between standard error and effect size resulting for some other reason, e.g., a
difference in study populations or protocols in the smaller studies. Furthermore, if there is
publication bias in this data set, it does not appear to account completely for the finding of an
increased NHL risk.
There are fewer studies on liver cancer than for kidney cancer and NHL. Of nine studies,
all of them cohort studies, in which there is a high likelihood of TCE exposure in individual
study subjects (e.g., based on job-exposure matrices or biomarker monitoring) and which met, to
a sufficient degree, the standards of epidemiologic design and analysis in a systematic review
(Anttila et al., 1995; Axelson et al., 1994; Boice et al., 1999; Boice et al., 2006b; Greenland et
al., 1994; Hansen et al., 2001; Morgan et al., 1998; Raaschou-Nielsen et al., 2003; Radican et al.,
2008), most reported relative risk estimates for liver and gallbladder cancer between 0.5 and 2.0
for overall exposure to TCE. Relative risk estimates were generally based on small numbers of
cases or deaths, with the result of wide confidence intervals on the estimates, except for one
study (Raaschou-Nielsen et al., 2003). This study has almost 6 times more cancer cases than the
next largest study and observed a statistically significant elevated liver and gallbladder cancer
risk with overall TCE exposure (RR = 1.35 [95% CI: 1.03, 1.77]). Ten additional studies were
given less weight because of their lesser likelihood of TCE exposure and other design limitations
that would decrease statistical power and study sensitivity (see Sections 4.1 and 4.5.2).
Consistency of the association between TCE exposure and liver cancer is further
supported by the results of meta-analyses. These meta-analyses found a statistically significant
increased summary relative risk estimate for liver and biliary tract cancer of 1.29 (95% CI: 1.07,
1. 56) with overall TCE exposure. Although there was no evidence of heterogeneity or
publication bias and the summary estimate was fairly insensitive to the use of alternative relative
risk estimates, the statistical significance of the summary estimate depends heavily on the one
large study by Raaschou-Nielsen et al. (2003). However, there were fewer adequate studies
available for meta-analysis of liver cancer (9 versus 17 for NHL and 15 for kidney), leading to
lower statistical power, even with pooling. Moreover, liver cancer is comparatively rarer, with
This document is a draft for review purposes only and does not constitute Agency policy.
724 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
age-adjusted incidences roughly half or less those for kidney cancer or NHL; thus, fewer liver
cancer cases are generally observed in individual cohort studies.
4.11.2.1.3. (b) Strength of the observed association
In general, the observed associations between TCE exposure and cancer are modest, with
relative risks or odds ratios for overall TCE exposure generally less than 2.0, and higher relative
risks or odds ratios for high exposure categories. Among the highest statistically significant
relative risks were those reported for kidney cancer in the studies by Henschler et al. (1995)
(7.97 [95% CI: 2.59, 8.59]) and Vamvakas et al. (1998) (10.80 [95% CI: 3.36, 34.75]). As
discussed in Section 4.5.3, risk magnitude in both studies is highly uncertain due, in part, to
possible selection biases, and neither was included in the meta-analyses. However, the findings
of these studies were corroborated, though with lower reported relative risks, by later studies
which overcame many of their deficiencies, such as Briining et al. (2003) (2.47 [95% CI: 1.36,
4.49]), Charbotel et al. (2006; 2009) (2.16 [95% CI: 1.02, 4.60] for the high cumulative exposure
group), and Moore et al. (2010) (2.05 [95% CI: 1.13, 3.73] for high confidence assessment of
TCE). In addition, the very high apparent exposure in the subjects of Henschler et al. (1995) and
Vamvakas et al. (1998) may have contributed to their reported relative risks being higher than
those in other studies. Exposures in most population case-control studies are of lower overall
TCE intensity compared to exposures in Briining et al. (2003) and Charbotel et al. (2006; 2009),
and, as would be expected, observed relative risk estimates are lower (1.24 [95% CI: 1.03,
1.49]), Pesch et al. (2000b); 1.30 [95% CI: 0.9, 1.9], Dosemeci et al. (1999)). A few high-quality
cohort and case-control studies reported statistically significant relative risks of approximately
2.0 with highest exposure, including Zhao et al. (2005) (4.9 [95% CI: 1.23, 19.6] for high TCE
score), Raaschou-Nielsen et al. (2003) (1.7 [95% CI: 1.1, 2.4] for >5 year exposure duration,
subcohort with higher exposure]), Charbotel et al. (2006) (2.16 [95% CI: 1.02, 4.60] for high
cumulative exposure and 2.73 [95% CI: 1.06, 7.07] for high cumulative exposure plus peaks) and
Moore et al. (2010) (2.23 [95% CI: 1.07, 4.64] for high cumulative exposure and 2.41 [95% CI:
1.05, 5.56] for high average intensity TCE exposure).
Among the highest statistically significant relative risks reported for NHL were those of
Hansen et al. (2001) (3.1 [95% CI: 1.3, 6.1]), Hardell et al. (1994) (7.2 [95% CI: 1.3, 42]), the
latter a case-control study whose magnitude of risk is uncertain because of self-reported
occupational TCE exposure. A similar magnitude of risk was reported in Purdue et al. (2011) for
highest exposure (3.3 [95% CI: 1.1, 10.1], >234,000 ppm-hour, and 7.9 [95% CI: 1.8, 34.3],
This document is a draft for review purposes only and does not constitute Agency policy.
725 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
>360 ppm-hour/week). Observed relative risk estimates for liver cancer and overall TCE
exposure are generally more modest.
The strength of association between TCE exposure and cancer is modest with overall
TCE exposure. Large relative risk estimates are considered strong evidence of causality;
however, a modest risk does not preclude a causal association and may reflect a lower level of
exposure, an agent of lower potency, or a common disease with a high background level (U.S.
EPA, 2005c). Modest relative risk estimates have been observed with several well-established
human carcinogens such as benzene and secondhand smoke. Chance cannot explain the
observed association between TCE and cancer; statistically significant associations are found in a
number of the studies that contribute greater weight to the overall evidence, given their design
and statistical analysis approaches. In addition, other known or suspected risk factors cannot
fully explain the observed elevations in kidney cancer relative risks. All kidney cancer case-
control studies included adjustment for possible confounding effects of smoking, and some
studies included body mass index, hypertension, and coexposure to other occupational agents
such as cutting or petroleum oils. Cutting oils and petroleum oils, known as metalworking
fluids, have not been associated with kidney cancer (Mirer, 2010; NIOSH, 1998), and potential
confounding by this occupational co-exposure is unable to explain the observed association with
TCE. Additionally, the associations between kidney cancer and TCE exposure remained in these
studies after statistical adjustment for possible known and suspected confounders. Charbotel et
al. (2005) observed a nonstatistically significantly kidney cancer risk with exposure to TCE
adjusted for cutting or petroleum oil exposures (1.96 [95% CI: 71, 5.37] for the high-cumulative
exposure group and 2.63 [95% CI: 0.79, 8,83] for high-exposure group with peaks).
All kidney cancer case-control studies adjusted for smoking except the Moore et al.
(2010) study, which reported that smoking did not significantly change the overall association
with TCE exposure. Although direct examination of smoking and other suspected kidney cancer
risk factors is usually not possible in cohort studies, confounding is less likely in Zhao et al.
(2005), given their use of an internal referent group and adjustment for socioeconomic status, an
indirect surrogate for smoking, and other occupational exposures. In addition, the magnitude of
the lung cancer risk in Raaschou-Nielsen et al. (2003) suggests a high smoking rate is unlikely
and cannot explain their finding on kidney cancer. Last, a meta-analysis of the nine cohort
studies that reported kidney cancer risks found a summary relative risk estimate for lung cancer
of 0.96 (95% CI: 0.76, 1.21) for overall TCE exposure and 0.96 (95% CI: 0.72, 1.27) for the
highest exposure group. These observations suggest that confounding by smoking is not an
alternative explanation for the kidney cancer meta-analysis results.
This document is a draft for review purposes only and does not constitute Agency policy.
726 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Few risk factors are recognized for NHL, with the exception of viruses and suspected
factors such as immunosuppression or smoking, which are associated with specific NHL
subtypes. Associations between NHL and TCE exposure are based on groupings of several NHL
subtypes. Three of the seven NHL case-control studies adjusted for age, sex and smoking in
statistical analyses (Miligi et al., 2006; Wang et al., 2009) two others adjusted for age, sex and
education (Cocco et al., 2010; Purdue et al., 2011), and the other three case-control studies
adjusted for age only or age and sex (Hardell et al., 1994; Nordstrom et al., 1998; Persson and
Fredrikson, 1999). Like for kidney cancer, direct examination of possible confounding in cohort
studies is not possible. The use of internal controls in some of the cohort studies is intended to
reduce possible confounding related to lifestyle differences, including smoking habits, between
exposed and referent subjects.
Heavy alcohol use and viral hepatitis are established risk factors for liver cancer, with
severe obesity and diabetes characterized as a metabolic syndrome associated with liver cancer.
Only cohort studies for liver cancer are available, and they were not able to consider these
possible risk factors.
4.11.2.1.4. (c) Specificity of the observed association
Specificity is generally not as relevant as other aspects forjudging causality. As stated in
the EPA Guidelines for Carcinogen Risk Assessment (2005), based on our current understanding
that many agents cause cancer at multiple sites, and cancers have multiple causes, the absence of
specificity does not detract from evidence for a causal effect. Evidence for specificity could be
provided by a biological marker in tumors that was specific to TCE exposure. There is some
evidence suggesting particular VHL mutations in kidney tumors may be caused by TCE, but
uncertainties in these data preclude a definitive conclusion.
4.11.2.1.5. (d) Temporal relationship of the observed association
Each cohort study was evaluated for the adequacy of the follow-up period to account for
the latency of cancer development. The studies with the greatest weight based on study design
characteristics (e.g., those used in the meta-analysis) all had adequate follow-up to assess
associations between TCE exposure and cancer. Therefore, the findings of those studies are
consistent with a temporal relationship.
This document is a draft for review purposes only and does not constitute Agency policy.
727 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
4.11.2.1.6. (e) Biological gradient (exposure-response relationship)
Exposure-response relationships are examined in the TCE epidemiologic studies only to a
limited extent. Many studies examined only overall "exposed" versus "unexposed" groups and
did not provide exposure information by level of exposure. Others do not have adequate
exposure assessments to confidently distinguish between levels of exposure. For example, many
studies used duration of employment as an exposure surrogate; however, this is a poor exposure
metric given subjects may have differing exposure intensity with similar exposure duration
(NRC, 2006).
Three studies of kidney cancer reported a statistically significant trend of increasing risk
with increasing TCE exposure, Zhao et al. (2005) (p = 0.023 for trend with TCE score),
Charbotel et al. (2005; 2007) (p = 0.04 for trend with cumulative TCE exposure) and Moore et
al. (2010) (p = 0.02 for trend with cumulative TCE exposure). Charbotel et al. (2007) was
specifically designed to examine TCE exposure and had a high-quality exposure assessment and
the Moore et al. (2010) exposure assessment considered detailed information on jobs using
solvents. Zhao et al. (2005) also had a relatively well-designed exposure assessment. A positive
trend was also observed in one other study (Raaschou-Nielsen et al. (2003), with employment
duration).
Biological gradient is further supported by meta-analyses for kidney cancer using only
the highest exposure groups and accounting for possible reporting bias, which yielded a higher
summary relative risk estimate (1.58 [95% CI: 1.28, 1.96]) than for overall TCE exposure (1.27
[95% CI: 1.13, 1.43]). Although this analysis uses a subset of studies in the overall TCE
exposure analysis, the finding of higher risk in the highest exposure groups, where such groups
were available, is consistent with a trend of increased risk with increased exposure.
The NHL case-control study of Purdue et al. (2011) reported a statistically significant
trend with TCE exposure (p = 0.02 for trend with average-weekly TCE exposure), and NHL risk
in Boice et al. (1999) appeared to increase with increasing exposure duration (p = 0.20 for
routine-intermittent exposed subjects). The borderline trend with TCE intensity in the case-
control studies of Wang et al. (2009) (p = 0.06) and Purdue et al. (2011) (p = 0.08 for trend with
cumulative TCE exposure) is consistent with their findings for average weekly TCE exposure.
As with kidney cancer, further support was provided by meta-analyses using only the highest
exposure groups, which yielded a higher summary relative risk estimate (1.43 [95% CI: 1.13,
1.82]) than for overall TCE exposure (1.23 [95% CI: 1.07, 1.42]). For liver cancer, the meta-
analyses using only the highest exposure groups yielded a lower, and nonstatistically significant,
summary estimate (1.28 [95% CI: 0.93, 1.77]) than for overall TCE exposure (1.29 [95% CI:
1.07, 1.56]). There were no case-control studies on liver cancer and TCE, and the cohort studies
This document is a draft for review purposes only and does not constitute Agency policy.
728 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
generally had few liver cancer cases, making it more difficult to assess exposure-response
relationships. The one large study (Raaschou-Nielsen et al., 2003) used only duration of
employment, which is an inferior exposure metric.
4.11.2.1.7. (f) Biological plausibility
TCE metabolism is similar in humans, rats, and mice and results in reactive metabolites.
TCE is metabolized in multiple organs and metabolites are systemically distributed. Several
oxidative metabolites produced primarily in the liver, including CH, TCA and DCA, are rodent
hepatocarcinogens. Two other metabolites, DCVC and DCVG, which can be produced and
cleared by the kidney, have shown genotoxic activity, suggesting the potential for
carcinogenicity. Kidney cancer, NHL, and liver cancer have all been observed in rodent
bioassays (see below). The laboratory animal data for liver and kidney cancer are the most
robust, corroborated in multiple studies, sexes, and strains, although each has only been reported
in a single species and the incidences of kidney cancer are quite low. Lymphomas were only
reported to be statistically significantly elevated in a single study in mice, but one additional
mouse study reported elevated lymphoma incidence and one rat study reported elevated leukemia
incidence. In addition, there is some evidence both in humans and laboratory animals for kidney,
liver and immune system noncancer toxicity from TCE exposure. Several hypothesized modes
of action have been presented for the rodent tumor findings, although there are insufficient data
to support any one mode of action, and the available evidence does not preclude the relevance of
the hypothesized modes of action to humans. Activation of macrophages, natural killer cells,
and cytokine production (e.g., tumor necrosis factor), may also play an etiologic role in
carcinogenesis, and so the immune-related effects of TCE should also be considered. In
addition, the decreased in lymphocyte counts and subsets, including CD4+ T cells, and decreased
lymphocyte activation seen in TCE-exposed workers (Lan et al., 2010) also support the
biological plausibility of a role of TCE exposure in NHL.
4.11.2.1.8. (g) Coherence
Coherence is defined as consistency with the known biology. As discussed under
biological plausibility, the observance of kidney and liver cancer, and NHL in humans is
consistent with the biological processing and toxicity of TCE.
This document is a draft for review purposes only and does not constitute Agency policy.
729 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
4.11.2.1.9. (h) Experimental evidence (from human populations)
Few experimental data from human populations are available on the relationship between
TCE exposure and cancer. The only study of a "natural experiment" (i.e., observations of a
temporal change in cancer incidence in relation to a specific event) notes that childhood
leukemia cases appeared to be more evenly distributed throughout Woburn, MA, after closure of
the two wells contaminated with trichloroethylene and other organic solvents (MDPH, 1997b).
4.11.2.1.10. (i) Analogy
Exposure to structurally related chlorinated solvents such as tetrachloroethylene and
dichloromethane have also been associated with kidney, lymphoid, and liver tumors in human,
although the evidence for TCE is considered stronger.
4.11.2.1.11. Conclusion
In conclusion, based on the weight-of-evidence analysis for kidney cancer and in
accordance with EPA guidelines, TCE is characterized as "carcinogenic to humans." This
hazard descriptor is used when there is convincing epidemiologic evidence of a causal
association between human exposure and cancer. Convincing evidence is found in the
consistency of the kidney cancer findings. The consistency of increased kidney cancer relative
risk estimates across a large number of independent studies of different designs and populations
from different countries and industries provides compelling evidence given the difficulty, a
priori, in detecting effects in epidemiologic studies when the relative risks are modest, the
cancers are relatively rare, and therefore, individual studies have limited statistical power. This
strong consistency argues against chance, bias, and confounding as explanations for the elevated
kidney cancer risks. In addition, statistically significant exposure-response trends are observed
in high-quality studies. These studies were designed to examine kidney cancer in populations
with high TCE exposure intensity. These studies addressed important potential confounders and
biases, further supporting the observed associations with kidney cancer as causal. In a meta-
analysis of the 15 studies that met the inclusion criteria, a statistically significant summary
relative risk estimate was observed for overall TCE exposure (RRm: 1.27 [95% CI: 1.13, 1.43]).
The summary relative risk estimate was greater for the highest TCE exposure groups (RRm: 1.58
[95% CI: 1.28, 1.96]; n = 13 studies). Meta-analyses investigating the influence of individual
studies and the sensitivity of the results to alternate relative risk estimate selections found the
This document is a draft for review purposes only and does not constitute Agency policy.
730 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
summary relative risk estimates to be highly robust. Furthermore, there was no indication of
publication bias or significant heterogeneity. It would require a substantial amount of negative
data from informative studies (i.e., studies having a high likelihood of TCE exposure in
individual study subjects and which meet, to a sufficient degree, the standards of epidemiologic
design and analysis in a systematic review) to contradict this observed association.
The evidence is less convincing for NHL and liver cancer. While the evidence is strong
for NHL, issues of (nonstatistically significant) study heterogeneity, potential publication bias,
and weaker exposure-response results contribute greater uncertainty. The evidence is more
limited for liver cancer mainly because only cohort studies are available and most of these
studies have small numbers of cases. In addition to the body of evidence described above
pertaining to kidney cancer, NHL, and liver cancer, the available epidemiologic studies also
provide suggestive evidence of an association between TCE exposure and other types of cancer,
including bladder, esophageal, prostate, cervical, breast, and childhood leukemia, breast.
Differences between these sets of data and the data for kidney cancer, NHL, and liver cancer are
observations are from fewer numbers of studies, a mixed pattern of observed risk estimates and
the general absence of exposure-response data from the studies using a quantitative TCE-specific
cumulative exposure measure.
4.11.2.1.12. Summary of Evidence for Trichloroethylene (TCE) Carcinogenicity in Rodents
Additional evidence of TCE carcinogenicity consists of increased incidences of tumors
reported in multiple chronic bioassays in rats and mice. In total, this database identifies some of
the same target tissues of TCE carcinogenicity also seen in epidemiological studies, including the
kidney, liver, and lymphoid tissues.
Of particular note is the site-concordant finding of TCE-induced kidney cancer in rats. In
particular, low, but biologically and sometimes statistically significant, increases in the incidence
of kidney tumors were observed in multiple strains of rats treated with TCE by either inhalation
or corn oil gavage (Maltoni et al., 1986; NTP, 1988, 1990). For instance, Maltoni et al. (1986)
reported that although only 4/130 renal adenocarcinomas in rats in the highest dose group, these
tumors had never been observed in over 50,000 Sprague-Dawley rats (untreated, vehicle-treated,
or treated with different chemicals) examined in previous experiments in the same laboratory. In
addition, the gavage study by NCI (1976) and two inhalation studies by Henschler et al. (1980),
and Fukuda et al. (1983) each observed one renal adenoma or adenocarcinoma in some dose
groups and none in controls. The largest (but still small) incidences were observed in treated
male rats, only in the highest dose groups. However, given the small numbers, an effect in
This document is a draft for review purposes only and does not constitute Agency policy.
731 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
females cannot be ruled out. Several studies in rats were limited by excessive toxicity,
accidental deaths, or deficiencies in reporting (NCI, 1976; NTP, 1988, 1990). Individually,
therefore, these studies provide only suggestive evidence of renal carcinogenicity. Overall,
given the rarity of these types of tumors in the rat strains tested and the repeated similar results
across experiments and strains, these studies taken together support the conclusion that TCE is a
kidney carcinogen in rats, with males being more sensitive than females. No other tested
laboratory species (i.e., mice and hamsters) have exhibited increased kidney tumors, although
high incidences of kidney toxicity have been reported in mice (Maltoni et al., 1986; NCI, 1976;
NTP, 1990). The GSH-conjugation-derived metabolites suspected of mediating TCE-induced
kidney carcinogenesis have not been tested in a standard 2-year bioassay, so their role cannot be
confirmed definitively. However, it is clear that GSH conjugation of TCE occurs in humans and
that the human kidney contains the appropriate enzymes for bioactivation of GSH conjugates.
Therefore, the production of the active metabolites thought to be responsible for kidney tumor
induction in rats likely occurs in humans.
Statistically significant increases in TCE-induced liver tumors have been reported in
multiple inhalation and gavage studies with male Swiss mice and B6C3F1 mice of both sexes
(Anna et al., 1994; Bull et al., 2002; Herren-Freund et al., 1987; Maltoni et al., 1986; NCI, 1976;
NTP, 1990). In female Swiss mice, on the other hand, Fukuda et al. (1983), in CD-I (ICR,
Swiss-derived) mice, and Maltoni et al. (1986) both reported small, nonsignificant increases at
the highest dose by inhalation. Henschler et al. (1984; 1980) reported no increases in either sex
of Han:NMRI (also Swiss-derived) mice exposed by inhalation and ICR/HA (Swiss) mice
exposed by gavage. However, the inhalation study (Henschler et al., 1980) had only 30 mice per
dose group and the gavage study (Henschler et al., 1984) had dosing interrupted due to toxicity.
Studies in rats (Henschler et al., 1980; Maltoni et al., 1986; NCI, 1976; NTP, 1988, 1990) and
hamsters (Henschler et al., 1980) did not report statistically significant increases in liver tumor
induction with TCE treatment. However, several studies in rats were limited by excessive
toxicity or accidental deaths (NCI, 1976; NTP, 1988, 1990), and the study in hamsters only had
30 animals per dose group. These data are inadequate for concluding that TCE lacks
hepatocarcinogenicity in rats and hamsters, but are indicative of a lower potency in these species.
Moreover, it is notable that a few studies in rats reported low incidences (too few for statistical
significance) of very rare biliary- or endothelial-derived tumors in the livers of some treated
animals (Fukuda et al., 1983; Henschler et al., 1980; Maltoni et al., 1986). Further evidence for
the hepatocarcinogenicity of TCE is derived from chronic bioassays of the TCE oxidative
metabolites CH, TCA, and DCA in mice (e.g., Bull et al., 1990; DeAngelo et al., 1996;
DeAngelo et al., 2008; DeAngelo et al., 1999; George et al., 2000; Leakey et al., 2003a) (Leakey
This document is a draft for review purposes only and does not constitute Agency policy.
732 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
et al., 2003a), all of which reported hepatocarcinogenicity. Very limited testing of these TCE
metabolites has been done in rats, with a single experiment reported in both Richmond et al.
(1995) and DeAngelo et al. (1996) finding statistically significant DCA-induced
hepatocarcinogenicity. With respect to TCA, DeAngelo et al. (1997), often cited as
demonstrating lack of hepatocarcinogenicity in rats, actually reported elevated adenoma
multiplicity and carcinoma incidence from TCA treatment. However, statistically, the role of
chance could not be confidently excluded because of the low number of animals per dose group
(20-24 per treatment group at final sacrifice). Overall, TCE and its oxidative metabolites are
clearly carcinogenic in mice, with males more sensitive than females and the B6C3F1 strain
appearing to be more sensitive than the Swiss strain. Such strain and sex differences are not
unexpected, as they appear to parallel, qualitatively, differences in background tumor incidence.
Data in other laboratory animal species are limited. Thus, except for DCA, which is
carcinogenic in rats, inadequate evidence exists to evaluate the hepatocarcinogenicity of these
compounds in rats or hamsters. However, to the extent that there is hepatocarcinogenic potential
in rats, TCE is clearly less potent in the strains tested in this species than in B6C3F1 and Swiss
mice.
Additionally, there is more limited evidence for TCE-induced lymphatic cancers in rats
and mice, lung tumors in mice, and testicular tumors in rats. With respect to the lymphomas,
Henschler et al. (1980) reported statistically significant increases in lymphomas in female
Han:NMRI mice treated via inhalation. While Henschler et al. (1980) suggested these
lymphomas were of viral origin specific to this strain, subsequent studies reported increased
lymphomas in female B6C3F1 mice treated via corn oil gavage (NTP, 1990) and leukemias in
male Sprague-Dawley and female August rats (Maltoni et al., 1986; NTP, 1988). However,
these tumors had relatively modest increases in incidence with treatment, and were not reported
to be increased in other studies. With respect to lung tumors, rodent bioassays have
demonstrated a statistically significant increase in pulmonary tumors in mice following chronic
inhalation exposure to TCE (Fukuda et al., 1983; Maltoni et al., 1988; Maltoni et al., 1986).
Pulmonary tumors were not reported in other species tested (i.e., rats and hamsters; (Fukuda et
al., 1983; Henschler et al., 1980; Maltoni et al., 1988; Maltoni et al., 1986)). Chronic oral
exposure to TCE led to a nonstatistically significant increase in pulmonary tumors in mice but,
again, not in rats or hamsters (Henschler et al., 1984; Maltoni et al., 1986; NCI, 1976; NTP,
1988, 1990; Van Duuren et al., 1979). A lower response via oral exposure would be consistent
with a role of respiratory metabolism in pulmonary carcinogenicity. Finally, increased testicular
(interstitial cell and Leydig cell) tumors have been observed in rats exposed by inhalation and
gavage (Maltoni et al., 1986; NTP, 1988, 1990). Statistically significant increases were reported
This document is a draft for review purposes only and does not constitute Agency policy.
733 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
in Sprague-Dawley rats exposed via inhalation (Maltoni et al., 1986) and Marshall rats exposed
via gavage (NTP, 1988). In three rat strains, ACI, August, and F344/N, a high (>75%) control
rate of testicular tumors was observed, limiting the ability to detect a treatment effect (NTP,
1988, 1990).
In summary, there is clear evidence for TCE carcinogenicity in rats and mice, with
multiple studies showing TCE to cause tumors at multiple sites. The apparent lack of site
concordance across laboratory animal species may be due to limitations in design or conduct in a
number of rat bioassays and/or genuine interspecies differences in sensitivity. Nonetheless, these
studies have shown carcinogenic effects across different strains, sexes, and routes of exposure,
and site-concordance is not necessarily expected for carcinogens. Of greater import is the
finding that there is site-concordance between the main cancers observed in TCE-exposed
humans and those observed in rodent studies—in particular, cancers of the kidney, liver, and
lymphoid tissues.
4.11.2.1.13. Summary of Additional Evidence on Biological Plausibility
Additional evidence from toxicokinetic, toxicity, and mechanistic studies supports the
biological plausibility of TCE carcinogenicity in humans.
4.11.2.1.14. Toxicokinetics
As described in Section 3, there is no evidence of major qualitative differences across
species in TCE absorption, distribution, metabolism, and excretion. In particular, available
evidence is consistent with TCE being readily absorbed via oral, dermal, and inhalation
exposures, and rapidly distributed to tissues via systemic circulation. Extensive in vivo and in
vitro data show that mice, rats, and humans all metabolize TCE via two primary pathways:
oxidation by CYPs and conjugation with glutathione via GSTs. Several metabolites and
excretion products from both pathways, including TCA, DCA, TCOH, TCOG, NAcDCVC, and
DCVG, have been detected in blood and urine from exposed humans was well as from at least
one rodent species. In addition, the subsequent distribution, metabolism, and excretion of TCE
metabolites are qualitatively similar among species. Therefore, humans possess the metabolic
pathways that produce the TCE metabolites thought to be involved in the induction of rat kidney
and mouse liver tumors, and internal target tissues of both humans and rodents experience a
similar mix of TCE and metabolites.
This document is a draft for review purposes only and does not constitute Agency policy.
734 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
As addressed in further detail elsewhere (see Sections 3 and 5), examples of quantitative
interspecies differences in toxicokinetics include differences in partition coefficients, metabolic
capacity and affinity in various tissues, and plasma binding of the metabolite TCA. These and
other differences are addressed through PBPK modeling, which also incorporates physiological
differences among species (see Section 3.5), and are accounted for in the PBPK model-based
dose-response analyses (see Section 5). Importantly, these quantitative differences affect only
interspecies extrapolations of carcinogenic potency, and do not affect inferences as to the
carcinogenic hazard for TCE. In addition, available data on toxicokinetic differences do not
appear sufficient to explain interspecies differences in target sites of TCE carcinogenicity
(discussed further in Section 5: Dose-Response Assessment).
4.11.2.1.15. Toxicity and mode of action
Many different MO As have been proposed for TCE-induced carcinogenesis. With
respect to genotoxicity, although it appears unlikely that TCE, as a pure compound, causes point
mutations, there is evidence for TCE genotoxicity with respect to other genetic endpoints, such
as micronucleus formation (see Section 4.2.1.4.4). In addition, as discussed further below,
several TCE metabolites have tested positive in genotoxicity assays. The MOA conclusions for
specific target organs in laboratory animals are summarized below. Only in the case of the
kidney is it concluded that the data are sufficient to support a particular MOA being operative.
However, the available evidence do not indicate that qualitative differences between humans and
test animals would preclude any of the hypothesized key events in rodents from occurring in
humans.
For the kidney, the predominance of positive genotoxicity data in the database of
available studies of TCE metabolites derived from GSH conjugation (in particular DCVC, see
Section 4.2.5), together with toxicokinetic data consistent with their systemic delivery to and in
situ formation in the kidney, supports the conclusion that a mutagenic MOA is operative in
TCE-induced kidney tumors (see Section 4.4.7.1). Relevant data include demonstration of
genotoxicity in available in vitro assays of GSH conjugation metabolites and reported kidney-
specific genotoxicity after in vivo administration of TCE or DCVC. Mutagenicity is a well-
established cause of carcinogenicity. While supporting the biological plausibility of this
hypothesized MOA, available data on the VHL gene in humans or transgenic animals do not
conclusively elucidate the role of VHL mutation in TCE-induced renal carcinogenesis.
Cytotoxicity and compensatory cell proliferation, also presumed to be mediated through
metabolites formed after GSH-conjugation of TCE, have also been suggested to play a role in the
This document is a draft for review purposes only and does not constitute Agency policy.
73 5 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
MOA for renal carcinogenesis, as high incidences of nephrotoxicity have been observed in
animals at doses that also induce kidney tumors. Human studies have reported markers for
nephrotoxicity at current occupational exposures, although data are lacking at lower exposures.
Toxicity is observed in both mice and rats, in some cases with nearly 100% incidence in all dose
groups, but kidney tumors are only observed at low incidences in rats at the highest tested doses.
Therefore, nephrotoxicity alone appears to be insufficient, or at least not rate-limiting, for rodent
renal carcinogenesis, since maximal levels of toxicity are reached before the onset of tumors. In
addition, nephrotoxicity has not been shown to be necessary for kidney tumor induction by TCE
in rodents. In particular, there is a lack of experimental support for causal links, such as
compensatory cellular proliferation or clonal expansion of initiated cells, between nephrotoxicity
and kidney tumors induced by TCE. Furthermore, it is not clear if nephrotoxicity is one of
several key events in a MOA, if it is a marker for an "upstream" key event (such as oxidative
stress) that may contribute independently to both nephrotoxicity and renal carcinogenesis, or if it
is incidental to kidney tumor induction. Moreover, while toxicokinetic differences in the GSH
conjugation pathway, along with their uncertainty, are addressed through PBPK modeling, no
data suggest that any of the proposed key events for TCE-induced kidney tumors rats are
precluded in humans. Therefore, TCE-induced rat kidney tumors provide additional support for
the convincing human evidence of TCE-induced kidney cancer, with mechanistic data supportive
of a mutagenic MOA.
The strongest data supporting the hypothesis of a mutagenic MOA in either the lung or
the liver are those demonstrating the genotoxicity of CH (see Section 4.2.4), which is produced
in these target organs as a result of oxidative metabolism of TCE. It has been suggested that CH
mutagenicity is unlikely to be the cause of TCE hepatocarcinogenicity because the
concentrations required to elicit these responses are several orders of magnitude higher that
achieved in vivo (Moore and Harrington-Brock, 2000). However, it is not clear how much of a
correspondence is to be expected from concentrations in genotoxicity assays in vitro and
concentrations in vivo, as reported in vivo CH concentrations are in whole liver homogenate
while in vitro concentrations are in culture media. The use of i.p. administration, which leads to
an inflammatory response, in many other in vivo genotoxicity assays in the liver and lung
complicates the comparison with carcinogenicity data. Also, it is difficult with the available data
to assess the contributions from genotoxic effects of CH along with those from the genotoxic and
nongenotoxic effects of other oxidative metabolites (e.g., DCA and TCA). Therefore, while data
are insufficient to conclude that a mutagenic MOA mediated by CH is operant, a mutagenic
MOA in the liver or lung, either mediated by CH or by some other oxidative metabolite of TCE,
cannot be ruled out.
This document is a draft for review purposes only and does not constitute Agency policy.
736 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
A second MOA hypothesis for TCE-induced liver tumors involves activation of the
PPARa receptor. Clearly, in vivo administration of TCE leads to activation of PPARa in rodents
and likely does so in humans as well (based on in vitro data for TCE and its oxidative
metabolites). However, the evidence as a whole does not support the view that PPARa is the
sole operant MOA mediating TCE hepatocarcinogenesis. Although metabolites of TCE activate
PPARa, the data on the subsequent elements in the hypothesized MOA (e.g., gene regulation,
cell proliferation, apoptosis, and selective clonal expansion), while limited, indicate significant
differences between PPARa agonists such as Wy-14643 and TCE or its metabolites. For
example, compared with other agonists, TCE induces transient as opposed to persistent increases
in DNA synthesis; increases (or is without effect on), as opposed to decreases, apoptosis; and
induces a different H-ras mutation frequency or spectrum. These data support the view that
mechanisms other than PPARa activation may contribute to these effects; besides PPARa
activation, the other hypothesized key events are nonspecific, and available data (e.g., using
knockout mice) do not indicate that they are solely or predominantly dependent on PPARa. A
second consideration is whether certain TCE metabolites (e.g., TCA) that activate PPARa are the
sole contributors to its carcinogenicity. As summarized above (see Section 4.11.1.3), TCA is not
the only metabolite contributing to the observed noncancer effects of TCE in the liver. Other
data also suggest that multiple metabolites may also contribute to the hepatic carcinogenicity of
TCE. Liver phenotype experiments, particularly those utilizing immunostaining for c-Jun,
support a role for both DCA and TCA in TCE-induced tumors, with strong evidence that TCA
cannot solely account for the characteristics of TCE-induced tumors (e.g., Bull et al., 2002). In
addition, H-ras mutation frequency and spectrum of TCE-induced tumors more closely
resembles that of spontaneous tumors or of those induced by DCA, and were less similar in
comparison to that of TCA-induced tumors. The heterogeneity of TCE-induced tumors is similar
to that observed to be induced by a diversity carcinogens including those that do not activate
PPARa, and to that observed in human liver cancer. Taken together, the available data indicate
that, rather than being solely dependent on a single metabolite (TCA) and/or molecular target
(PPARa) multiple TCE metabolites and multiple toxicity pathways contribute to TCE-induced
liver tumors.
Other considerations as well as new data published since the NRC (2006) review are also
pertinent to the liver tumor MOA conclusions. It is generally acknowledged that, qualitatively,
there are no data to support the conclusion that effects mediated by the PPARa receptor that
contribute to hepatocarcinogenesis would be biologically precluded in humans (Klaunig et al.,
2003; NRC, 2006). It has, on the other hand, been argued that due to quantitative toxicokinetic
and toxicodynamic differences, the hepatocarcinogenic effects of chemicals activating this
This document is a draft for review purposes only and does not constitute Agency policy.
737 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
receptor are "unlikely" to occur in humans (Klaunig et al., 2003; NRC, 2006); however, several
lines of evidence strongly undermine the confidence in this assertion. With respect to
toxicokinetics, as discussed above, quantitative differences in oxidative metabolism are
accounted for in PBPK modeling of available in vivo data, and do not support interspecies
differences of a magnitude that would preclude hepatocarcinogenic effects based on
toxicokinetics alone. With respect to the MOA proposed by Klaunig et al. (2003), recent
experiments have demonstrated that PPARa activation and the sequence of key events in the
hypothesized MOA are not sufficient to induce hepatocarcinogenesis (Yang et al., 2007).
Moreover, the demonstration that the PPARa agonist DEHP induces tumors in PPARa-null mice
supports the view that the events comprising the hypothesized MOA are not necessary for liver
tumor induction in mice by this PPARa agonist (Ito et al., 2007). Therefore, several lines of
evidence, including experiments published since the NRC (2006) review, call into question the
scientific validity of using the PPARa MOA hypothesis as the basis for evaluating the relevance
to human carcinogenesis of rodent liver tumors (Guyton et al., 2009).
In summary, available data support the conclusion that the MOA for TCE-induced liver
tumors in laboratory animals is not known. However, a number of qualitative similarities exist
between observations in TCE-exposed mice and what is known about the etiology and induction
of human hepatocellular carcinomas. Polyploidization, changes in glycogen storage, inhibition
of GST-zeta, and aberrant DNA methylation status, which have been observed in studies of mice
exposed to TCE or its oxidative metabolites, are all either clearly related to human
carcinogenesis or are areas of active research as to their potential roles (PPARa activation is
discussed below). The mechanisms by which TCE exposure may interact with known risk
factors for human hepatocellular carcinomas are not known. However, available data do not
suggest that TCE exposure to mice results in liver tumors that are substantially different in terms
of their phenotypic characteristics either from human hepatocellular carcinomas or from rodent
liver tumors induced by other chemicals.
Comparing various other, albeit relatively nonspecific, tumor characteristics between
rodent species and humans provides additional support to the biologic plausibility of TCE
carcinogenicity. For example, in the kidney and the liver, the higher incidences of background
and TCE-induced tumors in male rats and mice, respectively, as compared to females parallels
the observed higher human incidences in males for these cancers (Ries et al., 2008). For the
liver, while there is a lower background incidence of liver tumors in humans than in rodents, in
the United States there is an increasing occurrence of liver cancer associated with several factors,
including viral hepatitis, higher survival rates for cirrhosis, and possibly diabetes (reviewed in
El-Serag, 2007). In addition, Leakey et al. (2003b) reported that increased body weight in
This document is a draft for review purposes only and does not constitute Agency policy.
73 8 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
B6C3F1 mice is strongly associated with increased background liver tumor incidences, although
the mechanistic basis for this risk factor in mice has not been established. Nonetheless, it is
interesting that recent epidemiologic studies have suggested obesity, in addition to associated
disorders such as diabetes and metabolic syndrome, as a risk factor for human liver cancer (El-
Serag, 2007; El-Serag and Rudolph, 2007). Furthermore, the phenotypic and morphologic
heterogeneity of tumors seen in the human liver is qualitatively similar to descriptions of mouse
liver tumors induced by TCE exposure, as well as those observed from exposure to a variety of
other chemical carcinogens. These parallels suggest similar pathways (e.g., for cell signaling) of
carcinogenesis may be active in mice and humans and support the qualitative relevance of mouse
models of liver to human liver cancer.
For mouse lung tumors, MO A hypotheses have centered on TCE metabolites produced
via oxidative metabolism in situ. As discussed above, the hypothesis that the mutagenicity of
reactive intermediates or metabolites (e.g., CH) generated during CYP metabolism contributes to
lung tumors cannot be ruled out, although available data are inadequate to conclusively support
this MO A. An alternative MOA has been posited involving other effects of such oxidative
metabolites, particularly CH, including cytotoxicity and regenerative cell proliferation.
Experimental support for this alternative hypothesis remains limited, with no data on proposed
key events in experiments of duration 2 weeks or longer. While the data are inadequate to
support this MOA hypothesis, the data also do not suggest that any proposed key events would
be biologically plausible in humans. Furthermore, the focus of the existing MOA hypothesis
involving cytotoxicity has been CH, and, as summarized above (see Section 4.11.1.5), other
metabolites may contribute to respiratory tract noncancer toxicity or carcinogenicity. In sum, the
MOA for mouse lung tumors induced by TCE is not known.
A MOA subsequent to in situ oxidative metabolism, whether involving mutagenicity,
cytotoxicity, or other key events, may also be relevant to other tissues where TCE would
undergo CYP metabolism. For instance, CYP2E1, oxidative metabolites, and protein adducts
have been reported in the testes of rats exposed to TCE, and, in some rat bioassays, TCE
exposure increased the incidence of rat testicular tumors. However, inadequate data exist to
adequately define a MOA hypothesis for this tumor site.
4.11.3. Characterization of Factors Impacting Susceptibility
As discussed in more detail in Section 4.10, there is some evidence that certain
populations may be more susceptible to exposure to TCE. Factors affecting susceptibility
examined include lifestage, gender, genetic polymorphisms, race/ethnicity, preexisting health
status, and lifestyle factors and nutrition status.
This document is a draft for review purposes only and does not constitute Agency policy.
739 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Examination of early lifestages includes exposures such as transplacental transfer
(Beppu, 1968; Ghantous et al., 1986; Helliwell and Hutton, 1950; Laham, 1970; Withey and
Karpinski, 1985) and breast milk ingestion (Fisher et al., 1997; Fisher et al., 1990; Hamada and
Tanaka, 1995; Pellizzari et al., 1982), early lifestage-specific toxicokinetics, PBPK models
(Fisher et al., 1989; Fisher et al., 1990), and differential outcomes in early lifestages such as
developmental cardiac defects. Although there is more information on susceptibility to TCE
during early lifestages than on susceptibility during later lifestages or for other populations with
potentially increased susceptibility, there remain a number of uncertainties and data gaps
regarding children's susceptibility. Improved PBPK modeling for using childhood parameters
for early lifestages as recommended by the NRC (2006), and validation of these models will aid
in determining how variations in metabolic enzymes affect TCE metabolism. In particular, the
NRC states that it is prudent to assume children need greater protection than adults, unless
sufficient data are available to justify otherwise (NRC, 2006). Because the weight of evidence
supports a mutagenic MOA for TCE carcinogenicity in the kidney (see Section 4.4.7), and there
is an absence of chemical-specific data to evaluate differences in carcinogenic susceptibility,
early-life susceptibility should be assumed and the ADAFs should be applied, in accordance with
the Supplemental Guidance (discussed further in Section 5).
Fewer data are available on later lifestages, although there is suggestive evidence to
indicate that older adults may experience increased adverse effects than younger adults (Mahle et
al., 2007; Rodriguez et al., 2007). In general, more studies specifically designed to evaluate
effects in early and later lifestages are needed in order to more fully characterize potential life
stage-related TCE toxicity.
Examination of gender-specific susceptibility includes toxicokinetics, PBPK models
(Fisher et al., 1998), and differential outcomes. Gender differences observed are likely due to
variation in physiology and exposure.
Genetic variation likely has an effect on the toxicokinetics of TCE. In particular,
differences in CYP2E1 activity may affect susceptibility of TCE due to effects on production of
toxic metabolites (Kim and Ghanayem, 2006; Lipscomb et al., 1997; Povey et al., 2001; Yoon et
al., 2007). GST polymorphisms could also play a role in variability in toxic response (Briining et
al., 1997a; Wiesenhiitter et al., 2007), as well as other genotypes, but these have not been
sufficiently tested. Differences in genetic polymorphisms related to the metabolism of TCE have
also been observed among various race/ethnic groups (Inoue et al., 1989; Sato et al., 1991b).
Preexisting diminished health status may alter the response to TCE exposure. Individuals
with increased body mass may have an altered toxicokinetic response (Clewell et al., 2000;
Davidson and Beliles, 1991; Lash et al., 2000a; McCarver et al., 1998; Monster et al., 1979;
This document is a draft for review purposes only and does not constitute Agency policy.
740 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Sato, 1993; Sato et al., 1991b) resulting in changes the internal concentrations of TCE or in the
production of toxic metabolites. Other conditions, including diabetes and hypertension, are risk
factors for some of the same health effects that have been associated with TCE exposure, such as
renal cell carcinoma. However, the interaction between TCE and known risk factors for human
diseases is not known, and further evaluation of the effects due to these factors is needed.
Lifestyle and nutrition factors examined include alcohol consumption, tobacco smoking,
nutritional status, physical activity, and socioeconomic status. In particular, alcohol intake has
been associated with metabolic inhibition (altered CYP2E1 expression) of TCE in both humans
and experimental animals (Bardodej and Vyskocil, 1956; Barret et al., 1984; Kaneko et al.,
1994a; Larson and Bull, 1989; McCarver et al., 1998; Muller et al., 1975; Nakajima et al., 1988;
Nakajima et al., 1992a; Nakajima et al., 1990; Okino et al., 1991; Sato, 1993; Sato et al., 1991a;
Sato and Nakajima, 1985; Sato et al., 1980, 1981, 1983; Stewart et al., 1974; White and Carlson,
1981b). In addition, such factors have been associated with increased baseline risks for health
effects associated with TCE, such as kidney cancer (e.g., smoking) and liver cancer (e.g., alcohol
consumption). However, the interaction between TCE and known risk factors for human
diseases is not known, and further evaluation of the effects due to these factors is needed.
In sum, there is some evidence that certain populations may be more susceptible to
exposure to TCE. Factors affecting susceptibility examined include lifestage, gender, genetic
polymorphisms, race/ethnicity, preexisting health status, and lifestyle factors and nutrition status.
However, except in the case of toxicokinetic variability characterized using the PBPK model
described in Section 3.5, there are inadequate chemical-specific data to quantify the degree of
differential susceptibility due to such factors.
This document is a draft for review purposes only and does not constitute Agency policy.
741 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
5. DOSE-RESPONSE ASSESSMENT
5.1. DOSE-RESPONSE ANALYSES FOR NONCANCER ENDPOINTS
Because of the large number of noncancer health effects associated with trichloroethylene
(TCE) exposure and the large number of studies reporting on these effects, a screening process,
described below, was used to reduce the number of endpoints and studies to those that would
best inform the selection of the critical effects for the inhalation reference concentration (RfC)
and oral reference dose (RfD). 16 The screening process helped identify the more sensitive
endpoints for different types of effects within each health effect domain (e.g., different target
systems) and provided information on the exposure levels that could contribute to the most
sensitive effects, used for the RfC and RfD, as well as to additional noncancer effects as
exposure increases. These more sensitive endpoints were also used to investigate the impacts of
pharmacokinetic uncertainty and variability.
The general process used to derive the RfD and RfC was as follows (see Figure 5-1):
(1) Consider all studies described in Section 4 that report adverse noncancer health effects or
markers for such effects and provide quantitative dose-response datal7.
(2) Consider for each study/endpoint possible points of departure (PODs) on the basis of
applied dose, with the order of preference being first a benchmark dose (BMD)18 derived
from empirical modeling of the dose-response data, then a no-observed-adverse-effect
level (NOAEL), and lastly a lowest-observed-adverse-effect level (LOAEL).
(3) Adjust each POD by endpoint/study-specific "uncertainty factors" (UFs), accounting for
uncertainties and adjustments in the extrapolation from the study conditions to conditions
of human exposure, to derive candidate RfCs (cRfCs) or RfDs (cRfDs) intended to be
protective for each endpoint (individually) on the basis of applied dose.
16 In U.S. EPA noncancer health assessments, the RfC (RfD) is an estimate (with uncertainty spanning perhaps an
order of magnitude) of a continuous inhalation (daily oral) exposure to the human population (including sensitive
subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime. It can be derived
from a NOAEL, LOAEL, or benchmark concentration (dose), with uncertainty factors generally applied to reflect
limitations of the data used.
17 Adequate dose-response data comprise, at a minimum, one exposure group and an appropriate control group,
from which one can derive a LOAEL (or a NOAEL, if evidence of the effect is available from some other
comparable study).
18 More precisely, it is the BMDL, i.e., the (one-sided) 95% lower confidence bound on the dose corresponding to
the benchmark response for the effect that is used as the POD.
This document is a draft for review purposes only and does not constitute Agency policy.
1 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
(4) Array the cRfCs and cRfDs across the following health effect domains: (1) neurotoxic
effects; (2) systemic (body weight) and organ toxicity (kidney, liver) effects;
(3) immunotoxic effects; (4) reproductive effects; and (5) developmental effects.
13)
All
studies
(2^ Points of
Departure
~ (POPs)
1
Apply
'Uncertainty
Factors
Candidate RfC
(cRfCs) &
candidate RfDs
(cRfDs)
[applied dose]
Lowest
values within
each domain
©
Candidate
critical effects/
studies,
cRfCs & cRfDs
Apply PBPK
model;
Update UFs
PBPK-based
candidate RfCs
(p-cRfDs) &
candidate RfDs,
(p-cRfDs)
^C onsider and evaluate most sensitive
r
estimates across domains and their
uncertainties
RfC and RfD for
noncancer effects
Figure 5-1. Flow-chart of the process used to derive the RfD and RfC for
noncancer effects.
(5) Select as candidate critical effects those endpoints with the lowest cRfCs or cRfDs,
within each of these effect domains, taking into account the confidence in each estimate.
When there are alternative estimates available for a particular endpoint, preference is
given to studies whose design characteristics (e.g., species, statistical power, exposure
level(s) and duration, endpoint measures) are better suited for determining the most
sensitive human health effects of chronic TCE exposure.
(6) For each candidate critical effect selected in step 5, use, to the extent possible, the
physiologically based pharmarcokinetic (PBPK) model developed in Section 3.5 to
calculate an internal dose POD (idPOD) for plausible internal dose-metrics that were
selected on the basis of what is understood about the role of different TCE metabolites in
toxicity and the mode of action (MOA) for toxicity.
This document is a draft for re\'iew purposes only and does not constitute Agency policy.
2 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
(7) For each idPOD for each candidate critical effect, use the PBPK model to estimate
interspecies and within-human pharmacokinetic variability (or just within-human
variability for human-based PODs). The results of this calculation are 99th percentile
estimates of the human equivalent concentration and human equivalent dose (HEC99 and
HED99) for each candidate critical effect. 19
(8) Adjust each HEC99 or HED99 by endpoint/study-specific UFs (which, due to the use of
the PBPK model, may differ from the UFs used in step 3) to derive a PBPK model-based
candidate RfCs (p-cRfC) and RfD (p-cRfD) for each candidate critical effect.
(9) Characterize the uncertainties in the cRfCs, cRfDs, p-cRfCs, and p-cRfDs, with the
inclusion of quantitative uncertainty analyses of pharmacokinetic uncertainty and
variability as derived from the Bayesian population analysis using the PBPK model.
(10) Evaluate the most sensitive cRfCs, p-cRfCs, cRfDs, and p-cRfDs, taking into account the
confidence in the estimates, to arrive at an RfC and RfD for TCE.
In contrast to the approach used in most assessments, in which the RfC and RfD are each
based on a single critical effect, the final RfC and RfD for TCE were based on multiple critical
effects that resulted in very similar candidate RfC and RfD values at the low end of the full range
of values. This approach was taken here it is considered to provide more robust estimates of the
RfC and RfD and because it highlights the multiple effects that are yielding very similar
candidate values. The results of this process are summarized in the sections below, with
technical details presented in Appendix F.
5.1.1. Modeling Approaches and Uncertainty Factors for Developing Candidate Reference
Values Based on Applied Dose
This section summarizes the general methodology used with all the TCE studies and
endpoints for developing cRfCs and cRfDs on the basis of applied dose. A detailed discussion of
the application of these approaches to the studies and endpoints for each health effect domain
follows in the next section (see Section 5.1.2).
Standard adjustments20 were made to the applied doses to obtain continuous inhalation
exposures and daily average oral doses over the study exposure period (see Appendix F for
19 The choice of the 99th percentile is discussed in Section 5.1.3.2.
20Discontinuous exposures (e.g., gavage exposures once a day, 5 days/week, or inhalation exposures for
5 days/week, 6 hours/day) were adjusted to the continuous exposure yielding the same cumulative exposure. For
inhalation studies, these adjustments are equivalent to those recommended by U.S. EPA (1994b) for deriving a
This document is a draft for review purposes only and does not constitute Agency policy.
3 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
details), except for effects for which there was sufficient evidence that the effect was more
closely associated with administered exposure level (e.g., changes in visual function). The PODs
based on applied dose in the following sections and in Appendix F are presented in terms of the
adjusted doses (except where noted).
As described above, wherever possible,21 benchmark dose modeling was conducted to
obtain benchmark dose lower bounds (BMDLs) to serve as PODs for the cRfCs and cRfDs.
Note that not all quantitative dose-response data are amenable to benchmark dose modeling. For
example, while nonnumerical data (e.g., data presented in line or bar graphs rather than in tabular
form) were considered for developing LOAELs or NOAELs, they were not used for benchmark
dose modeling. In addition, sometimes the available models used do not provide an adequate fit
to the data. For the benchmark dose modeling for this assessment, the EPA's BenchMark Dose
Software (BMDS), which is freely available at www.epa.gov/ncea/bmds. was used. For
dichotomous responses, the log-logistic, multistage, and Weibull models were fitted. This subset
of BMDS dichotomous models was used to reduce modeling demands, and these particular
models were selected because, as a group, they have been found to be capable of describing the
great majority of dose-response data sets, and specifically for some TCE data sets (Falk Filipsson
and Victorin, 2003). For continuous responses, the distinct models available in BMDS—the
power, polynomial, and Hill models—were fitted. For some reproductive and developmental
data sets, two nested models (the nested logistic and the Rai and Van Ryzin models in BMDS22)
were fitted to examine and account for potential intralitter correlations. Models with
unconstrained power parameters <1 were considered when the dose-response relationship
appeared supralinear, but these models often yield very low BMDL estimates and there was no
situation in which an unconstrained model with a power parameter <1 was selected for the data
sets modeled here. In most cases, a constrained model or the Hill model provided an adequate fit
to such a dose-response relationship. In a few cases, the highest-dose group was dropped to
obtain an improved fit to the lower-dose groups. See Appendix F for further details on model
fitting and parameter constraints.
human equivalent concentration for a Category 3 gas for which the blood:air partition coefficient in laboratory
animals is greater than that in humans (see Section 3.1 for discussion of the TCE blood:air partition coefficient).
21 An exception was for the systemic effect of decreased body weight, which was observed in multiple chronic
studies. Dose-response data were available, but the resources were not invested into modeling these data because
the endpoint appeared a priori to be less sensitive than others and was not expected to be a critical effect.
22The BMDS vl.4 module for the National Center for Toxicological Research model failed with the TCE data sets.
This document is a draft for review purposes only and does not constitute Agency policy.
4 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
After fitting these models to the data sets, the following procedure for model selection was
applied. First, models were rejected if the p-w alue for goodness of fit was <0.10.23 Second,
models were rejected if they did not appear to adequately fit the low-dose region of the
dose-response relationship, based on an examination of graphical displays of the data and scaled
residuals. If the BMDL estimates from the remaining models were "sufficiently close" (with a
criterion of within twofold for "sufficiently close"), then the model with the lowest Akaike
Information Criteria (AIC) was selected. 24 If the BMDL estimates from the remaining models
are not sufficiently close, some model dependence is assumed. With no clear biological or
statistical basis to choose among them, the lowest BMDL was chosen as a reasonable
conservative estimate, unless the lowest BMDL appeared to be an outlier, in which case further
judgments were made. Additionally, for continuous models, constant variance models were used
for model parsimony unless the p-w alue for the test of homogenous variance was <0.10, in which
case the modeled variance models were considered.
For benchmark response (BMR) selection, statistical and biological considerations were
taken into account. For dichotomous responses, our general approach was to use 10% extra risk
as the BMR for borderline or minimally adverse effects and either 5% or 1% extra risk for
adverse effects, with 1% reserved for the most severe effects. For continuous responses, the
preferred approach for defining the BMR is to use a preestablished cut-point for the minimal
level of change in the endpoint at which the effect is generally considered to become biologically
significant (e.g., there is substantial precedence for using a 10% change in weight for organ and
body weights and a 5% change in weight for fetal weight). In the absence of a well-established
cut-point, a BMR of 1 (control) standard deviation (SD) change from the control mean, or
0.5 SD for effects considered to be more serious, was generally selected. For one neurological
effect (traverse time), a doubling (i.e., twofold change) was selected because the control SD
appeared unusually small.
After the PODs were determined for each study/endpoint, UFs were applied to obtain the
cRfCs and cRfDs. Uncertainty factors are used to address differences between study conditions
and conditions of human environmental exposure (U.S. EPA, 2002c). These include
23In a few cases in which none of the models fit the data with p> 0.10, linear models were selected on the basis of
an adequate visual fit overall.
24 Akaike Information Criteria—a measure of information loss from a dose-response model that can be used to
compare a set of models. Among a specified set of models, the model with the lowest AIC is considered the "best."
If two or more models share the lowest AIC, an average of the BMDLs could be used, but averaging was not used in
this assessment because for the one occasion in which models shared the lowest AIC, a selection was made based on
visual fit.
This document is a draft for review purposes only and does not constitute Agency policy.
5 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
(a) Extrapolating from laboratory animals to humans: If a POD is derived from
experimental animal data, it is divided by an UF to reflect pharmacokinetic and
pharmacodynamic differences that may make humans more sensitive than laboratory
animals. For oral exposures, the standard value for the interspecies UF is 10, which
breaks down (approximately) to a factor of three for pharmacokinetic differences
(which is removed if the PBPK model is used) and a factor of three for
pharmacodynamic differences. For inhalation exposures, ppm equivalence across
species is generally assumed, in which case pharmacokinetic differences are
considered to be negligible, and the standard value used for the interspecies UF is 3,
which is ascribed to pharmacodynamic differences. These standard values were used
for all the cRfCs and cRfDs based on laboratory animal data in this assessment.
(b) Human (intraspecies) variability: RfCs and RfDs apply to the human population,
including sensitive subgroups, but studies rarely examine sensitive humans. Sensitive
humans could be adversely affected at lower exposures than a general study
population; consequently, PODs from general-population studies are divided by an
UF to address sensitive humans. Similarly, the animals used in most laboratory
animal studies are considered to be "typical" or "average" responders, and the human
(intraspecies) variability UF is also applied to PODs from such studies to address
sensitive subgroups. The standard value for the human variability UF is 10, which
breaks down (approximately) to a factor of three for pharmacokinetic variability
(which is removed if the PBPK model is used) and a factor of three for
pharmacodynamic variability. This standard value was used for all the PODs in this
assessment with the exception of the PODs for a few immunological effects that were
based on data from a sensitive (autoimmune-prone) mouse strain; for those PODs, an
UF of 3 was used for human variability.
(c) Uncertainty in extrapolating from subchronic to chronic exposures: 25 RfCs and RfDs
apply to lifetime exposure, but sometimes the best (or only) available data come from
less-than-lifetime studies. Lifetime exposure can induce effects that may not be
apparent or as large in magnitude in a shorter study; consequently, a dose that elicits a
specific level of response from a lifetime exposure may be less than the dose eliciting
the same level of response from a shorter exposure period. Thus, PODs based on
subchronic exposure data are generally divided by a subchronic-to-chronic UF, which
has a standard value of 10. If there is evidence suggesting that exposure for longer
time periods does not increase the magnitude of an effect, a lower value of three or
one might be used. For some reproductive and developmental effects, chronic
exposure is that which covers a specific window of exposure that is relevant for
eliciting the effect, and subchronic exposure would correspond to an exposure that is
notably less than the full window of exposure.
25 Rodent studies exceeding 90 days of exposure are considered chronic, and rodent studies with 4 weeks to 90 days
of exposure are considered subchronic (see http://www.epa. gov/iris/hefo gloss.him).
This document is a draft for review purposes only and does not constitute Agency policy.
6 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
(d) Uncertainty in extrapolating from LOAELs to NOAELs: PODs are intended to be
estimates of exposure levels without appreciable risk under the study conditions so
that, after the application of appropriate UFs for interspecies extrapolation, human
variability, and/or duration extrapolation, the absence of appreciable risk is conveyed
to the RfC or RfD exposure level to address sensitive humans with lifetime exposure.
Under the NOAEL/LOAEL approach to determining a POD, however, adverse
effects are sometimes observed at all study doses. If the POD is a LOAEL, it is
divided by an UF to better estimate a NOAEL. The standard value for the
LOAEL-to-NOAEL UF is 10, although sometimes a value of three is used if the
effect is considered minimally adverse at the response level observed at the LOAEL
or even one if the effect is an early marker for an adverse effect. For one POD in this
assessment, a value of 30 was used for the LOAEL-to-NOAEL UF because the
incidence rate for the adverse effect was >90% at the LOAEL.
(e) Additional database uncertainties: A database UF of 1, 3 or 10 is used to reflect the
potential for deriving an underprotective toxicity value as a result of an incomplete
characterization of the chemical's toxicity. No database UF was used in this
assessment. See Section 5.1.4.1 for additional discussion of the uncertainties
associated with the overall database for TCE.
5.1.2. Candidate Critical Effects by Effect Domain
A large number of endpoints and studies were considered within each of the five health
effect domains. A comprehensive list of all endpoints/studies that were considered for
developing cRfCs and cRfDs is shown in Tables 5-1-5-5. These tables also summarize the
PODs for the various study endpoints, the UFs applied, and the resulting cRfCs or cRfDs.
Inhalation and oral studies are presented together so that the extent of the available data, as well
as concordance or lack thereof in the responses across routes of exposure, is evident. In addition,
the PBPK model developed in Section 3.5 will be applied to each candidate critical effect to
develop a POD based on internal dose (idPOD); and subsequent extrapolation of the idPOD to
pharmacokinetically sensitive humans is performed for both inhalation and oral human
exposures, regardless of the route of exposure in the original study.
The sections below discuss the cRfCs and cRfDs developed from the effects and studies
identified in the hazard characterization (see Section 4) that were suitable for the derivation of
reference values (i.e., that provided quantitative dose-response data). Because the general
approach for applying UFs was discussed above, the sections below only discuss the selection of
particular UFs when there are study characteristics that require additional judgment as to the
appropriate UF values and possible deviations from the standard values usually assigned.
This document is a draft for review purposes only and does not constitute Agency policy.
7 DRAFT—DO NOT CITE OR QUOTE
-------
5.1.2.1.1. Candidate Critical Neurological Effects on the Basis of Applied Dose
1 As summarized in Section 4.11.1.1, both human and experimental animal studies have
2 associated TCE exposure with effects on several neurological domains. The strongest
3 neurological evidence of hazard is for changes in trigeminal nerve function or morphology and
4 impairment of vestibular function. There is also evidence for effects on motor function; changes
5 in auditory, visual, and cognitive function or performance; structural or functional changes in the
6 brain; and neurochemical and molecular changes. Studies with numerical dose-response
7 information are summarized in Table 5-1, with their corresponding cRfCs or cRfDs shown in
8 Table 5-2. Because impairment of vestibular function occurs at higher exposures, such changes
9 were not considered candidate critical effects; but the other neurological effect domains are
10 represented. For trigeminal nerve effects, cRfC estimates based on two human studies are in a
This document is a draft for review purposes only and does not constitute Agency policy.
8 DRAFT—DO NOT CITE OR QUOTE
-------
H
tr
V
a-
o
o
c
3
3
fa
~1
fa
tf
3*
<
T3
C
>3
o
V
a>
VI
o
3
vf
fa
3
O-
O-
o
v
3
O
r-K
o
o
3
v
C
r-K
OQ
3
o
O
o"
Table. 5-1 Summary of studies of neurological effects suitable for dose-response assessment
Effect type
Study reference
Species, strain (if
appl.), sex, number
used for dose-response
assessment
Exposure(s) used for
dose-response
assessment
Endpoint(s) used for dose-response
assessment
Section 4
Section/Table
Trigeminal nerve effects
Section 4.3.1
Mhiri et al. (2004)
Human phosphate
industry workers (23
exposed, 23 controls)
Inhalation: Exposure
ranged from 50-150
ppm, for 6 hr/d for at
least 2 yr
Increased trigeminal somatosensory
evoked potential latency
Table 4-20
Ruijtenetal. (1991)
Human mail printing
workers (31 exposed,
28 controls)
Inhalation: Mean
cumulative exposure:
704 ppm x yrs; mean
exposure duration: 16
yr
Increased latency in masseter reflex
Table 4-20
Barret etal. (1992)
Rat, Sprague-Dawley,
female, 7/group
Oral: 0, 2,500 mg/kg; 1
dose/d, 5 d/wk, 10 wk
Increased internode length and fiber
diameter in class A fibers of the
trigeminal nerve observed with TCE
treatment; changes in fatty acid
composition.
Table 4-21
Auditory effects
Section 4.3.2
Rebert et al. (1991)
Rat, Long Evans, male,
10/group
Inhalation: 0, 1,600,
and 3,200 ppm; 12 h/d,
12 wks
Significant decreases in B AER
amplitude and an increase in latency
of appearance of the initial peak (PI).
Table 4-23
Albee et al. (2006)
Rat, Fischer 344, male
and female,
10/sex/group
Inhalation: 0, 250, 800,
2,500 ppm; 6 h/d, 5
d/wk, 13 wk
Mild frequency specific hearing
deficits; focal loss of cochlear hair
cells.
Table 4-23
Crofton and Zhao (1997)
Rat, Long Evans, male,
8-10/group
Inhalation: 0, 800,
1,600, 2,400, and 3,200
ppm; 6 h/d, 5 d/wk, 13
wk
Increased auditory thresholds as
measured by BAERs for the 16 kHz
tone
Table 4-23
-------
H
tr
V
a-
o
o
c
3
3
fa
~1
fa
tf
3*
<
T3
C
>3
o
V
a>
VI
o
3
vf
fa
3
O-
O-
o
v
3
O
r-K
o
o
3
v
C
r-K
OQ
3
o
O
o"
Table. 5-1 Summary of studies of neurological effects suitable for dose-response assessment (continued)
Effect type
Study reference
Species, strain (if
appl.), sex, number
used for dose-response
assessment
Exposure(s) used for
dose-response
assessment
Endpoint(s) used for dose-response
assessment
Section 4
Section/Table
Psychomotor effects
Section 4.3.6
Waseemetal. (2001)
Rat, Wistar, male,
8/group
Inhalation: 0, 376 ppm
for up to 180 d; 4 h/d, 5
d/wk
Changes in locomotor activity.
Table 4-31
Nunes et al. (2001)
Rat, Sprague-Dawley,
male, 10/group
Oral: 0, 2,000
mg/kg/day; 7 d
Increased foot splay.
Table 4-30
Moseretal. (1995)
Rat, Fischer 344,
female, 8/dose
Oral: 0, 150, 500,
1,500, and 5,000 mg/kg,
1 dose
Neuro-muscular impairment
Table 4-30
0, 50, 150, 500, and
1,500 mg/kg/day, 14 d
Increased rearing activity
Table 4-30
Visual function effects
Section 4.3.4
Blainetal. (1994)
Rabbit, New Zealand
albino, male, 6-8/group
Inhalation: 0, 350, 700
ppm; 4 h/d, 4 d/wk, 12
wk
Weekly electroretinograms (ERGs)
and oscillatory potentials (OPs).
Table 4-26
Cognitive effects
Section 4.3.5 and
4.3.6
Kulig et al. (1987)
Rat, Wistar, male,
8/dose
Inhalation: 0, 500,
1,000, and 1,500 ppm;
16 h/d, 5 d/wk, 18 wk
Increased time in 2-choice visual
discrimination test.
Table 4-31
Isaacson et al. (1990)
Rat, Sprague Dawley,
male weanlings,
12/dose
Oral: (1) 0 mg/kg/day, 8
wk
(2) 47 mg/kg/day, 4 wk
+ 0 mg/kg/day, 4 wk
(3) 47 mg/kg/day, 4 wk
+ 0 mg/kg/day, 2 wk +
24 mg/kg/day, 2 wk
Demyelination of hippocampus
Table 4-28
-------
H
tr
V
a-
o
o
c
3
3
fa
~1
fa
tf
3*
<
T3
C
>3
o
V
a>
VI
o
3
vf
fa
3
O-
O-
o
v
3
O
r-K
o
o
3
v
C
r-K
OQ
3
o
O
o"
Table. 5-1 Summary of studies of neurological effects suitable for dose-response assessment (continued)
Effect type
Study reference
Species, strain (if
appl.), sex, number
used for dose-response
assessment
Exposure(s) used for
dose-response
assessment
Endpoint(s) used for dose-response
assessment
Section 4
Section/Table
Mood and sleep disorders
Section 4.3.7
Albee et al. (2006)
Rat, Fischer 344, male
and female,
10/sex/group
Inhalation: 0, 250, 800,
2,500 ppm; 6 h/d, 5
d/wk, 13 wk
Increased handling reactivity
Table 4-33
Arito et al. (1994)
Rat, Wistar, male,
5/group
Inhalation: 0,50, 100,
and 300 ppm; 8 h/d, 5
d/wk, for 6 wk
Significant decreases in wakefulness.
Table 4-33
Other neurological effects
Section 4.3.9
Kjellstrand et al. (1987)
Rat, Sprague-Dawley,
female
0, 300 ppm, 24 h/d, 24
d
Sciatic nerve regeneration was
inhibited.
Table 4-36
Mouse, NMRI, male
0, 150, or 300 ppm, 24
h/d, 24 d
Sciatic nerve regeneration was
inhibited.
Table 4-36
Gash et al. (2008)
Rat, Fischer 344, male,
9/group
Oral: 0 and 1,000
mg/kg; 5 d/wk, 6 wk
Degeneration of dopamine-containing
neurons in substantia nigra.
Table 4-35
Table 5-2. Neurological effects in studies suitable for dose-response assessment, and corresponding cRfCs and
cRfDs
Effect type
Supporting studies
Species
POD
type
PODa
UFSC
UFis
UFh
UFloael
UFdb
TJF b
comp
cRfC
(ppm)
cRfD
(mg/kg/day)
Effect; comments
Trigeminal nerve effects
-------
Mliiri et al. (2004)
Human
LOAEL
40
1
1
10
10
1
100
0.40
Abnormal trigeminal somatosensory
evoked potentials; preferred POD based
on middle of reported range of
50-150 ppm.
Human
LOAEL
6
1
1
10
10
1
100
0.06
Alternate POD based on U-TCA and Ikeda
et al. (1972).
Ruijtenet al. (1991)
Human
LOAEL
14
1
1
10
3
1
30
0.47
Trigeminal nerve effects; POD based on
mean cumulative exposure and mean
duration UFloael = 3 due to early marker
effect and minimal degree of change.
Barret et al. (1992)
Rat
LOAEL
1,800
10
10
10
10
1
10,000c
0.18
Morphological changes; uncertain
adversity; some effects consistent with
demyelination.
Auditory effects
Rebert et al. (1991)
Rat
NOAEL
800
10
3
10
1
1
300
2.7
Albee et al. (2006)
Rat
NOAEL
140
10
3
10
1
1
300
0.47
Crofton and Zhao
(1997)
Rat
BMDL
274
10
3
10
1
1
300
0.91
Preferred, due to better dose-response
data, amenable to BMD modeling.
BMR = lOdB absolute change.
Psychomotor effects
Waseem et al. (2001)
Rat
LOAEL
45
1
3
10
3
1
0.45
Changes in locomotor activity; transient,
minimal degree of adversity; no effect
reported in same study for oral exposures
(210 mg/kg/day).
Nunes et al. (2001)
Rat
LOAEL
2,000
10
10
10
3
1
3,000
0.67
t Foot splaying; minimal adversity.
-------
H
tr
V
a-
o
o
c
3
3
fa
~1
fa
»
3*
<
T3
C
>3
o
V
a>
VI
o
3
vf
fa
3
O-
O-
o
v
3
O
f—K
o
o
3
v
C
rHK
CIQ
3
o
O
o"
Table 5-2. Neurological effects in studies suitable for dose-response assessment, and corresponding cRfCs and cRfDs
(continued)
Effect type
Supporting studies
Species
POD
type
PODa
UFSC
UFis
UFh
UF|oae|
UFdb
TTI7 b
' comp
cRfC
(ppm)
cRfD
(mg/kg/day)
Effect; comments
Moser et al. (1995)
Rat
BMDL
248
3
10
10
1
1
300
0.83
t # rears (standing on hindlimbs);
BMR = 1 SD change.
Rat
NOAEL
500
3
10
10
1
1
300
1.7
t Severity score for neuromuscular
changes.
Visual function effects
Blainet al. (1994)
Rabbit
LOAEL
350
10
3
10
10
1
3,000
0.12
POD not adjusted to continuous exposure
because visual effects more closely
associated with administered exposure.
Cognitive effects
Kuligetal. (1987)
Rat
NOAEL
500
1
3
10
1
1
30
17
t time in 2-choice visual discrim. test; test
involves multiple systems but largely visual
so not adjusted to continuous exposure.
Isaacson et al. (1990)
Rat
LOAEL
47
10
10
10
10
1
10,000c
0.0047
Demyelination in hippocampus.
Mood and sleep disorders
Albee et al. (2006)
Rat
NOAEL
140
10
3
10
1
1
300
0.47
Hyperactivity.
Arito et al. (1994)
Rat
LOAEL
12
3
3
10
10
1
1,000
0.012
Changes in wakefulness.
Other neurological effects
Kjellstrand et al.
Rat
LOAEL
300
10
3
10
10
1
3,000
0.10
i regeneration of sciatic nerve.
(1987)
Mouse
LOAEL
150
10
3
10
10
1
3,000
0.050
i regeneration of sciatic nerve.
Gash et al. (2008)
Rat
LOAEL
710
10
10
10
10
1
10,000c
0.071
Degeneration of dopaminergic neurons.
a Adjusted to continuous exposure unless otherwise noted. For inhalation studies, adjustments yield a POD that is a human equivalent concentration as
recommended for a Category 3 gas in EPA (1994) in the absence of PBPK modeling. Same units as cRfC (ppm) or cRfD (mg/kg/day).
b Product of individual uncertainty factors.
-------
Table 5-2. Neurological effects in studies suitable for dose-response assessment, and corresponding cRfCs and cRfDs
(continued)
°EPA's report on the RfC and RfD processes (U.S. EPA, 2002) recommends not deriving reference values with a composite UF of greater than 3,000; however,
composite UFs exceeding 3,000 are considered here because the derivation of the cRfCs and cRfDs is part of a screening process and the subsequent application
of the PBPK model for candidate critical effects will reduce the values of some of the individual UFs.
UFSC = subchronic-to-chronic UF; UF1S = interspecies UF; UFh = human variability UF; UFloael = LOAEL-to-NOAEL UF; UFdb = database UF.
Shaded studies/endpoints were selected as candidate critical effects/studies.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
similar range of 0.4-0.5 ppm (Mhiri et al., 2004; Ruijten et al., 1991). There remains some
uncertainty as to the exposure characterization, as shown by the use of an alternative POD for
Mhiri et al. (2004) based on urinary trichloroacetic acid (TCA) resulting in a fivefold smaller
cRfC. However, the overall confidence in these estimates is relatively high because they are
based on humans exposed under chronic or nearly chronic conditions. Other human studies (e.g.,
Barret et al., 1984), while indicative of hazard, did not have adequate exposure information for
quantitative estimates of an inhalation POD. A cRfD of 0.2 mg/kg/day was developed from the
only oral study demonstrating trigeminal nerve changes, a subchronic study in rats (Barret et al.,
1992). This estimate required multiple extrapolations with a composite uncertainty factor of
10,000.26
For auditory effects, a high confidence cRfC of about 0.7 ppm was developed based on
BMD modeling of data from Crofton and Zhao (1997); and cRfCs developed from two other
auditory studies (Albee et al., 2006; Rebert et al., 1991) were within about fourfold. No oral data
were available for auditory effects. For psychomotor effects, the available human studies (e.g.,
Rasmussen et al., 1983) did not have adequate exposure information for quantitative estimates of
an inhalation POD. However, a relatively high confidence cRfC of 0.5 ppm was developed from
a study in rats (Waseem et al., 2001). Two cRfDs within a narrow range of 0.7-1.7 mg/kg/day
were developed based on two oral studies reporting psychomotor effects (Moser et al., 1995;
Nunes et al., 2001), although varying in degree of confidence.
For the other neurological effects, the estimated cRfCs and cRfDs were more uncertain,
as there were fewer studies available for any particular endpoint, and the PODs from several
studies required more adjustment to arrive at a cRfC or cRfD. However, the endpoints in these
studies also tended to be indicative of more sensitive effects and, therefore, they need to be
considered. The lower cRfCs fall in the range 0.01-0.1 ppm and were based on effects on visual
function in rabbits (Blain et al., 1994), wakefulness in rats (Arito et al., 1994), and regeneration
of the sciatic nerve in mice and rats (Kjellstrand et al., 1987). Of these, altered wakefulness
(Arito et al., 1994) has both the lowest POD and the lowest cRfC. There is relatively high
confidence in this study, as it shows a clear dose-response trend, with effects persisting
postexposure. For the subchronic-to-chronic UF, a value of 3 was used because, even though it
was just a 6-week study, there was no evidence of a greater impact on wakefulness following
6 weeks of exposure than there was following 2 weeks of exposure at the LOAEL, although
26U.S. EPA's report on the RfC and RfD processes (U.S. EPA. 2002c) recommends not deriving reference values
with a composite UF of greater than 3,000; however, composite UFs exceeding 3,000 are considered here because
the derivation of the cRfCs and cRfDs is part of a screening process and the subsequent application of the PBPK
model for candidate critical effects will reduce the values of some of the individual UFs.
This document is a draft for review purposes only and does not constitute Agency policy.
15 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
there was an effect of repeated exposure on the postexposure period impacts of higher exposure
levels. The cRfDs, in the range 0.005-0.07, were based on demyelination in the hippocampus
(Isaacson et al., 1990) and degeneration of dopaminergic neurons (Gash et al., 2008), both in
rats. In both these cases, adjusting for study design characteristics led to a composite uncertainty
factor of 10,000,27 so the confidence in these cRfDs is lower. However, no other studies of these
effects are available.
In summary, although there is high confidence both in the hazard and in the cRfCs and
cRfDs for trigeminal nerve, auditory, or psychomotor effects, the available data suggest that the
more sensitive indicators of TCE neurotoxicity are changes in wakefulness, regeneration of the
sciatic nerve, demyelination in the hippocampus and degeneration of dopaminergic neurons.
Therefore, these more sensitive effects are considered the candidate critical effects for
neurotoxicity, albeit with more uncertainty in the corresponding cRfCs and cRfDs. Of these
more sensitive effects, for the reasons discussed above, there is greater confidence in the changes
in wakefulness reported by Arito et al. (1994). In addition, trigeminal nerve effects are
considered a candidate critical effect because this is the only type of neurological effect for
which human data are available, and the POD for this effect is similar to that from the most
sensitive rodent study (Arito et al., 1994, for changes in wakefulness). Between the two human
studies of trigeminal nerve effects, Ruijten et al. (1991) is preferred for deriving noncancer
reference values because its exposure characterization is considered more reliable.
5.1.2.1.2. Candidate Critical Kidney Effects on the Basis of Applied Dose
As summarized in Section 4.11.1.2, multiple lines of evidence support TCE
nephrotoxicity in the form of tubular toxicity, mediated predominantly through the glutathione
(GSH) conjugation product dichlorovinyl cysteine (DCVC). Available human studies, while
providing evidence of hazard, did not have adequate exposure information for quantitative
estimates of PODs. Several studies in rodents, some of chronic duration, have shown
histological changes, nephropathy, or increased kidney/body weight ratios. Studies with
numerical dose-response information are summarized in Table 5-3, with their corresponding
cRfCs or cRfDs shown in Table 5-4.
27U.S. EPA's report on the RfC and RfD processes (U.S. EPA. 2002c) recommends not deriving reference values
with a composite UF of greater than 3,000; however, composite UFs exceeding 3,000 are considered here because
the derivation of the cRfCs and cRfDs is part of a screening process and the subsequent application of the PBPK
model for candidate critical effects will reduce the values of some of the individual UFs.
This document is a draft for review purposes only and does not constitute Agency policy.
16 DRAFT—DO NOT CITE OR QUOTE
-------
H
tr
V
a-
o
o
c
3
3
fa
~1
fa
tf
<
T3
C
>3
o
V
VI
O
3
vf
fa
3
O-
O-
o
v
3
O
r-K
o
o
3
v
C
r-K
OQ
3
o
O
o"
Table 5-3. Summary of studies of kidney, liver, and body weight effects suitable for dose-response assessment
Effect type
Study reference
Species, strain (if
appl.), sex, number
used for dose-response
assessment
Exposure(s) used for
dose-response
assessment
Endpoint(s) used for dose-response
assessment
Section 4
Section/Table
Histological changes in kidney
Section 4.4.4
Maltoni (1986)
Rat, Sprague-Dawley,
M, 116-124/group
Inhalation: 0, 100, 300,
600 ppm, 7 h/d, 5 d/wk,
104 wk exposure,
observed for lifespan
Meganucleocytosis
Table 4-49,
Table 4-43
NTP (1990)
Rat, F344/A'. male and
female, 48-50/group
Oral: 0, 500, 1,000
mg/kg/day, 5 d/wk, 103
wk
Cytomegaly and karyomegaly
Table 4-45,
Table 4-44
NCI (1976)
Mouse, B6C3F1,
female, 20-50/group
Oral: 0, 869, 1,739
mg/kg/day, 5 d/wk,
TWA during exposure
period (78 wk),
observed for 90 wk
Toxic nephrosis
Table 4-46,
Table 4-44
NTP (1988)
Rat, Marshall, F,
44-50/group
Oral: 0, 500, 1,000
mg/kg/day, 5 d/wk, 104
wk
Toxic nephropathy
Table 4-47,
Table 4-44
| kidney/body weight ratio
Section 4.4.4
Kjellstrand et al., (1983a)
Mouse, NMRI, M,
10-20/group
Inhalation: 0 (air), 37,
75, 150, 225, 300, 450,
900, 1,800, 3,600 ppm;
continuous and
intermittent exposures
for 30-120 d
Increased kidney/body weight ratio
Table 4-43
Woolhiser et al. (2006)
Rat, Sprague-Dawley,
F, 16/group
Inhalation: 0, 100, 300,
and 1,000 ppm, 6 hr/d,
5 d/wk, for 4 wk
Increased kidney/body weight ratio
Table 4-43
-------
H
tr
V
a-
o
o
c
3
3
fa
~1
fa
tf
<
T3
C
>3
o
V
VI
O
3
vf
fa
3
O-
O-
o
v
3
O
r-K
o
o
3
v
C
r-K
OQ
3
o
O
Table 5-3. Summary of studies of kidney, liver, and body weight effects suitable for dose-response
assessment (continued)
Effect type
Study reference
Species, strain (if
appl.), sex, number
used for dose-response
assessment
Exposure(s) used for
dose-response
assessment
Endpoint(s) used for dose-response
assessment
Section 4
Section/Table
| liver/body weight ratio
Section 4.5.4.1
Kjellstrand et al. (1983a)
Mouse, NMRI, M,
10-20/group
Inhalation: 0 (air), 37,
75, 150, 225, 300, 450,
900, 1,800, 3,600 ppm;
continuous and
intermittent exposures
for 30-120 d
Increased liver/body weight ratio
Table 4-58
Woolhiser et al. (2006)
Rat, Sprague-Dawley,
F, 16/group
Inhalation: 0, 100, 300,
and 1,000 ppm, 6 hr/d,
5 d/wk, for 4 wk
Increased liver/body weight ratio
Table 4-58
Buben and O'Flaherty (1985)
Mouse, Swiss-Cox,
12-15/group
Oral: 0, 100, 200, 400,
800, 1,600, 2,400, and
3,200 mg/kg/day,
5 d/wk for 6 wk
Increased liver/body weight ratio
Table 4-54
Decreased body weight
NTP (1990)
Mouse, B6C3F1, M,
48-50/group
Oral: 0, 1,000
mg/kg/day, 5 d/wk, 103
wk
Decreased body weight.
NA
NCI (1976)
Rat, Osborn-Mendel, M
and F, 20-50/group
Oral: 0, 549, 1,097
mg/kg/day, 5 d/wk,
TWA during exposure
period (78 wk),
observed at 110 wk
Decreased body weight.
NA
O
-------
H
tr
V
a-
o
o
c
3
3
fa
~1
fa
<
T3
C
>3
o
V
VI
O
3
vf
fa
3
O-
O-
o
v
3
O
f—K
o
o
3
v
C
rHK
CIQ
3
o
O
o"
Table 5-4. Kidney, liver, and body weight effects in studies suitable for dose-response assessment, and
corresponding cRfCs and cRfDs
Effect type
Supporting studies
Species
POD
type
PODa
UFSC
UFis
UFh
UFloael
UFdb
TJF b
* comp
cRfC
(ppm)
cRfD
(mg/kg/day)
Effect; comments
Histological changes in kidney
Maltoni (1986)
Rat
BMDL
40.2
1
3
10
1
1
30
1.3
meganucleocytosis; BMR = 10% extra risk
Rat
BMDL
34
1
10
10
1
1
100
0.34
meganucleocytosis; BMR = 10% extra risk
NTP (1990)
Rat
LOAEL
360
1
10
10
10
1
1,000
0.36
cytomegaly and karyomegaly; considered
minimally adverse, but UFloael =10 due to
high response rate (>98%) at LOAEL; also
in mice, but use NCI (1976) for that
species
NCI (1976)
Mouse
LOAEL
620
1
10
10
30
1
3,000
0.21
toxic nephrosis; UFloael = 30 due to >90%
response at LOAEL for severe effect
NTP (1988)
Rat
BMDL
9.45
1
10
10
1
1
100
0.0945
toxic nephropathy; female Marshall (most
sensitive sex/strain); BMR = 5% extra risk
| kidney/body weight ratio
Kjellstrand et al.
(1983a)
Mouse
BMDL
34.7
1
3
10
1
1
30
1.2
BMR = 10% increase; 30 d, but 120 d a
120 ppm not more severe so UFsc = 1;
results are for males, which were slightly
more sensitive, and yielded better fit to
variance model
Woolliiser et al. (2006)
Rat
BMDL
15.7
1
3
10
1
1
30
0.52
BMR = 10% increase; UFsc = 1 based on
Kjellstrand et al. (1983b) result
| liver/body weight ratio
Kjellstrand et al.
(1983a)
Mouse
BMDL
21.6
1
3
10
1
1
30
0.72
BMR = 10% increase; UFsc = 1 based on
not more severe at 4 months
Woolliiser et al. (2006)
Rat
BMDL
25.2
1
3
10
1
1
30
0.84
BMR = 10% increase; UFsc = 1 based on
Kjellstrand et al. (1983b) result
Buben and O'Flaherty
(1985)
Mouse
BMDL
81.5
1
10
10
1
1
100
0.82
BMR = 10% increase; UFsc = 1 based on
Kjellstrand et al. (1983b) result
-------
Table 5-4. Kidney, liver, and body weight effects in studies suitable for dose-response assessment, and corresponding cRfCs and
cRfDs (continued)
Effect type
Supporting studies
Species
POD
type
PODa
UFSC
UFis
UFh
UFloael
UFdb
UF b
comp
cRfC
(ppm)
cRIi)
(mg/kg/day)
Effect; comments
Histological changes in kidney
NTP (1990)
Mouse
LOAEL
710
1
10
10
10
1
1,000
0.71
NCI (1976)
Rat
LOAEL
360
1
10
10
10
1
1,000
0.36
Reflects several, but not all, strains/sexes.
a Adjusted to continuous exposure unless otherwise noted. For inhalation studies, adjustments yield a POD that is a human equivalent concentration as
recommended for a Category 3 gas in U.S. EPA (1994b) in the absence of PBPK modeling. Same units as cRfC (ppm) or cRfD (mg/kg/day).
b Product of individual uncertainty factors.
UFSC = subchronic-to-chronic UF; UF1S = interspecies UF; UFh = human variability UF; UFloael = LOAEL-to-NOAEL UF; UFdb = database UF.
Shaded studies/endpoints were selected as candidate critical effects/studies.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
The cRfCs developed from three suitable inhalation studies, one reporting
meganucleocytosis in rats (Maltoni et al., 1986) and two others reporting increased kidney
weights in mice (Kjellstrand et al., 1983a) and rats (Woolhiser et al., 2006),28 are in a narrow
range of 0.5-1.3 ppm. All three utilized BMD modeling and, thus, take into account statistical
limitations of the Woolhiser et al. (2006) and Kjellstrand et al. (1983a) studies, such as
variability in responses or the use of low numbers of animals in the experiment. The response
used for kidney weight increases was the organ weight as a percentage of body weight, to
account for any commensurate decreases in body weight, although the results did not generally
differ much when absolute weights were used instead. Although the two studies reporting
kidney weight changes were subchronic, longer-term experiments by Kjellstrand et al. (1983a)
did not report increased severity, so no subchronic-to-chronic uncertainty factor was used in the
derivation of the cRfC. The high response level of 73% at the lowest dose for
meganucleocytosis in the chronic study of Maltoni et al. (1986) implies more uncertainty in the
low-dose extrapolation. However, it is the only inhalation study that includes histopathological
analysis, and it uses relatively high numbers of animals per dose group.
The suitable oral studies give cRfDs within a narrow range of 0.09-0.4 mg/kg/day, as
shown in Table 5-4, although the degree of confidence in the cRfDs varies considerably. For
cRfDs based on National Toxicology Program (NTP, 1990) and National Cancer Institute (NCI,
1976) chronic studies in rodents, extremely high response rates of >90% precluded BMD
modeling. An UF of 10 was applied for extrapolation from a LOAEL to a NOAEL in the NTP
(1990) study because the effect (cytomegaly and karyomegaly), although minimally adverse, was
observed at such a high incidence. An UF of 30 was applied for extrapolation from a LOAEL to
a NOAEL in the NCI (1976) study because of the high incidence of a clearly adverse effect
(toxic nephrosis). There is more confidence in the cRfDs based on meganucleocytosis reported
in Maltoni et al. (1986) and toxic nephropathy NTP (1988), as BMD modeling was used to
estimate BMDLs. Because these two oral studies measured somewhat different endpoints, but
both were sensitive markers of nephrotoxic responses, they were considered to have similarly
strong weight from a hazard perspective. For meganucleocytosis, a BMR of 10% extra risk was
selected because the effect was considered to be minimally adverse. For toxic nephropathy, a
BMR of 5% extra risk was used because toxic nephropathy is a severe toxic effect. This BMR
required substantial extrapolation below the observed responses (about 60%); however, the
response level seemed warranted for this type of effect and the ratio of the BMD to the BMDL
28Woolhiser et al. (20061 is an Organisation for Economic Co-operation and Development guideline
immunotoxicity study performed by the Dow Chemical Company, certified by Dow as conforming to Good
Laboratory Practices as published by the U.S. EPA for the Toxic Substances Control Act.
This document is a draft for review purposes only and does not constitute Agency policy.
21 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
was not large (1.56). Thus, from a dose-response extrapolation perspective, there is more
confidence in Maltoni et al. (1986). However, the effect observed in NTP (1988) is more severe
and therefore also merits consideration.
In summary, there is high confidence in the hazard and moderate confidence in the cRfCs
and cRfDs for histopathological and weight changes in the kidney. These effects are considered
to be candidate critical effects for several reasons. First, they appear to be the most sensitive
indicators of toxicity that are available for the kidney. In addition, as discussed in Section 3.5,
some pharmacokinetic data indicate substantially more production of GSH-conjugates thought to
mediate TCE kidney effects in humans relative to rats and mice, although there is uncertainty in
these data due to possible analytic errors. As discussed above, several studies are considered
reliable for developing cRfCs and cRfDs for these endpoints. For histopathological changes, in
general, the most sensitive were selected as candidate critical studies. These were the only
available inhalation study (Maltoni et al., 1986), the Maltoni et al. (1986) and NTP (1988) oral
studies in rats, and the NCI (1976) oral study in mice. For oral studies in rats, Maltoni et al.
(1986) was considered in addition to NTP (1988), despite its having a higher cRfD, because of
the much greater degree of low-dose extrapolation necessary for NTP (1988) and the excessive
mortality present in that study. While the NCI (1976) study has even greater uncertainty, as
discussed above, with a high response incidence at the POD that necessitates greater low-dose
extrapolation, it is included to add a second species to the set of candidate critical effects. For
kidney weight changes, both available studies were chosen as candidate critical studies.
5.1.2.1.3. Candidate Critical Liver Effects on the Basis of Applied Dose
As summarized in Section 4.11.1.3, while there is only limited epidemiologic evidence of
TCE hepatotoxicity, TCE clearly leads to liver toxicity in laboratory animals, likely through its
oxidative metabolites. Available human studies contribute to the overall weight of evidence of
hazard, but did not have adequate exposure information for quantitative estimates of PODs. In
rodent studies, TCE causes a wide array of hepatotoxic endpoints: increased liver weight, small
transient increases in DNA synthesis, changes in ploidy, cytomegaly, increased nuclear size, and
proliferation of peroxisomes. Increased liver weight (hepatomegaly, or specifically increased
liver/body weight ratio) has been the most studied endpoint across a range of studies in both
sexes of rats and mice, with a variety of exposure routes and durations. Hepatomegaly was
selected as the critical liver effect for multiple reasons. First, it has been consistently reported in
multiple studies in rats and mice following both inhalation and oral routes of exposure. In
addition, it appears to accompany the other hepatic effects at the doses tested, and hence
This document is a draft for review purposes only and does not constitute Agency policy.
22 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
constitutes a hepatotoxicity marker of similar sensitivity to the other effects. Finally, in several
studies, there are good dose-response data for BMD modeling.
As shown in Table 5-4, cRfCs for hepatomegaly developed from the two most suitable
subchronic inhalation studies (Kjellstrand et al., 1983a; Woolhiser et al., 2006), while in different
species (rats and mice, respectively), are both based on similar PODs derived from BMD
modeling, have the same composite uncertainty factor of 30, and result in similar cRfC estimates
of about 0.8 ppm. The cRfD for hepatomegaly developed from the oral study of Buben and
O'Flaherty (1985) in mice also was based on a POD derived from BMD modeling and resulted
in a cRfD estimate of 0.8 mg/kg/day. Among the studies reporting liver weight changes
(reviewed in Section 4.5 and Appendix E), this study had by far the most extensive
dose-response data. The response used in each case was the liver weight as a percentage of body
weight, to account for any commensurate decreases in body weight, although the results did not
generally differ much when absolute weights were used instead.
There is high confidence in all these candidate reference values. BMD modeling takes
into account statistical limitations such as variability in response or low numbers of animals and
standardizes the response rate at the POD. Although the studies were subchronic, hepatomegaly
occurs rapidly with TCE exposure, and the degree of hepatomegaly does not increase with
chronic exposure (Kjellstrand et al., 1983a), so no subchronic-to-chronic uncertainty factor was
used.
In summary, there is high confidence both in the hazard and the cRfCs and cRfDs for
hepatomegaly. Hepatomegaly also appears to be the most sensitive indicator of toxicity that is
available for the liver and is therefore considered a candidate critical effect. As discussed above,
several studies are considered reliable for developing cRfCs and cRfDs for this endpoint, and,
since they all indicated similar sensitivity but represented different species and/or routes of
exposure, were all considered candidate critical studies.
5.1.2.1.4. Candidate Critical Body Weight Effects on the Basis of Applied Dose
The chronic oral bioassays NCI (1976) and NTP (1990) reported decreased body weight
with TCE exposure, as shown in Table 5-4. However, the lowest doses in these studies were
quite high, even on an adjusted basis (see PODs in Table 5-4). These were not considered
critical effects because they are not likely to be the most sensitive noncancer endpoints, and were
not considered candidate critical effects.
This document is a draft for review purposes only and does not constitute Agency policy.
23 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
5.1.2.1.5. Candidate Critical Immunological Effects on the Basis of Applied Dose
As summarized in Section 4.11.1.4, the human and experimental animal studies of TCE
and immune-related effects provide strong evidence for a role of TCE in autoimmune disease
and in a specific type of generalized hypersensitivity syndrome, while there are fewer data
pertaining to immunosuppressive effects. Available human studies, while providing evidence of
hazard, did not have adequate exposure information for quantitative estimates of PODs. Several
studies in rodents were available on autoimmune and immunosuppressive effects that were
adequate for deriving cRfCs and cRfDs. Studies with numerical dose-response information are
summarized in Table 5-5, with their corresponding cRfCs or cRfDs summarized in Table 5-6.
For decreased thymus weights, a cRfD from the only suitable study (Keil et al., 2009) is
0.00035 mg/kg/day based on results from nonautoimmune-prone B6C3F1 mice, with a
composite uncertainty factor of 1,000 for a POD that is a LOAEL (the dose-response relationship
is sufficiently supralinear that attempts at BMD modeling did not result in adequate fits to these
data). Thymus weights were not significantly affected in autoimmune prone mice in the same
study, consistent with the results reported by Kaneko et al. (2000) in autoimmune-prone mice. In
addition, Keil et al. (2009) and Peden-Adams et al. (2008) reported that for several
immunotoxicity endpoints associated with TCE, the autoimmune-prone strain appeared to be less
sensitive than the nonautoimmune prone B6C3F1 strain. In rats, Woolhiser et al. (2006) reported
no significant change in thymus weights in the Sprague-Dawley (S-D) strain. These data are
consistent with normal mice being sensitive to this effect as compared to autoimmune-prone
mice or S-D rats, so the results of Keil et al. (2009) are not necessarily discordant with the other
studies.
For autoimmune effects, the cRfC from the only suitable inhalation study (Kaneko et al.,
2000) is 0.07 ppm. This study reported changes in immunoreactive organs (i.e., liver and spleen)
in autoimmune-prone mice. BMD modeling was not feasible, so a LOAEL was used as the
POD. The standard value of 10 was used for the LOAEL-to-NOAEL UF because the
inflammation was reported to include sporadic necrosis in the hepatic lobules at the LOAEL, so
this was considered an adverse effect. A value of 3 was used for the human (intraspecies)
variability UF because the effect was induced in autoimmune-prone mice, a sensitive mouse
strain for such an effect. The cRfDs from the oral studies (Cai et al., 2008; Griffin et al., 2000;
Keil et al., 2009) spanned about a 100-fold range from 0.004-0.5 mg/kg/day. Each of the studies
used different markers for autoimmune effects, which may explain the over 100-fold range of
PODs (0.4-60 mg/kg/day). The most sensitive endpoint, reported by Keil et al. (2009), was
increases in anti-dsDNA and anti-ssDNA antibodies, early markers for systemic lupus
erythematosus (SLE), in B6C3Flmice exposed to the lowest tested dose of 0.35 mg/kg/day,
This document is a draft for review purposes only and does not constitute Agency policy.
24 DRAFT—DO NOT CITE OR QUOTE
-------
1 yielding a cRfD of 0.004 mg/kg/day. Therefore, the results of Keil et al. (2009) are not
2 discordant with the higher PODs and cRfDs derived from the other oral studies that examined
3 more frank autoimmune effects.
This document is a draft for review purposes only and does not constitute Agency policy.
25 DRAFT—DO NOT CITE OR QUOTE
-------
H
tr
V
a-
o
o
c
3
3
fa
~1
fa
tf
3*
<
T3
C
>3
o
V
a>
VI
o
3
vf
fa
3
O-
O-
o
v
3
O
r-K
o
o
3
v
C
r-K
OQ
3
o
O
o"
Table 5-5. Summary of studies of immunological effects suitable for dose-response assessment
Effect type
Study reference
Species, strain (if
appl.), sex, number
used for dose-response
assessment
Exposure(s) used for
dose-response
assessment
Endpoint(s) used for dose-response
assessment
Section 4
Section/Table
I thymus weight
Section 4.6.2.3
Keil et al. (2009)
Mouse, B6C3F1,
Female, 10/group
Oral: 0, 1,400, or
14,000 ppb TCE (0,
0.35, or 3.5 mg/kg/day),
27 wk
Decreased thymus weights; decrease
in thymus cellularity
Table 4-78
Autoimmunity
Section 4.6.2.3
Kaneko et al, (2000)
5/group
Inhalation: 0, 500,
1,000, or 2,000 ppm
TCE, 4 h/d, 6 d/wk, 8
wk
Liver inflammation, splenomegaly and
hyperplasia of lymphatic follicles
Table 4-78
Keil et al. (2009)
Mouse, B6C3F1,
Female, 10/group
Oral: 0, 1,400, or
14,000 ppb TCE (0,
0.35, or 3.5 mg/kg/day),
27 wk
Increased anti-dsDNA and
anti-ssDNA antibodies
Table 4-78
Griffin et al. (2000)
Mouse, MRL +/+,
Female, 8/group
Oral: 0,21, 100, or 400
mg/kg/day, 32 wk
Various signs of autoimmune hepatitis
(serology, ex vivo assays of cultured
splenocytes, clinical and
histopathologic findings)
Table 4-78
Cai et al. (2008)
Mouse, MRL +/+,
Female, 5/group
Oral: 0 or 60
mg/kg/day, 48 wk
Hepatic necrosis; hepatocyte
proliferation; leukocyte infiltrate in the
liver, lungs, and kidneys;
Table 4-78
-------
Table 5-5. Summary of studies of immunological effects suitable for dose-response assessment
(continued)
Effect type
Study reference
Species, strain (if
appl.), sex, number
used for dose-response
assessment
Exposure(s) used for
dose-response
assessment
Endpoint(s) used for dose-response
assessment
Section 4
Section/Table
Immunosuppression
Section 4.6.2.1
Woolhiser et al. (2006)
Rat, Sprague-Dawley,
female, 16/group
Inhalation: 0, 100, 300,
or 1,000 ppm, 6 h/d, 5
d/wk, 4-wk
Decreased plaque-forming cell assay
response.
Table 4-76
Sanders et al. (1982b)
Mouse, CD-I, Female,
7-25/group
Oral: 0,0.1, 1.0, 2.5, or
5.0 mg/mL (0, 18, 217,
393, or 660 mg/kg/day,
from Tucker et al.,
1982), 4 or 6 mo
Decreased humoral immunity,
cell-mediated immunity, and bone
marrow stem cell colonization.
Table 4-76
-------
H
tr
V
a-
o
o
c
3
3
fa
~1
fa
»
3*
<
T3
C
>3
o
V
a>
VI
o
3
vf
fa
3
O-
O-
o
v
3
O
f—K
o
o
3
v
C
rHK
CIQ
3
o
O
o"
Table 5-6. Immunological effects in studies suitable for dose-response assessment, and corresponding cRfCs
and cRfDs
Effect type
Supporting studies
Species
POD
type
PODa
UFSC
UFis
UFh
UF|oae|
UFdb
TTI7 b
' comp
cRfC
(ppm)
cRIi)
(mg/kg/day)
Effect; comments
J. thymus weight
Keil et al. (2009)
Mouse
LOAEL
0.35
1
10
10
10
1
1,000
0.00035
i thymus weight; corresponding decrease in total
thymic cellularity reported at 10 x higher dose
Autoimmunity
Kaneko et al.. (2000)
Mouse
(MRL-lpr/
Ipr)
LOAEL
70
10
3
3
10
1
1,000
0.070
Changes in immunoreactive oigans—liver (incl.
sporadic necrosis in hepatic lobules), spleen
UFh = 3 due to autoimmune-prone mouse
Keil et al. (2009)
Mouse
LOAEL
0.35
1
10
10
1
1
100
0.0035
t anti-dsDNA and anti-ssDNA Abs (early markers
for SLE) (B6C3F1 mouse); UFloael = 1 due to
early marker
Griffin et al. (2000)
Mouse
(MRL+/+
)
BMDL
13.4
1
10
3
1
1
30
0.45
Various signs of autoimmune hepatitis;
BMR = 10% extra risk for > minimal effects
Cai et al. (2008)
Mouse
(MRL+/+
)
LOAEL
60
1
10
3
10
1
300
0.20
Inflammation in liver, kidney, lungs, and pancreas,
which may lead to SLE-like disease; UFh= 3 due
to autoimmune-prone mouse; UFloael = 10 since
some hepatic necrosis
Immunosuppression
Woolhiser et al. (2006)
Rat
BMDL
31.2
10
3
10
1
1
300
0.10
IPFC response; BMR = 1 SD change
Sanders et al. (1982b)
Mouse
NOAF.L
190
1
10
10
1
1
100
1.9
i humoral response to sRBC; largely transient
during exposure
Mouse
LOAEL
18
1
10
10
3
1
300
0.060
i stern cell bone marrow recolonization
(sustained); females more sensitive
Mouse
LOAEL
18
1
10
10
3
1
300
0.060
I cell-mediated response to sRBC (largely transient
during exposure); females more sensitive
a Adjusted to continuous exposure unless otherwise noted. For inhalation studies, adjustments yield a POD that is a human equivalent concentration as
recommended for a Category 3 gas in U.S. EPA (1994b) in the absence of PBPK modeling. Same units as cRfC (ppm) or cRfD (mg/kg/day).
b Product of individual uncertainty factors.
UFSC = subchronic-to-chronic UF; UF1S = interspecies UF; UFh = human variability UF; UFloael = LOAEL-to-NOAEL UF; UFdb = database UF.
Shaded studies/endpoints were selected as candidate critical effects/studies.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
For immunosuppressive effects, the only suitable inhalation study (Woolhiser et al.,
2006) gave a cRfC of 0.08 ppm. The cRfDs from the only suitable oral study (Sanders et al.,
1982b) ranged from 0.06 mg/kg/day to 2 mg/kg/day, based on different markers for
immunosuppression. Woolhiser et al. (2006) reported decreased plaque-forming cell (PFC)
response in rats. Data from Woolhiser et al. (2006) were amenable to BMD modeling, but there
is notable uncertainty in the modeling. First, it is unclear what should constitute the cut-point for
characterizing the change as minimally biologically significant, so a BMR of 1 control SD
change was used. In addition, the dose-response relationship is supralinear, and the highest
exposure group was dropped to improve the fit to the low-dose data points. Nonetheless, the
uncertainty in the BMD modeling is no greater than the uncertainty inherent in the use of a
LOAEL or NOAEL. The more sensitive endpoints reported by Sanders et al. (1982b), both of
which were in female mice exposed to a LOAEL of 18 mg/kg/day TCE in drinking water for
4 months, were decreased cell-mediated response to sheep red blood cells (sRBC) and decreased
stem cell bone recolonization, a sign of impaired bone marrow function. The cRfD based on
these endpoints is 0.06 mg/kg/day, with a LOAEL-to-NOAEL UF of 3 because, although the
immunosuppressive effects may not be adverse in and of themselves, multiple effects were
observed suggesting potentially less resilience to an insult requiring an immunological response.
In summary, there is high qualitative confidence for TCE immunotoxicity and moderate
confidence in the cRfCs and cRfDs that can be derived from the available studies. Decreased
thymus weight reported at relatively low exposures in nonautoimmune-prone mice is a clear
indicator of immunotoxicity (Keil et al., 2009), and is therefore considered a candidate critical
effect. A number of studies have also reported changes in markers of immunotoxicity at
relatively low exposures. Therefore, among markers for autoimmune effects, the more sensitive
measures of autoimmune changes in liver and spleen (Kaneko et al., 2000)and increased
anti-dsDNA and anti-ssDNA antibodies (Keil et al., 2009) are considered the candidate critical
effects. Similarly, for markers of immunosuppression, the more sensitive measures of decreased
PFC response (Woolhiser et al., 2006), decreased stem cell bone marrow recolonization, and
decreased cell-mediated response to sRBC (both from Sanders et al., 1982b) are considered the
candidate critical effects.
5.1.2.1.6. Candidate Critical Respiratory Tract Effects on the Basis of Applied Dose
As summarized in Section 4.11.1.5, available data are suggestive of TCE causing
respiratory tract toxicity, based primarily on short-term studies in mice and rats. However, these
studies are generally at high inhalation exposures and over durations of less than 2 weeks. Thus,
This document is a draft for review purposes only and does not constitute Agency policy.
29 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
these were not considered critical effects because such data are not necessarily indicators of
longer-term effects at lower exposure and are not likely to be the most sensitive noncancer
endpoints for chronic exposures. Therefore, cRfCs and cRfDs were not developed for them.
5.1.2.1.7. Candidate Critical Reproductive Effects on the Basis of Applied Dose
As summarized in Section 4.11.1.6, both human and experimental animal studies have
associated TCE exposure with adverse reproductive effects. The strongest evidence of hazard is
for effects on sperm and male reproductive outcomes, with evidence from multiple human
studies and several experimental animal studies. There is also substantial evidence for effects on
the male reproductive tract and male serum hormone levels, as well as evidence for effects on
male reproductive behavior. There are fewer data and more limited support for effects on female
reproduction. Studies with numerical dose-response information are summarized in Table 5-7,
with their corresponding cRfCs or cRfDs summarized in Table 5-8.
5.1.2.1.8. Male reproductive effects (effects on sperm and reproductive tract)
A number of available studies have reported functional and structural changes in sperm
and male reproductive organs and effects on male reproductive outcomes following TCE
exposure (see Table 5-8). A cRfC of 0.014 ppm was derived based on hyperzoospermia reported
in the available human study (Chia et al., 1996), but there is substantial uncertainty in this
estimate due to multiple issues.29 Among the rodent inhalation studies, the cRfC of 0.2 ppm
based on increased abnormal sperm in the mouse reported by Land et al. (1981) is considered
relatively reliable because it is based on BMD modeling rather than a LOAEL or NOAEL.
However, increased sperm abnormalities do not appear to be the most sensitive effect, as
29Mean exposure estimates for the exposure groups were limited because they were defined in terms of ranges and
because they were based on mean urinary TCA (mg/g creatinine). There is substantial uncertainty in the conversion
of urinary TCA to TCE exposure level (see discussion of Mhiri et al. (20041. for neurotoxicity, above). In addition,
there was uncertainty about the adversity of the effect being measured. While rodent evidence supports effects of
TCE on sperm, and hyperzoospermia has reportedly been associated with infertility, the adversity of the
hyperzoospermia (i.e., high sperm density) outcome measured in the Chia et al. (1996) study is unclear.
Furthermore, the cut-point used to define hyperzoospermia in this study (i.e., >120 million sperm per mL ejaculate)
is lower than some other reported cut-points, such as 200 and 250 million sperm/mL. A BMR of 10% extra risk was
used on the assumption that this is a minimally adverse effect, but biological significance of this effect level is
unclear.
This document is a draft for review purposes only and does not constitute Agency policy.
3 0 DRAFT—DO NOT CITE OR QUOTE
-------
1 Kumar et al. (2001b; 2000a; 2000b) reported a similar POD to be a LOAEL for reported multiple
2 effects on sperm and testes, as well as altered testicular enzyme markers in the rat. Although
3 there are greater uncertainties associated with the cRfC of 0.02 ppm for this effect and a
4 composite UF of 3,000 was applied to the POD, the uncertainties are generally typical of those
5 encountered in RfC derivations.
This document is a draft for review purposes only and does not constitute Agency policy.
31 DRAFT—DO NOT CITE OR QUOTE
-------
H
tr
V
a-
o
o
c
3
3
fa
~1
fa
tf
3*
<
T3
C
>3
o
V
a>
VI
o
3
vf
fa
3
O-
O-
o
v
3
O
r-K
o
o
3
v
C
r-K
OQ
3
o
O
o"
Table 5-7. Summary of studies of reproductive effects suitable for dose-response assessment
Effect type
Study reference
Species, strain (if
appl.), sex, number
used for dose-response
assessment
Exposure(s) used for
dose-response
assessment
Endpoint(s) used for dose-response
assessment
Section 4
Section/Table
Effects on sperm, male reproductive
outcomes
Sections
4.8.1.1-4.8.1.2
Chiaetal. (1996)
Human, 85 men (37 low
exposure, 48 high
exposure)
Inhalation: Mean
personal air TCE: 29.6
ppm; MeanU-TCA:
22.4 mg/g creatinine
Decreased normal sperm morphology
and hyperzoospermia
Table 4-85
Land et al. (1981)
Mouse, C57BlxC3H
(Fl), M, 5 or 10/group
Inhalation: 0, 200,
2,000 ppm, 4 h/d, 5 d
exposure, 23 d rest
Increased percent morphologically
abnormal epididymal sperm
Table 4-86
Kan et al. (2007)
Mouse, CD-I, male,
4/group
Inhalation: 0 or 1,000
ppm, 6 h/d, 5 d/wk, 4
wk
Abnormalities of the head and tail in
sperm located in the epididymal
lumen.
Table 4-86
Xu et al. (2004)
Mouse, CD-I, male,
4-27/group
Inhalation: 0 or 1,000
ppm, 6 h/d, 5 d/wk, 6
wk
Decreased in vitro sperm-oocyte
binding and in vivo fertilization.
Table 4-86
Kumar et al. (2000b)
Rat, Wistar, male,
12-13/group
Inhalation: 0 or 376
ppm, 4 h/d, 5 d/wk,
2-10 wk exposed, 2-8
wk unexposed.
Multiple sperm effects.
Table 4-86
Kumar et al., (2001b)
Rat, Wistar, male,
6/group
Inhalation: 0 or 376
ppm, 4 h/d, 5 d/wk, 24
wk
Multiple sperm effects, increasing
severity from 12-24 weeks exposure.
Table 4-86
-------
H
tr
V
a-
o
o
c
3
3
fa
~1
fa
tf
3*
<
T3
C
>3
o
V
a>
VI
o
3
vf
fa
3
O-
O-
o
v
3
O
r-K
o
o
3
v
C
r-K
OQ
3
o
O
o"
Table 5-7. Summary of studies of reproductive effects suitable for dose-response assessment (continued)
Effect type
Study reference
Species, strain (if
appl.), sex, number
used for dose-response
assessment
Exposure(s) used for
dose-response
assessment
Endpoint(s) used for dose-response
assessment
Section 4
Section/Table
George et al. (1985)
Mouse, CD-I, male and
female, 20
pairs/treatment group;
40 controls/sex
Oral: 0, 173, 362, or
737 mg/kg/day,
Breeders exposed 1 wk
premating, then for 13
wk; pregnant females
exposed throughout
gestation (i.e., 18 wk
total)
Decreased sperm motility in F0 and F1
males.
Table 4-87
DuTeaux et al., (2004b)
Rat, Sprague-Dawley,
male, 3/group, or
Simonson albino (UC
Davis), male, 3/group
Oral: 0, 143, or 270
mg/kg/day, 14 d
Decreased ability of sperm to fertilize
oocytes collected from untreated
females. Oxidative damage to sperm
membrane in head and mid-piece.
Table 4-87
Male reproductive tract effects
Section 4.8.1.2
Forkert et al. (2002)
Mouse, CD-I, male,
6/group
Inhalation: 0 or 1,000
ppm, 6 h/d,5 d/wk, 19 d
over 4 wk
Sloughing of epididymal epithelial
cells.
Table 4-86
Kan et al. (2007)
Mouse, CD-I, male,
4/group
Inhalation: 0 or 1,000
ppm, 6 h/d,5 d/wk,
1-4 wk
Degeneration and sloughing of
epididymal epithelial cells (more
severe by 4 wks). Vesiculation in
cytoplasm, disintegration of
basolateral cell membranes, sloughing
of epithelial cells.
Table 4-86
Kumar et al. (2000b)
Rat, Wistar, male,
12-13/group
Inhalation: 0 or 376
ppm, 4 h/d, 5 d/wk,
2-10 wk exposed,
2-8 wk unexposed
Smaller, necrotic spermatogenic
tubules.
Table 4-86
-------
H
tr
V
a-
o
o
c
3
3
fa
~1
fa
tf
3*
<
T3
C
>3
o
V
a>
VI
o
3
vf
fa
3
O-
O-
o
v
3
O
r-K
o
o
3
v
C
r-K
OQ
3
o
O
o"
Table 5-7. Summary of studies of reproductive effects suitable for dose-response assessment (continued)
Effect type
Study reference
Species, strain (if
appl.), sex, number
used for dose-response
assessment
Exposure(s) used for
dose-response
assessment
Endpoint(s) used for dose-response
assessment
Section 4
Section/Table
Kumar et al. (2001b)
Rat, Wistar, male,
6/group
Inhalation: 0 or 376
ppm, 4 h/d, 5 d/wk,
24 wk
Decreased testes weight, numbers of
spermatogenic cells and spermatids,
testes atrophy, smaller tubules devoid
of spermatocytes and spermatids,
hyperplastic Leydig cells, altered
testicular enzyme markers. Increasing
severity from 12-24 weeks exposure.
Table 4-86
George et al. (1985)
Mouse, CD-I, male and
female, 20
pairs/treatment group;
40 controls/sex
Oral: 0, 173, 362, or
737 mg/kg/day,
Breeders exposed 1 wk
premating, then for
13 wk; pregnant
females exposed
throughout gestation
(i.e., 18 wk total)
Decreased testes and seminal vesicle
weights in F0.
Table 4-87
George et al. (1986)
Rat, F334, males and
female, 20
pairs/treatment group,
40 controls/sex
Oral: 0, 72, 186, or 389
mg/kg/day (estimated),
Breeders exposed 1 wk
premating, then for
13 wk; pregnant
females exposed
throughout gestation
(i.e., 18 wk total)
Increased testes and epididymis
weights in F0.
Table 4-87
Female maternal weight gain
Section 4.8.3.2
Carney et al. (2006)
Rat, Sprague-Dawley,
females, 27 dams/group
Inhalation: 0, 50, 150,
or 600 ppm, 6 h/d; GD
6-20
Decreased BW gain on GD 6-9
Table 4-96
Schwetz et al. (1975)
Rat, Sprague-Dawley,
female, 20-35/group
Inhalation: 0 or 300
ppm, 7 h/d; GD 6-15
Decreased BW gain on GD 6-9
Table 4-96
-------
H
tr
V
a-
o
o
c
3
3
fa
~1
fa
tf
3*
<
T3
C
>3
o
V
a>
VI
o
3
vf
fa
3
O-
O-
o
v
3
O
r-K
o
o
3
v
C
r-K
OQ
3
o
O
o"
Table 5-7. Summary of studies of reproductive effects suitable for dose-response assessment (continued)
Effect type
Study reference
Species, strain (if
appl.), sex, number
used for dose-response
assessment
Exposure(s) used for
dose-response
assessment
Endpoint(s) used for dose-response
assessment
Section 4
Section/Table
Narotsky et al. (1995)
Rat, Fischer 344,
females, 8-12
dams/group
Oral: 0, 10.1, 32, 101,
320, 475, 633, 844, or
1,125 mg/kg/day, GD
6-15
Decreased BW gain on GD 6-8 and
6-20.
Table 4-98
Mansonetal. (1984)
Rat, Long-Evans,
female, 23-25/group
Oral: 0, 10, 100, or
1,000 mg/kg/day, 6
wks: 2 wk premating,
1 wk mating period, GD
1-21
Decreased gestation B W gain.
Table 4-87
George et al. (1986)
Rat, F334, males and
female, 20
pairs/treatment group,
40 controls/sex
Oral: 0, 72, 186, or 389
mg/kg/day (estimated),
Breeders exposed 1 wk
premating, then for
13 wk; pregnant
females exposed
throughout gestation
(i.e., 18 wk total)
Decreased term and postpartum dam
BW inF0 andFl.
Table 4-87
Female reproductive outcomes
Section 4.8.3.2
Narotsky et al. (1995)
Rat, Fischer 344,
females, 8-12
dams/group
Oral: 0, 10.1, 32, 101,
320, 475, 633, 844, or
1,125 mg/kg/day, GD
6-15
Delayed parturition.
Table 4-98
Reproductive behavior
Section 4.8.1.2
Zenick et al. (1984)
Rat, Long-Evans, male,
10/group
Oral: 0, 10, 100, or
1,000 mg/kg/day,
5 d/wk, 6 wk exposure,
4 wks recovery
Impaired copulatory performance.
Table 4-87
-------
H
tr
V
a-
o
o
c
3
3
fa
~1
fa
tf
3*
<
T3
C
>3
o
V
a>
VI
o
3
vf
fa
3
O-
O-
o
v
3
O
r-K
o
o
X/l
Table 5-7. Summary of studies of reproductive effects suitable for dose-response assessment (continued)
Effect type
Study reference
Species, strain (if
appl.), sex, number
used for dose-response
assessment
Exposure(s) used for
dose-response
assessment
Endpoint(s) used for dose-response
assessment
Section 4
Section/Table
George et al. (1986)
Rat, F334, males and
female, 20
pairs/treatment group,
40 controls/sex
Oral: 0, 72, 186, or 389
mg/kg/day (estimated),
Breeders exposed 1 wk
premating, then for
13 wk; pregnant
females exposed
throughout gestation
(i.e., 18 wk total)
Decreased F0 mating in cross-over
mating trials.
Table 4-87
Reproductive effects from exposure
to both sexes
Section 4.8.1.2
George et al. (1986)
Rat, F334, males and
female, 20
pairs/treatment group,
40 controls/sex
Oral: 0, 72, 186, or 389
mg/kg/day (estimated),
Breeders exposed 1 wk
premating, then for
13 wk; pregnant
females exposed
throughout gestation
(i.e., 18 wk total)
Decreased F0 litters/pair and live F1
pups/litter.
Table 4-87
c
r-K
OQ
3
o
O
O
-------
H
tr
Table 5-8. Reproductive effects in studies suitable for dose-response assessment, and corresponding cRfCs and
cRfDs
Effect type
Supporting studies
Species
POD
type
PODa
UFSC
UFis
UFh
UF|oae|
UFdb
TTI7 b
' comp
cRfC
(ppm)
cRfD
(mg/kg/day)
Effect; comments
Effects on sperm, male reproductive outcomes
Cilia etal. (1996)
Human
BMDL
1.43
10
1
10
1
1
100
0.014
Hyperzoospennia; exposure estimates
based on U-TCA from Ikeda et al.
(1972); BMR = 10% extra risk
Land et al. (1981)
Mouse
BMDL
46.9
10
3
10
1
1
300
0.16
t abnormal sperm; BMR = 0.5 SD
Kan etal. (2007)
Mouse
LOAEL
180
10
3
10
10
1
3,000
0.060
t abnormal sperm; Land et al. (1981)
cRfC preferred due to BMD modeling
Xu et al. (2004)
Mouse
LOAEL
180
10
3
10
10
1
3,000
0.060
I fertilization
Kumar et al. (2001b;
2000b)
Rat
LOAEL
45
10
3
10
10
1
3,000
0.015
Multiple sperm effects, increasing
severity from 12-24 weeks
Rat
LOAEL
45
1
3
10
10
1
300
0.15
Pre- and postimplantation losses; UFsc
= 1 due to exposure covered time
period for sperm development; higher
response for preimplantation losses
George et al. (1985)
Mouse
NOAF.L
362
1
10
10
1
1
100
3.6
i sperm motility
DuTeaux et al., (2004a)
Rat
LOAEL
141
10
10
10
10
1
10,000c
0.014
i ability of sperm to fertilize in vitro
Male reproductive tract effects
Forkert et al. (2002);
Kan etal. (2007)
Mouse
LOAEL
180
10
3
10
10
1
3,000
0.060
Effects on epididymis epithelium
Kumar et al. (2001b;
2000b)
Rat
LOAEL
45
10
3
10
10
1
3,000
0.015
Testes effects, altered testicular
enzyme markers, increasing severity
from 12-24 weeks
George et al. (1985)
Mouse
NOAF.L
362
1
10
10
1
1
100
3.6
i testis/seminal vesicle weights
George et al. (1986)
Rat
NOAF.L
186
1
10
10
1
1
100
1.9
t testis/epididymis weights
-------
H
tr
V
a-
o
o
c
3
3
fa
~1
fa
»
3*
<
T3
C
>3
o
V
a>
VI
o
3
vf
fa
3
O-
O-
o
v
3
O
f—K
o
o
3
v
C
rHK
CIQ
3
o
O
o"
Table 5-8. Reproductive effects in studies suitable for dose-response assessment, and corresponding cRfCs and
cRfDs (continued)
Effect type
Supporting studies
Species
POD
type
PODa
UFSC
UFis
UFh
UF|oae|
UFdb
TTI7 b
' comp
cRfC
(ppm)
cRfD
(mg/kg/day)
Effect; comments
Female maternal weight gain
Carney et al. (2006)
Rat
BMDL
10.5
1
3
10
1
1
30
0.35
I BW gain; BMR = 10% decrease
Schwetz et al. (1975)
Rat
LOAEL
88
1
3
10
10
1
300
0.29
i mat BW; Carney et al. (2006) cRfC
preferred due to BMD modeling
Narotsky et al. (1995)
Rat
BMDL
108
1
10
10
1
1
100
1.1
I BW gain; BMR = 10% decrease
Mansonet al. (1984)
Rat
NOAF.I.
100
1
10
10
1
1
100
1.0
I BW gain; Narotsky et al. (1995)
preferred due to BMD modeling
(different strain)
George et al. (1986)
Rat
NOAF.I.
186
1
10
10
1
1
100
1.9
i postpartum BW; Narotsky et al.
(1995) cRfD preferred due to BMD
modeling
Female reproductive outcomes
Narotsky et al. (1995)
Rat
LOAEL
475
1
10
10
10
1
1,000
0.48
Delayed parturition
Reproductive behavior
Zenick et al. (1984)
Rat
NOAF.I.
100
1
10
10
1
1
100
1.0
i copulatory performance in males
George et al. (1986)
Rat
LOAEL
389
1
10
10
10
1
1,000
0.39
i mating (both sexes exposed)
Reproductive effects from exposure to both sexes
George et al. (1986)
Rat
BMDL
179
1
10
10
1
1
100
1.8
i # litters/pair; BMR = 0.5 SD
Rat
BMDL
152
1
10
10
1
1
100
1.5
I live pups/litter; BMR = 0.5 SD
a Adjusted to continuous exposure unless otherwise noted. For inhalation studies, adjustments yield a POD that is a human equivalent concentration as recommended for a
Category 3 gas in U.S. EPA (1994b) in the absence of PBPK modeling. Same units as cRfC (ppm) or cRfD (mg/kg/day).
b Product of individual UFs.
c EPA's report on the RfC and RfD processes (U.S. EPA, 2002c) recommends not deriving reference values with a composite UF of greater than 3,000; however, composite UFs
exceeding 3,000 are considered here because the derivation of the cRfCs and cRfDs is part of a screening process and the subsequent application of the PBPK model for
candidate critical effects will reduce the values of some of the individual UFs.
UFSC = subchronic-to-chronic UF; UFis = interspecies UF; UFh = human variability UF; UFioaei = LOAEL-to-NOAEL UF; UFdb = database UF.
-------
This document is a draft for review purposes only and does not constitute Agency policy.
on TVT» AT7T MAT /^TTF1 /~\T> AT TATT
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Standard values of 3, 10, and 10 were used for the interspecies UF, the human variability
UF, and the LOAEL-to-NOAEL UF, respectively. In addition, although the study would have
qualified as a chronic exposure study based on its duration of 24 weeks (i.e., >10% of lifetime),
statistically significant decreases in testicular weight and in sperm count and motility were
already observed from subchronic exposure (12 weeks) to the same TCE exposure concentration
and these effects became more severe after 24 weeks of exposure. Moreover, several testicular
enzyme markers associated with spermatogenesis and germ cell maturation had significantly
altered activities after 12 weeks of exposure, with more severe alterations at 24 weeks, and
histological changes were also observed in the testes at 12 weeks, with the testes being severely
deteriorated by 24 weeks. Thus, since the single exposure level used was already a LOAEL from
subchronic exposure, and the testes were even more seriously affected by longer exposures, a
subchronic-to-chronic UF of 10 was applied.30 Note that for the cRfC derived for pre and
postimplantation losses reported by Kumar et al. (2000b), the subchronic-to-chronic UF was not
applied because the exposure covered the time period for sperm development. This cRfC was
0.2 ppm, similar to that derived from Land et al. (1981) based on BMD modeling of increases in
abnormal sperm.
At a higher inhalation POD, Xu et al. (2004) reported decreased fertilization following
exposure in male mice, and Forkert et al. (2002) and Kan et al. (2007) reported effects on the
epididymal epithelium in male mice. Kan et al. (2007) reported degenerative effects on the
epididymis as early as 1 week into exposure that became more severe at 4 weeks of exposure
when the study ended; increases in abnormal sperm were also observed. As with the cRfC
developed from the Kumar et al. (2001b; 2000a; 2000b), a composite UF of 3,000 was applied to
these data, but the uncertainties are again typical of those encountered in RfC derivations.
Standard values of 3 for the interspecies UF, 10 for the human variability UF, 10 for the
LOAEL-to-NOAEL UF, and 10 for the subchronic-to-chronic UF were applied to each of the
study PODs.
Among the oral studies, cRfDs derived for decreased sperm motility and changes in
reproductive organ weights in rodents reported by George et al. (George et al., 1985, 1986) were
relatively high (2-4 mg/kg/day), and these effects were not considered candidate critical effects.
The remaining available oral study of male reproductive effects is DuTeaux et al. (2004a), which
reported decreased ability of sperm from TCE-exposed rats to fertilize eggs in vitro. This effect
30 Alternatively, the value of the LOAEL-to-NOAEL UF could have been increased above 10 to reflect the extreme
severity of the effects at the LOAEL after 24 weeks; however, the comparison of the 12-week and 24-week results
gives such a clear depiction of the progression of the effects, it was more compelling to frame the issue as a
subchronic-to-chronic extrapolation issue.
This document is a draft for review purposes only and does not constitute Agency policy.
40 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
occurred in the absence of changes in combined testes/epididymes weight, sperm concentration
or motility, or histological changes in the testes or epididymes. DuTeaux et al. (2004a)
hypothesize that the effect is due to oxidative damage to the sperm. A LOAEL was used as the
POD, and the standard UF values of 10 were used for each of the UFs, i.e., the
subchronic-to-chronic UF (14-day study; substantially less than the 70-day time period for sperm
development), the interspecies UF for oral exposures, the human variability UF, and the
LOAEL-to-NOAEL UF. The resulting composite UF was 10,000,31 and this yielded a cRfD of
0.01 mg/kg/day. The excessive magnitude of the composite UF, however, highlights the
uncertainty in this estimate.
In summary, there is high qualitative confidence for TCE male reproductive tract toxicity
and lower confidence in the cRfCs and cRfDs that can be derived from the available studies.
Relatively high PODs are derived from several studies reporting less sensitive endpoints (George
et al., 1985, 1986; 1981), and correspondingly higher cRfCs and cRfDs suggest that they are not
likely to be critical effects. The studies reporting more sensitive endpoints also tend to have
greater uncertainty. For the human study by Chia et al. (1996), as discussed above, there are
uncertainties in the characterization of exposure and the adversity of the effect measured in the
study. For the Kumar et al. (2001b; 2000a; 2000b), Forkert et al. (2002) and Kan et al. (2007)
studies, the severity of the sperm and testes effects appears to be continuing to increase with
duration even at the end of the study, so it is plausible that a lower exposure for a longer duration
may elicit similar effects. For the DuTeaux et al. (2004a) study, there is also duration- and
low-dose extrapolation uncertainty due to the short duration of the study in comparison to the
time period for sperm development as well as the lack of a NOAEL at the tested doses. Overall,
even though there are limitations in the quantitative assessment, there remains sufficient
evidence to consider these to be candidate critical effects.
5.1.2.1.9. Other reproductive effects
With respect to female reproductive effects, several studies reporting decreased maternal
weight gain were suitable for deriving candidate reference values (see Table 5-8). The cRfCs
from the two inhalation studies (Carney et al., 2006; Schwetz et al., 1975) yielded virtually the
same estimate (0.3-0.4 ppm), although the Carney et al. (2006) result is preferred due to the use
31U.S. EPA's report on the RfC and RfD processes (U.S. EPA. 200201 recommends not deriving reference values
with a composite UF of greater than 3,000; however, composite UFs exceeding 3,000 are considered here because
the derivation of the cRfCs and cRfDs is part of a screening process and the subsequent application of the PBPK
model for candidate critical effects will reduce the values of some of the individual UFs.
This document is a draft for review purposes only and does not constitute Agency policy.
41 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
of BMD modeling, which obviates the need for the 10-fold LOAEL-to-NOAEL UF used for
Schwetz et al. (1975) (the other UFs, with a product of 30, were the same). The cRfDs for this
endpoint from the three oral studies were within twofold of each other (1.1-1.9 mg/kg/day), with
the same composite UFs of 100. The most sensitive estimate of Narotsky et al. (1995) is
preferred due to the use of BMD modeling and the apparent greater sensitivity of the rat strain
used.
With respect to other reproductive effects, the most reliable cRfD estimates of about
2 mg/kg/day, derived from BMD modeling with composite UFs of 100, are based on decreased
litters/pair and decreased live pups/litter in rats reported in the continuous breeding study of
George et al. (1986). Both of these effects were considered severe adverse effects, so a BMR of
a 0.5 control SD shift from the control mean was used. Somewhat lower cRfDs of
0.4-1 mg/kg/day were derived based on delayed parturition in females (Narotsky et al., 1995),
decreased copulatory performance in males (Zenick et al., 1984), and decreased mating for both
exposed males and females in cross-over mating trials (George et al., 1986), all with composite
UFs of 100 or 1,000 depending on whether a LOAEL or NOAEL was used.
In summary, there is moderate confidence both in the hazard and the cRfCs and cRfDs
for reproductive effects other than the male reproductive effects discussed previously. While
there are multiple studies suggesting decreased maternal body weight with TCE exposure, this
systemic change may not be indicative of more sensitive reproductive effects. None of the
estimates developed from other reproductive effects is particularly uncertain or unreliable.
Therefore, delayed parturition (Narotsky et al., 1995) and decreased mating (George et al.,
1986), which yielded the lowest cRfDs, were considered candidate critical effects. These effects
were also included so that candidate critical reproductive effects from oral studies would not
include only that reported by DuTeaux et al. (2004a), from which deriving the cRfD entailed a
higher degree of uncertainty.
5.1.2.1.10. Candidate Critical Developmental Effects on the Basis of Applied Dose
As summarized in Section 4.11.1.7, both human and experimental animal studies have
associated TCE exposure with adverse developmental effects. Weakly suggestive epidemiologic
data and fairly consistent experimental animal data support TCE exposure posing a hazard for
increased prenatal or postnatal mortality and decreased pre or postnatal growth. In addition,
congenital malformations following maternal TCE exposure have been reported in a number of
epidemiologic and experimental animal studies. There is also some support for TCE effects on
neurological and immunological development. Available human studies, while indicative of
This document is a draft for review purposes only and does not constitute Agency policy.
42 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
hazard, did not have adequate exposure information for quantitative estimates of PODs, so only
experimental animal studies are considered here. Studies with numerical dose-response
information are summarized in Table 5-9, with their corresponding cRfCs or cRfDs summarized
in Table 5-10.
For pre and postnatal mortality and growth, a cRfC of 0.06 ppm for resorptions,
decreased fetal weight, and variations in skeletal development indicative of delays in ossification
was developed based on the single available (rat) inhalation study considered (Healy et al., 1982)
and utilizing the composite UF of 300 for an inhalation POD that is a LOAEL. The cRfDs for
pre and postnatal mortality derived from oral studies were within about a 10-fold range of
0.4-5 mg/kg/day, depending on the study and specific endpoint assessed. Of these, the estimate
based on Narotsky et al. (1995) rat data was both the most sensitive and most reliable cRfD. The
dose response for increased full-litter resorptions from this study is based on BMD modeling.
Because of the severe nature of this effect, a BMR of 1% extra risk was used. The ratio of the
resulting BMD to the BMDL was 5.7, which is on the high side, but given the severity of the
effect and the low background response, a judgment was made to use 1% extra risk.
Alternatively, a 10% extra risk could have been used, in which case the POD would have been
considered more analogous to a LOAEL than a NOAEL, and a LOAEL-to-NOAEL UF of 10
would have been applied, ultimately resulting in the same cRfD estimate. The cRfDs for altered
pre and postnatal growth developed from the oral studies ranged about 10-fold from
0.8-8 mg/kg/day, all utilizing the composite UFs for the corresponding type of POD. The cRfDs
for decreased fetal weight, both of which were based on NOAELs, were consistent, being about
twofold apart (George et al., 1985; Narotsky et al., 1995). The cRfD based on postnatal growth
at 21 days, reported in George et al. (1986), was lower and is preferred because it was based on
BMD modeling. A BMR of 5% decrease in weight was used for postnatal growth at 21 days
because decreases in weight gain so early in life were considered similar to effects on fetal
weight.
For congenital defects, there is relatively high confidence in the cRfD for eye defects in
rats reported in Narotsky et al. (1995), derived using a composite UF of 100 for BMD modeling
in a study of duration that encompasses the full window of eye development. However, the most
sensitive developmental effect by far was heart malformations in the rat reported by
Johnson et al. (2003), yielding a cRfD estimate of 0.0002 mg/kg/day, also with a composite UF
of 100. As discussed in detail in Section 4.8 and summarized in Section 4.11.1.7, although this
study has important limitations, the overall weight of evidence supports an effect of TCE on
cardiac development, and this is the only study of heart malformations available for conducting
dose-response analysis. Individual data were kindly provided by Dr. Johnson (personal
This document is a draft for review purposes only and does not constitute Agency policy.
43 DRAFT—DO NOT CITE OR QUOTE
-------
1 communication from Paula Johnson, University of Arizona, to Susan Makris, EPA, 25 August
2 2008), and, for analyses for which the pup was the unit of measure, BMD modeling was done
This document is a draft for review purposes only and does not constitute Agency policy.
44 DRAFT—DO NOT CITE OR QUOTE
-------
Table 5-9. Summary of studies of developmental effects suitable for dose-response assessment
5^
cs
si
3
a
>!
&
o
fS"
sj
§
>!
Grj
O
a
a
a
§•
>!
o
H §
I ^
81
w
2; ^
o
H
o;
a
"T3
H S.
W|
O •
o
c
o
H
W
Effect type
Study reference
Species, strain (if
appl.), sex, number
used for dose-response
assessment
Exposure(s) used for
dose-response
assessment
Endpoint(s) used for dose-response
assessment
Section 4
Section/Table
Pre and postnatal mortality
Section 4.8.1.2 and
4.8.3.2
George et al. (1985)
Mouse, CD-I, male and
female, 20
pairs/treatment group;
40 controls/sex
Oral: 0, 173, 362, or
737 mg/kg/day,
Breeders exposed 1 wk
premating, then for
13 wk; pregnant
females exposed
throughout gestation
(i.e., 18 wk total)
Increase perinatal mortality (PND
0-21)
Table 4-87
Narotsky et al. (1995)
Rat, Fischer 344,
females, 8-12
dams/group
Oral: 0, 10.1, 32, 101,
320, 475, 633, 844, or
1,125 mg/kg/day, GD
6-15
Increased resorptions, prenatal loss,
and postnatal mortality
Table 4-98
Manson et al. (1984)
Rat, Long-Evans,
female, 23-25/group
Oral: 0, 10, 100, or
1,000 mg/kg/day, 6 wk:
2 wk premating, 1 wk
mating period, GD
1-21
Increased neonatal deaths on PND 1,
10, and 14.
Table 4-87
Healy et al. (1982)
Rat, Wistar, females,
31-32 dams/group
Inhalation: 0 or 100
ppm, 4 h/d; GD 8-21
Increased resorptions.
Table 4-96
Pre and postnatal growth
Section 4.8.3.2
Healy etal. (1982)
Rat, Wistar, females,
31-32 dams/group
Inhalation: 0 or 100
ppm, 4 h/d; GD 8-21
Decreased fetal weight, increased
bipartite or absent skeletal ossification
centers
Table 4-96
-------
Table 5-9. Summary of studies of developmental effects suitable for dose-response assessment
(continued)
5^
cs
si
3
a
>!
£
G\
o
fS"
sj
^ 8
>!
Grj
O
a
a
a
§•
>!
o
H §
I ^
81
w
2; ^
o
H
o;
a
"T3
H S.
W|
O •
o
c
o
H
W
Effect type
Study reference
Species, strain (if
appl.), sex, number
used for dose-response
assessment
Exposure(s) used for
dose-response
assessment
Endpoint(s) used for dose-response
assessment
Section 4
Section/Table
Narotsky et al. (1995)
Rat, Fischer 344,
females, 8-12
dams/group
Oral: 0, 10.1, 32, 101,
320, 475, 633, 844, or
1,125 mg/kg/day, GD
6-15
Decreased pup BW on PND 1 and 6.
Table 4-98
George et al. (1985)
Mouse, CD-I, male and
female, 20
pairs/treatment group;
40 controls/sex
Oral: 0, 173, 362, or
737 mg/kg/day,
Breeders exposed 1 wk
premating, then for
13 wk; pregnant
females exposed
throughout gestation
(i.e., 18 wk total)
Decreased live birth weights, PND 4
pup body weights.
Table 4-87
George et al. (1986)
Rat, F334, males and
female, 20
pairs/treatment group,
40 controls/sex
Oral: 0, 72, 186, or 389
mg/kg/day (estimated),
Breeders exposed 1 wk
premating, then for
13 wk; pregnant
females exposed
throughout gestation
(i.e., 18 wk total)
Decreased F1 BW on PND 4-80.
Table 4-87
Congenital defects
Section 4.8.3.2
Narotsky et al. (1995)
Rat, Fischer 344,
females, 8-12
dams/group
Oral: 0, 10.1, 32, 101,
320, 475, 633, 844, or
1,125 mg/kg/day, GD
6-15
Increased incidence of eye defects.
Table 4-98
Johnson et al. (2003)
Rat, Sprague-Dawley,
female, 9-13/group, 55
in control group
Oral: 0, 0.00045, 0.048,
0.218, or 129
mg/kg/day), GD 0-22
Increased percentage of abnormal
hearts; increased percentage of litters
with abnormal hearts.
Table 4-76
-------
Table 5-9. Summary of studies of developmental effects suitable for dose-response assessment
(continued)
5^
cs
si
3
a
>!
£
-J
o
fS"
sj
- O
>!
Grj
O
a
a
a
§•
>!
o
H §
I ^
81
w
2; ^
o
H
o;
a
"T3
H S.
W|
O •
o
c
o
H
W
Effect type
Study reference
Species, strain (if
appl.), sex, number
used for dose-response
assessment
Exposure(s) used for
dose-response
assessment
Endpoint(s) used for dose-response
assessment
Section 4
Section/Table
Developmental neurotoxicity
Sections 4.3.8.2
and 4.8.3.2
George et al. (1986)
Rat, F334, males and
female, 20
pairs/treatment group,
40 controls/sex
Oral: 0, 72, 186, or 389
mg/kg/day (estimated),
Breeders exposed 1 wk
premating, then for
13 wk; pregnant
females exposed
throughout gestation
(i.e., 18 wk total)
Decreased locomotor, as assessed by
increased time required for pups to
cross the first grid in open-field
testing.
Table 4-34 and
Table 4-98
Fredriksson et al. (1993)
Mouse, NMRI, male
pups, 12 pups from 3-4
different litters/group
Oral: 0, 50, or 290
mg/kg/day, PND 10-16
Decreased rearing activity on PND 60.
Table 4-34 and
Table 4-98
Taylor etal. (1985)
Rat, Sprague-Dawley,
females, no.
dams/group not
reported
Oral: 0,312, 625, or
1,250 mg/L (0, 45, 80,
or 140 mg/kg/day
estimated), dams (and
pups) exposed from
14 d prior to mating
until end of lactation
Increased exploratory behavior in 60-
and 90-d old male rats (offspring).
Table 4-34 and
Table 4-98
Isaacson and Taylor (1989)
Rat, Sprague-Dawley,
females, 6 dams/group
Oral: 0, 4.0, or 8.1 mg/d
(0, 15, or 32 mg/kg/day
estimated)3, dams (and
pups) exposed from
14 d prior to mating
until end of lactation.
Decreased myelinated fibers in the
stratum lacunosum-moleculare of
pups; decreased myelin in the
hippocampus.
Table 4-34 and
Table 4-98
Developmental immunotoxicity
Section 4.8.3.2
-------
Table 5-9. Summary of studies of developmental effects suitable for dose-response assessment
(continued)
Effect type
Study reference
Species, strain (if
appl.), sex, number
used for dose-response
assessment
Exposure(s) used for
dose-response
assessment
Endpoint(s) used for dose-response
assessment
Section 4
Section/Table
Peden-Adams et al. (2006)
Mouse, B6C3F1, dams
and
both sexes offspring, 5
dams/group; 5-7
pups/group at 3 wks;
4-5 pups/sex/group at 8
wks
Oral: 0, 1,400, or
14,000 ppb in water (0,
0.37, or 3.7 mg/kg/day
estimated), parental
mice and/or offspring
exposed during mating,
and from GD 0 thru
3 or 8 wk of age
Suppressed PFC responses in males
and in females. Delayed
hypersensitivity response increased at
8 wks of age in females. Splenic cell
population decreased in 3 wk old
pups. Increased thymic T-cells at 8
wks of age. Delayed hypersensitivity
response increased at 8 wks of age in
males and females
Table 4-98
aThe Isaacson and Taylor (1989) and Taylor et al. (1985) studies report different doses despite identical study designs and administered concentrations, both
studies taking TCE degradation into account. Taylor et al. (1985) report total consumption of 646 mg, 1102 mg, and 1991 mg TCE for rats exposed to 312, 625,
and 1250 mg TCE/L drinking water, respectively. Dividing by the 56 days of exposure and the average 250 g per rat for female SD rats of those ages yields
estimated doses of roughly 45, 80, and 140 mg/kg/day, respectively. Isaacson and Taylor (1989) report average doses of TCE of 4.0 and 8.1 mg/day
corresponding to exposures of 312 and 625 mg TCE/L drinking water, respectively. Dividing by the average 250 g per rat yields estimated doses of 16 and 32
mg/kg/day, respectively. Thus the estimated doses for Taylor et al. (1985) are nearly 3-times higher than those for Isaacson and Taylor (1989), for reasons
unknown.
-------
Table 5-10. Developmental effects in studies suitable for dose-response assessment, and corresponding cRfCs
and cRfDs
o-
Cb
s
>!
O5
<
cS'
ss
^ O
&n
>3
o
5S
a
s
o-
o
H §
I
81
2;
o
H
O,
•3
"T3
H S.
ffl.K"
O •
o
c
o
H
W
Effect type
Supporting studies
Species
POD
type
PODa
UFSC
UFis
UFh
UFloael
UFdb
TTI7 b
' comp
cRfC
(ppm)
cRfD
(mg/kg/day)
Effect; comments
Pre and postnatal mortality
George et al. (1985)
Mouse
NOAEL
362
1
10
10
1
1
100
3.6
t perinatal mortality
Narotsky et al. (1995)
Rat
LOAEL
475
1
10
10
10
1
1,000
0.48
Postnatal mortality; Manson et al.
(1984) cRfD preferred for same
endpoint due to NOAEL vs. LOAEL
Mansonet al. (1984)
Rat
NOAEL
100
1
10
10
1
1
100
1.0
t neonatal death
Healy et al. (1982)
Rat
LOAEL
17
1
3
10
10
1
300
0.057
Resorptions
Narotsky et al. (1995)
Rat
BMDL
469
1
10
10
1
1
100
4.7
Prenatal loss; BMR = 1% extra risk
Rat
BMDL
32.2
1
10
10
1
1
100
0.32
Resorptions; BMR = 1% extra risk
Pre and postnatal growth
Healy etal. (1982)
Rat
LOAEL
17
1
3
10
10
1
300
0.057
i fetal weight; skeletal effects
Narotsky et al. (1995)
Rat
NOAEL
844
1
10
10
1
1
100
8.4
i fetal weight
George et al. (1985)
Mouse
NOAEL
362
1
10
10
1
1
100
3.6
i fetal weight
George et al. (1986)
Rat
BMDL
79.7
1
10
10
1
1
100
0.80
i BW at d21; BMR = 5% decrease
Congenital defects
Narotsky et al. (1995)
Rat
BMDL
60
1
10
10
1
1
100
0.60
Eye defects; low BMR (1%), but
severe effect and lowbkgd. rate (<1%)
Johnson et al. (2003)
Rat
BMDL
0.0146
1
10
10
1
1
100
0.00015
Heart malformations (litters);
BMR = 10% extra risk (only -1/10
from each litter affected); highest-dose
group (1,000-fold higher than next
highest) dropped to improve model fit.
-------
Table 5-10. Developmental effects in studies suitable for dose-response assessment, and corresponding cRfCs
and cRfDs (continued)
o-
Cb
3
>!
o
Cr
>s
>s
S:
^3
ss
^3
o
>3
o
a
a
s
o-
>3
a
o
c>
o
s
O
o
2!
H
O,
•3
"T3
H S.
ffl.K"
O •
o
c
o
H
W
Effect type
Supporting studies
Species
POD
type
PODa
UFSC
UFis
UFh
UF|oae|
UFdb
TTI7 b
' comp
cRfC
(ppm)
cRfD
(mg/kg/day)
Effect; comments
Rat
BMDL
0.0207
1
10
10
1
1
100
0.00021
Heart malformations (pups);
BMR = 1% extra risk; preferred due to
accounting for intralitter effects via
nested model and pups being the unit
of measure; highest-dose group
(1,000-fold higher than next highest)
dropped to improve model fit
Developmental neurotoxicity
George et al. (1986)
Rat
BMDL
72.6
1
10
10
1
1
100
0.73
I locomotor activity; BMR = doubling
of traverse time; results from females
(males similar with BMDL = 92)
Fredriksson et al.
(1993)
Mouse
LOAEL
50
3
10
10
10
1
3,000
0.017
I rearing postexposure; pup gavage
dose; No effect at tested doses on
locomotion behavior; UFsc = 3
because exposure only during
PND 10-16
Taylor et al. (1985)
Rat
LOAEL
45
1
10
10
10
1
1,000
0.045
t exploration postexposure; estimated
dam dose; Less sensitive than Isaacson
and Taylor (1989), but included
because exposure is preweaning, so
can utilize PBPK model
Isaacson and Taylor
(1989)
Rat
LOAEL
16
1
10
10
10
1
1,000
0.016
I myelination in hippocampus;
estimated dam dose
Developmental immunotoxicity
-------
Table 5-10. Developmental effects in studies suitable for dose-response assessment, and corresponding cRfCs
and cRfDs (continued)
Effect type
Supporting studies
Species
POD
type
PODa
UFSC
UFis
UFh
UF|oae|
UFdb
TTI7 k
1 comp
cRfC
(ppm)
cRIi)
(mg/kg/day)
Effect; comments
Peden-Adams et al.
(2006)
Mouse
LOAEL
0.37
1
10
10
10
1
1,000
0.00037
I PFC, |DTH; POD is estimated dam
dose (exposure throughout gestation
and lactation + to 3 or 8 wks of age);
UF LOAEL = 10 since | DTH and
also multiple immuno. effects
"Adjusted to continuous exposure unless otherwise noted. For inhalation studies, adjustments yield a POD that is a human equivalent concentration as
recommended for a Category 3 gas in U.S. EPA (1994b) in the absence of PBPK modeling. Same units as cRfC (ppm) or cRfD (mg/kg/day).
bProduct of individual uncertainty factors.
UFSC = subchronic-to-chronic UF; UF1S = interspecies UF; UFh = human variability UF; UFloael = LOAEL-to-NOAEL UF; UFdb = database UF.
Shaded studies/endpoints were selected as candidate critical effects/studies.
-------
This document is a draft for review purposes only and does not constitute Agency policy.
CO TVT» AT7T MAT /^TTF1 /~\T> AT TATT
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
using nested models because accounting for the intralitter correlation improved model fit. For
these latter analyses, a 1% extra risk of a pup having a heart malformation was used as the BMR
because of the severity of the effect, since, for example, some of the types of malformations
observed could have been fatal. The ratio of the resulting BMD to the BMDL was about three.
For developmental neurotoxicity, the cRfD estimates based on the four oral studies span a
wide range from 0.02-0.8 mg/kg/day. The most reliable estimate, with a composite UF of 100,
is based on BMD modeling of decreased locomotor activity in rats reported in George et al.
(1986), although a nonstandard BMR of a twofold change was selected because the control SD
appeared unusually small. The cRfDs developed for decreased rearing postexposure in mice
(Fredriksson et al., 1993), increased exploration postexposure in rats (Taylor et al., 1985) and
decreased myelination in the hippocampus of rats (Isaacson and Taylor, 1989), while being more
than 10-fold lower, are all within a threefold range of 0.02-0.05 mg/kg/day. Importantly, there
is some evidence from adult neurotoxicity studies of TCE causing demyelination, so there is
additional biological support for the latter effect. There is greater uncertainty in the Fredriksson
et al. (1993), the cRfD for which utilized a subchronic-to-chronic UF of three rather than one,
because exposure during postnatal day (PND) 10-16 does not cover the full developmental
window (Rice and Barone, 2000). The cRfDs derived from Taylor et al. (1985) and (Isaacson
and Taylor, 1989) used the composite UF of 1,000 for a POD that is a LOAEL. While there is
greater uncertainty in these endpoints, none of the uncertainties is particularly high, and they also
appear to be more sensitive indicators of developmental neurotoxicity than that from George
et al. (1986).
A cRfD of 0.0004 mg/kg/day was developed from the study (Peden-Adams et al., 2006)
that reported developmental immunotoxicity. The main effects observed were significantly
decreased PFC response and increased delayed-type hypersensitivity. The data on these effects
were kindly provided by Dr. Peden-Adams (personal communication from Margie
Peden-Adams, Medical University of South Carolina, to Jennifer Jinot, EPA, 26 August 2008);
however, the dose-response relationships were sufficiently supralinear that attempts at BMD
modeling did not result in adequate fits to these data. Thus, the LOAEL was used as the POD.
Although decreased PFC response may not be considered adverse in and of itself, a
LOAEL-to-NOAEL UF of 10 was used because of the increased delayed-type hypersensitivity at
the same dose. While there is uncertainty in this estimate, it is notable that decreased PFC
response was also observed in an immunotoxicity study in adult animals (Woolhiser et al., 2006),
lending biological plausibility to the effect.
In summary, there is moderate-to-high confidence both in the hazard and the cRfCs and
cRfDs for developmental effects of TCE. It is also noteworthy that the PODs for the more
This document is a draft for review purposes only and does not constitute Agency policy.
53 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
sensitive developmental effects were similar to or, in most cases, lower than the PODs for the
more sensitive reproductive effects, suggesting that developmental effects are not a result of
paternal or maternal toxicity. Among inhalation studies, cRfCs were only developed for effects
in rats reported in Healy et al. (1982), so the effects of resorptions, decreased fetal weight, and
delayed skeletal ossification were considered candidate critical developmental effects. Because
resorptions were also reported in oral studies, the most sensitive (rat) oral study (and most
reliable for dose-response analysis) of Narotsky et al. (1995) was also selected as a candidate
critical study for this effect. The confidence in the oral studies and candidate reference values
developed for more sensitive endpoints is more moderate, but still sufficient for consideration as
candidate critical effects. The most sensitive endpoints by far are the increased fetal heart
malformations in rats reported by Johnson et al. (2003) and the developmental immunotoxicity in
mice reported by Peden-Adams et al. (2006), and these are both considered candidate critical
effects. Neurodevelopmental effects are a distinct type among developmental effects. Thus, the
next most sensitive endpoints of decreased rearing postexposure in mice (Fredriksson et al.,
1993), increased exploration postexposure in rats (Taylor et al., 1985) and decreased myelination
in the hippocampus of rats (Isaacson and Taylor, 1989) are also considered candidate critical
effects.
5.1.2.1.11. Summary of cRfCs, cRfDs, and Candidate Critical Effects
An overall summary of the cRfCs, cRfDs, and candidate critical effects across the health
effect domains is shown in Tables 5-11-5-12. These tables present, for each type of noncancer
effect, the relative ranges of the cRfC and cRfD developed for the different endpoints. The
candidate critical effects selected above for each effect domain are shown in bold. As discussed
above, these effects were generally selected to represent the most sensitive endpoints, across
species where possible. From these candidate critical effects, candidate reference values based
on internal dose-metrics from the PBPK model (p-cRfCs and p-cRfDs) were developed where
possible. Application of the PBPK model is discussed in the next section.
5.1.3. Application of Physiologically Based Pharmacokinetic (PBPK) Model to Inter- and
Intraspecies Extrapolation for Candidate Critical Effects
For the candidate critical effects, the use of PBPK modeling of internal doses could
justify, where appropriate, replacement of the uncertainty factors for pharmacokinetic inter and
intraspecies extrapolation. For more details on PBPK modeling used to estimate levels of
This document is a draft for review purposes only and does not constitute Agency policy.
54 DRAFT—DO NOT CITE OR QUOTE
-------
1 dose-metrics corresponding to different exposure scenarios in rodents and humans, as well as a
2 qualitative discussion of the uncertainties and limitations of the model, see Section 3.5.
This document is a draft for review purposes only and does not constitute Agency policy.
5 5 DRAFT—DO NOT CITE OR QUOTE
-------
Table 5-11. Ranges of cRfCs based on applied dose for various noncancer effects associated with inhalation
TCE exposure
cRfC range
(ppm)
Neurological
Systemic/organ-specific
Immunological
Reproductive
Developmental
10-100
Impaired visual discrimination
(rat)
1-10
Kidney
meganucleocytosis
(rat)
| kidney weight (mouse)
0.1-1
Ototoxicity (rat)
Hyperactivity (rat)
Changes in locomotor activity
(rat)
Trigeminal nerve effects
(human)
Impaired visual function
(rabbit)
I regeneration of sciatic
nerve (rat)
| liver weight (rat)
| liver weight (mouse)
| kidney weight (rat)
I PFC response (rat)
i maternal body weight gain
(rat)
t abnormal sperm (mouse)
pre/postimplantation losses
(male rat exp)
0.01-0.1
I regeneration of sciatic
nerve (mouse)
Disturbed wakefulness (rat)
Autoimmune changes
(MRL—lpr/lpr
mouse)
Effects on epididymis
epithelium (mouse)
I fertilization (male mouse
exp)
Testes and sperm effects (rat)
Hyperzoospermia (human)
Resorptions (female rat)
I fetal weight (rat)
Skeletal effects (rat)
Endpoints in bold were selected as candidate critical effects (see Sections 5.1.2.1-5.1.2.8).
-------
Table 5-12. Ranges of cRfDs based on applied dose for various noncancer effects associated with oral TCE
exposure
cRfD range
(mg/kg/day)
Neurological
Systemic/organ-specific
Immunological
Reproductive
Developmental
1-10
t neuromuscular changes
(rat)
I BW (mouse)
i humoral response to
sRBC (mouse)
i testis/seminal vesicle
weight (mouse)
i sperm motility (mouse)
t testis/epididymis weight
(rat)
i litters/pair (rat)
i live pups/litter (rat)
i BW gain (rat)
i copulatory performance
(rat)
i fetal weight (rat)
Prenatal loss (rat)
i fetal weight (mouse)
t neonatal mortality (mouse,
rat)
0.1-1
t # rears (rat)
t foot splaying (rat)
Trigeminal nerve effect
(rat)
| liver weight (mouse)
j BW (mouse)
| BW (rat)
Toxic nephropathy (other
rat strains/sexes and
mouse)
Meganucleocytosis (male
Sprague-Dawley rat)
Signs of autoimmune
hepatitis (MRL +/+
mouse)
Inflamm. in various tissues
(MRL +/+ mouse)
Delayed parturition (rat)
I mating (rat)
| BW at PND21 (rat)
I locomotor activity (rat)
Eye defects (rat)
Resorptions (rat)
0.01-0.1
Degeneration of
dopaminergic
neurons (rat)
Toxic nephropathy (female
Marshall rat)
I cell-mediated response
to sRBC (mouse)
I stem cell bone marrow
recolonization (mouse)
I ability of sperm to
fertilize (rat)
| exploration (postexp.)
(rat)
I rearing (postexp.)
(mouse)
I myelination in
hippocampus (rat)
0.001-0.01
Demyelination in
hippocampus (rat)
t anti-dsDNA and
anti-ssDNA Abs (early
marker for SLE)
(mouse)
io'-o.ooi
I thymus weight (mouse)
Immunotox (J, PFC, |
DTH) (B6C3F1 mouse)
Heart malformations (rat)
Endpoints in bold were selected as candidate critical effects (see Sections 5.1.2.1-5.1.2.8).
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Quantitative analyses of the PBPK modeling uncertainties and their implications for
dose-response assessment, utilizing the results of the Bayesian analysis of the PBPK model, are
discussed separately in Section 5.1.4.
5.1.3.1.1. Selection of Dose-metrics for Different Endpoints
One area of scientific uncertainty in noncancer dose-response assessment is the
appropriate scaling between rodent and human doses for equivalent responses. As discussed
above, the interspecies UF of 10 is usually thought of as a product of two factors of
(approximately) three each for pharmacokinetics and pharmacodynamics. In this assessment,
EPA's cross-species scaling methodology, grounded in general principles of allometric variation
of biologic processes, is used for describing pharmacokinetic equivalence (Allen and Fisher,
1993; Allen et al., 1987; Crump et al., 1989; "Supplementary data for TCE assessment: Rat
population example," 2011; U.S. EPA, 1992, 2005c) U.S. EPA 2011. Briefly, in the absence of
adequate information to the contrary, the methodology determines pharmacokinetic equivalence
across species through equal average lifetime concentrations or AUCs of the toxicant. Thus, in
cases where the PBPK model can predict internal concentrations of the active moiety, equivalent
daily AUCs are assumed to address cross-species pharmacokinetics, and the interspecies UF is
reduced to 3 to account for the remaining pharmacodynamic factor.
In the absence of directly estimated AUCs, the cross-species scaling methodology
assumes that, unless there is evidence to the contrary ("Supplementary data for TCE assessment:
Rat population example," 2011; U.S. EPA, 1992, 2005c)
(1) The production of the active moiety(ies) is proportional to dose
(2) The clearance of the active moiety(ies) scales allometrically by body weight to the %
power; and
(3) The tissue distribution is equal across species
Under these assumptions, for oral exposures, pharmacokinetic equivalence of AUCs between
animals to humans is expressed on the basis of mg/kgyYd, not mg/kg/day ("body-weight
scaling"). For inhalation exposures, pharmacokinetic equivalence would be on the basis of
equivalent air concentrations, since the alveolar ventilation rate (which determines dose, for a
constant air concentration) scales approximately by body weight to the % power, cancelling out
the assumed scaling dependence of clearance.
This document is a draft for review purposes only and does not constitute Agency policy.
5 8 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
However, when one or more metabolites are thought to be the toxicologically active
compound(s), it is often the case that a PBPK model can predict the rate of production of the
active moiety(ies) (i.e., the rate of metabolism) but cannot predict AUCs due to lack of data to
inform clearance. In this case, assumption (1) above can be replaced by the PBPK model, while
the other two cross-species scaling methodology assumptions are retained. The resulting
pharmacokinetic equivalence can therefore be expressed on the basis of rate of
metabolism/kgy7d.32 Thus, in cases where the PBPK model can predict the rate of production of
the active metabolite(s), equivalent daily amounts metabolized through the appropriate pathway
per unit body weight to the 3/4 power are assumed to address cross-species pharmacokinetics, and
the interspecies UF is reduced to 3 to account for the remaining pharmacodynamic factor.
In addition, in some cases when AUCs cannot be estimated, there are data to replace
assumption (2), above, that the clearance of the active moiety(ies) scales allometrically by body
weight to the 3/4 power. Often, this is considered for toxicity associated with local (in situ)
production of "reactive" metabolites whose concentrations cannot be directly measured in the
target tissue. In such a case, an alternative approach of scaling the rate of local metabolism by
target tissue mass, rather than body weight to the 3/4 power, is appropriate if the metabolites are
sufficiently reactive and are cleared by "spontaneous" deactivation (i.e., changes in chemical
structure without the need of biological influences). In particular, use of this alternative scaling
approach requires evidence that (1) the active moiety or moieties do not leave the target tissue in
appreciable quantities (i.e., are cleared primarily by in situ transformation to other chemical
species and/or binding to/reactions with cellular components); and (2) the clearance of the active
moieties from the target tissue is governed by biochemical reactions whose rates are independent
of body weight (e.g., purely chemical reactions). If these conditions are met, equivalent daily
amounts metabolized through the appropriate pathway per unit target tissue mass are assumed to
address cross-species pharmacokinetics, and the interspecies UF is reduced to 3 to account for
the remaining pharmacodynamic factor.
32 Consider a circulating stable metabolite X. Under a one-compartment model, at steady-state, the production ofX
will be equal to the clearance of A', so that
Rmet = VdX BW xQfX kcj,
where Rmet = rate of production ofX (mg/time), Vd = fractional volume of distribution, BW= body weight,
Cx = concentration of X and kd = clearance of X in units of 1/time. Then, for the concentration Cx to be equivalent
between experimental animals (A) and humans (Ily.
CX = [RmeJBW X kcl X Vd]H= [RmeJBW X kcl X Vd\A.
Under the cross-species scaling methodology, it is assumed that Vd is the same across species, so
[Rme/BW x kj\,, = [RmJBW x /v-t/|A, Next, under the cross-species scaling methodology, kd (with units of 1/time) is
assumed to scale according to BWI/4 ^^¦P2005c| U.S. EPA 2011, leading to:
Rmet(H)IBWH3/4 = Rmet(A)/BWA
This document is a draft for review purposes only and does not constitute Agency policy.
59 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Finally, there is the case where local metabolism, rather than systemically delivered
metabolite(s), is thought to be involved in toxicity, but there are inadequate data to determine
either the rate of local metabolism or its clearance. In this case, assumption (1) above can be
replaced by the assumption that local metabolism will be proportional to blood concentration.
Because tissue blood flow approximately scales allometrically by body weight to the 3/4 power,
combining this with assumptions (2) and (3) above will lead to the AUC of the parent compound
in blood as an appropriate surrogate for local metabolism. Thus, in this case, equivalent daily
AUCs of the parent compound are assumed to address cross-species pharmacokinetics, and the
interspecies UF is reduced to 3 to account for the remaining pharmacodynamic factor.
To summarize, the internal dose-metric for addressing cross-species pharmacokinetics is
based on the Agency's cross-species scaling methodology. The preferred dose-metric under this
methodology is equivalent daily AUC of the active moiety (parent compound or metabolite).
For metabolites, in cases where the rate of production, but not the rate of clearance, of the active
moiety can be estimated, the preferred dose-metric is the rate of metabolism (through the
appropriate pathway) scaled by body weight to the 3/4 power. If there are sufficient data to
consider the active metabolite moiety(ies) "reactive" and cleared through nonbiological
processes, then the preferred dose-metric is the rate of metabolism (through the appropriate
pathway) scaled by the tissue mass. Finally, if local metabolism is thought to be involved, but
cannot be estimated with the available data, then the AUC of the parent compound in blood is
considered an appropriate surrogate and thus the preferred dose-metric.
These dose-metrics were then also used in addressing the pharmacokinetic component,
UFh-pk, of the UF for human (intraspecies) variability. Because all the dose-metrics used for
TCE were for adults, and the dose-metrics are not very sensitive to the plausible range of adult
body weight, for convenience the body weight 3/4 scaling used for interspecies extrapolation was
retained for characterization of human variability. However, it should be emphasized that this
intraspecies characterization is of pharmacokinetics only, and not pharmacodynamics.
In general, an attempt was made to use tissue-specific dose-metrics representing
particular pathways or metabolites identified from available data on the role of metabolism in
toxicity for each endpoint (discussed in more detail below). The selection was limited to
dose-metrics for which uncertainty and variability could be adequately characterized by the
PBPK model (see Section 3.5). For most endpoints, sufficient information on the role of
metabolites or MOA was not available to identify likely relevant dose-metrics, and more
"upstream" metrics representing either parent compound or total metabolism had to be used.
The "primary" or "preferred" dose-metric referred to in subsequent tables has the greater
biological support for its involvement in toxicity, whereas "alternative" dose-metrics are those
This document is a draft for review purposes only and does not constitute Agency policy.
60 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
that may also be plausibly involved (discussed further below). A discussion of the dose-metrics
selected for particular noncancer endpoints follows.
5.1.3.1.2. Kidney toxicity (meganucleocytosis, increased kidney weight, toxic nephropathy)
As discussed in Sections 4.4.6-4.4.7, there is sufficient evidence to conclude that
TCE-induced kidney toxicity is caused predominantly by GSH conjugation metabolites either
produced in situ in or delivered systemically to the kidney. As discussed in Section 3.3.3.2,
bioactivation of S-dichlorovinyl glutathione (DCVG), DCVC, and
A'-acetyl-S-fl ,2-dichlrovinyl)-L-cysteine (NAcDCVC) within the kidney, either by beta-lyase,
flavin mono-oxygenase (FMO), or cytochrome P450 (CYP), produces reactive species, any or all
of which may cause nephrotoxicity. Therefore, multiple lines of evidence support the conclusion
that renal bioactivation of DCVC is the preferred basis for internal dose extrapolations for
TCE-induced kidney toxicity. However, uncertainties remain as to the relative contribution from
each bioactivation pathway; and quantitative clearance data necessary to calculate the
concentration of each species are lacking. Moreover, the estimates of the amount bioactivated
are indirect, derived from the difference between overall GSH conjugation flux and NAcDCVC
excretion (see Section 3.5.7.3.1).
Under the cross-species scaling methodology, the rate of renal bioactivation of DCVC
would be scaled by body weight to the 3/4 power. However, it is necessary to consider whether
there are adequate data to support use of the alternative scaling by target tissue mass. For the
beta-lyase pathway, Dekant et al. (1988) reported in trapping experiments that the postulated
reactive metabolites decompose to stable (unreactive) metabolites in the presence of water.
Moreover, the necessity of a chemical trapping mechanism to detect the reactive metabolites
suggests a very rapid reaction such that it is unlikely that the reactive metabolites leave the site
of production. Therefore, these data support the conclusion that, for this bioactivation pathway,
clearance is chemical in nature and hence species-independent. If this were the only
bioactivation pathway, then scaling by kidney weight would be supported. With respect to the
FMO bioactivation pathway, Sausen and Elfarra (1991) reported that after direct dosing of the
postulated reactive sulfoxide (DCVC sulfoxide), the sulfoxide was detected as an excretion
product in bile. These data suggest that reactivity in the tissue to which the sulfoxide was
delivered (the liver, in this case) is insufficient to rule out a significant role for enzymatic or
other biologically mediated systemic clearance. Therefore, according to the criteria outlined
above, for this bioactivation pathway, the data support scaling the rate of metabolism by body
weight to the 3/4 power. For P450-mediated bioactivation producing NAcDCVC sulfoxide, the
only relevant data on clearance are from a study of the structural analogue to DCVC,
This document is a draft for review purposes only and does not constitute Agency policy.
61 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
fluoromethyl-2,2-difluoro-l-(trifluoromethyl)vinyl ether (FDVE; Sheffels et al., 2004), which
reported that the postulated reactive sulfoxide was detected in urine. This suggests that the
sulfoxide is sufficiently stable to be excreted by the kidney and supports the scaling of the rate of
metabolism by body weight to the 3/4 power.
Therefore, because the contributions to TCE-induced nephrotoxicity from each possible
bioactivation pathway are not clear, and the scaling by body weight to the 3/4 power is supported
for two of the identified three bioactivation pathways, it is decided here to scale the DCVC
bioactivation rate by body weight to the 3/4 power. The primary internal dose-metric for
TCE-induced kidney toxicity is thus, the weekly rate of DCVC bioactivation per unit body
weight to the 3/4 power (ABioactDCVCBW34 |mg/kg '/week|) However, it should be noted
that due to the larger relative kidney weight in rats as compared to humans, scaling by kidney
weight instead of body weight to the 3/4 power would only change the quantitative interspecies
extrapolation by about twofold,33 so the sensitivity of the results to the scaling choice is
relatively small. In addition, quantitative estimates for this dose-metric are only available in rats
and humans, and not in mice. Accordingly, this metric was only used for extrapolating results
from rat toxicity studies.
An alternative dose-metric that also involves the GSH conjugation pathway is the amount
of GSH conjugation scaled by the 3/4 power of body weight (AMetGSHBW34 |mg/kg '/week|)
This dose-metric uses the total flux of GSH conjugation as the toxicologically-relevant dose,
and, thus, incorporates any direct contributions from DCVG and DCVC, which are not addressed
in the DCVC bioactivation metric. The rationale for scaling by body weight to the 3/4 power
rather than target tissue mass is the same as above. Because of the lack of availability of the
DCVC bioactivation dose-metric in mice, the GSH conjugation metric is used as the primary
dose-metric for the nephrotoxicity endpoint in studies of mice.
Another alternative dose-metric is the total amount of TCE metabolism (oxidation and
GSH conjugation together) scaled by the 3/4 power of body weight (TotMetabBW34
[mg/kg/4/week]). This dose-metric uses the total flux of TCE metabolism as the toxicologically
relevant dose, and, thus, incorporates the possible involvement of oxidative metabolites, acting
either additively or interactively, in addition to GSH conjugation metabolites in nephrotoxicity
(see Section 4.4.6). However, this dose-metric is given less weight than those involving GSH
conjugation because, as discussed in Sections 4.4.6, the weight of evidence supports the
conclusion that GSH conjugation metabolites play a predominant role in nephrotoxicity. The
33The range of the difference is 2.1-2.4-fold using the posterior medians for the relative kidney weight in rats and
humans from the PBPK model described in Section 3.5 (see Table 3-37), and body weights of 0.3-0.4 kg for rats
and 60-70 kg for humans.
This document is a draft for review purposes only and does not constitute Agency policy.
62 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
rationale for scaling by body weight to the 3/4 power rather than target tissue mass is the same as
above.
5.1.3.1.3. Liver weight increases (hepatomegaly)
As discussed in Section 4.5.6, there is substantial evidence that oxidative metabolism is
involved in TCE hepatotoxicity, based primarily on similarities in noncancer effects with a
number of oxidative metabolites of TCE (e.g., chloral hydrate [CH], TCA, and dichloroacetic
acid [DCA]). While TCA is a stable, circulating metabolite, CH and DCA are relatively
short-lived, although enzymatically cleared (see Section 3.3.3.1). As discussed in Section
4.5.6.2.1, there is substantial evidence that TCA alone does not adequately account for the
hepatomegaly induced by TCE; therefore, unlike in previous dose-response analyses (Barton and
Clewell, 2000; Clewell and Andersen, 2004), the AUC of TCA in plasma or in liver were not
considered as dose-metrics. However, there are inadequate data across species to quantify the
dosimetry of CH and DCA, and other intermediates of oxidative metabolism (such as TCE-oxide
or dichloroacetylchloride) may be involved in hepatomegaly. Thus, due to uncertainties as to the
active moiety(ies), but given the strong evidence associating TCE liver effects with oxidative
metabolism in the liver, hepatic oxidative metabolism is the preferred basis for internal dose
extrapolations of TCE-induced liver weight increases.
Under the cross-species scaling methodology, the rate of hepatic oxidative metabolism
would be scaled by body weight to the 3/4 power. However, it is necessary to consider whether
there are adequate data to support use of the alternative scaling by target tissue mass. Several of
the oxidative metabolites are stable and systemically available, and several of those that are
cleared rapidly are metabolized enzymatically, so, according to the criteria discussed above,
there are insufficient data to support the conclusions that the active moiety or moieties do not
leave the target tissue in appreciable quantities and are cleared by mechanisms whose rates are
independent of body weight.
Therefore, the primary internal dose-metric for TCE-induced liver weight changes is
selected to be the weekly rate of hepatic oxidation per unit body weight to the 3/4 power
(AMetLivlBW34 [mg/kgyYweek]). The use of this dose-metric is also supported by the analysis
in Section 4.5.6.2.1 showing much more consistency in the dose-response relationships for
TCE-induced hepatomegaly across studies and routes of exposure using this metric and the total
oxidative metabolism dose-metric (discussed below) as compared to the AUC of TCE in blood.
It should be noted that due to the larger relative liver weight in mice as compared to humans,
scaling by liver weight instead of body weight to the 3/4 power would only change the
This document is a draft for review purposes only and does not constitute Agency policy.
63 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
quantitative interspecies extrapolation by about fourfold,34 so the sensitivity of the results to the
scaling choice is relatively modest.
It is also known that the lung has substantial capacity for oxidative metabolism, with
some proportion of the oxidative metabolites produced there entering systemic circulation. Thus,
it is possible that extrahepatic oxidative metabolism can contribute to TCE-induced
hepatomegaly. Therefore, the total amount of oxidative metabolism of TCE scaled by the
3/4 power of body weight (TotOxMetabBW34 [mg/kg/4/week]) was selected as an alternative
dose-metric (the rationale for the body weight to the 3/4 power scaling is analogous to that for
hepatic oxidative metabolism, above).
5.1.3.1.4. Developmental toxicity—heart malformations
As discussed in Section 4.8.3.2.1, several studies have reported that the prenatal exposure
to TCE oxidative metabolites TCA or DCA also induces heart malformations, suggesting that
oxidative metabolism is involved in TCE-induced heart malformations. However, there are
inadequate data across species to quantify the dosimetry of DCA, and it is unclear if other
products of TCE oxidative metabolism are involved. Therefore, the total amount of oxidative
metabolism of TCE scaled by the 3/4 power of body weight (TotOxMetabBW34 [mg/kgyYweek])
was selected as the primary dose-metric. The rationale for the scaling by body weight to the
3/4 power is analogous to that for hepatic oxidative metabolism, above.
An alternative dose-metric that is considered here is the AUC of TCE in (maternal) blood
(AUCCBld [mg-hour/L/day]). The placenta is a highly perfused tissue, and TCE is known to
cross the placenta to the fetus, with rats showing similar (within twofold) maternal and fetal
blood TCE concentrations (see Section 3.2). This dose-metric accounts for the possible roles
either of local metabolism or of TCE itself.
5.1.3.1.5. Reproductive toxicity—decreased ability of sperm to fertilize oocytes
The decreased ability of sperm to fertilize oocytes observed by DuTeaux et al. (2004a)
occurred in the absence of changes in combined testes/epididymes weight, sperm concentration
or motility, or histological changes in the testes or epididymes. However, there was evidence of
34The range of the difference is 3.5-3.9-fold using the posterior medians for the relative liver weight in mice and
humans from the PBPK model described in Section 3.5 (see Table 3-37), and body weights of 0.03-0.04 kg for mice
and 60-70 kg for humans.
This document is a draft for review purposes only and does not constitute Agency policy.
64 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
oxidative damage to the sperm, and DuTeaux et al. (2003) previously reported the ability of the
rat epididymis and efferent ducts to metabolize TCE oxidatively. Based on this evidence,
DuTeaux et al. (2004a) hypothesize that the decreased ability to fertilize is due to oxidative
damage to the sperm from local metabolism. Thus, the primary dose-metric for this endpoint is
selected to be the AUC of TCE in blood (AUCCBld [mg-hour/L/day]), based on the assumption
that in situ oxidation of systemically-delivered TCE (the flow rate of which scales as body
weight to the % power) is the determinant of toxicity.
Because metabolites causing oxidative damage may be delivered systemically to the
target tissue, an alternative dose-metric that is considered here is total oxidative metabolism of
TCE scaled by the % power of body weight (TotOxMetabBW34 [mg/kg3/7day]). The rationale
for the scaling by body weight to the % power is analogous to that for hepatic oxidative
metabolism, above. Because oxidative metabolites make up the majority of TCE metabolism,
total metabolism gives very similar results (within 1.2-fold) to total oxidative metabolism and is
therefore not included as a dose-metric.
5.1.3.1.6. Other reproductive and developmental effects and neurological effects and
immunologic effects
For all other candidate critical endpoints listed in Tables 5-11-5-12, including
developmental effects other than heart malformations and reproductive effects other than
decreased ability of sperm to fertilize, there is insufficient information for site-specific
determinations of an appropriate dose-metric. While TCE metabolites and/or metabolizing
enzymes have been reported in some of these tissues (e.g., male reproductive tract), their general
roles in toxicity in the respective tissues have not been established. The choice of total
metabolism as the primary dose-metric is based on the observation that, in general, TCE toxicity
is associated with metabolism rather than the parent compound. It is acknowledged that there is
no compelling evidence that definitively establishes one metric as more plausible than the other
in any particular case. Nonetheless, as a general inference in the absence of specific data, total
metabolism is viewed as more likely to be involved in toxicity than the concentration of TCE
itself.
Therefore, given that the majority of the toxic and carcinogenic responses in many tissues
to TCE appears to be associated with metabolism, the primary dose-metric is selected to be total
metabolism of TCE scaled by the % power of body weight (TotMetabBW34 [mg/kgyYd]). The
rationale for the scaling by body weight to the % power is analogous to that for the other
metabolism dose-metrics, above. Because oxidative metabolites make up the majority of TCE
This document is a draft for review purposes only and does not constitute Agency policy.
65 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
metabolism, total oxidative metabolism gives very similar results (within 1.2-fold) to total
metabolism and is therefore not included as a dose-metric.
An alternative dose-metric that is considered here is the AUC of TCE in blood
(AUCCBld [mg-hour/L/day]). This dose-metric would account for the possible role of local
metabolism, which is determined by TCE delivered in blood via systemic circulation to the target
tissue (the flow rate of which scales as body weight to the 3/4 power), and the possible role of
TCE itself. This dose-metric would also be most applicable to tissues that have similar
tissue:blood partition coefficients across and within species.
Because the PBPK model described in Section 3.5 did not include a fetal compartment,
the maternal internal dose-metric is taken as a surrogate for developmental effects in which
exposure was before or during pregnancy (Fredriksson et al., 1993; Johnson et al., 2003;
Narotsky et al., 1995; Taylor et al., 1985). This was considered reasonable because TCE and the
major circulating metabolites (TCA and trichloroethanol [TCOH]) appear to cross the placenta
(see Sections 3.2, 3.3, and 4.10 (Fisher et al., 1989; Ghantous et al., 1986)), and maternal
metabolizing capacity is generally greater than that of the fetus (see Section 4.10). In the cases
where exposure continues afterbirth (Isaacson and Taylor, 1989; Peden-Adams et al., 2006), no
PBPK model-based internal dose was used. Because of the complicated fetus/neonate dosing
that includes transplacental, lactational, and direct (if dosing continues postweaning) exposure,
the maternal internal dose is no more accurate a surrogate than applied dose in this case.
5.1.3.1.7. Methods for Inter- and Intraspecies Extrapolation Using Internal Doses35
As shown in Figures 5-2 and 5-3, the
general approach taken to use the
internal dose-metrics in deriving
HECs and HEDs was to first apply
the rodent PBPK model to get rodent
values for the dose-metrics
35An alternative approach (e.g.. Clewell et al.. 20021 applies the UFs to the internal dose prior to using the human
PBPK model to derive a human exposure level. As noted by Barton and Clewell (20001 for previous TCE PBPK
models, because the human PBPK model for TCE is linear for all the dose metrics over very broad dose and
concentration ranges, essentially identical results would be obtained using this alternative approach. Specifically,
for all the primary dose metrics, the difference in the two approaches is less than 2-fold, with the results from the
critical studies differing by <0.1%. For some studies using AUCBld as an alternative dose metric, the difference
ranged from 3- to 7-fold. Overall, use of the alternative approach would not significantly change the noncancer
dose-response assessment of TCE, and the derived RfC and RfD would be identical.
This document is a draft for review purposes only and does not constitute Agency policy.
66 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
corresponding to the applied doses in
a study reporting noncancer effects.
The idPOD is then obtained either
directly from the internal dose
corresponding to the applied dose
LOAEL or NOAEL, or by
dose-response modeling of responses
with respect to the internal doses to
derive a BMDL in terms of internal
dose. Separately, the human PBPK
model is run for a range of
continuous exposures from 10 1 to
"3
2x10 ppm or mg/kg/day to obtain
the relationship between human
exposure and internal dose for the
same dose-metric used for the
rodent. The human equivalent
exposure (HEC or HED)
corresponding to the idPOD is
derived by interpolation. It should
be noted that median values of
dose-metrics were used for rodents,
whereas both median and 99th
percentile values were used for
humans. As discussed in
Section 3.5, the rodent population
model characterizes study-to-study
variation, while, within a study,
animals with the same
sex/species/strain combination were
assumed to be identical
pharmacokinetically and represented
by the group average (typically the
only data reported).
This document is a draft for review purposes only and does not constitute Agency policy.
67 DRAFT—DO NOT CITE OR QUOTE
-------
distribution (combined
ncertainty and variability)
Rodent
non-cancer
Rodent
model
parameters
study
experimental
paradigm
ipistribution
PBPK
model
Rodent
non-cancer
study
responses
Rodent
internal
dose
edian
Dose-Response Model
or
LOAEUNOAEL
idPOD (internal
dose unit) =
BMDL or
LOAEL or
NOAEL
0.1-2000 ppm
in air or
0.1-2000
mg/kg-d
continuous
exposure
^distribution
Human
model
parameters
PBPK
model
Human
Internal dose
as function of
applied dose
[distribution (separate
iuncertainty and variability)
Overall
median
Overall
, 99th percentile
invert functions of dose
or concentration
"Typical"
human internal
dose as
function
of applied
dose
"Sensitive"
human internal
dose as
function
of applied
dose
"Typical"
human
equivalent
dose or
concentration
_S_
"Sensitive"
human
equivalent
dose or
concentration
HEC5Q or
hed50
(replaces
POD/UF
is-adj
HECgg or
hed99
[replaces
POD/(UFis.adj*UFh.pk)]
1
2
3
4
5
Figure 5-2. Flow-chart for dose-response analyses of rodent noncancer
effects using PBPK model-based dose-metrics. Square nodes indicate point
values, circle nodes indicate distributions, and the inverted triangle indicates
a (deterministic) functional relationship.
This document is a draft for review purposes only and does not constitute Agency policy.
68 DRAFT—DO NOT CITE OR QUOTE
-------
Human internal
dose
Rodent internal
dose
Oocerta«*y&
variably
distribu
Uncertainty &
variability
distribution
idPOD |
, Lower 99th
percentile
Human inhalation
^exposure (ppm)
=HECc
Study dose groups
LOAELI
NOAEL
Human internal
dose
Uncertainty &
variability
distribution
\
Lower 99th
percentile
Human oral exposure
' (mg/kg/d)
=HED
99
1 Figure 5-3. Schematic of combined interspecies, intraspecies, and
2 route-to-route extrapolation from a rodent study LOAEL or NOAEL. In the
3 case where BMD modeling is performed, the applied dose values are replaced
4 by the corresponding median internal dose estimate, and the idPOD is the
5 modeled BMDL in internal dose units.
6
This document is a draft for review purposes only and does not constitute Agency policy.
69 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Therefore, use of median dose-metric values can be interpreted as assuming that the animals in
the noncancer toxicity study were all "typical" animals and the idPOD is for a rodent that is
pharmacokinetically "typical." In practice, the use of median or mean internal doses for rodents
did not make much difference except when the uncertainty in the rodent dose-metric was high.
The impact of the uncertainty in the rodent PBPK dose-metrics is analyzed quantitatively in
Section 5.1.4.2.
The human population model characterizes individual-to-individual variation, in addition
to its uncertainty. The "median" value for the HEC or HED was calculated as a point of
comparison but was not actually used for derivation of candidate reference values. Because the
RfC and RfD are intended to characterize the dose below which a sensitive individual would
likely not experience adverse effects, the overall 99th percentile of the combined uncertainty and
variability distribution was used for deriving the HEC and HED (denoted HEC99 and HED99)
from each idPOD.36 As shown in Figures 5-2 and 5-3, the HEC99 or HED99 replaces the quantity
PODl{UFis_adj x UFh-pk) in the calculation of the RfC or RfD, i.e., the pharmacokinetic
components of the UFs representing interspecies extrapolation and human interindividual
variability.
As calculated, the extrapolated HEC99 and HED99 can be interpreted as being the dose or
exposure for which there is 99% likelihood that a randomly selected individual will have an
internal dose less than or equal to the idPOD derived from the rodent study. By contrast, the
HEC50 and HED50 can be interpreted as being the dose or exposure for which there is 50%
likelihood that a randomly selected individual will have an internal dose less than or equal to the
idPOD derived from the rodent study. Values of HEC99 or HED99 are shown for each study and
dose-metric considered in Tables 5-13-5-18. In addition, values of HEC50 or HED50 are shown
for comparison, to give a sense of the difference between the median and the 99% confidence
bound for combined uncertainty and variability. The separate contributions of uncertainty and
36While for uncertainty, a 95th percentile is often selected by convention, there is no explicit guidance on the
selection of the percentile for human toxicokinetic variability. Ideally, all sources of uncertainty and variability
would be included, and percentile selected that is more in line with the levels of risk at which cancer dose-response
is typically characterized (e.g., 106 to 104) along with a level of confidence. However, only toxicokinetic
uncertainty and variability is assessed quantitatively. Because the distribution here incorporates both uncertainty
and variability simultaneously, a percentile higher than the 95th (a conventional choice for uncertainty only) was
selected. However, percentiles greater than the 99th are likely to be progressively less reliable due to the unknown
shape of the tail of the input uncertainty and variability distributions for the PBPK model parameters (which were
largely assumed to be normal or lognormal), and the fact that only 42 individuals were incorporated in the PBPK
model for characterization of uncertainty and inter-individual variability (see Section 3.5). This concern is
somewhat ameliorated because the candidate reference values also incorporate use of UFs to account for inter- and
intraspecies toxicodynamic sensitivity.
This document is a draft for review purposes only and does not constitute Agency policy.
70 DRAFT—DO NOT CITE OR QUOTE
-------
Table 5-13. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on PBPK modeled
internal dose-metrics) for candidate critical neurological effects
s
>!
u
Or
>s
<
cS'
ss
^ o
c-n
>3
o
5S
a
s
o-
o
H §
I
81
2;
" a
•3
"T3
H S.
ffl.K"
O •
o
c
o
H
W
H
O,
Effect type
Candidate critical studies
Species
POD
type
hec50
or
HED50
POD,
HEC99,
or
HED99a
UF
u 1 sc
UFls
UFh
UFloael
UFdb
UF b
^ ¦ comi)
cRfC or
p-cRfC
(PPm)
cRtD or
p-cRfD
(mg/kg/day)
Candidate critical effect; comments
[dose-metric]
Trigeminal nerve effects
Ruijtenetal. (1991)
Human
LOAEL
14
1
1
10
3
1
30
0.47
Trigeminal nerve effects
HEC
14
5.3
1
1
3
3
1
10
0.53
[TotMetabBW34]
HEC
14
8.3
1
1
3
3
1
10
0.83
[AUCCBld]
HED
7.4
7.3
1
1
3
3
1
10
0.73
[TotMetabBW34] (route-to-route)
HED
59
14
1
1
3
3
1
10
1.4
[AUCCBld] (route-to-route)
Cognitive effects
Isaacson et al. (1990)
Rat
LOAEL
47
10
10
10
10
1
10,000c
0.0047
demyelination in hippocampus
HED
9.4
9.2
10
3
3
10
1
1,000
0.0092
[TotMetabBW34]
HED
31
4.3
10
3
3
10
1
1,000
0.0043
[AUCCBld]
HEC
18
7.1
10
3
3
10
1
1,000
0.0071
[TotMetabBW34] (route-to-route)
HEC
3.8
2.3
10
3
3
10
1
1,000
0.0023
[AUCCBld] (route-to-route)
Mood and sleep disorders
Arito et al. (1994)
Rat
LOAEL
12
3
3
10
10
1
1,000
0.012
Changes in wakefulness
HEC
13
4.8
3
3
3
10
1
300
0.016
[TotMetabBW34]
HEC
15
9.0
3
3
3
10
1
300
0.030
[AUCCBld]
HED
6.6
6.5
3
3
3
10
1
300
0.022
[TotMetabBW34] (route-to-route)
HED
65
15
3
3
3
10
1
300
0.051
[AUCCBld] (route-to-route)
-------
Table 5-13. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on PBPK modeled internal
dose-metrics) for candidate critical neurological effects (continued)
Effect type
Candidate critical studies
Species
POD
type
hec50
or
HED50
POD,
HEC99,
or
HED99a
UFj,.
UFls
UFh
UFloael
UFdb
TTTT k
^ 1 com [)
cRfC or
p-cRfC
(ppm)
cRfD or
p-cRfD
(mg/kg/da
y)
Candidate critical effect; comments
[dose-metric]
Other neurological effects
Kjellstrand et al. (1987)
Rat
LOAEL
300
10
3
10
10
1
3,000
0.10
J, regeneration of sciatic nerve
HEC
274
93
10
3
3
10
1
1,000
0.093
[TotMetabBW34]
HEC
487
257
10
3
3
10
1
1,000
0.26
[AUCCBld]
HED
110
97
10
3
3
10
1
1,000
0.097
[TotMetabBW34] (route-to-route)
HED
436
142
10
3
3
10
1
1,000
0.14
[AUCCBld] (route-to-route)
Mouse
LOAEL
150
10
3
10
10
1
3,000
0.050
J, regeneration of sciatic nerve
HEC
378
120
10
3
3
10
1
1,000
0.12
[TotMetabBW34]
HEC
198
108
10
3
3
10
1
1,000
0.11
[AUCCBld]
HED
145
120
10
3
3
10
1
1,000
0.12
[TotMetabBW34] (route-to-route)
HED
237
76
10
3
3
10
1
1,000
0.076
[AUCCBld] (route-to-route)
Gash et al. (2008)
Rat
LOAEL
710
10
10
10
10
1
10,000c
0.071
degeneration of dopaminergic neurons
HED
56
53
10
3
3
10
1
1,000
0.053
[TotMetabBW34]
HED
571
192
10
3
3
10
1
1,000
0.19
[AUCCBld]
HEC
126
47
10
3
3
10
1
1,000
0.047
[TotMetabBW34] (route-to-route)
HEC
679
363
10
3
3
10
1
1,000
0.36
[AUCCBld] (route-to-route)
"Applied dose POD adjusted to continuous exposure unless otherwise noted. POD, HEC99, and HED99 have same units as cRfC (ppm) or cRfD (mg/kg/day).
bProduct of individual uncertainty factors, rounded to 3, 10, 30, 100, 300, 1,000, 3,000, or 10,000 [see Footnote c below].
°EPA's report on the RfC and RfD processes (U.S. EPA, 2002c) recommends not deriving reference values with a composite UF of greater than 3,000; however,
composite UFs exceeding 3,000 are considered here because the derivation of the cRfCs and cRfDs is part of a screening process and the application of the
PBPK model for candidate critical effects reduces the values of some of the individual UFs for the p-cRfCs and p-cRfDs.
UFSC = subchronic-to-chronic UF; UF1S = interspecies UF; UFh = human variability UF; UFioaei = LOAEL-to-NOAEL UF; UFdb = database UF.
Shaded rows represent the p-cRfC or p-cRfD using the preferred PBPK model dose-metric.
-------
Table 5-14. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on PBPK modeled
internal dose-metrics) for candidate critical kidney effects
s
>!
o
O5
<
cS'
ss
<1 ^
o
c-n
>3
o
5S
a
s
o-
o
H §
I
81
2;
" a
•3
"T3
H S.
ffl.K"
O •
o
c
o
H
W
H
O,
Effect type
Candidate critical
studies
Species
POD
type
HECjo
or
hed50
POD,
HEC99,
or
HED99a
UFSC
UFis
UFh
UFloael
UFdb
TJF b
* comp
cRfCor
p-cRfC
(ppm)
cRI'D or
p-cRfD
(mg/kg/day)
Candidate critical effect;
comments [dose-metric]
Histological changes in kidney
Maltoni (1986)
(inlialation)
Rat
BMDL
40.2
1
3
10
1
1
30
1.3
meganucleocytosis; BMR = 10%
HEC
0.28
0.038
1
3
3
1
1
10
0.0038
[ABioactDCVCBW34]
HEC
0.45
0.058
1
3
3
1
1
10
0.0058
[AMetGSHBW34]
HEC
39
15.3
1
3
3
1
1
10
1.5
[TotMetabBW34]
HED
0.22
0.023
1
3
3
1
1
10
0.0023
[ABioactDCVCBW34]
(route-to-route)
HED
0.35
0.036
1
3
3
1
1
10
0.0036
[AMetGSHBW34] (route-to-route)
HED
19
19
1
3
3
1
1
10
1.9
[TotMetabBW34] (route-to-route)
NCI (1976)
Mouse
LOAEL
620
1
10
10
30
1
3,000
0.21
toxic neplirosis
HED
2.9
0.30
1
3
3
30
1
300
0.00101
[AMetGSHBW34]
HED
51
48
1
3
3
30
1
300
0.160
[TotMetabBW34]
HEC
3.9
0.50
1
3
3
30
1
300
0.00165
[AMetGSHBW34] (route-to-route)
HEC
113
42
1
3
3
30
1
300
0.140
[TotMetabBW34] (route-to-route)
-------
Table 5-14. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on PBPK modeled internal
dose-metrics) for candidate critical kidney effects (continued)
s
>!
o
Or
>s
<
cS'
ss
+*¦ o
&n
rs
Zn
O
S
s
o-
o
H §
I
81
2;
" a
•3
"T3
H S.
ffl.K"
O •
p
o
c
o
H
W
H
O,
Effect type
Candidate critical
studies
Species
POD
type
HEC50
or
hed50
POD,
HEC99,
or
HED99a
UFSC
UFis
UFh
UFloael
UFdb
UFcomp
b
cRfCor
p-cRfC
(ppm)
cRI'D or
p-cRfD
(mg/kg/day)
Candidate critical effect;
comments [dose-metric]
NTP (1988)
Rat
BMDL
9.45
1
10
10
1
1
100
0.0945
toxic nephropathy; BMR = 5%;
female Marshall (most sensitive
sex/strain)
HED
0.033
0.0034
1
3
3
1
1
10
0.00034
[ABioactDCVCBW34]
HED
0.053
0.0053
1
3
3
1
1
10
0.00053
[AMetGSHBW34]
HED
0.75
0.74
1
3
3
1
1
10
0.074
[TotMetabBW34]
HEC
0.042
0.0056
1
3
3
1
1
10
0.00056
[ABioactDCVCBW34]
(route-to-route)
HEC
0.067
0.0087
1
3
3
1
1
10
0.00087
[AMetGSHBW34] (route-to-route)
HEC
1.4
0.51
1
3
3
1
1
10
0.051
[TotMetabBW34] (route-to-route)
Histological changes in kidney (continued)
Maltoni (1986) (oral)
Rat
BMDL
34
1
10
10
1
1
100
0.34
meganucleocytosis; BMR = 10%
HED
0.15
0.015
1
3
3
1
1
10
0.0015
[ABioactDCVCBW34]
HED
0.25
0.025
1
3
3
1
1
10
0.0025
[AMetGSHBW34]
HED
11
11
1
3
3
1
1
10
0.11
[TotMetabBW34]
HEC
0.19
0.025
1
3
3
1
1
10
0.0025
[ABioactDCVCBW34]
(route-to-route)
HEC
0.31
0.041
1
3
3
1
1
10
0.0041
[AMetGSHBW34] (route-to-route)
HEC
22
8.5
1
3
3
1
1
10
0.85
[TotMetabBW34] (route-to-route)
-------
Table 5-14. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on PBPK modeled internal
dose-metrics) for candidate critical kidney effects (continued)
Effect type
Candidate critical
studies
Species
POD
type
HEC50
or
hed50
POD,
HEC99,
or
HED99a
UFSC
UFis
UFh
UFloael
UFdb
UFcomp
b
cRfCor
p-cRfC
(ppm)
cRI'D or
p-cRfD
(mg/kg/day)
Candidate critical effect;
comments [dose-metric]
| kidney/body weight ratio
Kjellstrand et al.
(1983b)
Mouse
BMDL
34.7
1
3
10
1
1
30
1.2
BMR = 10%
HEC
0.88
0.12
1
3
3
1
1
10
0.012
[AMetGSHBW34]
HEC
52
21
1
3
3
1
1
10
2.1
[TotMetabBW34]
HED
0.69
0.070
1
3
3
1
1
10
0.0070
[AMetGSHBW34] (route-to-route)
HED
25
25
1
3
3
1
1
10
2.5
[TotMetabBW34] (route-to-route)
Woolhiser et al.
(2006)
Rat
BMDL
15.7
1
3
10
1
1
30
0.52
BMR = 10%
HEC
0.099
0.013
1
3
3
1
1
10
0.0013
[ABioactDCVCBW34]
HEC
0.17
0.022
1
3
3
1
1
10
0.0022
[AMetGSHBW34]
HEC
29
11
1
3
3
1
1
10
1.1
[TotMetabBW34]
HED
0.078
0.0079
1
3
3
1
1
10
0.00079
[ABioactDCVCBW34]
(route-to-route)
HED
0.13
0.013
1
3
3
1
1
10
0.0013
[AMetGSHBW34] (route-to-route)
HED
14
14
1
3
3
1
1
10
1.4
[TotMetabBW34] (route-to-route)
2! fc,
1—1 s
H S.
ffl.K"
O •
p
o
c
o
H
W
a Applied dose POD adjusted to continuous exposure unless otherwise noted. POD, HEC99, and HED99 have same units as cRfC or cRfD.
bProduct of individual uncertainty factors, rounded to 3, 10, 30, 100, 300, 1,000, or 3,000.
UFSC = subchronic-to-chronic UF; UF1S = interspecies UF; UFh = human variability UF; UFloael = LOAEL-to-NOAEL UF; UFdb = database UF.
Shaded rows represent the p-cRfC or p-cRfD using the preferred PBPK model dose-metric.
-------
Table 5-15. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on PBPK modeled
internal dose-metrics) for candidate critical liver effects
Effect type
Candidate critical
studies
Species
POD
type
HECjo
or
hed50
POD,
HEC99,
or
HED99a
UFSC
UFis
UFh
UFloael
UFdb
UF b
* comp
cRfCor
p-cRfC
(ppm)
cRI'D or
p-cRfD
(mg/kg/day)
Candidate critical effect; comments
[dose-metric]
| liver/body weight ratio
Kjellstrand et al.
(1983b)
Mouse
BMDL
21.6
1
3
10
1
1
30
0.72
BMR = 10% increase
HEC
25
9.1
1
3
3
1
1
10
0.91
[AMetLivlBW34]
HEC
75
24.9
1
3
3
1
1
10
2.5
[TotOxMetabBW34]
HED
9.0
7.9
1
3
3
1
1
10
0.79
[AMetLivlBW34] (route-to-route)
HED
32
25.7
1
3
3
13
1
10
2.6
[TotOxMetabBW34] (route-to-route)
Woolhiser et al.
(2006)
Rat
BMDL
25
1
3
10
1
1
30
0.83
BMR = 10% increase
HEC
53
19
1
3
3
1
1
10
1.9
[AMetLivlBW34]
HEC
46
16
1
3
3
1
1
10
1.6
[TotOxMetabBW34]
HED
19
16
1
3
3
1
1
10
1.6
[AMetLivlBW34] (route-to-route)
HED
20
17
1
3
3
1
1
10
1.7
[TotOxMetabBW34] (route-to-route)
Buben and OFlaherty
(1985)
Mouse
BMDL
82
1
10
10
1
1
100
0.82
BMR = 10% increase
HED
12
10
1
3
3
1
1
10
1.0
[AMetLivlBW34]
HED
15
13
1
3
3
1
1
10
1.3
[TotOxMetabBW34]
HEC
32
11
1
3
3
1
1
10
1.1
[AMetLivlBW34] (route-to-route)
HEC
34
11
1
3
3
1
1
10
1.1
[TotOxMetabBW34] (route-to-route)
O
c
o
H
W
a Applied dose POD adjusted to continuous exposure unless otherwise noted. POD, HEC99, and HED99 liave same units as cRfC (ppm) or cRfD (mg/kg/day).
bProduct of individual uncertainty factors, rounded to 3, 10, 30, 100, 300, 1,000, or 3,000.
UFSC = subchronic-to-chronic UF; UF1S = interspecies UF; UFh = human variability UF; UFioaei = LOAEL-to-NOAEL UF; UFdb = database UF.
Shaded rows represent the p-cRfC or p-cRfD using the preferred PBPK model dose-metric.
-------
Table 5-16. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on PBPK modeled
internal dose-metrics) for candidate critical immunological effects
s
>!
<1
u
Or
>s
<
cS'
ss
^1 o
c-n
>3
o
5S
a
s
o-
o
H §
I
81
2;
" a
•3
"T3
H S.
ffl.K"
O •
o
c
o
H
W
H
O,
Effect type
Candidate critical
studies
Species
POD
type
HECjo
or
hed50
POD,
HEC99,
or
HED99a
UFSC
UFis
UFh
UFloael
UFdb
TJF b
^ * comp
cRfCor
p-cRfC
(ppm)
cRI'D or
p-cRfD
(mg/kg/day)
Candidate critical effect; comments
[dose-metric]
I thymus weight
Keil et al. (2009)
Mouse
LOAEL
0.35
1
10
10
10
1
1,000
0.00035
i thymus weight
HED
0.049
0.048
1
3
3
10
1
100
0.00048
[TotMetabBW34]
HED
0.20
0.016
1
3
3
10
1
100
0.00016
[AUCCBld]
HEC
0.092
0.033
1
3
3
10
1
100
0.00033
[TotMetabBW34] (route-to-route)
HEC
0.014
0.0082
1
3
3
10
1
100
0.000082
[AUCCBld] (route-to-route)
Autoimmunity
Kaneko et al. (2000)
Mouse
LOAEL
70
10
3
3
10
1
1,000
0.070
Changes in iimnunoreactive organs -
liver (including sporadic necrosis in
hepatic lobules), spleen; UFh = 3 due to
autoimmune-prone mouse
HEC
97
37
10
3
1
10
1
300
0.12
[TotMetabBW34]
HEC
121
69
10
3
1
10
1
300
0.23
[AUCCBld]
HED
44
42
10
3
1
10
1
300
0.14
[TotMetabBW34] (route-to-route)
HED
181
57
10
3
1
10
1
300
0.19
[AUCCBld] (route-to-route)
Keil et al. (2009)
Mouse
LOAEL
0.35
1
10
10
1
1
100
0.0035
t anti-dsDNA and anti-ssDNA Abs
(early markers for SLE)
HED
0.049
0.048
1
3
3
1
1
10
0.0048
[TotMetabBW34]
HED
0.20
0.016
1
3
3
1
1
10
0.0016
[AUCCBld]
HEC
0.092
0.033
1
3
3
1
1
10
0.0033
[TotMetabBW34] (route-to-route)
HEC
0.014
0.0082
1
3
3
1
1
10
0.00082
[AUCCBld] (route-to-route)
-------
Table 5-16. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on PBPK modeled internal
dose-metrics) for candidate critical immunological effects (continued)
Effect type
Candidate critical
studies
Species
POD
type
HEC50
or
hed50
POD,
hec99,
or
HED99a
UFSC
UFis
UFh
UFloael
UFdb
UF b
* comp
cRfCor
p-cRfC
(ppm)
cRI'D or
p-cRfD
(mg/kg/day)
Candidate critical effect; comments
[dose-metric]
Immunosuppression
Woolhiser et al.
(2006)
Rat
BMDL
24.9
10
3
10
1
1
300
0.083
I PFC response; BMR = 1 SD change;
dropped highest dose
HEC
29
11
10
3
3
1
1
100
0.11
[TotMetabBW34]; all does groups
HEC
263
140
10
3
3
1
1
100
1.4
[AUCCBld] ; all does groups
HED
14
14
10
3
3
1
1
100
0.14
[TotMetabBW34] (route-to-route); all
does groups
HED
282
91
10
3
3
1
1
100
0.91
[AUCCBld] (route-to-route); all does
groups
Sanders et al. (1982a)
Mouse
LOAEL
18
1
10
10
3
1
300
0.060
i stem cell bone marrow
recolonization (sustained); j
cell-mediated response to sRBC
(largely transient during exposure);
females more sensitive
HED
2.5
2.5
1
3
3
3
1
30
0.083
[TotMetabBW34]
HED
8.8
0.84
1
3
3
3
1
30
0.028
[AUCCBld]
HEC
4.8
1.7
1
3
3
3
1
30
0.057
[TotMetabBW34] (route-to-route)
HEC
0.73
0.43
1
3
3
3
1
30
0.014
[AUCCBld] (route-to-route)
O
c
o
H
W
"Applied ose POD adjusted to continuous exposure unless otherwise noted. POD, HEC99, and HED99 have same units as cRfC (ppm) or cRfD (mg/kg/day).
bProduct of individual uncertainty factors, rounded to 3, 10, 30, 100, 300, 1,000, or 3,000.
UFSC = subchronic-to-chronic UF; UF1S = interspecies UF; UFh = human variability UF; UFioaei = LOAEL-to-NOAEL UF; UFdb = database UF.
Shaded rows represent the p-cRfC or p-cRfD using the preferred PBPK model dose-metric.
-------
Table 5-17. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on PBPK modeled
internal dose-metrics) for candidate critical reproductive effects
s
>!
O
Or
>s
<
cS'
ss
<1 ^
^ O
c-n
>3
o
5S
a
s
o-
o
H §
I
81
2;
" a
•3
"T3
H S.
ffl.K"
O •
o
c
o
H
W
H
O,
Effect type
Candidate critical
studies
Species
POD
type
HECjo
or
hed50
POD,
HEC99,
or
HED99a
UFSC
UFis
UFh
UFloael
UFdb
TJF b
* comp
cRfCor
p-cRfC
(ppm)
cRI'D or
p-cRfD
(mg/kg/day)
Candidate critical effect; comments
[dose-metric]
Effects on sperm, male reproductive outcomes
Cilia etal. (1996)
Human
BMDL
1.4
10
1
10
1
1
100
0.014
Hyperzoospermia; BMR = 10% extra
risk
HEC
1.4
0.50
10
1
3
1
1
30
0.0017
[TotMetabBW34]
HEC
1.4
0.83
10
1
3
1
1
30
0.0028
[AUCCBld]
HED
0.74
0.73
10
1
3
1
1
30
0.024
[TotMetabBW34] (route-to-route)
HED
15
1.6
10
1
3
1
1
30
0.053
[AUCCBld] (route-to-route)
Xu et al. (2004)
Mouse
LOAEL
180
10
3
10
10
1
3,000
0.060
I fertilization
HEC
190
67
10
3
3
10
1
1,000
0.067
[TotMetabBW34]
HEC
321
170
10
3
3
10
1
1,000
0.17
[AUCCBld]
HED
80
73
10
3
3
10
1
1,000
0.073
[TotMetabBW34] (route-to-route)
HED
324
104
10
3
3
10
1
1,000
0.10
[AUCCBld] (route-to-route)
Kumar et al (2000b)
Rat
LOAEL
45
10
3
10
10
1
3,000
0.015
Multiple sperm effects, increasing
severity from 12-24 weeks
HEC
32
13
10
3
3
10
1
1,000
0.013
[TotMetabBW34]
HEC
91
53
10
3
3
10
1
1,000
0.053
[AUCCBld]
HED
16
16
10
3
3
10
1
1,000
0.016
[TotMetabBW34] (route-to-route)
HED
157
49
10
3
3
10
1
1,000
0.049
[AUCCBld] (route-to-route)
-------
Table 5-17. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on PBPK modeled internal
dose-metrics) for candidate critical reproductive effects (continued)
s
>!
00
o
o
Or
>s
<
cS'
ss
o
c-n
>3
o
5S
a
s
o-
o
H §
I
81
2;
" a
•3
"T3
H S.
ffl.K"
O •
o
c
o
H
W
H
O,
Effect type
Candidate critical
studies
Species
POD
type
HECjo
or
HEDjo
POD,
HEC99,
or
HED99a
UFSC
UFis
UFh
UF|oae|
UFdb
TTI7 b
' comp
cRfCor
p-cRfC
(ppm)
cRI'D or
p-cRfD
(mg/kg/day)
Candidate critical effect;
comments [dose-metric]
DuTeaux et al.
(2004b)
Rat
LOAEL
141
10
10
10
10
1
10,000c
0.014
i ability of sperm to fertilize in vitro
HED
66
16
10
3
3
10
1
1,000
0.016
[AUCCBld]
HED
65
42
10
3
3
10
1
1,000
0.042
[TotOxMetabBW34]
HEC
16
9.3
10
3
3
10
1
1,000
0.0093
[AUCCBld] (route-to-route)
HEC
160
43
10
3
3
10
1
1,000
0.043
[TotOxMetabBW34] (route-to-route)
Male reproductive tract effects
Forkert et al. (2002);
Kan etal. (2007)
Mouse
LOAEL
180
10
3
10
10
1
3,000
0.060
Effects on epididymis epithelium
HEC
190
67
10
3
3
10
1
1,000
0.067
[TotMetabBW34]
HEC
321
170
10
3
3
10
1
1,000
0.17
[AUCCBld]
HED
80
73
10
3
3
10
1
1,000
0.073
[TotMetabBW34] (route-to-route)
HED
324
104
10
3
3
10
1
1,000
0.10
[AUCCBld] (route-to-route)
Kumar et al. (2000b)
(2001a)
Rat
LOAEL
45
10
3
10
10
1
3,000
0.015
Testes effects, testicular enzyme
markers, increasing severity from
12-24 weeks
HEC
32
13
10
3
3
10
1
1,000
0.013
[TotMetabBW34]
HEC
91
53
10
3
3
10
1
1,000
0.053
[AUCCBld]
HED
16
16
10
3
3
10
1
1,000
0.016
[TotMetabBW34] (route-to-route)
HED
157
49
10
3
3
10
1
1,000
0.049
[AUCCBld] (route-to-route)
Female reproductive outcomes
Narotsky et al.
(1995) '
Rat
LOAEL
475
1
10
10
10
1
1,000
0.48
Delayed parturition
HED
47
44
1
3
3
10
1
100
0.44
[TotMetabBW34]
HED
350
114
1
3
3
10
1
100
1.1
[AUCCBld]
HEC
98
37
1
3
3
10
1
100
0.37
[TotMetabBW34] (route-to-route)
HEC
363
190
1
3
3
10
1
100
1.9
[AUCCBld] (route-to-route)
-------
Table 5-17. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on PBPK modeled internal
dose-metrics) for candidate critical reproductive effects (continued)
Effect type
Candidate critical
studies
Species
POD
type
HECjo
or
HEDjo
POD,
hec99,
or
HED99a
UFSC
UFis
UFh
UF|oae|
UFdb
TTI7 b
' comp
cRfCor
p-cRfC
(ppm)
cRI'D or
p-cRfD
(mg/kg/day)
Candidate critical effect;
comments [dose-metric]
Reproductive behavior
George et al. (1986)
Rat
LOAEL
389
1
10
10
10
1
1,000
0.39
i mating (both sexes exposed)
HED
85
77
1
3
3
10
1
100
0.77
[TotMetabBW34]
HED
167
52
1
3
3
10
1
100
0.52
[AUCCBld]
HEC
204
71
1
3
3
10
1
100
0.71
[TotMetabBW34] (route-to-route)
HEC
103
60
1
3
3
10
1
100
0.60
[AUCCBld] (route-to-route)
"Applied dose POD adjusted to continuous exposure unless otherwise noted. POD, HEC99, and HED99 have same units as cRfC (ppm) or cRfD (mg/kg/day).
bProduct of individual uncertainty factors, rounded to 3, 10, 30, 100, 300, 1,000, 3,000, or 10,000 (see footnote [c] below).
°EPA's report on the RfC and RfD processes (U.S. EPA, 2002c) recommends not deriving reference values with a composite UF of greater than 3,000; however,
composite UFs exceeding 3,000 are considered here because the derivation of the cRfCs and cRfDs is part of a screening process and the application of the
PBPK model for candidate critical effects reduces the values of some of the individual UFs for the p-cRfCs and p-cRfDs.
UFSC = subchronic-to-chronic UF; UF1S = interspecies UF; UFh = human variability UF; UFioaei = LOAEL-to-NOAEL UF; UFdb = database UF.
Shaded rows represent the p-cRfC or p-cRfD using the preferred PBPK model dose-metric.
-------
Table 5-18. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on PBPK modeled
internal dose-metrics) for candidate critical developmental effects
s
>!
00
o
Or
>s
<
cS'
ss
o
&n
rs
Zn
O
S
s
o-
o
H §
I
81
2;
" a
•3
"T3
H S.
ffl.K"
O •
p
o
c
o
H
W
H
O,
Effect type
Candidate critical
studies
Species
POD
type
HEQo or
hed50
POD,
HEC99,
or
HED99a
UFSC
UFis
UFh
UFloael
UFdb
TJF b
* comp
cRfCor
p-cRfC
(ppm)
cRI'D or
p-cRfD
(mg/kg/day)
Candidate critical effect;
comments [dose-metric]
Pre and postnatal mortality
Healyetal. (1982)
Rat
LOAEL
17
1
3
10
10
1
300
0.057
Resorptions
HEC
16
6.2
1
3
3
10
1
100
0.062
[TotMetabBW34]
HEC
23
14
1
3
3
10
1
100
0.14
[AUCCBld]
HED
8.7
8.5
1
3
3
10
1
100
0.085
[TotMetabBW34] (route-to-route)
HED
73
20
1
3
3
10
1
100
0.20
[AUCCBld] (route-to-route)
Narotsky et al. (1995)
Rat
BMDL
32.2
1
10
10
1
1
100
0.32
Resorptions; BMR = 1% extra
risk
HED
29
28
1
3
3
1
1
10
2.8
[TotMetabBW34]
HED
95
29
1
3
3
1
1
10
2.9
[AUCCBld]
HEC
57
23
1
3
3
1
1
10
2.3
[TotMetabBW34] (route-to-route)
HEC
40
24
1
3
3
1
1
10
2.4
[AUCCBld] (route-to-route)
Pre and postnatal growth
Healyetal. (1982)
Rat
LOAEL
17
1
3
10
10
1
300
0.057
I fetal weight; skeletal effects
HEC
16
6.2
1
3
3
10
1
100
0.062
[TotMetabBW34]
HEC
23
14
1
3
3
10
1
100
0.14
[AUCCBld]
HED
8.7
8.5
1
3
3
10
1
100
0.085
[TotMetabBW34] (route-to-route)
HED
73
20
1
3
3
10
1
100
0.20
[AUCCBld] (route-to-route)
-------
Table 5-18. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on PBPK modeled internal
dose-metrics) for candidate critical developmental effects (continued)
s
>!
00
o
O5
<
cS'
ss
o
c-n
>3
o
5S
a
s
o-
o
H §
I
81
2;
" a
•3
"T3
H S.
ffl.K"
O •
o
c
o
H
W
H
O,
Effect type
Candidate critical
studies
Species
POD
type
HEC50 or
hed50
POD,
hec99,
or
HED99a
UFSC
UFis
UFh
UFloael
UFdb
UF b
* comp
cRfCor
p-cRfC
(ppm)
cRI'D or
p-cRfD
(mg/kg/day)
Candidate critical effect;
comments [dose-metric]
Congenital defects
Johnson et al. (2003)
Rat
BMDL
0.0207
1
10
10
1
1
100
0.00021
Heart malformations (pups);
BMR =1% extra risk;
highest-dose group (1,000-fold
higher than next highest) dropped
to improve model fit
HED
0.0058
0.0052
1
3
3
1
1
10
0.00052
[TotOxMetabBW34]
HED
0.019
0.0017
1
3
3
1
1
10
0.00017
[AUCCBld]
HEC
0.012
0.0037
1
3
3
1
1
10
0.00037
[TotOxMetabBW34]
(route-to-route)
HEC
0.0016
0.00093
1
3
3
1
1
10
0.000093
[AUCCBld] (route-to-route)
Developmental neurotoxicity
Fredriksson et al.
(1993)
Mouse
LOAEL
50
3
10
10
10
1
3,000
0.017
I rearing postexposure; pup
gavage dose
HED
4.2
4.1
3
3
3
10
1
300
0.014
[TotMetabBW34]
HED
27
3.5
3
3
3
10
1
300
0.012
[AUCCBld]
HEC
8.0
3.0
3
3
3
10
1
300
0.010
[TotMetabBW34] (route-to-route)
HEC
3.1
1.8
3
3
3
10
1
300
0.0061
[AUCCBld] (route-to-route)
Taylor et al. (1985)
Rat
LOAEL
45
1
10
10
10
1
1,000
0.045
t exploration postexposure;
estimated dam dose
HED
11
11
1
3
3
10
1
100
0.11
[TotMetabBW34]
HED
30
4.1
1
3
3
10
1
100
0.041
[AUCCBld]
HEC
22
8.4
1
3
3
10
1
100
0.084
[TotMetabBW34] (route-to-route)
-------
Table 5-18. cRfCs and cRfDs (based on applied dose) and p-cRfCs and p-cRfDs (based on PBPK modeled internal
dose-metrics) for candidate critical developmental effects (continued)
ss
S
>!
eP
>s
>s
<
^3
ss
o
o
53
a
§•
>)
s
o
ri
o
s
Effect type
Candidate critical
studies
Species
POD
type
HEC50 or
hed50
POD,
hec99,
or
HED99a
UFSC
UFis
UFh
UFloael
UFdb
UF b
* comp
cRfCor
p-cRfC
(ppm)
cRI'D or
p-cRfD
(mg/kg/day)
Candidate critical effect;
comments [dose-metric]
HEC
3.7
2.2
1
3
3
10
1
100
0.022
[AUCCBld] (route-to-route)
Isaacson and Taylor
(1989)
Rat
LOAEL
16
1
10
10
10
1
1,000
0.016
I myelination in hippocampus;
estimated dam dose
Developmental immunotoxicity
Peden-Adams et al.
(2006)
Mouse
LOAEL
0.37
1
10
10
10
1
1,000
0.00037
i PFC, fDTH; POD is estimated
dam dose (exposure throughout
gestation and lactation + to 3 or 8
wks of age)
00
o
p
H
o
o
2;
•3
"T3
H S.
ffl.R-
O '
P
o
c
o
H
W
H
O,
"Applied dose POD adjusted to continuous exposure unless otherwise noted. POD, HEC99, and HED99 have same units as cRfC (ppm) or cRfD (mg/kg/day).
bProduct of individual uncertainty factors, rounded to 3, 10, 30, 100, 300, 1,000, or 3,000.
UFSC = subchronic-to-chronic UF; UF1S = interspecies UF; UFh = human variability UF; UFioaei = LOAEL-to-NOAEL UF; UFdb = database UF.
Shaded rows represent the p-cRfC or p-cRfD using the preferred PBPK model dose-metric or, in the cases where the PBPK model was not used, the cRfD or
cRfC based on applied dose.
=5
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
variability in the human PBPK model are analyzed quantitatively, along with the uncertainty in
the rodent PBPK dose-metrics as mentioned above, in Section 5.1.4.2.
Because they are derived from rodent internal dose estimates, the HEC and HED are
derived in the same manner independent of the route of administration of the original rodent
study. Therefore, a route-to-route extrapolation from an oral (inhalation) study in rodents to a
HEC (HED) in humans is straight-forward. As shown in Tables 5-13-5-18, route-to-route
extrapolation was performed for a number of endpoints with low cRfCs and cRfDs to derive
p-cRfDs and p-cRfCs.
5.1.3.1.8. Results and Discussion of p-RfCs and p-RfDs for Candidate Critical Effects
Tables 5-13-5-18 present the p-cRfCs and p-cRfDs developed using the PBPK internal
dose-metrics, along with the cRfCs and cRfDs based on applied dose for comparison, for each
health effect domain.
The greatest impact of using the PBPK model was, as expected, for kidney effects, since
as discussed in Sections 3.3 and 3.5, some toxicokinetic data indicate substantially more GSH
conjugation of TCE and subsequent bioactivation of GSH-conjugates in humans relative to rats
or mice. In addition, as discussed in Sections 3.3 and 3.5, the available in vivo data indicate high
interindividual variability in the amount of TCE conjugated with GSH. The overall impact is
that the p-cRfCs and p-cRfDs based on the preferred dose-metric of bioactivated DCVC are
300- to 400-fold lower than the corresponding cRfCs and cRfDs based on applied dose. As
shown in Figure 3-20 in Section 3.5, for this dose-metric there is about a 30- to 100-fold
difference (depending on exposure route and level) between rats and humans in the "central
estimates" of interspecies differences for the fraction of TCE that is bioactivated as DCVC. The
uncertainty in the human central estimate is only on the order of twofold (in either direction),
while that in the rat central estimate is substantially greater, about 10-fold (in either direction).
In addition, the interindividual variability about the human median estimate is on the order of
10-fold (in either direction). However, as noted in Section 3.3.3.2, there are a number of
discrepancies in estimates for the extent of GSH conjugation that may be related to different
analytical methods, and it is possible that GSH conjugation data to which the PBPK model was
calibrated overestimated the extent of DCVG formation by a substantial amount. Thus, there
remain significant uncertainties in the human estimates of GSH conjugation derived from the
PBPK model. Moreover, the estimates of the amount bioactivated are indirect, derived from the
difference between overall GSH conjugation flux and NAcDCVC excretion (see Section
3.5.7.3.1). Therefore, while there is a high degree of confidence in the nephrotoxic hazard posed
This document is a draft for review purposes only and does not constitute Agency policy.
85 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
by TCE, there is less confidence in the p-cRfCs and p-RfDs derived using GSH conjugation
dose-metrics for these effects.
In addition, in two cases in which BMD modeling was employed, using internal
dose-metrics led to a sufficiently different dose-response shape so as to change the resulting
reference value by greater than fivefold. For the Woolhiser et al. (2006) decreased PFC
response, this occurred with the AUC of TCE in blood dose-metric, leading to a p-cRfC 17-fold
higher than thecRfC based on applied dose. However, the model fit for this effect using this
metric was substantially worse than the fit using the preferred metric of Total oxidative
metabolism. Moreover, whereas an adequate fit was obtained with applied dose only with the
highest-dose group dropped, all the dose groups were included when the total oxidative
metabolism dose-metric was used while still resulting in a good model fit. Therefore, it appears
that using this metric resolves some of the low-dose supralinearity in the dose-response curve.
Nonetheless, the overall impact of the preferred metric was minimal, as the p-cRfC based on the
Total oxidative metabolism metric was less than 1.4-fold larger than the cRfC based on applied
dose. The second case in which BMD modeling based on internal doses changed the candidate
reference value by more than fivefold was for resorptions reported by Narotsky et al. (1995).
Here, the p-cRfDs were seven- to eightfold larger than the corresponding cRfD based on applied
dose. However, for applied dose there is substantial uncertainty in the low-dose curvature of the
dose-response curve. This uncertainty persisted with the use of internal dose-metrics, so the
BMD remains somewhat uncertain (see figures in Appendix F).In the remaining cases, which
generally involved the "generic" dose-metrics of total metabolism and AUC of TCE in blood, the
p-cRfCs and p-cRfDs were within fivefold of the corresponding cRfC or cRfD based on applied
dose, with the vast majority within threefold. This suggests that the standard UFs for inter and
intraspecies pharmacokinetic variability are fairly accurate in capturing these differences for
these TCE studies.
5.1.4. Uncertainties in cRfCs and cRfDs
5.1.4.1.1. Qualitative Uncertainties
An underlying assumption in deriving reference values for noncancer effects is that the
dose-response relationship for these effects has a threshold. Thus, a fundamental uncertainty is
the validity of that assumption. For some effects, in particular effects on very sensitive processes
(e.g., developmental processes) or effects for which there is a nontrivial background level and
even small exposures may contribute to background disease processes in more susceptible
This document is a draft for review purposes only and does not constitute Agency policy.
86 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
people, a practical threshold (i.e., a threshold within the range of environmental exposure levels
of regulatory concern) may not exist.
Nonetheless, under the assumption of a threshold, the desired exposure level to have as a
reference value is the maximum level at which there is no appreciable risk for an adverse effect
in (nonnegligible) sensitive subgroups (of humans). However, because it is not possible to know
what this level is, "uncertainty factors" are used to attempt to address quantitatively various
aspects, depending on the data set, of qualitative uncertainty.
First there is uncertainty about the "point of departure" for the application of UFs.
Conceptually, the POD should represent the maximum exposure level at which there is no
appreciable risk for an adverse effect in the study population under study conditions (i.e., the
threshold in the dose-response relationship). Then, the application of the relevant UFs is
intended to convey that exposure level to the corresponding exposure level for sensitive human
subgroups exposed continuously for a lifetime. In fact, it is again not possible to know that
exposure level even for a laboratory study because of experimental limitations (e.g., the power to
detect an effect, dose spacing, measurement errors, etc.), and crude approximations like the
NOAEL or a BMDL are used. If a LOAEL is used as the POD, the LOAEL-to-NOAEL UF is
applied as an adjustment factor to get a better approximation of the desired exposure level
(threshold), but the necessary extent of adjustment is unknown.
If a BMDL is used as the POD, there are uncertainties regarding the appropriate
dose-response model to apply to the data, but these should be minimal if the modeling is in the
observable range of the data. There are also uncertainties about what BMR to use to best
approximate the desired exposure level (threshold, see above). For continuous endpoints, in
particular, it is often difficult to identify the level of change that constitutes the "cut-point" for an
adverse effect. Sometimes, to better approximate the desired exposure level, a BMR somewhat
below the observable range of the data is selected. In such cases, the model uncertainty is
increased, but this is a trade-off to reduce the uncertainty about the POD not being a good
approximation for the desired exposure level.
For each of these types of PODs, there are additional uncertainties pertaining to
adjustments to the administered exposures (doses). Typically, administered exposures (doses)
are converted to equivalent continuous exposures (daily doses) over the study exposure period
under the assumption that the effects are related to concentration x time, independent of the daily
(or weekly) exposure regimen (i.e., a daily exposure of 6 hours to 4 ppm is considered equivalent
to 24 hours of exposure to 1 ppm). However, the validity of this assumption is generally
unknown, and, if there are dose-rate effects, the assumption of C x t equivalence would tend to
bias the POD downwards. Where there is evidence that administered exposure better correlates
This document is a draft for review purposes only and does not constitute Agency policy.
87 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
to the effect than equivalent continuous exposure averaged over the study exposure period (e.g.,
visual effects), administered exposure was not adjusted. For the PBPK analyses in this
assessment, the actual administered exposures are taken into account in the PBPK modeling, and
equivalent daily values (averaged over the study exposure period) for the dose-metrics are
obtained (see above, Section 5.1.3.2). Additional uncertainties about the PBPK-based estimates
include uncertainties about the appropriate dose-metric for each effect, although for some effects
there was better information about relevant dose-metrics than for others (see Section 5.1.3.1).
Furthermore, as discussed in Section 3.3.3.2, there remains substantial uncertainty in the
extrapolation of GSH conjugation from rodents to humans due to limitations in the available
data.
Second, there is uncertainty about the UFs. The human variability UF is to some extent
an adjustment factor because for more sensitive people, the dose-response relationship shifts to
lower exposures. However, there is uncertainty about the extent of the adjustment required, i.e.,
about the distribution of human susceptibility. Therefore, in the absence of data on a more
sensitive population(s) or on the distribution of susceptibility in the general population, an UF of
10 is generally used, in part for pharmacokinetic variability and in part for pharmacodynamic
variability. The PBPK analyses in this assessment attempt to account for the pharmacokinetic
portion of human variability using human data on pharmacokinetic variability. A quantitative
uncertainty analysis of the PBPK-derived dose-metrics used in the assessment is presented in
Section 5.1.4.2 below. There is still uncertainty regarding the susceptible subgroups for TCE
exposure and the extent of pharmacodynamic variability.
If the data used to determine a particular POD are from laboratory animals, an
interspecies extrapolation UF is used. This UF is also to some extent an adjustment factor for the
expected scaling for toxicologically-equivalent doses across species (i.e., according to body
weight to the 3/4 power for oral exposure). However, there is also uncertainty about the true
extent of interspecies differences for specific noncancer effects from specific chemical
exposures. Often, the "adjustment" component of this UF has been attributed to
pharmacokinetics, while the "uncertainty" component has been attributed to pharmacodynamics,
but as discussed above in Section 5.1.3.1, this is not the only interpretation supported. For oral
exposures, the standard value for the interspecies UF is 10, which can be viewed as breaking
down (approximately) to a factor of three for the "adjustment" (nominally pharmacokinetics) and
a factor of three for the "uncertainty" (nominally pharmacodynamics). For inhalation exposures,
no adjustment across species is generally assumed for fixed air concentrations (ppm
equivalence), and the standard value for the interspecies UF is 3 reflects "uncertainty"
(nominally pharmacodynamics only). The PBPK analyses in this assessment attempt to account
This document is a draft for review purposes only and does not constitute Agency policy.
88 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
for the "adjustment" portion of interspecies extrapolation using rodent pharmacokinetic data to
estimate internal doses for various dose-metrics. With respect to the "uncertainty" component,
quantitative uncertainty analyses of the PBPK-derived dose-metrics used in the assessment are
presented in Section 5.1.4.2 below. However, these only address the pharmacokinetic
uncertainties in a particular dose-metric, and there is still uncertainty regarding the true
dose-metrics. Nor do the PBPK analyses address the uncertainty in either cross-species
pharmacodynamic differences (i.e., about the assumption that equal doses of the appropriate
dose-metric convey equivalent risk across species for a particular endpoint from a specific
chemical exposure) or in cross-species pharmacokinetic differences not accounted for by the
PBPK model dose-metrics (e.g., departures from the assumed interspecies scaling of clearance of
the active moiety, in the cases where only its production is estimated). A value of 3 is typically
used for the "uncertainty" about cross-species differences, and this generally represents true
uncertainty because it is usually unknown, even after adjustments have been made to account for
the expected interspecies differences, whether humans have more or less susceptibility, and to
what degree, than the laboratory species in question.
If only subchronic data are available, the subchronic-to-chronic UF is to some extent an
adjustment factor because, if the effect becomes more severe with increasing exposure, then
chronic exposure would shift the dose-response relationship to lower exposures. However, the
true extent of the shift is unknown.
Sometimes a database UF is also applied to address limitations or uncertainties in the
database. The overall database for TCE is quite extensive, with studies for many different types
of effects, including 2-generation reproductive studies, as well as neurological, immunological,
and developmental immunological studies. In addition, there were sufficient data to develop a
reliable PBPK model to estimate route-to-route extrapolated doses for some candidate critical
effects for which data were only available for one route of exposure. Thus, there is a high degree
of confidence that the TCE database was sufficient to identify some sensitive endpoints.
5.1.4.1.2. Quantitative Uncertainty Analysis of Physiologically Based Pharmacokinetic
(PBPK) Model-Based Dose-metrics for Lowest-Observed-Adverse-Effect Level (LOAEL)
or No-Observed-Adverse-Effect Level (NOAEL)-Based Points of Departure (PODs)
The Bayesian analysis of the PBPK model for TCE generates distributions of uncertainty
and variability in the internal dose-metrics that can be readily used for characterizing the
uncertainty and variability in the PBPK model-based derivations of the HEC and HED.
However, in the primary analysis, a number of simplifications are made including (1) use of
This document is a draft for review purposes only and does not constitute Agency policy.
89 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
median estimates for rodent internal doses and (2) expressing the "sensitive human" HEC and
HED in terms of combined uncertainty and variability. Therefore, a 2-dimensional quantitative
uncertainty and variability analysis is performed, the objective of which is to characterize the
impact of these assumptions.
As shown in Figure 5-4, the overall approach taken for the uncertainty analysis is similar
to that used for the point estimates except for the carrying through of separate uncertainty and
variability distributions throughout the analysis. In particular, to address simplification
(1), above, the distribution of rodent internal dose estimates is carried through; and to address
simplification (2), above, uncertainty and variability distributions in human internal dose
estimates are kept distinct.
Because of a lack of tested software and limitations of time and resources, this analysis
was not performed for idPODs based on BMD modeling, and was only performed for idPODs
derived from a LOAEL or NOAEL. However, for those endpoints for which BMD modeling
was performed, for the purposes of this uncertainty analysis, an alternative idPOD was used
based on the study LOAEL or NOAEL.
In brief, the methodology involves an iterative process of sampling from three separate
distributions - the uncertainty distribution of rodent PBPK model parameters, the uncertainty
distribution of human population PBPK parameters, and the variability distribution of human
individual PBPK model parameters - the latter two of which are related hierarchically. For a
sample from the rodent parameter distribution, the corresponding idPOD is calculated. Then, an
individual is sampled from a human population distribution, which itself is sampled from the
uncertainty distribution of population parameters. For this individual, a human equivalent
exposure (HEC or HED) corresponding to the idPOD is derived by interpolation. Taking
multiple individuals from this population, a HEC or HED corresponding to the median and
99th percentile individuals is then derived. Repeating this process (starting again with a sample
from the rodent distribution) results in two distributions (both reflecting uncertainty): one of
"typical" individuals represented by the distribution of population medians, and one of
"sensitive" individuals represented by the distribution of an upper percentile of the population
(e.g., 99th percentile). This uncertainty reflects both uncertainty in the rodent internal dose and
uncertainty in the human population parameters. Thus, for selected quantiles of the population
and level of confidence (e.g., Xth percentile individual at Yth% confidence), the interpretation is
that at the resulting HEC or HED, there is Y% confidence that X% of the population has an
internal dose less than that of the rodent in the toxicity study.
As shown in Tables 5-19-5-23, the HEC99 and HED99 derived using the rodent median
dose-metrics and the combined uncertainty and variability in human dose-metrics is generally
This document is a draft for review purposes only and does not constitute Agency policy.
90 DRAFT—DO NOT CITE OR QUOTE
-------
[distribution (combined
incertainty and variability)
Rodent
non-cancer
study
experimental
paradigm
Rodent
model
parameters
^distribution
fixed
PBPK
model
Rodent
non-cancer
study
responses
Rodent
internal
dose
istribution
fixed
Dose-Response Model
or
LOAEUNOAEL
.distribution
idPOD
LOAEL
orNOAEL
(internal
dose unit
0.1-2000 ppm
in air or
0.1-2000
mg/kg-d
continuous
exposure
fixed
Human
model
parameters
PBPK
model
invert functions of dose
or concentration
istribution
[distribution (separate
ncertainty and variability)
Human
internal
dose
{distribution of functions
f dose or concentration
Human
dose or
concent ratio r
Uncertainty distribution\^at idPOD
of population median
Uncertainty distribution
of population 99th
percentile
Typical
human
equivalent
Sensitive
Human
equivalent
1
2
3
4
distribution distribution
near (within 1.3-fold of) the median confidence level estimate of the HEC and HED for the
99th percentile individual. Therefore, the interpretation is that there is about 50% confidence that
human exposure at the HEC99 or HED99 will, in 99% of the human population, lead to an internal
This document is a draft for review purposes only and does not constitute Agency policy.
91 DRAFT—DO NOT CITE OR QUOTE
-------
1 Figure 5-4. Flow-chart for uncertainty analysis of HECs and HEDs derived
2 using PBPK model-based dose-metrics. Square nodes indicate point values,
3 circle nodes indicate distributions, and the inverted triangle indicates a
4 (deterministic) functional relationship.
5
This document is a draft for review purposes only and does not constitute Agency policy.
92 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 5-19. Comparison of "sensitive individual" HECs or HEDs for
2 neurological effects based on PBPK modeled internal dose-metrics at
3 different levels of confidence and sensitivity, at the NOAEL or LOAEL
4
Candidate critical effect
Candidate critical study
(species)
POD
type
Ratio
HEC/DS0:
hec/d99
HECy or HEDa
[Dose-metric]
*
II
V©
a: =99,
median
a: =99,
951cb
Neurological
Trigeminal nerve effects
Ruijtenet al. (1991) (human)
HEC
2.62
5.4
5.4
2.6
[TotMetabBW34]
HEC
1.68
8.3
8.3
4.9
[AUCCBld]
HED
1.02
7.3
7.2
3.8
[TotMetabBW34] (rtr)
HED
4.31
14
16
8.0
[AUCCBld] (rtr)
Demyelination in hippocampus
Isaacson et al. (1990) (rat)
HED
1.02
9.21
9.20
7.39
[TotMetabBW34]
HED
7.20
4.29
5.28
2.52
[AUCCBld]
HEC
2.59
7.09
6.77
4.94
[TotMetabBW34] (rtr)
HEC
1.68
2.29
2.42
0.606
[AUCCBld] (rtr)
Changes in wakefulness
Anto et al. (1994) (rat)
HEC
2.65
4.79
4.86
2.37
[TotMetabBW34]
HEC
1.67
9
9.10
4.63
[AUCCBld]
HED
1.02
6.46
6.50
3.39
[TotMetabBW34] (rtr)
HED
4.25
15.2
18.0
8.33
[AUCCBld] (rtr)
i regeneration of sciatic nerve
Kjellstrand et al. (1987) (rat)
HEC
2.94
93.1
93.6
38.6
[TotMetabBW34]
HEC
1.90
257
266
114
[AUCCBld]
HED
1.13
97.1
96.8
43.4
[TotMetabBW34] (rtr)
HED
3.08
142
147
78.0
[AUCCBld] (rtr)
i regeneration of sciatic nerve
Kjellstrand et al. (1987)
(mouse)
HEC
3.16
120
125
48.8
[TotMetabBW34]
HEC
1.84
108
111
59.7
[AUCCBld]
HED
1.21
120
121
57.0
[TotMetabBW34] (rtr)
HED
2.13
75.8
79.1
53.4
[AUCCBld] (rtr)
Degeneration of dopaminergic
neurons
Gash et al. (2008) (rat)
HED
1.06
53
53.8
17.1
[TotMetabBW34]
HED
2.98
192
199
94.7
[AUCCBld]
HEC
2.70
46.8
47.9
14.2
[TotMetabBW34] (rtr)
5
6 HEC99 = the 99th percentile of the combined human uncertainty and variability distribution of continuous exposure
7 concentrations that lead to the (fixed) median estimate of the rodent internal dose at the POD.
8 HEC99.median (or HEC99 95icb) = the median (or 95th percentile lower confidence bound) estimate of the uncertainty
9 distribution of continuous exposure concentrations for which the 99th percentile individual has an internal dose
10 less than the (uncertain) rodent internal dose at the POD.
11 rtr = route-to-route extrapolation using PBPK model and the specified dose-metric.
12 Shaded rows denote results for the primary dose-metric.
13
This document is a draft for review purposes only and does not constitute Agency policy.
93 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 5-20. Comparison of "sensitive individual" HECs or HEDs for kidney
2 and liver effects based on PBPK modeled internal dose-metrics at different
3 levels of confidence and sensitivity, at the NOAEL or LOAEL
4
Candidate critical effect
Candidate critical study
(species)
POD
type
Ratio
HEC/DS0:
hec/d99
HECV or HEDV
[Dose-metric]
*
II
V©
a: =99,
median
a: =99,
951cb
Kidney
Meganucleocytosis [NOAEL] *
Maltoni et al. (1986) (rat
inhalation)
HEC
7.53
0.0233
0.0260
0.00366
[ABioactDCVCBW34]
HEC
7.70
0.0364
0.0411
0.00992
[AMetGSHBW34]
HEC
2.57
8.31
7.97
4.03
[TotMetabBW34]
HED
9.86
0.0140
0.0156
0.00216
[ABioactDCVCBW34]
(rtr)
HED
9.83
0.0223
0.0242
0.00597
[AMetGSHBW34] (rtr)
HED
1.02
10.6
10.7
5.75
[TotMetabBW34] (rtr)
Toxic nephrosis
NCI (1976) (mouse)
HED
9.51
0.30
0.32
0.044
[AMetGSHBW34]
HED
1.05
48
48.9
16.2
[TotMetabBW34]
HEC
7.78
0.50
0.514
0.0703
[AMetGSHBW34] (rtr)
HEC
2.67
42
43.5
13.7
[TotMetabBW34] (rtr)
Toxic nephropathy [LOAEL]*
NTP (1988) (rat)
HED
9.75
0.121
0.126
0.0177
[ABioactDCVCBW34]
HED
9.64
0.193
0.210
0.0379
[AMetGSHBW34]
HED
1.03
33.1
33.1
11.1
[TotMetabBW34]
HEC
7.55
0.201
0.204
0.0269
[ABioactDCVCBW34]
(rtr)
HEC
7.75
0.314
0.353
0.0676
[AMetGSHBW34] (rtr)
HEC
2.59
28.2
27.2
8.77
[TotMetabBW34] (rtr)
Meganucleocytosis [NOAEL] *
Maltoni et al. (1986) (rat oral)
HED
9.85
0.0133
0.0145
0.00158
[ABioactDCVCBW34]
HED
9.86
0.0214
0.0249
0.00366
[AMetGSHBW34]
HED
1.02
8.7
8.57
4.95
[TotMetabBW34]
HEC
7.55
0.022
0.0249
0.00256
[ABioactDCVCBW34]
(rtr)
HEC
7.71
0.0349
0.0424
0.00615
[AMetGSHBW34] (rtr)
HEC
2.60
6.66
6.31
3.70
[TotMetabBW34] (rtr)
t kidney/body weight ratio
[NOAEL]*
Kjellstrand et al. (1983a)
(mouse)
HEC
7.69
0.111
0.103
0.00809
[AMetGSHBW34]
HEC
2.63
34.5
33.7
13.5
[TotMetabBW34]
HED
9.78
0.068
0.00641
0.00497
[AMetGSHBW34] (rtr)
HED
1.03
39.9
39.2
17.9
[TotMetabBW34] (rtr)
This document is a draft for review purposes only and does not constitute Agency policy.
94 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
Table 5-20. Comparison of "sensitive individual" HECs or HEDs for kidney
and liver effects based on PBPK modeled internal dose-metrics at different
levels of confidence and sensitivity, at the NOAEL or LOAEL (continued)
Candidate critical effect
Candidate critical study
(species)
POD
type
Ratio
HEC/DS0:
hec/d99
HECv or HEDy
[Dose-metric]
*
II
V©
a: =99,
median
a: =99,
951cb
t kidney/body weight ratio
[NOAEL]3
Woolhiser et al. (2006) (rat)
HEC
7.53
0.0438
0.0481
0.00737
[ABioactDCVCBW34]
HEC
7.70
0.0724
0.0827
0.0179
[AMetGSHBW34]
HEC
2.54
16.1
15.2
7.56
[TotMetabBW34]
HED
9.84
0.0264
0.0282
0.00447
[ABioactDCVCBW34]
(rtr)
HED
9.81
0.0444
0.0488
0.0111
[AMetGSHBW34] (rtr)
HED
1.02
19.5
19.2
10.5
[TotMetabBW34] (rtr)
Liver
t liver/body weight ratio
[LOAEL]3
Kjellstrand et al. (1983a)
(mouse)
HEC
2.85
16.2
16.3
6.92
[AMetLivlBW34]
HEC
3.63
40.9
38.1
15.0
[TotOxMetabBW34]
HED
1.16
14.1
14.1
5.85
[AMetLivlBW34] (rtr)
HED
1.53
40.1
39.4
17.9
[TotOxMetabBW34]
(rtr)
t liver/body weight ratio
[NOAEL]3
Woolhiser et al. (2006) (rat)
HEC
2.86
20.7
21.0
11.0
[AMetLivlBW34]
HEC
2.94
18.2
17.1
8.20
[TotOxMetabBW34]
HED
1.20
17.8
17.7
9.94
[AMetLivlBW34] (rtr)
HED
1.21
19.6
19.3
10.5
[TotOxMetabBW34]
(rtr)
t liver/body weight ratio
[LOAEL]3
Buben and O'Flaherty (1985)
(mouse)
HED
1.14
8.82
8.95
4.17
[AMetLivlBW34]
HED
1.14
9.64
9.78
5.28
[TotOxMetabBW34]
HEC
2.80
10.1
9.97
4.83
[AMetLivlBW34] (rtr)
HEC
3.13
7.83
7.65
4.23
[TotOxMetabBW34]
(rtr)
3BMDL used for p-cRfC or p-cRfD, but LOAEL or NOAEL (as noted) used for uncertainty analysis.
HEC99 = the 99th percentile of the combined human uncertainty and variability distribution of continuous exposure
concentrations that lead to the (fixed) median estimate of the rodent internal dose at the POD.
HEC99 median (or HEC99,95icb) = the median (or 95th percentile lower confidence bound) estimate of the uncertainty
distribution of continuous exposure concentrations for which the 99th percentile individual has an internal dose
less than the (uncertain) rodent internal dose at the POD.
rtr = route-to-route extrapolation using PBPK model and the specified dose-metric.
Shaded rows denote results for the primary dose-metric.
This document is a draft for review purposes only and does not constitute Agency policy.
95 DRAFT—DO NOT CITE OR QUOTE
-------
Table 5-21. Comparison of "sensitive individual" HECs or HEDs for
immunological effects based on PBPK modeled internal dose-metrics at
different levels of confidence and sensitivity, at the NOAEL or LOAEL
Candidate critical effect
Candidate critical study
(species)
POD
type
Ratio
HEC/DS0:
hec/d99
HECV or HEDV
[Dose-metric]
*
II
V©
a: =99,
median
a: =99,
951cb
Immunological
Changes in iimnunoreactive
organs—liver (including
sporadic necrosis in hepatic
lobules), spleen
Kaneko et al. (2000) (mouse)
HEC
2.65
36.7
38.3
16.0
[TotMetabBW34]
HEC
1.75
68.9
70.0
37.1
[AUCCBld]
HED
1.04
42.3
43.3
21.3
[TotMetabBW34] (rtr)
HED
3.21
56.5
59.0
39.8
[AUCCBld] (rtr)
t anti-dsDNA and anti-ssDNA
Abs (early markers for SLE); J,
thymus weight
Keil et al. (2009) (mouse)
HED
1.02
0.0482
0.0483
0.0380
[TotMetabBW34]
HED
12.1
0.0161
0.0189
0.00363
[AUCCBld]
HEC
2.77
0.0332
0.0337
0.0246
[TotMetabBW34] (rtr)
HEC
1.69
0.00821
0.00787
0.00199
[AUCCBld] (rtr)
| PFC response [NOAEL]3
Woolhiser et al. (2006) (rat)
HEC
2.54
16.1
15.2
7.56
[TotMetabBW34]
HEC
1.73
59.6
60.1
26.2
[AUCCBld]
HED
1.02
19.5
19.2
10.5
[TotMetabBW34] (rtr)
HED
3.21
52
55.9
33.0
[AUCCBld] (rtr)
i stem cell bone marrow
recolonization; J, cell-mediated
response to sRBC
Sanders et al. (1982b)
(mouse)
HED
1.02
2.48
2.48
1.94
[TotMetabBW34]
HED
10.5
0.838
0.967
0.187
[AUCCBld]
HEC
2.77
1.72
1.75
1.28
[TotMetabBW34] (rtr)
HEC
1.68
0.43
0.412
0.103
[AUCCBld] (rtr)
5
6 aBMDL used for p-cRfC or p-cRfD, but LOAEL or NOAEL (as noted) used for uncertainty analysis.
8 HEC99 = the 99th percentile of the combined human uncertainty and variability distribution of continuous exposure concentrations that lead to the
9 (fixed) median estimate of the rodent internal dose at the POD.
10 HEC99 median (or HECgg.gsicb) = the median (or 95th percentile lower confidence bound) estimate of the uncertainty distribution of continuous
11 exposure concentrations for which the 99th percentile individual has an internal dose less than the (uncertain) rodent internal dose at the POD.
12 rtr = route-to-route extrapolation using PBPK model and the specified dose-metric.
13 Shaded rows denote results for the primary dose-metric.
14
This document is a draft for review purposes only and does not constitute Agency policy.
96 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 5-22. Comparison of "sensitive individual" HECs or HEDs for
2 reproductive effects based on PBPK modeled internal dose-metrics at
3 different levels of confidence and sensitivity, at the NOAEL or LOAEL
4
Candidate critical effect
Candidate critical study
(species)
POD
type
Ratio
HEC/DS0:
hec/d99
HECV or HEDV
[Dose-metric]
*
II
V©
a: =99,
median
a: =99,
951cb
Reproductive
Hyperzoospermia
Cilia et al. (1996) (human)
HEC
2.78
0.50
0.53
0.25
[TotMetabBW34]
HEC
1.68
0.83
0.83
0.49
[AUCCBld]
HED
1.02
0.73
0.71
0.37
[TotMetabBW34] (rtr)
HED
9.69
1.6
2.0
0.92
[AUCCBld] (rtr)
I fertilization
Xu et al. (2004) (mouse)
HEC
2.85
66.6
72.3
26.6
[TotMetabBW34]
HEC
1.89
170
171
97.1
[AUCCBld]
HED
1.09
73.3
76.9
32.9
[TotMetabBW34] (rtr)
HED
3.11
104
109
67.9
[AUCCBld] (rtr)
Multiple sperm effects,
testicular enzyme markers
Kumar et al. (2001a; 2000b)
(rat)
HEC
2.53
12.8
12.2
6.20
[TotMetabBW34]
HEC
1.72
53.2
54.4
23.2
[AUCCBld]
HED
1.02
15.8
15.7
8.60
[TotMetabBW34] (rtr)
HED
3.21
48.8
52.6
30.6
[AUCCBld] (rtr)
i ability of sperm to fertilize in
vitro
DuTeaux et al. (2004b) (rat)
HED
4.20
15.6
18.1
4.07
[AUCCBld]
HED
1.57
41.7
41.9
32.0
[TotOxMetabBW34]
HEC
1.67
9.3
10.1
2.09
[AUCCBld] (rtr)
HEC
3.75
42.5
55.6
39.1
[TotOxMetabBW34] (rtr)
Effects on epididymis
epithelium
Forkert et al. (2002); Kan
et al. (2007) (mouse)
HEC
2.85
66.6
72.3
26.6
[TotMetabBW34]
HEC
1.89
170
171
97.1
[AUCCBld]
HED
1.09
73.3
76.9
32.9
[TotMetabBW34] (rtr)
HED
3.11
104
109
67.9
[AUCCBld] (rtr)
Testes effects
Kumar et al. (2001a; 2000b)
(rat)
HEC
2.53
12.8
12.2
6.20
[TotMetabBW34]
HEC
1.72
53.2
54.4
23.2
[AUCCBld]
HED
1.02
15.8
15.7
8.60
[TotMetabBW34] (rtr)
HED
3.21
48.8
52.6
30.6
[AUCCBld] (rtr)
Delayed parturition
Narotsky et al. (1995) (rat)
HED
1.06
44.3
43.9
15.1
[TotMetabBW34]
HED
3.07
114
119
47.7
[AUCCBld]
HEC
2.66
36.9
35.3
11.6
[TotMetabBW34] (rtr)
HEC
1.91
190
197
48.1
[AUCCBld] (rtr)
5
This document is a draft for review purposes only and does not constitute Agency policy.
97 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 5-22. Comparison of "sensitive individual" HECs or HEDs for
2 reproductive effects based on PBPK modeled internal dose-metrics at
3 different levels of confidence and sensitivity, at the NOAEL or LOAEL
4 (continued)
5
Candidate critical effect
Candidate critical study
(species)
POD
type
Ratio
HEC/DS0:
hec/d99
HECv or HEDy
[Dose-metric]
*
II
V©
a: =99,
median
a: =99,
951cb
i mating (both sexes exposed)
George et al. (1986) (rat)
HED
1.10
77.4
77.1
34.2
[TotMetabBW34]
HED
3.21
51.9
55.8
14.7
[AUCCBld]
HEC
2.86
71.1
70.0
29.5
[TotMetabBW34] (rtr)
HEC
1.73
59.5
63.3
8.14
[AUCCBld] (rtr)
6
7 HEC99 = the 99th percentile of the combined human uncertainty and variability distribution of continuous exposure
8 concentrations that lead to the (fixed) median estimate of the rodent internal dose at the POD.
9 HEC99 median (or HEC99,95icb) = the median (or 95th percentile lower confidence bound) estimate of the uncertainty
10 distribution of continuous exposure concentrations for which the 99th percentile individual has an internal dose
11 less than the (uncertain) rodent internal dose at the POD.
12 rtr = route-to-route extrapolation using PBPK model and the specified dose-metric.
13 Shaded rows denote results for the primary dose-metric.
14
This document is a draft for review purposes only and does not constitute Agency policy.
98 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 5-23. Comparison of "sensitive individual" HECs or HEDs for
2 developmental effects based on PBPK modeled internal dose-metrics at
3 different levels of confidence and sensitivity, at the NOAEL or LOAEL
4
Candidate critical effect
Candidate critical study
(species)
POD
type
Ratio
HEC/D50:
HEC/D99
HEC y or HED.y
[Dose-metric]
X
II
v.O
A" =95,
median
A" =95,
951cb
Developmental
Resorptions
Healy et al. (1982) (rat)
HEC
2.58
6.19
6.02
3.13
[TotMetabBW34]
HEC
1.69
13.7
13.9
7.27
[AUCCBld]
HED
1.02
8.5
8.50
4.61
[TotMetabBW34] (rtr)
HED
3.68
19.7
22.4
11.5
[AUCCBld] (rtr)
Resorptions [LOAEL]3
Narotsky et al. (1995) (rat)
HED
1.06
44.3
43.9
15.1
[TotMetabBW34]
HED
3.07
114
119
47.7
[AUCCBld]
HEC
2.66
36.9
35.3
11.6
[TotMetabBW34] (rtr)
HEC
1.91
190
197
48.1
[AUCCBld] (rtr)
J. fetal weight; skeletal effects
Healy et al. (1982) (rat)
HEC
2.58
6.19
6.02
3.13
[TotMetabBW34]
HEC
1.69
13.7
13.9
7.27
[AUCCBld]
HED
1.02
8.5
8.50
4.61
[TotMetabBW34] (rtr)
HED
3.68
19.7
22.4
11.5
[AUCCBld] (rtr)
Heart malformations (pups)
[LOAEL]3
Johnson et al. (2003) (rat)
HED
1.02
0.012
0.012
0.0102
[TotOxMetabBW34]
HED
11.6
0.00382
0.00476
0.00112
[AUCCBld]
HEC
2.75
0.00848
0.00866
0.00632
[TotOxMetabBW34] (rtr)
HEC
1.70
0.00216
0.00221
0.000578
[AUCCBld] (rtr)
J. rearing postexposure
Fredriksson et al. (1993)
(mouse)
HED
1.02
4.13
4.19
2.22
[TotMetabBW34]
HED
7.69
3.46
4.21
0.592
[AUCCBld]
HEC
2.71
2.96
2.96
1.48
[TotMetabBW34] (rtr)
HEC
1.68
1.84
1.81
0.302
[AUCCBld] (rtr)
t exploration postexposure
Taylor et al. (1985) (rat)
HED
1.02
10.7
10.7
8.86
[TotMetabBW34]
HED
7.29
4.11
5.08
1.16
[AUCCBld]
HEC
2.57
8.36
7.94
5.95
[TotMetabBW34] (rtr)
HEC
1.68
2.19
2.31
0.580
[AUCCBld] (rtr)
5
6 aBMDL used for p-cRfC or p-cRfD, but LOAEL or NOAEL (as noted) used for uncertainty analysis.
7 HEC99 = the 99th percentile of the combined human uncertainty and variability distribution of continuous exposure
8 concentrations that lead to the (fixed) median estimate of the rodent internal dose at the POD.
9 HEC99.median (or HEC99 95icb) = the median (or 95th percentile lower confidence bound) estimate of the uncertainty
10 distribution of continuous exposure concentrations for which the 99th percentile individual has an internal dose
11 less than the (uncertain) rodent internal dose at the POD.
12 rtr = route-to-route extrapolation using PBPK model and the specified dose-metric.
13 Shaded rows denote results for the primary dose-metric.
This document is a draft for review purposes only and does not constitute Agency policy.
99 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
dose less than or equal to that in the subjects (rodent or human) exposed at the POD in the
corresponding study.
In several cases, the uncertainty, as reflected in the ratio between the 95% and 50%
confidence bounds on the 99th percentile individual, was rather high (e.g., >5-fold), and reflected
primarily uncertainty in the rodent internal dose estimates, discussed previously in Section 3.5.7.
The largest uncertainties (ratios between 95% to 50% confidence bounds of 8- to 10-fold) were
for kidney effects in mice using the AMetGSHBW34 dose-metric (Kjellstrand et al., 1983a; NCI,
1976). More moderate uncertainties (ratios between 95% to 50% confidence bounds of five- to
eightfold) were evident in some oral studies using the AUCCBld dose-metric (Fredriksson et al.,
1993; George et al., 1986; Keil et al., 2009; Sanders et al., 1982b), as well as in studies reporting
kidney effects in rats in which the ABioactDCVCBW34 or AMetGSHBW34 dose-metrics were
used (Maltoni et al., 1986; NTP, 1988; Woolhiser et al., 2006). Therefore, in these cases, a POD
that is protective of the 99th percentile individual at a confidence level higher than 50% could be
as much as an order of magnitude lower.
For comparison, Tables 5-19 and 5-23 also show the ratios of the overall 50th percentile
to the overall 99th percentile HECs and HEDs, reflecting combined human uncertainty and
variability at the median study/endpoint idPOD. The smallest ratios (up to 1.2-fold) are for total,
oxidative, and hepatic oxidative metabolism dose-metrics from oral exposures, due to the large
hepatic first-pass effect resulting in virtually all of the oral intake being metabolized before
systemic circulation. Conversely, the large hepatic first-pass results in high variability in the
blood concentration of TCE following oral exposures, with ratios up to 12-fold at low exposures
(e.g., 90 vs. 99% first-pass would result in amounts metabolized differing by about 10% but TCE
blood concentrations differing by about 10-fold). From inhalation exposures, there is moderate
variability in these metrics, about two- to threefold. For GSH conjugation and bioactivated
DCVC, however, variability is high (eight- to 10-fold) for both exposure routes, which follows
from the incorporation in the PBPK model analysis of the data from Lash et al. (1999a) showing
substantial interindividual variability in GSH conjugation in humans.
Finally, it is important to emphasize that this analysis only addresses pharmacokinetic
uncertainty and variability, so other aspects of extrapolation addressed in the UFs (e.g., LOAEL
to NOAEL, subchronic to chronic, and pharmacodynamic differences), discussed above, are not
included in the level of confidence.
This document is a draft for review purposes only and does not constitute Agency policy.
100 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
5.1.5. Summary of Noncancer Reference Values
5.1.5.1.1. Preferred Candidate Reference Values (cRfCs, cRfD, p-cRfCs and p-cRfDs) for
Candidate Critical Effects
The candidate critical effects that yielded the lowest p-cRfC or p-cRfD for each type of
effect, based on the primary dose-metric, are summarized in Tables 5-24 (p-cRfCs) and 5-25
(p-cRfDs). These results are extracted from Tables 5-13-5-18. In cases where a route-to-route
extrapolated p-cRfC (p-cRfD) is lower than the lowest p-cRfC (p-cRfD) from an inhalation
(oral) study, both values are presented in the table. In addition, if there is greater than usual
uncertainty associated with the lowest p-cRfC or p-cRfD for a type of effect, then the endpoint
with the next lowest value is also presented. Furthermore, given those selections, the same sets
of critical effects and studies are displayed across both tables, with the exception of two oral
studies for which route-to-route extrapolation was not performed. Tables 5-24 and 5-25 are
further summarized in Tables 5-26 and 5-27 to present the overall preferred p-cRfC and p-cRfD
for each type of noncancer effect. The purpose of these summary tables is to show the most
sensitive endpoints for each type of effect and the apparent relative sensitivities (based on
reference value estimates) of the different types of effects.
For neurological, kidney, immunological, and developmental effects, the lowest p-cRfCs
were derived from oral studies by route-to-route extrapolation. This appears to be a function of
the lack of comparable inhalation studies for many effects studied via the oral exposure route, for
which there is a larger database of studies. For the liver and reproductive effects, inhalation
studies yielded a p-cRfC lower than the lowest route-to-route extrapolated p-cRfC for that type
of effect. Conversely, the lowest p-cRfDs were derived from oral studies with the exception of
reproductive effects, for which route-to-route extrapolation from an inhalation study in humans
also yielded among the lowest p-cRfDs. The only effect for which there were comparable
studies for comparing a p-cRfC from an inhalation study with a p-cRfC estimated by
route-to-route extrapolation from an oral study was increased liver weight in the mouse. The
primary dose-metric of amount of TCE oxidized in the liver yielded similar p-cRfCs of 1.0 and
1.1 ppm for the inhalation result and the route-to-route extrapolated result, respectively (see
Table 5-15).
As can be seen in these tables, the most sensitive types of effects (the types with the
lowest p-cRfCs and p-cRfDs) appear to be developmental, kidney, and immunological (adult and
developmental) effects, and then neurological and reproductive effects, in that order. Lastly, the
liver effects have p-cRfC and p-cRfD values that are about 3Vi orders of magnitude higher than
those for developmental, kidney, and immunological effects.
This document is a draft for review purposes only and does not constitute Agency policy.
101 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
This document is a draft for review purposes only and does not constitute Agency policy.
102 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 5-24. Lowest p-cRfCs or cRfCs for different effect domains
2
p-cRfC or cRfC in ppm
(composite uncertainty factor)
Effect domain
Effect type
Candidate critical effect
(Species/Critical Study)
Preferred
dose-metrica
Default
methodology
Alternative
dose-met rics/stu dies
(Tables 5-8-5-13)
Neurologic
Trigeminal nerve
effects
Trigeminal nerve effects
(human/Ruijtenetal., 1991)
0.54
(10)
0.47
(30)
0.83
(10)
Cognitive effects
Demyelination in hippocampus
(rat/Isaacson et al., 1990)
0.0071
(1,000)
[rtr]
0.0023
(1,000)
Mood/sleep
changes
Changes in wakefulness
(rat/Arito et al., 1994)
0.016
(300)
0.012
(1,000)
0.030
(300)
Kidney
Histological
changes
Toxic nephropathy
(rat/NTP, 1988)
0.00056
(10)
[rtr]
0.00087-1.3
(10-300)
Toxic nephrosis
(mouse/NCI, 1976)
0.0017
(300)
[rtr]
Meganucleocytosis
(rat/Maltoni et al., 1986)
0.0025
(10)
[rtr]
t kidney weight
t kidney weight
(rat/Woolhiser et al., 2006)
0.0013
(10)
0.52
(30)
0.0022-2.1
(10-30)
Liver
t liver weight
t liver weight
(mouse/Kjellstrand et al., 1983a)
0.91
(10)
0.72
(30)
0.83-2.5
(10-30)
Immunologic
i thymus weight
I thymus weight
(mouse/Keil et al., 2009)
0.00033
(100)
[rtr]
0.000082
(100)
Immuno-suppressio
n
i stem cell recolonization
(mouse/Sanders et al., 1982b)
0.057
(30)
[rtr]
0.014-1.4
(30-100)
Decreased PFC response
(rat/Woolhiser et al., 2006)
0.11
(100)
0.083
(300)
Autoimmunity
t anti-dsDNA and anti-ssDNA Abs
(mouse/Keil et al., 2009)
0.0033
(10)
[rtr]
0.00082-0.23
(10-300)
Autoimmune organ changes
mouse/Kaneko et al. (2000)
0.12
(300)
0.070
(1,000)
This document is a draft for review purposes only and does not constitute Agency policy.
103 DRAFT—DO NOT CITE OR QUOTE
-------
Table 5-24. Lowest p-cRfCs or cRfCs for different effect domains
(continued)
p-cRfC or cRfC in ppm
(composite uncertainty factor)
Effect domain
Effect type
Candidate critical effect
(Species/Critical Study)
Preferred
dose-metric
a
Default
methodology
Alternative
dose-metrics/studie
s
(Tables 5-8-5-13)
Reproductive
Effects on sperm
and testes
i ability of sperm to fertilize
(rat/DuTeaux et al., 2004b)
0.0093
(1,000)
[rtr]
0.028-0.17
(30-1,000)
Multiple effects
(rat/200la; Kumar et al., 2000b)
0.013
(1,000)
0.015
(3,000)
Hyperzoospermia
(human/Chia et al., 1996)b
0.017
(30)
0.014
(100)
Developmental
Congenital defects
Heart malformations
(rat/Johnson et al., 2003)
0.00037
(10)
[rtr]
0.000093
(10)
Develop, neurotox.
I rearing postexposure
(rat/Fredrikssonetal., 1993)
0.028
(300)
[rtr]
0.0077-0.084
(100-300)
Pre/postnat
al mortality/growth
Resorptions/J, fetal weight/
skeletal effects
(rat/Healy et al., 1982)
0.062
(100)
0.057
(300)
0.14-2.4
(10-100)
1
2 aThe critical effects/studies and p-cRfCs used to derive the RfC are in bold; supporting effects/studies and p-cRfCs
3 in italics.
4 bGreater than usual degree of uncertainty (see Section 5.1.2).
5
6 rtr = route-to-route extrapolated result.
7
This document is a draft for review purposes only and does not constitute Agency policy.
104 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 5-25. Lowest p-cRfDs or cRfDs for different effect domains
2
p-cRfD or cRfD in mg/kg/day
(composite uncertainty factor)
Effect domain
Effect type
Candidate critical effect
(Species/Critical Study)
Preferred
dose-metrica
Default
methodology
Alternative
dose-met rics/stu dies
(Tables 5-8-5-13)
Neurologic
Trigeminal nerve
effects
Trigeminal nerve effects
(human/Ruijtenetal., 1991)
0.73
(10)
[rtr]
1.4
(10)
Cognitive effects
Demyelination in hippocampus
(rat/Isaacson et al., 1990)
0.0092
(1,000)
0.0047
(10,000b)
0.0043
(1,000)
Mood/sleep
changes
Changes in wakefulness
(rat/Arito et al., 1994)
0.022
(300)
[rtr]
0.051
(300)
Kidney
Histological
changes
Toxic nephropathy
(rat/NTP, 1988)
0.00034
(10)
0.0945
(100)
0.00053-1.9
(10-300)
Toxic nephrosis
(mouse/NCI, 1976)
0.0010
(300)
Meganucleocytosis
(rat/Maltoni et al., 1986)
0.0015
(10)
0.34
(100)
t kidney weight
f kidney weight
(rat/Woolhiser et al., 2006)
0.00079
(10)
[rtr]
0.0013-2.5
(10)
Liver
t liver weight
t liver weight
(mouse/Kjellstrand et al., 1983a)
0.79
(10)
[rtr]
0.82-2.6
(10-100)
Immunologic
i thymus weight
I thymus weight
(mouse/Keil et al., 2009)
0.00048
(100)
0.00035
(1,000)
0.00016
(100)
Immuno-suppressio
n
i stem cell recolonization
(mouse/Sanders et al., 1982b)
0.083
(30)
0.060
(300)
0.028-0.91
(30-100)
Decreased PFC response
(rat/Woolhiser et al., 2006)
0.14
(100)
[rtr]
Autoimmunity
t anti-dsDNA and anti-ssDNA Abs
(mouse/Keil et al., 2009)
0.0048
(10)
0.0035
(100)
0.0016-0.19
(10-300)
Autoimmune organ changes
(Kaneko et al., 2000)
0.14
(300)
[rtr]
This document is a draft for review purposes only and does not constitute Agency policy.
105 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
Table 5-25. Lowest p-cRfDs or cRfDs for different effect domains
(continued)
Effect domain
Effect type
Candidate critical effect
(Species/Critical Study)
p-cRfD or cRfD in mg/kg/day
(composite uncertainty factor)
Preferred
dose-metric"
Default
methodology
Alternative
dose-met rics/stu dies
(Tables 5-8-5-13)
Reproductive
Effects on sperm
and testes
i ability of sperm to fertilize
(DuTeaux et al., 2004b)
0.016
(1,000)
0.014
(10,000b)
0.042-0.10
(30-1,000)
Multiple effects
(rat/200la; Kumar et al., 2000b)
0.016
(1,000)
[rtr]
Hyperzoospermia
(human/Chia et al., 1996)°
0.024
(30)
[rtr]
Developmental
Develop,
immunotox.
i PFC, t DTH
(rat/Peden-Adams et al., 2006)d
0.00037
(1,000)
Same as
preferred
-
Congenital defects
Heart malformations
(rat/Johnson et al., 2003)
0.00052
(10)
0.00021
(100)
0.00017
(10)
Develop, neurotox.
I rearing postexposure
(rat/Fredriksson et al., 1993)d
0.016
(1,000)
Same as
preferred
0.017-0.11
(100-3,000)
Pre/postnat
al mortality/growth
Resorptions/J, fetal weight/
skeletal effects
(rat/Healy et al., 1982)
0.085
(100)
[rtr]
0.70-2.9
(10-100)
a The critical effects/studies and p-cRfDs or cRfDs used to derive the RfD are in bold; supporting effects/studies and
p-cRfDs in italics.
bEPA's report on the RfC and RfD processes (U.S. EPA, 2002c) recommends not deriving reference values with a
composite UF of greater than 3,000; however, composite UFs exceeding 3,000 are considered here because the
derivation of the cRfCs and cRfDs is part of a screening process and the application of the PBPK model for
candidate critical effects reduces the values of some of the individual UFs for the p-cRfCs and p-cRfDs.
0 Greater than usual degree of uncertainty (see Section 5.1.2).
dNo PBPK model based analyses were done, so cRfD on the basis of applied dose only.
rtr = route-to-route extrapolated result (no value for default methodology).
This document is a draft for review purposes only and does not constitute Agency policy.
106 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 5-26. Lowest p-cRfCs for candidate critical effects for different types
2 of effect based on primary dose-metric
3
Type of effect
Effect
(primary dose-metric)
p-cRfC (ppm)
Neurological
Demyelination in hippocampus in rats
(T otMetabB W 34)
0.007 (rtr)
Kidney
Toxic nephropathy in rats
(ABioactDCVCBW34)
0.0006 (rtr)
Liver
Increased liver weight in mice
(AMetLi v 1B W3 4)
0.9
Immunological
Decreased thymus weight in mice
(T otMetabB W 34)
0.0003 (rtr)
Reproductive
Decreased ability of rat sperm to fertilize
(AUCCBld)
0.009 (rtr)a
Developmental
Heart malformations in rats
(T otOxMetabB W 34)
0.0004 (rtr)
4
5 a This value is supported by the p-cRfC value of 0.01 ppm for multiple testes and sperm effects from an inhalation
6 study in rats.
7 rtr = route-to-route extrapolated result.
8
9
10 Table 5-27. Lowest p-cRfDs for candidate critical effects for different types
11 of effect based on primary dose-metric
12
Type of effect
Effect
(primary dose-metric)
p-cRfD
(mg/kg/day)
Neurological
Demyelination in hippocampus in rats
(TotMetabBW34)
0.009
Kidney
Toxic nephropathy in rats
(ABioactDCVCBW34)
0.0003
Liver
Increased liver weight in mice
(AMetLivlBW34)
0.8 (rtr)
Immunological
Decreased thymus weight in mice
(TotMetabBW34)
0.0005
Reproductive
Decreased ability of rat sperm to fertilize (AUCCBld) and multiple
testes and sperm effects (TotMetabBW34)a
0.02
Developmental
Heart malformations in rats
(TotOxMetabBW34)
0.0005b
13 aEndpoints from two different studies yielded the same p-cRfD value.
This document is a draft for review purposes only and does not constitute Agency policy.
107 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
bThis value is supported by the cRfD value of 0.0004 mg/kg/day derived for developmental immunotoxicity effects
in mice (Peden-Adams et al., 2006); however, no PBPK analyses were done for this latter effect, so the value of
0.0004 mg/kg/day is based on applied dose,
rtr = route-to-route extrapolated result.
5.1.5.1.2. Reference Concentration
The goal is to select an overall RfC that is well supported by the available data (i.e.,
without excessive uncertainty given the extensive database) and protective for all the candidate
critical effects, recognizing that individual candidate RfC values are by nature somewhat
imprecise. The lowest candidate RfC values within each health effect category span a 3,000-fold
range from 0.0003-0.9 ppm (see Table 5-26). One approach to selecting a RfC would be to
select the lowest calculated value of 0.0003 ppm for decreased thymus weight in mice.
However, as can be seen in Table 5-24, three p-cRfCs are in the relatively narrow range of
0.0003-0.0006 ppm at the low end of the overall range. Given the somewhat imprecise nature of
the individual candidate RfC values, and the fact that multiple effects/studies lead to similar
candidate RfC values, the approach taken in this assessment is to select a RfC supported by
multiple effects/studies. The advantages of this approach, which is only possible when there is a
relatively large database of studies/effects and when multiple candidate values happen to fall
within a narrow range at the low end of the overall range, are that it leads to a more robust RfC
(less sensitive to limitations of individual studies) and that it provides the important
characterization that the RfC exposure level is similar for multiple noncancer effects rather than
being based on a sole explicit critical effect.
Table 5-28 and Table 5-29 summarize the PODs and UFs for the two critical and one
supporting studies/effects, respectively, corresponding to the p-cRfCs that have been chosen as
the basis of the RfC for TCE noncancer effects. Each of these lowest candidate p-cRfCs, ranging
from 0.0003-0.0006 ppm, for developmental, immunologic, and kidney effects, are values
derived from route-to-route extrapolation using the PBPK model. The lowest p-cRfC estimate
(for a primary dose-metric) from an inhalation studies is 0.001 ppm for kidney effects, which is
higher than the route-to-route extrapolated p-cRfC from the most sensitive oral study. For each
of the candidate RfCs, the PBPK model was used for inter- and intraspecies extrapolation, based
on the preferred dose-metric for each endpoint.
There is moderate confidence in the lowest p-cRfC for developmental effects (heart
malformations) (see Section 5.1.2.8) and the lowest p-cRfC estimate for immunological effects
(see Section 5.1.2.5), and these are considered the critical effects used for deriving the RfC. For
developmental effects, although the available study has important limitations, the overall weight
of evidence supports an effect of TCE on cardiac development. For immunological effects, there
This document is a draft for review purposes only and does not constitute Agency policy.
108 DRAFT—DO NOT CITE OR QUOTE
-------
1 is high confidence in the evidence for an immunotoxic hazard from TCE, but the available
2 dose-response data preclude application of BMD modeling.
3 For kidney effects (see Section 5.1.2.2), there is high confidence in the evidence for a
4 nephrotoxic hazard from TCE. Moreover, the lowest p-cRfC for kidney effects (toxic
5 nephropathy) is derived from a chronic study and is based on BMD modeling. However, as
6 Table 5-28. Summary of critical studies, effects, PODs, and UFs used to
7 derive the RfC
8
Keil et al. (2009)—Decreased thymus weight in female B6C3F1 mice exposed for 30 weeks by
drinking water.
• idPOD = 0.139 mg TCE metabolized/kgyYd, which is the PBPK model-predicted internal
dose at the applied dose LOAEL of 0.35 mg/kg/day (continuous) (no BMD modeling due
to inadequate model fit caused by supralinear dose-response shape) (see Appendix F,
Section F.6.4).
• HEC99 = 0.033 ppm (lifetime continuous exposure) derived from combined interspecies,
intraspecies, and route-to-route extrapolation using PBPK model.
• UFloaei =10 because POD is a LOAEL for an adverse effect.
• UFis = 3.16 because the PBPK model was used for interspecies extrapolation.
• UFh = 3.16 because the PBPK model was used to characterize human toxicokinetic
variability.
• p-cRfC = 0.033/100 = 0.00033 ppm (2 ^g/m3).
Johnson et al. (2003)—fetal heart malformations in S-D rats exposed from GD 1-22 by drinking
water.
• idPOD = 0.0142 mg TCE metabolized by oxidation/kg 7d, which is the BMDL from
BMD modeling using PBPK model-predicted internal doses, with highest-dose group
(1,000-fold higher than next highest-dose group) dropped, pup as unit of analysis, BMR
= 1% (due to severity of defects, some of which could have been fatal), and a nested
Log-logistic model to account for intralitter correlation (see Appendix F, Section F.6.5).
• HEC99 = 0.0037 ppm (lifetime continuous exposure) derived from combined interspecies,
intraspecies, and route-to-route extrapolation using PBPK model.
• UFis = 3.16 because the PBPK model was used for interspecies extrapolation.
• UFh = 3.16 because the PBPK model was used to characterize human toxicokinetic
variability.
• p-cRfC = 0.0037/10 = 0.00037 ppm (2 ^g/m3).
9
10 GD = gestation day.
11
12
13 Table 5-29. Summary of supporting studies, effects, PODs, and UFs for the
14 RfC
15
This document is a draft for review purposes only and does not constitute Agency policy.
109 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
NTP (1988)—Toxic nephropathy in female Marshall rats exposed for 104 weeks by oral gavage
(5 d/wk).
• idPOD = 0.0132 mg DCVC bioactivated/kg 7d, which is the BMDL from BMD
modeling using PBPK model-predicted internal doses, BMR = 5% (clearly toxic effect),
and log-logistic model (see Appendix F, Section F.6.1).
• HEC99 = 0.0056 ppm (lifetime continuous exposure) derived from combined interspecies,
intraspecies, and route-to-route extrapolation using PBPK model.
• UFis = 3.16 because the PBPK model was used for interspecies extrapolation.
• UFh = 3.16 because the PBPK model was used to characterize human toxicokinetic
variability.
• p-cRfC = 0.0056/10 = 0.00056 ppm (3 ^g/m3).
discussed in Section 3.3.3.2, there remains substantial uncertainty in the extrapolation of GSH
conjugation from rodents to humans due to limitations in the available data. In addition, the
p-cRfC for toxic nephropathy had greater dose-response uncertainty since the estimation of its
POD involved extrapolation from high response rates (>60%). Therefore, toxic nephropathy is
considered supportive but is not used as a primary basis for the RfC. The other sensitive
p-cRfCs for kidney effects in Table 5-19 were all within a factor of 5 of that for toxic
nephropathy; however, these values similarly relied on the uncertain interspecies extrapolation of
GSH conjugation.
As a whole, the estimates support a RfC of 0.0004 ppm (0.4 ppb or 2 (J,g/m ). This
estimate essentially reflects the midpoint between the similar p-cRfC estimates for the two
critical effects (0.00033 ppm for decreased thymus weight in mice and 0.00037 ppm for heart
malformations in rats), rounded to one significant figure. This estimate is also within a factor of
two of the p-cRfC estimate of 0.00006 ppm for the supporting effect of toxic nephropathy in rats.
Thus, there is robust support for a RfC of 0.0004 ppm provided by estimates for multiple effects
from multiple studies. The estimates are based on PBPK model-based estimates of internal dose
for interspecies, intraspecies, and route-to-route extrapolation, and there is sufficient confidence
in the PBPK model and support from mechanistic data for one of the dose-metrics
(TotOxMetabBW34 for the heart malformations). There is high confidence that
ABioactDCVCBW34 and AMetGSHBW34 would be appropriate dose-metrics for kidney
effects, but there is substantial uncertainty in the PBPK model predictions for these dose-metrics
in humans (see Section 5.1.3.1). Note that there is some human evidence of developmental heart
defects from TCE exposure in community studies (see Section 4.8.3.1.1) and of kidney toxicity
in TCE-exposed workers (see Section 4.4.1).
-3
In summary, the RfC is 0.0004 ppm (0.4 ppb or 2 (J,g/m ) based on route-to-route
extrapolated results from oral studies for the critical effects of heart malformations (rats) and
This document is a draft for review purposes only and does not constitute Agency policy.
110 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
immunotoxicity (mice). This RfC value is further supported by route-to-route extrapolated
results from an oral study of toxic nephropathy (rats).
5.1.5.1.3. Reference Dose
As with the RfC determination above, the goal is to select an overall RfD that is well
supported by the available data (i.e., without excessive uncertainty given the extensive database)
and protective for all the candidate critical effects, recognizing that individual candidate RfD
values are by nature somewhat imprecise. The lowest candidate RfD values within each health
effect category span a nearly 3,000-fold range from 0.0003-0.8 mg/kg/day (see Table 5-26).
One approach to selecting a RfC would be to select the lowest calculated value of 0.0003 ppm
for toxic nephropathy in rats. However, as can be seen in Table 5-25, multiple p-cRfDs or cRfDs
from oral studies are in the relatively narrow range of 0.0003-0.0008 mg/kg/day at the low end
of the overall range. Given the somewhat imprecise nature of the individual candidate RfD
values, and the fact that multiple effects/studies lead to similar candidate RfD values, the
approach taken in this assessment is to select a RfD supported by multiple effects/studies. The
advantages of this approach, which is only possible when there is a relatively large database of
studies/effects and when multiple candidate values happen to fall within a narrow range at the
low end of the overall range, are that it leads to a more robust RfD (less sensitive to limitations
of individual studies) and that it provides the important characterization that the RfD exposure
level is similar for multiple noncancer effects rather than being based on a sole explicit critical
effect.
Table 5-30 and Table 5-31 summarize the PODs and UFs for the three critical and
two supporting studies/effects, respectively, corresponding to the p-cRfDs or cRfDs that have
been chosen as the basis of the RfD for TCE noncancer effects. Two of the lowest p-cRfDs for
the primary dose-metrics—0.0008 mg/kg/day for increased kidney weight in rats and 0.0005
mg/kg/day for both heart malformations in rats and decreased thymus weights in mice—are
derived using the PBPK model for inter- and intraspecies extrapolation, and a third—
0.0003 mg/kg/day for increased toxic nephropathy in rats—is derived using the PBPK model for
inter- and intraspecies extrapolation as well as route-to-route extrapolation from an inhalation
study. The other of these lowest values—0.0004 mg/kg/day for developmental immunotoxicity
(decreased PFC response and increased delayed-type hypersensitivity) in mice—is based on
applied dose.
There is moderate confidence in the p-cRfDs for decreased thymus weights (see
Section 5.1.2.5) and heart malformations (see Section 5.1.2.8) and the cRfD for developmental
This document is a draft for review purposes only and does not constitute Agency policy.
111 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
immunological effects (see Section 5.1.2.8), and these effects are considered the critical effects
used for deriving the RfD. For developmental effects, although the available study has important
limitations, the overall weight of evidence supports an effect of TCE on cardiac development.
For adult and developmental immunological effects, there is high confidence in the evidence for
an immunotoxic hazard from TCE. However, the available dose-response data for
immunological effects preclude application of BMD modeling.
For kidney effects (see Section 5.1.2.2), there is high confidence in the evidence for a
nephrotoxic hazard from TCE. Moreover, the two lowest p-cRfDs for kidney effects (toxic
nephropathy and increased kidney weight) are both based on BMD modeling and one is derived
from a chronic study. However, as discussed in Section 3.3.3.2, there remains substantial
uncertainty in the extrapolation of GSH conjugation from rodents to humans due to limitations in
the available data. In addition, the p-cRfD value for toxic nephropathy had greater
dose-response uncertainty since the estimation of its POD involved extrapolation from high
response rates (>60%). Therefore, kidney effects are considered supportive but are not used as a
primary basis for the RfD.
Table 5-30. Summary of critical studies, effects, PODs, and UFs used to
derive the RfD
Keil et al. (2009)—Decreased thymus weight in female B6C3F1 mice exposed for 30 weeks by
drinking water.
• idPOD = 0.139 mg TCE metabolized/kgyYd, which is the PBPK model-predicted internal
dose at the applied dose LOAEL of 0.35 mg/kg/day (continuous) (no BMD modeling due
to inadequate model fit caused by supralinear dose-response shape) (see Appendix F,
Section F.6.4).
• HED99 = 0.048 mg/kg/day (lifetime continuous exposure) derived from combined
interspecies and intraspecies extrapolation using PBPK model.
• UFloaei =10 because POD is a LOAEL for an adverse effect.
• UFis = 3.16 because the PBPK model was used for interspecies extrapolation.
• UFh = 3.16 because the PBPK model was used to characterize human toxicokinetic
variability.
• p-cRfD = 0.048/100 = 0.00048 mg/kg/day.
This document is a draft for review purposes only and does not constitute Agency policy.
112 DRAFT—DO NOT CITE OR QUOTE
-------
Peden-Adams et al. (2006)—Decreased PFC response (3 and 8 weeks), increased delayed-type
hypersensitivity (8 weeks) in pups exposed from GD 0-3- or 8-weeks-of-age through drinking
water (placental and lactational transfer, and pup ingestion).
• POD = 0.37 mg/kg/day is the applied dose LOAEL (estimated daily dam dose) (no BMD
modeling due to inadequate model fit caused by supralinear dose-response shape). No
PBPK modeling was attempted due to lack of appropriate models/parameters to account
for complicated fetal/pup exposure pattern (see Appendix F, Section F.6.6).
• UFloaei =10 because POD is a LOAEL for multiple adverse effects.
• UFis = 10 for interspecies extrapolation because PBPK model was not used.
• UFh = 10 for human variability because PBPK model was not used.
• cRfD = 0.37/1,000 = 0.00037 mg/kg/day.
Johnson et al. (2003)—fetal heart malformations in S-D rats exposed from GD 1-22 by drinking
water
• idPOD = 0.0142 mg TCE metabolized by oxidation/kg 7d, which is the BMDL from
BMD modeling using PBPK model-predicted internal doses, with highest-dose group
(1,000-fold higher than next highest-dose group) dropped, pup as unit of analysis, BMR =
1% (due to severity of defects, some of which could have been fatal), and a nested
Log-logistic model to account for intralitter correlation (see Appendix F, Section F.6.5).
• HED99 = 0.0051 mg/kg/day (lifetime continuous exposure) derived from combined
interspecies and intraspecies extrapolation using PBPK model.
• UFis = 3.16 because the PBPK model was used for interspecies extrapolation.
• UFh = 3.16 because the PBPK model was used to characterize human toxicokinetic
variability.
• p-cRfD = 0.0051/10 = 0.00051 mg/kg/day.
1
2 GD = gestation day.
3
This document is a draft for review purposes only and does not constitute Agency policy.
113 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Table 5-31. Summary of supporting studies, effects, PODs, and UFs for the
RfD
NTP (1988)—Toxic nephropathy in female Marshall rats exposed for 104 weeks by oral gavage
(5 d/wk).
• idPOD = 0.0132 mg DCVC bioactivated/kg 7d, which is the BMDL from BMD
modeling using PBPK model-predicted internal doses, BMR = 5% (clearly toxic effect),
and Log-logistic model (see Appendix F, Section F.6.1).
• HED99 = 0.0034 mg/kg/day (lifetime continuous exposure) derived from combined
interspecies and intraspecies extrapolation using PBPK model.
• UFis = 3.16 because the PBPK model was used for interspecies extrapolation.
• UFh = 3.16 because the PBPK model was used to characterize human toxicokinetic
variability.
• p-cRfD = 0.0034/10 = 0.00034 mg/kg/day.
Woolhiser et al. (2006)—Increased kidney weight in female S-D rats exposed for 4 weeks by
inhalation (6 h/d, 5 d/wk).
• idPOD = 0.0309 mg DCVC bioactivated/kg3/i/d, which is the BMDL from BMD
modeling using PBPK model-predicted internal doses, BMR = 10%, and Hill model with
constant variance (see Appendix F, Section F.6.3).
• HED99 = 0.0079 mg/kg/day (lifetime continuous exposure) derived from combined
interspecies and intraspecies extrapolation using PBPK model.
• UFsc = 1 because Kjellstrand et al. (1983a) reported that in mice, kidney effects after
exposure for 120 d was no more severe than those after 30 d exposure.
• UFis = 3.16 because the PBPK model was used for interspecies extrapolation.
• UFh = 3.16 because the PBPK model was used to characterize human toxicokinetic
variability.
• p-cRfC = 0.0079/10 = 0.00079 mg/kg/day.
As a whole, the estimates support a RfD of 0.0005 mg/kg/day. This estimate is within
20% of the estimates for the critical effects—0.0004 mg/kg/day for developmental
immunotoxicity (decreased PFC and increased delayed-type hypersensitivity) in mice, and
0.0005 mg/kg/day both for heart malformations in rats and for decreased thymus weights in
mice. This estimate is also within approximately a factor of two of the supporting effect
estimates of 0.0003 mg/kg/day for toxic nephropathy in rats and 0.0008 mg/kg/day for increased
kidney weight in rats. Thus, there is strong, robust support for a RfD of 0.0005 mg/kg/day
provided by the concordance of estimates derived from multiple effects from multiple studies.
The estimates for kidney effects, thymus effects, and developmental heart malformations are
based on PBPK model-based estimates of internal dose for interspecies and intraspecies
extrapolation, and there is sufficient confidence in the PBPK model and support from
mechanistic data for one of the dose-metrics (TotOxMetabBW34 for the heart malformations).
This document is a draft for review purposes only and does not constitute Agency policy.
114 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
There is high confidence that ABioactDCVCBW34 would be an appropriate dose-metric for
kidney effects, but there is substantial uncertainty in the PBPK model predictions for this
dose-metric in humans (see Section 5.1.3.1). Note that there is some human evidence of
developmental heart defects from TCE exposure in community studies (see Section 4.8.3.1.1)
and of kidney toxicity in TCE-exposed workers (see Section 4.4.1).
In summary, the RfD is 0.0005 mg/kg/day based on the critical effects of heart
malformations (rats), adult immunological effects (mice), and developmental immunotoxicity
(mice), all from oral studies. This RfD value is further supported by results from an oral study
for the effect of toxic nephropathy (rats) and route-to-route extrapolated results from an
inhalation study for the effect of increased kidney weight (rats).
5.2. DOSE-RESPONSE ANALYSIS FOR CANCER ENDPOINTS
This section describes the dose-response analysis for cancer endpoints. Section 5.2.1
discusses the analyses of data from chronic rodent bioassays. Section 5.2.2 discusses the
analyses of human epidemiologic data. Section 5.2.3 discusses the choice of the preferred
inhalation unit risk and oral slope factor estimates, as well as the application of age-dependent
adjustment factors to the slope factor and unit risk estimates.
5.2.1. Dose-Response Analyses: Rodent Bioassays
This section describes the calculation of cancer slope factor and unit risk estimates based
on rodent bioassays. First, all the available studies (i.e., chronic rodent bioassays) were
considered, and those suitable for dose-response modeling were selected for analysis (see
Section 5.2.1.1). Then dose-response modeling using the linearized multistage model was
performed using applied doses (default dosimetry) as well as PBPK model-based internal doses
(see Section 5.2.1.2). Bioassays for which time-to-tumor data were available were analyzed
using poly-3 adjustment techniques and using a Multistage Weibull model. In addition, a cancer
potency estimate for different tumor types combined was derived from bioassays in which there
was more than one type of tumor response in the same sex and species. Slope factor and unit
risk estimates based on PBPK model-estimated internal doses were then extrapolated to human
population slope factor and unit risk estimates using the human PBPK model. From these results
(see Section 5.2.1.3), estimates from the most sensitive bioassay (i.e., that with the greatest slope
factor or unit risk estimate) for each combination of administration route, sex, and species, based
on the PBPK model-estimated internal doses, were considered as candidate slope factor or unit
This document is a draft for review purposes only and does not constitute Agency policy.
115 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
risk estimates for TCE. Uncertainties in the rodent-based dose-response analyses are described
in Section 5.2.1.4.
5.2.1.1.1. Rodent Dose-Response Analyses: Studies and Modeling Approaches
The rodent cancer bioassays that were identified for consideration for dose-response
analysis are listed in Tables 5-32 (inhalation bioassays) and 5-33 (oral bioassays) for each
sex/species combination. The bioassays selected for dose-response analysis are marked with an
asterisk; rationales for rejecting the bioassays that were not selected are provided in the
"Comments" columns of the tables. For the selected bioassays, the tissues/organs that exhibited
a TCE-associated carcinogenic response and for which dose-response modeling was performed
are listed in the "Tissue/Organ" columns.
Table 5-32. Inhalation bioassays
Study
Strain
Tissue/Organ
Comments
Female mice
Fukuda et al. (1983)a
Cij:CD-l (ICR)
Lung
Henschler et al. (1980)a
Han:NMRI
Lymphoma
Maltoni et al. (1986)a
B6C3F1
Liver, Lung
Maltoni et al. (1986)
Swiss
-
No dose-response
Male mice
Henschler et al. (1980)
Han:NMRI
-
No dose-response
Maltoni et al. (1986)
B6C3F1
Liver
Exp #BT306: excessive
fighting
Maltoni et al. (1986)
B6C3F1
Liver
Exp #BT306bis. Results
similar to Swiss mice
Maltoni et al. (1986)a
Swiss
Liver
Female rats
Fukuda et al. (1983)
Sprague-Dawley
-
No dose-response
Henschler et al. (1980)
Wistar
-
No dose-response
Maltoni et al. (1986)
Sprague-Dawley
-
No dose-response
Male rats
This document is a draft for review purposes only and does not constitute Agency policy.
116 DRAFT—DO NOT CITE OR QUOTE
-------
Henschler et al. (1980)
Wistar
-
No dose-response
Maltoni et al. (1986)a
Sprague-Dawley
Kidney, Ley dig
cell, Leukemia
1
2 Selected for dose-response analysis.
3
4 "No dose-response" = no tumor incidence data suitable for dose-response modeling.
5
This document is a draft for review purposes only and does not constitute Agency policy.
117 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 5-33. Oral bioassays
2
Study
Strain
Tissue/organ
Comments
Female mice
Henschler et al.
(1984)
Han:NMRI
-
Toxicity, no dose-response
NCI (1976)a
B6C3F1
Liver, lung, sarcomas
and lymphomas
NTP (1990)
B6C3F1
Liver, lung,
lymphomas
Single dose
Van Duuren et al.
(1979)
Swiss
Liver
Single dose, no dose-response
Male mice
Anna et al. (1994)
B6C3F1
Liver
Single dose
Bull et al. (2002)
B6C3F1
Liver
Single dose
Henschler et al.
(1984)
Han:NMRI
-
Toxicity, no dose-response
NCI (1976)a
B6C3F1
Liver
NTP (1990)
B6C3F1
Liver
Single dose
Van Duuren et al.
(1979)
Swiss
-
Single dose, no dose-response
Female rats
NCI (1976)
Osborne-Mendel
-
Toxicity, no dose-response
NTP (1988)
ACI
-
No dose-response
NTP (1988)a
August
Leukemia
NTP (1988)
Marshall
-
No dose-response
NTP (1988)
Osborne-Mendel
Adrenal cortex
Adenomas only
NTP (1990)
F344/jV
-
No dose-response
Male rats
NCI (1976)
Osborne-Mendel
-
Toxicity, no dose-response
NTP (1988)
ACI
-
No dose-response
NTP (1988)a
August
Subcutaneous tissue
sarcomas
NTP (1988)a
Marshall
Testes
This document is a draft for review purposes only and does not constitute Agency policy.
118 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
NTP (1988)a
Osborne-Mendel
Kidney
NTP (1990)a
F344IN
Kidney
a Selected for dose-response analysis.
"No dose-response" = no tumor incidence data suitable for dose-response modeling.
The general approach used was to model each sex/species/bioassay tumor response to
determine the most sensitive bioassay response (in terms of human equivalent exposure or dose)
for each sex/species combination. The various modeling approaches, model selection, and slope
factor and unit risk derivation are discussed below. Modeling was done using the applied dose
or exposure (default dosimetry) and several internal dose-metrics. The dose-metrics used in the
dose-response modeling are discussed in Section 5.2.1.2. Because of the large volume of
analyses and results, detailed discussions about how the data were modeled using the various
dosimetry and modeling approaches and results for individual data sets are provided in Appendix
G. The overall results are summarized and discussed in Section 5.2.1.3.
Most tumor responses were modeled using the multistage model in EPA's BMDS
(www.epa.gov/ncea/bmds). The multistage model is a flexible model, capable of fitting most
cancer bioassay data, and it is EPA's long-standing model for the modeling of such cancer data.
The multistage model has the general form
P(d)= 1 - exp ~{q0 + qxd + q2d2 + ... +qkdk^ (Eq. 5-1)
where P(d) represents the lifetime risk (probability) of cancer at dose and parameters q, > 0,
for i = 0, 1, ..., k. For each data set, the multistage model was evaluated for one stage and (n- 1)
stages, where n is the number of dose groups in the bioassay. A detailed description of how the
data were modeled, as well as tables of the dose-response input data and figures of the multistage
modeling results, is provided in Appendix G.
Only models with acceptable fit (p > 0.05) were considered. 37 If 1-parameter and
2-parameter models were both acceptable (in no case was there a 3-parameter model), the more
parsimonious model (i.e., the 1-parameter model) was selected unless the inclusion of the
37 When considering multiple types of model for noncancer effects, p > 0,10 is used. For cancer, there is a prior
preference for the multistage model, thus the p > 0.05 (which increases the probability of accepting the preferred
model).
This document is a draft for review purposes only and does not constitute Agency policy.
119 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
2nd parameter resulted in a statistically significant38 improvement in fit. If two different
1-parameter models were available (e.g., a 1-stage model and a 3-stage model with and P2
both equal to 0), the one with the best fit, as indicated by the lowest AIC value, was selected. If
the AIC values were the same (to three significant figures), then the lower-stage model was
selected. Visual fit and scaled chi-square residuals were also considered for confirmation in
model selection. For two data sets, the highest-dose group was dropped to improve the fit in the
lower dose range.
From the selected model for each data set, the maximum likelihood estimate (MLE) for
the dose corresponding to a specified level of risk (i.e., the benchmark dose, or BMD) and its
95% lower confidence bound (BMDL) were estimated.39 In most cases, the risk level, or BMR,
was 10% extra risk;40 however, in a few cases with low response rates, a BMR of 5%, or even
P/o, extra risk was used to avoid extrapolation above the range of the data. As discussed in
Section 4.4, there is sufficient evidence to conclude that a mutagenic MOA is operative for
TCE-induced kidney tumors, so linear extrapolation from the BMDL to the origin was used to
derive slope factor and unit risk estimates for this site. For all other tumor types, the available
evidence supports the conclusion that the MOA(s) for TCE-induced rodent tumors is unknown,
as discussed in Sections 4.5-4.10 and summarized in Section 4.11.2.3. Therefore, linear
extrapolation was also used based on the general principles outlined in EPA's Guidelines for
Carcinogen Risk Assessment (U .S. EPA, 2005c) and reviewed below in Section 5.2.1.4.1. Thus,
for all TCE-associated rodent tumors, slope factor and unit risk estimates are equal to
BMR/BMDL (e.g., 0.10/BMDLio for a BMR of 10%). See Section 5.2.1.3 for a summary of the
slope factor and unit risk estimates for each sex/species/bioassay/tumor type.
Some of the bioassays exhibited differential early mortality across the dose groups, and,
for three such male rat studies (identified with checkmarks in the "Time-to-tumor" column of
Table 5-34), analyses that take individual animal survival times into account were performed.
(For bioassays with differential early mortality occurring primarily before the time of the
1st tumor [or 52 weeks, whichever came first], the effects of early mortality were largely
accounted for by adjusting the tumor incidence for animals at risk, as described in Appendix G,
and the dose-response data were modeled using the regular multistage model, as discussed
above, rather than approaches that account for individual animal survival times.)
38Using a standard criterion for nested models, that the difference in -2 x log-likelihood exceeds 3.84 (the
95th percentile of %2 [1]).
39BMDS estimates confidence intervals using the profile likelihood method.
40Extra risk over the background tumor rate is defined as [P(d) - P(0)] / [1 - P(0)], where P{d) represents the
lifetime risk (probability) of cancer at dose d.
This document is a draft for review purposes only and does not constitute Agency policy.
120 DRAFT—DO NOT CITE OR QUOTE
-------
1 Two approaches were used to take individual survival times into account. First, EPA's
2 Multistage Weibull (MSW) software4l was used for time-to-tumor modeling. The Multistage
3 Weibull time-to-tumor model has the general form
4
41This software has been thoroughly tested and externally reviewed. In February 2009, it will become available on
U.S. EPA's Web site.
This document is a draft for review purposes only and does not constitute Agency policy.
121 DRAFT—DO NOT CITE OR QUOTE
-------
Table 5-34. Specific dose-response analyses performed and dose-metrics used
5^
cs
si
3
a
>!
£
N>
o
&
O
fS"
sj
>8
K) ^3
O
>!
Grj
O
a
a
a
§•
>!
o
H §
I ^
81
w
2; ^
¦ a
"T3
H S.
W|
O •
o
c
o
H
W
H
o;
Bioassay
Strain
Endpoint
Applied
dose
PBPK-based—primary
dose-metric
PBPK-based—
alternative
dose-met ric(s)
Time-to-
tumor
INHALATION
Female mice
Fukuda et al. (1983)
Crj:CD-l (ICR)
Lung adenomas and carcinomas
a/
AMetLngBW34
T otOxMetabB W3 4
AUCCBld
Henschler et al. (1980)
Han:NMRI
Lymphoma
a/
TotMetabBW34
AUCCBld
Maltoni et al. (1986)
B6C3F1
Liver hepatomas
V
AMetLivlBW34
T otOxMetabB W3 4
Lung adenomas and carcinomas
V
AMetLngBW34
T otOxMetabB W3 4
AUCCBld
Combined risk
V
Male mice
Maltoni et al. (1986)
Swiss
Liver hepatomas
V
AMetLivlBW34
T otOxMetabB W3 4
Female rats
None selected
Male rats
Maltoni et al. (1986)
Sprague-Dawley
Kidney adenomas and carcinomas
V
ABioactDCVCBW34
AMetGSHBW34
TotMetabBW34
Leydig cell tumors
V
TotMetabBW34
AUCCBld
Leukemias
V
TotMetabBW34
AUCCBld
Combined risk
V
-------
Table 5-34. Specific dose-response analyses performed and dose-metrics used (continued)
Bioassay
Strain
Endpoint
Applied
dose
PBPK-based—primary
dose-metric
PBPK-based—
alternative
dose-met ric(s)
Time-to-
tumor
ORAL
Female mice
NCI (1976)
B6C3F1
Liver carcinomas
a/
AMetLivlBW34
T otOxMetabB W3 4
Lung adenomas and carcinomas
a/
AMetLngBW34
T otOxMetabB W3 4
AUCCBld
Multiple sarcomas/lymphomas
V
TotMetabBW34
AUCCBld
Combined risk
V
Male mice
NCI (1976)
B6C3F1
Liver carcinomas
a/
AMetLivlBW34
T otOxMetabB W3 4
Female rats
NTP (1988)
August
Leukemia
a/
TotMetabBW34
AUCCBld
Male rats
NTP (1988)
August
Subcutaneous tissue sarcomas
a/
TotMetabBW34
AUCCBld
NTP (1988)
Marshall
Testicular interstitial cell tumors
V
TotMetabBW34
AUCCBld
a/
NTP (1988)
Osborne-Mendel
Kidney adenomas and carcinomas
V
ABioactDCVCBW34
AMetGSHBW34
TotMetabBW34
a/
NTP (1990)
F344IN
Kidney adenomas and carcinomas
V
ABioactDCVCBW34
AMetGSHBW34
TotMetabBW34
V
PBPK-based dose-metric abbreviations:
ABioactDCVCBW34 = Amount of DCVC bioactivated in the kidney per unit body weight'4 (mg DCVC/kgyYweek).
AMetGSHB W34 = Amount of TCE conjugated with GSH per unit body weight7' (mg TCE/kgyYweek).
AMetLivlBW34 = Amount of TCE oxidized per unit body weight'4 (mg TCE/kgyYweek).
AMetLngBW34 = Amount of TCE oxidized in the respiratory tract per unit body weight'4 (mg TCE/kg'/4/week).
AUCCBld = Area under the curve of the venous blood concentration of TCE (mg-hour/L/week).
TotMetabBW34 = Total amount of TCE metabolized per unit body weight' 4 (mg TCE/kg'/4/week).
TotOxMetabB W34 = Total amount of TCE oxidized per unit body weight'4 (mg TCE/kg'4/week).
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
P(d,t} = 1 - exp —{qQ + qxd + q2d2 + ... +qkdk) x (t — t0Y
(Eq. 5-2)
where P(d,t) represents the probability of a tumor by age t for dose d, and parameters z > 1,
to > 0, and q, > 0 for i = 0,1,...,&, where k = the number of dose groups; the parameter ^represents
the time between when a potentially fatal tumor becomes observable and when it causes death.
(All of our analyses used the model for incidental tumors, which has no to term.) Although the
fit of the MSW model can be assessed visually using the plot feature of the MSW software,
because there is no applicable goodness-of-fit statistic with a well-defined asymptotic
distribution, an alternative survival-adjustment technique, "poly-3 adjustment," was also applied
(Portier and Bailer, 1989). This technique was used to adjust the tumor incidence denominators
based on the individual animal survival times.42 The adjusted incidence data then served as
inputs for EPA's BMDS multistage model, and model (i.e., stage) selection was conducted as
already described above. Under both survival-adjustment approaches, BMDs and BMDLs were
obtained and slope factor and unit risks derived as discussed above for the standard multistage
model approach. See Appendix G for a more detailed description of the MSW modeling and for
the results of both
the MSW and poly-3 approaches for the individual data sets. A comparison of the results for the
three different data sets and the various dose-metrics used is presented in Section 5.2.1.3.
For bioassays that exhibited more than one type of tumor response in the same sex and
species (these studies have a row for "combined risk" in the "Endpoint" column of Table 5-34),
the cancer potency for the different tumor types combined was estimated. The combined tumor
risk estimate describes the risk of developing tumors for any (not all together) of the tumor types
that exhibited a TCE-associated tumor response; this estimate then represents the total excess
cancer risk. The model for the combined tumor risk is also multistage, with the sum of the
stage-specific multistage coefficients from the individual tumor models serving as the
stage-specific coefficients for the combined risk model (i.e., for each q,,
qi[combined\ = qn + <]i2 + ¦¦¦+ qik, where the q,s are the coefficients for the powers of dose and k is
the number of tumor types being combined) (Bogen, 1990; NRC, 1994). This model assumes
that the occurrences of two or more tumor types are independent. Although the resulting model
equation can be readily solved for a given BMR to obtain an MLE (BMD) for the combined risk,
42Each tumorless animal is weighted by its fractional survival time (number of days on study divided by 728 days,
the typical number of days in a two-year bioassay) raised to the power of three to reflect the fact that animals are at
greater risk of cancer at older ages. Animals with tumors are given a weight of one. The sum of the weights of all
the animals in an exposure group yields the effective survival-adjusted denominator.
This document is a draft for review purposes only and does not constitute Agency policy.
124 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
the confidence bounds for the combined risk estimate are not calculated by available modeling
software. Therefore, the confidence bounds on the combined BMD were estimated using a
Bayesian approach, computed using Markov chain Monte Carlo techniques and implemented
using the freely available WinBugs software (Spiegelhalter et al., 2003). Use of WinBugs for
derivation of a distribution of BMDs for a single multistage model has been demonstrated by
Kopylev et al. (2007), and this approach can be straightforwardly generalized to derive the
distribution of BMDs for the combined tumor load. For further details on the implementation of
this approach and for the results of the analyses, see Appendix G.
5.2.1.1.2. Rodent Dose-Response Analyses: Dosimetry
In modeling the applied doses (or exposures), default dosimetry procedures were applied
to convert applied rodent doses to human equivalent doses. Essentially, for inhalation exposures,
"ppm equivalence" across species was assumed. For oral doses, 3/4-power body-weight scaling
was used, with a default average human body weight of 70 kg. See Appendix G for more details
on the default dosimetry procedures.
In addition to applied doses, several internal dose-metrics were used in the dose-response
modeling for each tumor type. Use of internal dose-metrics in dose-response modeling is
described here briefly. For more details on the PBPK modeling used to estimate the levels of the
dose-metrics corresponding to different exposure scenarios in rodents and humans, as well as a
qualitative discussion of the uncertainties and limitations of the model, see Section 3.5; for a
more detailed discussion of how the dose-metrics were used in dose-response modeling, see
Appendix G. Quantitative analyses of the uncertainties and their implications for dose-response
assessment, utilizing the results of the Bayesian analysis of the PBPK model, are discussed
separately in Section 5.2.1.4.2.
5.2.1.1.3. Selection of dose-metrics for different tumor types
One area of scientific uncertainty in cancer dose-response assessment is the appropriate
scaling between rodent and human doses for equivalent responses. As discussed above, for
applied dose, the standard dosimetry assumptions for equal lifetime carcinogenic risk are, for
inhalation exposure, the same lifetime exposure concentration in air, and, for oral exposure, the
same lifetime daily dose scaled by body weight to the % power. In this assessment, the
cross-species scaling methodology, grounded in the principles of allometric variation of biologic
processes, is used for describing pharmacokinetic equivalence (Allen and Fisher, 1993; Allen et
This document is a draft for review purposes only and does not constitute Agency policy.
125 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
al., 1987; Crump et al., 1989; "Supplementary data for TCE assessment: Rat population
example," 2011; U.S. EPA, 1992, 2005c) U.S. EPA 2011. Briefly, in the absence of adequate
information to the contrary, the methodology determines pharmacokinetic equivalence across
species through equal average lifetime concentrations or AUCs of the toxicant. Thus, in cases
where the PBPK model can predict internal concentrations of the active moiety, equivalent daily
AUCs are assumed to address cross-species pharmacokinetics. For cancer assessments, there is
currently no adjustment for pharmacodynamic differences.
More detailed discussion of the cross-species scaling methodology, and its implications
for dose-metric selection, was presented for the noncancer dose-response analyses in Section
5.1.3.1, and those details are not repeated here.
To summarize, the preferred dose-metric under this methodology is equivalent daily
AUC of the active moiety (parent compound or metabolite). For metabolites, in cases where the
rate of production, but not the rate of clearance, of the active moiety can be estimated, the
preferred dose-metric is the rate of metabolism (through the appropriate pathway) scaled by body
weight to the % power. If there are sufficient data to consider the active metabolite moiety(ies)
"reactive" and cleared through nonbiological processes, then the preferred dose-metric is the rate
of metabolism (through the appropriate pathway) scaled by the tissue mass. Finally, if local
metabolism is thought to be involved but cannot be estimated with the available data, then the
AUC of the parent compound in blood is considered an appropriate surrogate and thus the
preferred dose-metric.
Generally, an attempt was made to use tissue-specific dose-metrics representing
particular pathways or metabolites identified from available data as having a likely role in the
induction of a tissue-specific cancer. Where insufficient information was available to establish
particular metabolites or pathways of likely relevance to a tissue-specific cancer, more general
"upstream" metrics representing either parent compound or total metabolism had to be used. In
addition, the selection of dose-metrics was limited to metrics that could be adequately estimated
by the PBPK model (see Section 3.5). The (PBPK-based) dose-metrics used for the different
tumor types are listed in Table 5-34. For each tumor type, the "primary" dose-metric referred to
in Table 5-34 is the metric representing the particular metabolite or pathway whose involvement
in carcinogenicity has the greatest biological support, whereas "alternative" dose-metrics
represent upstream metabolic pathways (or TCE distribution, in the case of AUCCBld) that may
be more generally involved.
This document is a draft for review purposes only and does not constitute Agency policy.
126 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
5.2.1.1.3.1.Kidney
As discussed in Sections 4.4.6-4.4.7, there is sufficient evidence to conclude that
TCE-induced kidney tumors in rats are primarily caused by GSH-conjugation metabolites either
produced in situ in or delivered systemically to the kidney. As discussed in Section 3.3.3.2,
bioactivation of these metabolites within the kidney, either by beta-lyase, FMO, or P450s,
produces reactive species. Therefore, multiple lines of evidence support the conclusion that
renal bioactivation of DCVC is the preferred basis for internal dose extrapolations of
TCE-induced kidney tumors. However, uncertainties remain as to the relative contributions from
each bioactivation pathway, and quantitative clearance data necessary to calculate the
concentration of each species are lacking. Moreover, the estimates of the amount bioactivated
are indirect, derived from the difference between overall GSH conjugation flux and NAcDCVC
excretion (see Section 3.5.7.3.1).
The rationales for the dose-metrics for kidney tumors are the same as for kidney
noncancer toxicity, discussed above in Section 5.1.3.1.1, and not repeated here. The primary
internal dose-metric for TCE-induced kidney tumors is the weekly rate of DCVC bioactivation
per unit body weight to the 3/4 power (ABioactDCVCBW34 |mg/kg '/week|) Due to the larger
relative kidney weight in rats as compared to humans, using the alternative scaling by kidney
weight instead of body weight to the 3/4 power would only change the quantitative interspecies
extrapolation by about twofold,43 so the sensitivity of the results to the scaling choice is
relatively small. An alternative dose-metric that also involves the GSH conjugation pathway is
the amount of GSH conjugation scaled by the 3/4 power of body weight (AMetGSHBW34
[mg/kg3/Vweek]). This dose-metric uses the total flux of GSH conjugation as the
toxicologically-relevant dose, and, thus, incorporates any direct contributions from DCVG and
DCVC, which are not addressed in the DCVC bioactivation metric. Another alternative
dose-metric is the total amount of TCE metabolism (oxidation and GSH conjugation together)
scaled by the 3/4 power of body weight (TotMetabBW34 |mg/kg '/week|) This dose-metric
uses the total flux of TCE metabolism as the toxicologically relevant dose, and, thus,
incorporates the possible involvement of oxidative metabolites, acting either additively or
interactively, in addition to GSH conjugation metabolites in nephrocarcinogenicity (see
Section 4.4.6). While there is no evidence that TCE oxidative metabolites can on their own
induce kidney cancer, some nephrotoxic effects attributable to oxidative metabolites (e.g.,
peroxisome proliferation) may modulate the nephrocarcinogenic potency of GSH metabolites.
43The range of the difference is 2.1-2.4-fold using the posterior medians for the relative kidney weight in rats and
humans from the PBPK model described in Section 3.5 (see Table 3-36) and body weights of 0.3-0.4 kg for rats and
60-70 kg for humans.
This document is a draft for review purposes only and does not constitute Agency policy.
127 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
However, this dose-metric is given less weight than those involving GSH conjugation because,
as discussed in Sections 4.4.6 and 4.4.7, the weight of evidence supports the conclusion that
GSH conjugation metabolites play a predominant role in nephrocarcinogenicity.
5.2.1.1.3.2.Liver
As discussed in Section 4.5.6, there is substantial evidence that oxidative metabolism is
involved in TCE hepatocarcinogenicity, based primarily on noncancer and cancer effects similar
to those observed with TCE being observed with a number of oxidative metabolites of TCE (e.g.,
CH, TCA, and DCA). While TCA is a stable, circulating metabolite, CH and DCA are relatively
short-lived, although enzymatically cleared (see Section 3.3.3.1). As discussed in Sections 4.5.6
and 4.5.7, there is now substantial evidence that TCA does not adequately account for the
hepatocarcinogenicity of TCE; therefore, unlike in previous dose-response analyses (Clewell and
Andersen, 2004; Rhomberg, 2000), the AUC of TCA in plasma and in liver were not considered
as dose-metrics. However, there are inadequate data across species to quantify the dosimetry of
CH and DCA, and other intermediates of oxidative metabolism (such as TCE-oxide or
dichloroacetylchloride) also may be involved in carcinogenicity. Thus, due to uncertainties as to
the active moiety(ies), but the strong evidence associating TCE liver effects with oxidative
metabolism in the liver, hepatic oxidative metabolism is the preferred basis for internal dose
extrapolations of TCE-induced liver tumors.
The rationales for the dose-metrics for liver tumors are the same as for liver noncancer
toxicity, discussed above in Section 5.1.3.1.2, and not repeated here. The primary internal
dose-metric for TCE-induced liver tumors is selected to be the weekly rate of hepatic oxidation
per unit body weight to the 3/4 power (AMetLivlBW34 |mg/kg '/week|) Due to the larger
relative liver weight in mice as compared to humans, scaling by liver weight instead of body
weight to the 3/4 power would only change the quantitative interspecies extrapolation by about
fourfold,44 so the sensitivity of the results to the scaling choice is relatively modest. The total
amount of oxidative metabolism of TCE scaled by the 3/4 power of body weight
(TotOxMetabBW34 [mg/kg/4/week]) was selected as an alternative dose-metric (the
justification for the body weight to the 3/4 power scaling is analogous to that for hepatic oxidative
metabolism, above). This dose-metric accounts for the possible additional contributions of
systemically delivered products of lung oxidation.
44The range of the difference is 3.5-3.9-fold using the posterior medians for the relative liver weight in mice and
humans from the PBPK model described in Section 3.5 (see Table 3-37) and body weights of 0.03-0.04 kg for mice
and 60-70 kg for humans.
This document is a draft for review purposes only and does not constitute Agency policy.
128 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
5.2.1.1.3.3.Lung
As discussed in Section 4.7.3, in situ oxidative metabolism in the respiratory tract may be
more important to lung toxicity than systemically delivered metabolites, at least as evidenced by
acute pulmonary toxicity. While chloral was originally implicated as the active metabolite,
based on either acute toxicity or mutagenicity of chloral and/or chloral hydrate, more recent
evidence suggests that other oxidative metabolites may also contribute to lung toxicity. These
data include the identification of dichloroacetyl lysine adducts in Clara cells (Forkert et al.,
2006), and the induction of pulmonary toxicity by TCE in CYP2El-null mice, which may
generate a different spectrum of oxidative metabolites as compared to wild-type mice
(respiratory tract tissue also contains P450s from the CYP2F family). Overall, the weight of
evidence supports the selection of respiratory tract oxidation of TCE as the preferred basis for
internal dose extrapolations of TCE-induced lung tumors. However, uncertainties remain as to
the relative contributions from different oxidative metabolites, and quantitative clearance data
necessary to calculate the concentration of each species are lacking.
Under the cross-species scaling methodology, the rate of respiratory tract oxidation
would be scaled by body weight to the % power. For chloral, as discussed in Section 4.7.3, the
reporting of substantial TCOH but no detectable chloral hydrate in blood following TCE
exposure from experiments in isolated, perfused lungs (Dalbey and Bingham, 1978) support the
conclusion that chloral does not leave the target tissue in substantial quantities, but that there is
substantial clearance by enzyme-mediated biotransformation. Dichloroacetyl chloride is a
relatively-short-lived intermediate from aqueous (nonenzymatic) decomposition of TCE-oxide
that can be trapped with lysine or degrade further to form DC A, among other products (Cai and
Guengerich, 1999). Cai and Guengerich (1999) reported a half-life of TCE-oxide under aqueous
conditions of 12 s at 23EC, a time-scale that would be shorter at physiological conditions (37EC)
and that includes formation of dichloroacetyl chloride as well as its decomposition. Therefore,
evidence for this metabolite suggests its clearance both is sufficiently rapid so that it would
remain at the site of formation and is nonenzymatically mediated so that its rate would be
independent of body weight. Other oxidative metabolites may also play a role, but, because they
have not been identified, no inferences can be made as to their clearance.
Therefore, because it is not clear what the contributions to TCE-induced lung tumors are
from different oxidative metabolites produced in situ and the scaling by body weight to the
3/4 power is supported for at least one of the possible active moieties, it was decided here to scale
the rate of respiratory tract tissue oxidation of TCE by body weight to the % power. The primary
This document is a draft for review purposes only and does not constitute Agency policy.
129 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
internal dose-metric for TCE-induced lung tumors is, thus, the weekly rate of respiratory tract
oxidation per unit body weight to the 3/4 power (AMetLngBW34 |mg/kg '/week|) It should be
noted that, due to the larger relative respiratory tract tissue weight in mice as compared to
humans, scaling by tissue weight instead of body weight to the 3/4 power would change the
quantitative interspecies extrapolation by less than twofold,45 so the sensitivity of the results to
the scaling choice is relatively small.
While there is substantial evidence that acute pulmonary toxicity is related to pulmonary
oxidative metabolism, for carcinogenicity, it is possible that, in addition to locally produced
metabolites, systemically-delivered oxidative metabolites also play a role. Therefore, total
oxidative metabolism scaled by the 3/4 power of body weight (TotOxMetabBW34
[mg/kgy4/week]) was selected as an alternative dose-metric (the justification for the body weight
to the 3/4 power scaling is analogous to that for respiratory tract oxidative metabolism, above).
Another alternative dose-metric considered here is the AUC of TCE in blood (AUCCBld
[mg-hour/L/week]). This dose-metric would account for the possibility that local metabolism is
determined primarily by TCE delivered in blood via systemic circulation to pulmonary tissue
(the flow rate of which scales as body weight to the 3/4 power), as assumed in previous PBPK
models, rather than TCE delivered in air via diffusion to the respiratory tract, as is assumed in
the PBPK model described in Section 3.5. However, as discussed in Section 3.5 and
Appendix A, the available pharmacokinetic data provide greater support for the updated model
structure. This dose-metric also accounts for the possible role of TCE itself in pulmonary
carcinogenicity (consistent with the assumption that the same average concentration of TCE in
blood will lead to a similar lifetime cancer risk across species).
5.2.1.1.3 .4.Other sites
For all other sites listed in Table 5-34, there is insufficient information for site-specific
determinations of appropriate dose-metrics. While TCE metabolites and/or metabolizing
enzymes have been reported in some of these tissues (e.g., male reproductive tract), their roles in
carcinogenicity for these specific sites have not been established. Although "primary" and
"alternative" dose-metrics are defined, they do not differ appreciably in their degrees of
plausibility.
45The range of the difference is 1.6-1.8-fold using the posterior medians for the relative respiratory tract tissue
weight in mice and humans from the PBPK model described in Section 3.5 (see Table 3-37), and body weights of
0.03-0.04 kg for mice and 60-70 kg for humans.
This document is a draft for review purposes only and does not constitute Agency policy.
13 0 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Given that the majority of the toxic and carcinogenic responses to TCE appear to be
associated with metabolism, total metabolism of TCE scaled by the 3/4 power of body weight was
selected as the primary dose-metric (TotMetabBW34 |mg/kg '/week|) This dose-metric uses
the total flux of TCE metabolism as the toxicologically-relevant dose, and, thus, incorporates the
possible involvement of any TCE metabolite in carcinogenicity.
An alternative dose-metric considered here is the AUC of TCE in blood. This
dose-metric would account for the possibility that the determinant of carcinogenicity is local
metabolism, governed primarily by TCE delivered in blood via systemic circulation to the target
tissue (the flow rate of which scales as body weight to the 3/4 power). This dose-metric also
accounts for the possible role of TCE itself in carcinogenicity (consistent with the assumption
that the same average concentration of TCE in blood will lead to a similar lifetime cancer risk
across species).
5.2.1.1.4. Methods for dose-response analyses using internal dose-metrics
As shown in Figure 5-5, the general approach taken for the use of internal dose-metrics in
dose-response modeling was to first apply the rodent PBPK model to obtain rodent values for the
dose-metrics corresponding to the applied doses in a bioassay. Then, dose-response modeling
for a tumor response was performed using the internal dose-metrics and the multistage model or
the survival-adjusted modeling approaches described above to obtain a BMD and BMDL in
terms of the dose-metric. On an internal dose basis, humans and rodents are presumed to have
similar lifetime cancer risks, and the relationship between human internal and external doses is
essentially linear at low doses up to 0.1 mg/kg/day or 0.1 ppm, and nearly linear up to
10 mg/kg/day or 10 ppm. Therefore, the BMD and BMDL were then converted human
equivalent doses (or exposures) using conversion ratios estimated from the human PBPK model
at 0.001 mg/kg/day or 0.001 ppm (see Table 5-35). Because the male and female conversions
differed by less than 11%, the human BMDLs were derived using the mean of the sex-specific
conversion factors (except for testicular tumors, for which only male conversion factors were
used). Finally, a slope factor or unit risk estimate for that tumor response was derived from the
human "BMDLs" as described above (i.e., BMR/BMDL). Note that the converted "BMDs" and
"BMDLs" are not actually human equivalent BMDs and BMDLs corresponding to the BMR
because the conversion was not made in the dose range of the BMD; the converted BMDs and
BMDLs are merely intermediaries to obtain a converted slope factor or unit risk estimate. In
addition, it should be noted that median values of dose-metrics were used for rodents, whereas
mean values were used for humans. Because the rodent population model characterizes
This document is a draft for review purposes only and does not constitute Agency policy.
131 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
study-to-study variation, animals of the same sex/species/strain combination within a study were
assumed to be identical. Therefore, use of median dose-metric values for rodents can be
interpreted as assuming that the animals in the bioassay were all "typical" animals and the
dose-response model is estimating a "risk to the typical rodent." In practice, the use of median
or mean internal doses for rodents did not make much difference except when the uncertainty in
the dose-metric was high (e.g., AMetLungBW34 dose-metric in mice). A quantitative analysis
of the impact of the uncertainty in the rodent PBPK dose-metrics is included in Section 5.2.1.4.2.
On the other hand, the human population model characterizes individual-to-individual variation.
Because the quantity of interest is the human population mean risk, the expected value
(averaging over the uncertainty) of the population mean (averaging over the variability)
dose-metric was used for the conversion
Figure 5-5. Flow-chart for dose-response analyses of rodent bioassays using
PBPK model-based dose-metrics. Square nodes indicate point values,
circular nodes indicate distributions, and the inverted triangles indicate a
(deterministic) functional relationship.
This document is a draft for review purposes only and does not constitute Agency policy.
132 DRAFT—DO NOT CITE OR QUOTE
-------
Bioassay
experimental
paradigm
^distribution
PBPK
model
Rodent
internal
dose
Bioassay
edian
Dose-
esponse
Model
[distribution (combined
iLincertainty and variability)
0.001 ppm
in air or
0.001 mg/kg-d
continuous
exposure
fixed <
Human
model
parameters
PBPK
model
BMD, BMDL
(internal
dose unit)
istribution
[distribution (separate
ncertainty and variability)
Human
internal
dose
[population
^[nean
nean
Site-specific
cancer unit
risk = BMR/
BMDL
(per internal
dose unit)
Expected
average
human
internal dose
per ppm or
per mg/kg-d
Human site-
specific cancer
unit risk
(per ppm or
per mg/kg-d)
Table 5-35. Mean PBPK model predictions for weekly internal dose in
humans exposed continuously to low levels of TCE via inhalation (ppm) or
orally (mg/kg/day)
This document is a draft for review purposes only and does not constitute Agency policy.
13 3 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Dose-metric
0.001 ppm
0.001 mg/kg/day
Female
Male
Female
Male
ABioactDCVCBW34
0.00324
0.00324
0.00493
0.00515
AMetGSHB W 3 4
0.00200
0.00200
0.00304
0.00318
AMetLivlBW34
0.00703
0.00683
0.0157
0.0164
AMetLngB W 3 4
0.00281
0.00287
6.60x10 5
6.08x10 5
AUCCBld
0.00288
0.00298
0.000411
0.000372
TotMetabBW34
0.0118
0.0117
0.0188
0.0196
T otOxMetabB W 3 4
0.00984
0.00970
0.0157
0.0164
See note to Table 5-34 for dose-metric abbreviations. Values represent the mean of the (uncertainty) distribution of
population means for each sex and exposure scenario, generated from Monte Carlo simulation of 500 populations of
500 individuals each.
to human slope factor or unit risks. Therefore, the extrapolated slope factor or unit risk estimates
can be interpreted as the expected "average risk" across the population based on rodent
bioassays.
5.2.1.1.5. Rodent Dose-Response Analyses: Results
A summary of the PODs and slope factor and unit risk estimates for each
sex/species/bioassay/tumor type is presented in Tables 5-36 (inhalation studies) and 5-37 (oral
studies). The PODs for individual tumor types were extracted from the modeling results in the
figures in Appendix G. For the applied dose (default dosimetry) analyses, the POD is the BMDL
from the male human ("M") BMDL entry at the top of the figure for the selected model; male
results were extracted because the default weight for males in the PBPK modeling is 70 kg,
which is the overall human weight in EPA's default dosimetry methods (for inhalation, male and
female results are identical). As described in Section 5.2.1.2 above, for internal dose-metrics,
male and female results were averaged, and the converted human "BMDLs" are not true BMDLs
because they were converted outside the linear range of the PBPK models. It can be seen in
Appendix G that the male and female results were similar for all the dose-metrics.
For two data sets, the highest dose (exposure) group was dropped to get a better fit when
using applied doses. This technique can improve the fit when the response tends to plateau with
increasing dose. Plateauing typically occurs when metabolic saturation alters the pattern of
This document is a draft for review purposes only and does not constitute Agency policy.
134 DRAFT—DO NOT CITE OR QUOTE
-------
"1
Table 5-36. Summary of PODs and unit risk estimates for each sex/species/bioassay/tumor type (inhalation)
S?
>!
rs
5
Grj
Si
fS"
_ St
U) *3
o
o
>!
>!
o
Si
s
§•
>!
a
o
H §
I >!
81
w
2; ^
H J
1—1 s
H S.
W|
O •
o
c
o
H
W
Study
Tumor type
BMR
PODs (ppm, in human equivalent exposures)3
Applied
dose
AUC
CBld
TotMetab
BW34
TotOxMetab
BW34
AMetLng
BW34
AMetLivl
BW34
AMetGSH
BW34
ABioact
DCVCBW34
Female mouse
Fukuda
Lung AD + CARC
0.1
26.3
55.5
31.3
38.8
Henschler
Lymphoma
0.1
11.0b
b
9.84
Maltoni
Lung AD + CARC
0.1
44.6
96.6
51.4
55.7
Liver
0.05
37.1
45.8
41.9
Combined
0.05
15.7
20.7
Male mouse
Maltoni
Liver
0.1
34.3
51
37.9
Male rat
Maltoni
Leukemia
0.05
28.2°
b
28.3
Kidney AD + CARC
0.01
22.7
13.7
0.197
0.121
Leydig cell
0.1
18.6°
d
18.1
Combined
0.01
1.44
1.37
Study
Tumor type
Unit risk estimate (ppm 1)e
Applied dose
AUC
CBld
TotMetab
BW34
TotOxMetab
BW34
AMetLng
BW34
AMetLivl
BW34
AMetGSH
BW34
ABioact
DCVCBW34
Female mouse
Fukuda
Lung AD + CARC
3.8 x 1(T3
1.8 x 10"3
3.2 x 10"3
2.6 x 10 3
Henschler
Lymphoma
9.1 x 1(T3
1.0 x 10 2
Maltoni
Lung AD + CARC
2.2 x 1CT3
1.0 x 10"3
1.9 x 10"3
1.8 x 10 3
Liver
1.3 x 1(T3
1.1 x 10"3
1.2 x 10 3
Combined
3.2 x 1CT3
2.4 x 10 3
-------
Table 5-36. Summary of PODs and unit risk estimates for each sex/species/bioassay/tumor type (inhalation)
(continued)
Study
Tumor type
Unit risk estimate (ppm 1)e
Applied dose
AUC
CBld
TotMetab
BW34
TotOxMetab
BW34
AMetLng
BW34
AMetLivl
BW34
AMetGSH
BW34
ABioact
DCVCBW34
Male mouse
Maltoni
Liver
2.9 x 1(T3
2.0 x 10"3
2.6 x 10 3
Male rat
Maltoni
Leukemia
1.8 x 1(T3
1.8 x 10 3
Kidney AD + CARC
4.4 x icr4
7.3 x 10"4
5.1 x 10"2
8.3 x 10 2
Leydig cell
5.4 x 1(T3
5.5 x 10 3
Combined
7.0 x icr3
7.3 x 10"3
Tor the applied doses, the PODs are BMDLs. However, for the internal dose-metrics, the PODs are not actually human equivalent BMDLs corresponding to the
BMR because the interspecies conversion does not apply to the dose range of the BMDL; the converted BMDLs are merely intermediaries to obtain a converted
unit risk estimate. The calculation that was done is equivalent to using linear extrapolation from the BMDLs in terms of the internal dose-metric to get a unit
risk estimate for low-dose risk in terms of the internal dose-metric and then converting that estimate to a unit risk estimate in terms of human equivalent
exposures. The PODs reported here are what one would get if one then used the unit risk estimate to calculate the human exposure level corresponding to a 10%
extra risk, but the unit risk estimate is not intended to be extrapolated upward out of the low-dose range, e.g., above 10 4 risk. In addition, for the internal
dose-metrics, the PODs are the average of the male and female human "BMDL" results presented in Appendix G.
bInadequate fit to control group, but the primary metric, TotMetabBW34, fits adequately.
Dropped highest-dose group to improve model fit.
inadequate overall fit.
"Unit risk estimate = BMR/POD. Results for the primary dose-metric are in bold.
AD = adenoma, CARC = carcinoma.
-------
"1
Table 5-37. Summary of PODs and slope factor estimates for each sex/species/bioassay/tumor type (oral)
S?
>!
rs
5
Grj
Si
fS"
_ St
U) *3
-J
o
o
>!
>!
o
Si
s
§•
>!
a
o
H §
I >!
81
w
2; ^
H J
1—1 s
H S.
W|
O •
02
c
o
H
W
Study
Tumor type
BMR
PODs (mg/kg/day, in human equivalent doses)3
Applied
dose
AUC
CBld
TotMetab
BW34
TotOxMetab
BW34
AMetLng
BW34
AMetLivl
BW34
AMetGSH
BW34
ABioact
DCVCBW34
Female mouse
NCI
Liver carc
0.1
26.5
17.6
14.1
Lung AD + CARC
0.1
41.1
682
24.7
757
Leukemias + sarcomas
0.1
43.1
733
20.6
Combined
0.05
7.43
5.38
Male mouse
NCI
Liver carc
0.1
8.23
4.34
3.45
Female rat
NTP (1988)
Leukemia
0.05
72.3
3,220
21.7
Male rat
NTP (1990)°
Kidney AD + CARC
0.1
32
11.5
0.471
0.292
NTP (1988)
Marshall
Testicular
0.1
3.95
167
1.41
August
Subcut sarcoma
0.05
60.2
2,560
21.5
Osborne-Mendef
Kidney AD + CARC
0.1
41.5
14.3
0.648
0.402
Female mouse
NCI
Liver carc
3.8 x 1(T3
5.7 x 10"3
7.1 x 10 3
Lung AD + CARC
2.4 x 1(T3
1.5 x 10"4
4.0 x 10"3
1.3 x 10 4
Leukemias + sarcomas
2.3 x 1(T3
1.4 x 10"4
4.9 x 10 3
Combined
6.7 x 1(T3
9.3 x 10 3
-------
Table 5-37. Summary of PODs and slope factor estimates for each sex/species/bioassay/tumor type (oral)
(continued)
Study
Tumor type
Slope factor estimate (mg/kg/day) 1)b
Applied dose
AUC
CBld
TotMetab
BW34
TotOxMetab
BW34
AMetLng
BW34
AMetLivl
BW34
AMetGSH
BW34
ABioact
DCVCBW34
Male mouse
NCI
Liver carc
1.2 x 1(T2
2.3 x 10"2
2.9 x 10 2
Female rat
NTP (1988)
Leukemia
6.9 x 1(T4
1.6 x 10"5
2.3 x 10 3
Male rat
NTP (1990)°
Kidney AD + CARC
1.6 x 1(T3
4.3 x 10"3
1.1 X 10"1
1.7 x 10 1
NTP (1988)
Marshall
Testicular
2.5 x 1(T2
6.0 x 10"4
7.1 x 10 2
August
Subcut sarcoma
8.3 x 1(T4
2.0 x 10"5
2.3 x 10 3
Osborne-Mendef
Kidney AD + CARC
2.4 x 1(T3
7.0 x 10"3
1.5 x 10"1
2.5 x 10 1
Tor the applied doses, the PODs are BMDLs. However, for the internal dose-metrics, the PODs are not actually human equivalent BMDLs corresponding to the
BMR because the interspecies conversion does not apply to the dose range of the BMDL; the converted BMDLs are merely intermediaries to obtain a converted
slope factor estimate. The calculation that was done is equivalent to using linear extrapolation from the BMDLs in terms of the internal dose-metric to get a
slope factor estimate for low-dose risk in terms of the internal dose-metric and then converting that estimate to a slope factor estimate in terms of human
equivalent doses. The PODs reported here are what one would get if one then used the slope factor estimate to calculate the human dose level corresponding to
a 10% extra risk, but the slope factor estimate is not intended to be extrapolated upward out of the low-dose range, e.g., above 10 4 risk. In addition, for the
internal dose-metrics, the PODs are the average of the male and female human "BMDL" results presented in Appendix G.
bSlope factor estimate = BMR/POD. Results for the primary dose-metric are in bold.
Using MSW adjusted incidences (see text and Table 5-38).
dUsing poly-3 adjusted incidences (see text and Table 5-38).
AD = adenoma, CARC = carcinoma.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
metabolite formation or when survival is impacted at higher doses, and it is assumed that these
high-dose responses are less relevant to low-dose risk. The highest-dose group was not dropped
to improve the fit for any of the internal dose-metrics because it was felt that if the dose-metric
was an appropriate reflection of internal dose of the reactive metabolite(s), then use of the
dose-metric should have ameliorated the plateauing in the dose-response relationship (note that
rd
survival-impacted data sets were addressed using survival adjustment techniques). For a 3 data
set (Henschler lymphomas), it might have helped to drop the highest exposure group, but there
were only two exposure groups, so this was not done. As a result, the selected model, although it
had an adequate fit overall, did not fit the control group very well (the model estimated a higher
background response than was observed); thus, the BMD and BMDL were likely overestimated
and the risk underestimated. The estimates from the NCI (1976) oral male mouse liver cancer
data set are also somewhat more uncertain because the response rate was extrapolated down from
a response rate of about 50% extra risk to the BMR of 10% extra risk.
Some general patterns can be observed in Tables 5-36 and 5-37. For inhalation, the unit
risk estimates for different dose-metrics were generally similar (within about 2.5-fold) for most
tumor types. The exception was for kidney cancer, where the estimates varied by over 2 orders
of magnitude, with the AMetGSHBW34 and ABioactDCVCBW34 metrics yielding the highest
estimates. This occurs because pharmacokinetic data indicate, and the PBPK model predicts,
substantially more GSH conjugation (as a fraction of intake), and hence subsequent
bioactivation, in humans relative to rats. The range of the risk estimates for individual tumor
types overall (across tumor types and dose-metrics) was encompassed by the range of estimates
across the dose-metrics for kidney cancer in the male rat, which was from 4.4 x 10 4 per ppm
(applied dose) to 8.3 x io~2 per ppm (ABioactDCVCBW34).
For oral exposure, the slope factor estimates are more variable across dose-metrics
because of first-pass effects in the liver (median estimates for the fraction of TCE metabolized in
one pass through the liver in mice, rats, and humans are >0.8). Here, the exception is for the risk
estimates for cancer of the liver itself, which are also within about a 2.5-fold range, because the
liver gets the full dose of all the metrics during that "first pass." For the other tumor types, the
range of estimates across dose-metrics varies from about 30-fold to over two orders of
magnitude, with the estimates based on AUCCBld and AMetLngBW34 being at the low end and
those based on AMetGSHBW34 and ABioactDCVCBW34 again being at the high end. For
AUCCBld, the PBPK model predicted the blood concentrations to scale more closely to body
weight rather than the 3/4 power of body weight, so the extrapolated human unit risks using this
dose-metric are smaller than those obtained by applied dose or other dose-metrics that included
3/4 power body weight scaling. For AMetLngBW34, pharmacokinetic data indicate, and the
This document is a draft for review purposes only and does not constitute Agency policy.
13 9 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
PBPK model predicts, that the human respiratory tract metabolizes a lower fraction of total TCE
intake than the mouse respiratory tract, so the extrapolated risk to humans based on this metric is
lower than that obtained using applied dose or other dose-metrics. Overall, the oral slope factor
estimates for individual tumor types ranged from 1.6 x 10 5 per mg/kg/day (female rat leukemia,
AUCCBld) to 2.5 x 10 1 per mg/kg/day (male Osborne-Mendel rat kidney,
ABioactDCVCBW34), a range of over 4 orders of magnitude. It must be recognized, however,
that not all dose-metrics are equally credible, and, as will be presented below, the slope factor
estimates for total cancer risk for the most sensitive bioassay response for each sex/species
combination using the primary (preferred) dose-metrics fall within a very narrow range.
Results for survival-adjusted analyses are summarized in Table 5-38. For the
time-independent (BMDS) multistage model, the risk estimates using poly-3 adjustment are
higher than those without poly-3 adjustment. This is to be expected because the
poly-3 adjustment decreases denominators when accounting for early mortality, and, for these
data sets, the higher-dose groups had greater early mortality. The difference was fairly modest
for the kidney cancer data sets (about 30% higher) but somewhat larger for the testicular cancer
data set (about 150% higher).
In addition, the MSW time-to-tumor model generated higher risk estimates than the
poly-3 adjustment technique. The MSW results were about 40% higher for the NTP F344 rat
kidney cancer data sets and about 60% higher for the NTP Osborne-Mendel rat kidney cancer
data sets. For the NTP Marshall rat testicular cancer data set, the discrepancies were greater; the
results ranged from about 100% to 180% higher for the different dose-metrics. As discussed in
Section 5.2.1.1, these two approaches differ in the way they take early mortality into account.
The poly-3 technique merely adjusts the tumor incidence denominators, using a constant power
3 of time, to reflect the fact that animals are at greater risk of cancer at older ages. The MSW
model estimates risk as a function of time (and dose), and it estimates the power (of time)
parameter for each data set.46 For the NTP F344 rat kidney cancer and NTP Marshall rat
testicular cancer data sets, the estimated power parameter was close to 3 in each case, ranging
from 3.0-3.7; for the NTP Osborne-Mendel rat kidney cancer data sets, however, the estimated
power parameter was about 10 for each of the dose-metrics, presumably reflecting the fact that
these were late-occurring tumors (the earliest occurred at 92 weeks). Using a higher power
46Conceptually, the approaches differ most when different tumor contexts (incidental or fatal) are considered,
because the poly-3 technique only accounts for time of death, while the MSW model can account for the tumor
context and attempt to estimate an induction time (tO), although this was not done for any of the data sets in this
assessment.
This document is a draft for review purposes only and does not constitute Agency policy.
140 DRAFT—DO NOT CITE OR QUOTE
-------
1 parameter than 3 in the poly-3 adjustment would give even less weight to nontumor-bearing
2 animals that die early and would, thus, increase the adjusted incidence even more in the
3 Table 5-38. Comparison of survival-adjusted results for 3 oral male rat data
4 sets"
5
Dose-metric
Adjustment
method
BMR
POD
(mg/kg/day)
BMD:BMDL
Slope factor estimate
(per mg/kg/day)
NTP (1990) F344 rat kidney AD + CARC
Applied dose
unadj BMDS
0.05
56.9
1.9
8.8 x 10~4
poly-3 BMDS
0.1
89.2
1.9
1.1 x KT3
MSW
0.05
32.0
2.6
1.6 x 10~3
TotMetabBW34
unadj BMDS
0.05
20.2
2.1
2.5 x 10~3
poly-3 BMDS
0.1
31.8
1.7
3.1 x KT3
MSW
0.05
11.5
3.1
4.3 x 10~3
AMetGSHBW34
unadj BMDS
0.05
0.841
1.9
5.9 x 10~2
poly-3 BMDS
0.1
1.32
1.9
7.6 x 10~2
MSW
0.05
0.471
2.4
1.1 x KT1
ABioactDCVCBW34
unadj BMDS
0.05
0.522
1.9
9.6 x 10~2
poly-3 BMDS
0.1
0.817
1.9
1.2 x 10"1
MSW
0.05
0.292
2.4
1.7 x 10"1
NTP (1988) Osborne-Mendel rat kidney AD + CARC
Applied dose
unadj BMDS
0.1
86.6
1.7
1.2 x 10~3
poly-3 BMDS
0.1
65.9
1.7
1.5 x 10~3
MSW
0.1
41.5
2.0
2.4 x 10~3
TotMetabBW34
unadj BMDS
0.1
30.4
1.7
3.3 x 10~3
poly-3 BMDS
0.1
23.1
1.7
4.3 x 10~3
MSW
0.1
14.3
2.0
7.0 x 10~3
AMetGSHBW34
unadj BMDS
0.1
1.35
1.7
7.4 x 10~2
poly-3 BMDS
0.1
1.03
1.7
9.7 x 10~2
MSW
0.1
0.648
2.0
1.5 x 10"1
ABioactDCVCBW34
unadj BMDS
0.1
0.835
1.7
1.2 x 10"1
poly-3 BMDS
0.1
0.636
1.7
1.6 x 10"1
MSW
0.1
0.402
2.0
2.5 x 10"1
NTP (1988) Marshall rat testicular tumors
Applied dose
unadj BMDS
0.1
9.94
1.4
1.0 x 10~2
poly-3 BMDS
0.1
3.95
1.5
2.5 x 10~2
MSW
0.1
1.64
5.2
6.1 x 10~2
AUCCBld
unadj BMDS
0.1
427
1.4
2.3 x 10~4
poly-3 BMDS
0.1
167
1.6
6.0 x 10~4
MSW
0.1
60.4
2.6
1.7 x 10~3
TotMetabBW34
unadj BMDS
0.1
3.53
4.3
2.8 x 10~2
poly-3 BMDS
0.1
1.41
1.5
7.1 x 10"2
MSW
0.1
0.73
9.4
1.4 x 10"1
6
7 Tor the applied doses, the PODs are BMDLs. However, for the internal dose-metrics, the PODs are not actually
8 human equivalent BMDLs corresponding to the BMR because the interspecies conversion does not apply to the
This document is a draft for review purposes only and does not constitute Agency policy.
141 DRAFT—DO NOT CITE OR QUOTE
-------
1 dose range of the BMDL; the converted BMDLs are merely intermediaries to obtain a converted slope factor
2 estimate. Results for the primary dose-metric are in bold.
3
4 AD = adenoma, CARC = carcinoma.
5
This document is a draft for review purposes only and does not constitute Agency policy.
142 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
highest-dose groups where the early mortality is most pronounced, increasing the slope factor
estimate. Nonetheless, as noted above, the MSW results were only about 60% higher for the
NTP Osborne-Mendel rat kidney cancer data sets for which MSW estimated a power parameter
of about 10.
In general, the risk estimates from the MSW model would be preferred because, as
discussed above, this model incorporates more information (e.g., tumor context) and estimates
the power parameter rather than using a constant value of three. From Table 5-38, it can be seen
that the results from MSW yielded higher BMD:BMDL ratios than the results from the
poly-3 technique. These ratios were only slightly higher and not unusually large for MSW
model analyses of the NTP (1988, 1990) kidney tumor estimates, and this, along with the
adequate fit (assessed visually) of the MSW model, supports using the slope factor estimates
from the MSW modeling of rat kidney tumor incidence. On the other hand, the BMD:BMDL
ratio was relatively large for the applied dose analysis and, in particular, for the preferred
dose-metric analysis (9.4-fold) of the NTP Marshall rat testicular tumor data set. Therefore, for
this endpoint, the poly-3-adjusted results were used, although they may underestimate risk
somewhat as compared to the MSW model.
In addition to the results from dose-response modeling of individual tumor types, the
results of the combined tumor risk analyses for the three bioassays in which the rodents exhibited
increased risks at multiple sites are also presented in Tables 5-36 and 5-37, in the rows labeled
"combined" under the column heading "Tumor Type." These results were extracted from the
detailed results in Appendix G. Note that, because of the computational complexity of the
combined tumor analyses, dose-response modeling was only done using applied dose and a
common upstream internal dose-metric, rather than using the different preferred dose-metrics for
each tumor type within a combined tumor analysis.
For the Maltoni female mouse inhalation bioassay, the combined tumor risk estimates are
bounded by the highest individual tumor risk estimates and the sums of the individual tumor
risks estimates (the risk estimates are upper bounds, so the combined risk estimate, i.e., the upper
bound on the sum of the individual central tendency estimates, should be less than the sum of the
individual upper bound estimates), as one would expect. The common upstream internal
dose-metric used for the combined analysis was TotOxMetabBW34, which is not the primary
metric for either of the individual tumor types. For the liver tumors, the primary metric was
AMetLivlBW34, but as can be seen in Table 5-36, it yields results similar to those for
TotOxMetabBW34. Likewise, for the lung tumors, the primary metric was AMetLngBW34,
which yields a unit risk estimate slightly smaller that for TotOxMetabBW34. Thus, the results of
This document is a draft for review purposes only and does not constitute Agency policy.
143 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
the combined analysis using TotOxMetabBW34 as a common metric is not likely to substantially
over- or underestimate the combined risk based on preferred metrics for each of the tumor types.
For the Maltoni male rat inhalation bioassay, the combined risk estimates are also
reasonably bounded, as expected. The common upstream internal dose-metric used for the
combined analysis was TotMetabBW34, which is the primary metric for two of the
three individual tumor types. However, as can be seen in Table 5-36, the risk estimate for the
preferred dose-metric for the third tumor type, ABioactDCVCBW34 for the kidney tumors, is
substantially higher than the risk estimates for the primary dose-metrics for the other two tumor
types and would dominate a combined tumor risk estimate across primary dose-metrics; thus, the
ABioactDCVCBW34-based kidney tumor risk estimate alone can reasonably be used to
represent the total cancer risk for the bioassay using preferred internal dose-metrics, although it
would underestimate the combined risk to some extent (e.g., the kidney-based estimate is
—2 — 2
8.3 x 10 per ppm; the combined estimate would be about 9x10 per ppm, rounded to
one significant figure).
For the third bioassay (NCI female mouse oral bioassay), the combined tumor risk
estimates are once again reasonably bounded. The common upstream internal dose-metric used
for the combined analysis was TotOxMetabBW34, which is not the primary metric for any of the
three individual tumor types but was considered to be the most suitable metric to apply as a basis
for combining risk across these different tumor types. The slope factor estimate for the lung
based on the primary dose-metric for that site becomes negligible compared to the estimates for
the other two tumor types (see Table 5-37). However, the slope factor estimates for the
remaining two tumor types are both somewhat underestimated using the TotOxMetabBW34
metric rather than the primary metrics for those tumors (the TotOxMetabBW34-based estimate
for leukemias + sarcomas, which is not presented in Table 5-30 because, in the absence of better
mechanistic information, more upstream metrics were used for that individual tumor type, is
4.1 x 10 per mg/kg/day). Thus, overall, the combined estimate based on TotOxMetabBW34 is
probably a reasonable estimate for the total tumor risk in this bioassay, although it might
overestimate risk slightly.
The most sensitive sex/species results are extracted from Tables 5-29 and 5-30 and
presented in Tables 5-39 (inhalation) and 5-40 (oral) below. The BMD:BMDL ratios for all the
results corresponding to the slope factor and unit risk estimates based on the preferred
dose-metrics ranged from 1.3-2.1. For inhalation, the most sensitive bioassay responses based
_3 _2
on the preferred dose-metrics ranged from 2.6 x 10 per ppm to 8.3 x 10 per ppm across the
sex/species combinations (with the exception of the female rat, which exhibited no apparent
This document is a draft for review purposes only and does not constitute Agency policy.
144 DRAFT—DO NOT CITE OR QUOTE
-------
1 TCE-associated response in the 3 available bioassays). For oral exposure, the most sensitive
2 bioassay responses
3
4 Table 5-39. Inhalation: most sensitive bioassay for each sex/species
5 combination"
6
Sex/species
Endpoint
(study)
Unit risk per ppm
Preferred
dose-metric
Default
methodology
Alternative
dose-metrics,
studies, or
endpoints
Female mouse
Lymphoma
(Henschler et al., 1980)
1.0 x 10~2
9.1 x 10~3
1 x 10~3 ~4 x 10~3
Male mouse
Liver hepatoma
(Maltoni et al., 1986)
2.6 x 10~3
2.9 x 10~3
2 x 10~3
Female rat
-
-
-
-
Male rat
Leukemia+
Kidney AD and CARC+
Leydig cell tumors
(Maltoni et al., 1986)
8.3 x 10~2
7.0 x 10~3
4 x 10~4~ 5 x 10~2
[individual site
results]
7
8 "Results extracted from Table 5-36.
9 AD = adenoma, CARC = carcinoma.
10
11
12 Table 5-40. Oral: most sensitive bioassay for each sex/species combination"
13
Sex/species
Endpoint
(Study)
Unit risk per mg/kg/day
Preferred
dose-metric
Default
methodology
Alternative
dose-metrics,
studies, or
endpoints
Female mouse
Liver CARC +
lung AD and CARC+
sarcomas + leukemias
(NCI, 1976)
9.3 x 10~3
6.7 x 10~3
1 x 10~4 ~ 7 x 10~3
[individual site
results]
Male mouse
Liver CARC
(NCI, 1976)
2.9 x 10~2
1.2 x 10~2
2 x 10~2
Female rat
Leukemia
(NTP, 1988)
2.3 x 10~3
6.9 x 10~4
2 x 10~5
This document is a draft for review purposes only and does not constitute Agency policy.
145 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Male rat
Kidney AD + CARC
2.5 x 10_1
2.4 x 10~3b
2 x 10~5 ~2 x 10_1
(NTP, 1988, Osborne-
Mendel)
"Results extracted from Table 5-37.
bMost sensitive male rat result using default methodology is 2.5 x 10 2 per mg/kg/day for NTP (1988) Marshall rat
testicular tumors.
AD = adenoma, CARC = carcinoma.
_3 _ i
based on the preferred dose-metrics ranged from 2.3 x 10 per mg/kg/day to 2.5 x 10 per
mg/kg/day across the sex/species combinations. For both routes of exposure, the most sensitive
sex/species response was (or was dominated by, in the case of the combined tumors in the male
rat by inhalation) male rat kidney cancer based on the preferred dose-metric of
ABioactDCVCBW34.
5.2.1.1.6. Uncertainties in Dose-Response Analyses of Rodent Bioassays
5.2.1.1.7. Qualitative discussion of uncertainties
All risk assessments involve uncertainty, as study data are extrapolated to make
inferences about potential effects in humans from environmental exposure. The largest sources
of uncertainty in the TCE rodent-based cancer risk estimates are interspecies extrapolation and
low-dose extrapolation. Some limited human (occupational) data from which to estimate human
cancer risk are available, and cancer risk estimates based on these data are developed in
Section 5.2.2 below. In addition, some quantitative uncertainty analyses of the interspecies
differences in pharmacokinetics were conducted and are presented in Section 5.2.1.4.2.
The rodent bioassay data offer conclusive evidence of carcinogenicity in both rats and
mice, and the available epidemiologic and mechanistic data support the relevance to humans of
the TCE-induced carcinogenicity observed in rodents. The epidemiologic data provide sufficient
evidence that TCE is "carcinogenic to humans" (see Section 4.11). There is even some evidence
of site concordance with the rodent findings, although site concordance is not essential to human
relevance and, in fact, is not observed across TCE-exposed rats and mice. The strongest
evidence in humans is for TCE-induced kidney tumors, with fairly strong evidence for
lymphomas and some lesser support for liver tumors; each of these tumor types has also been
observed in TCE rodent bioassays. Furthermore, the mechanistic data are supportive of human
relevance because, while the exact reactive species associated with TCE-induced tumors are not
known, the metabolic pathways for TCE are qualitatively similar for rats, mice, and humans (see
This document is a draft for review purposes only and does not constitute Agency policy.
146 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Section 3.3). The impact of uncertainties with respect to quantitative differences in TCE
metabolism is discussed in Section 5.2.1.4.2.
Typically, the cancer risk estimated is for the total cancer burden from all sites that demonstrate
an increased tumor incidence for the most sensitive experimental species and sex. It is expected
that this approach is protective of the human population, which is more diverse but is exposed to
lower exposure levels.
For the inhalation unit risk estimates, the preferred estimate from the most sensitive
_2
species and sex was the estimate of 8.3 x 10 per ppm for the male rat, which was based on
multiple tumors observed in this sex/species but was dominated by the kidney tumor risk
estimated with the dose-metric for bioactivated DCVC. This estimate was the high end of the
range of estimates (see Table 5-39) but was within an order of magnitude of other estimates,
such as the preferred estimate for the female mouse and the male rat kidney estimate based on
the GSH conjugation dose-metric, which provide additional support for an estimate of this
magnitude. The preferred estimate for the male mouse was about an order of magnitude and a
half lower. The female rat showed no apparent TCE-associated tumor response in the three
available inhalation bioassays; however, this apparent absence of response is inconsistent with
the observations of increased cancer risk in occupationally exposed humans and in female rats in
oral bioassays. In Section 5.2.2.2, an inhalation unit risk estimate based on the human data is
derived and can be compared to the rodent-based estimate.
For the oral slope factor estimate, the preferred estimate from the most sensitive species
and sex was the estimate of 2.5 x 10 1 per mg/kg/day, again for the male rat, based on the kidney
tumor risk estimated with the dose-metric for bioactivated DCVC. This estimate was at the high
end of the range of estimates (see Table 5-40) but was within an order of magnitude of other
estimates, such as the preferred male mouse estimate and the male rat kidney estimate based on
the GSH conjugation dose-metric, which provide additional support for an estimate of this
magnitude. The preferred estimates for the female mouse and the female rat were about another
order of magnitude lower. Some of the oral slope factor estimates based on the alternative
dose-metric of AUC for TCE in the blood were as much as three orders of magnitude lower, but
these estimates were considered less credible than those based on the preferred dose-metrics. In
Section 5.2.2.3, an oral slope factor estimate based on the human (inhalation) data is derived
using the PBPK model for route-to-route extrapolation; this estimate can be compared to the
rodent-based estimate.
Furthermore, the male rat kidney tumor estimates from the inhalation (Maltoni et al., 1986) and
oral (NTP, 1988) studies were consistent on the basis of internal dose using the dose-metric for
bioactivated DCVC. In particular, the linearly extrapolated slope (i.e., the BMR/BMDL) per unit
This document is a draft for review purposes only and does not constitute Agency policy.
147 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
of internal dose derived from Maltoni et al. (1986) male rat kidney tumor data was 2.4 x 10 1 per
weekly mg DCVC bioactivated per unit body weight74, while the analogous slope derived from
_2
NTP (1988) male rat kidney tumor data was 9.3 x 10 per weekly mg DCVC bioactivated per
unit body weight74 (MSW-modeled results), a difference of less than
threefold.47 These results also suggest that differences between routes of administration are
adequately accounted for by the PBPK model using this dose-metric.
Regarding low-dose extrapolation, a key consideration in determining what extrapolation
approach to use is the MOA(s). However, MOA data are lacking or limited for each of the
cancer responses associated with TCE exposure, with the exception of the kidney tumors (see
Section 4.11). For the kidney tumors, the weight of the available evidence supports the
conclusion that a mutagenic MOA is operative (see Section 4.4); this MOA supports linear
low-dose extrapolation. For the other TCE-induced tumors, the MOA(s) is unknown. When the
MOA(s) cannot be clearly defined, EPA generally uses a linear approach to estimate low-dose
risk (U.S. EPA, 2005c), based on the following general principles:
• A chemical's carcinogenic effects may act additively to ongoing biological processes,
given that diverse human populations are already exposed to other agents and have
substantial background incidences of various cancers.
• A broadening of the dose-response curve (i.e., less rapid fall-off of response with
decreasing dose) in diverse human populations and, accordingly, a greater potential for
risks from low-dose exposures (!!! INVALID CITATION !! !)is expected for two reasons:
First, even if there is a "threshold" concentration for effects at the cellular level, that
threshold is expected to differ across individuals. Second, greater variability in response
to exposures would be anticipated in heterogeneous populations than in inbred laboratory
species under controlled conditions (due to, e.g., genetic variability, disease status, age,
nutrition, and smoking status).
47For the Maltoni et al. (1986') male rat kidney tumors, the unit risk estimate of 8.3 x 10-2 per ppm using the
ABioactDCVCBW34 dose metric, from Table 5-36, is divided by the average male and female internal doses at
0.001 ppm, (0.0034/0.001), from Table 5-35, to yield a unit risk in internal dose units of 2.4 x 10-2. For the
NTP (1988) male rat kidney tumors, the unit risk estimate of 2.5 x 10-1 per mg/kg/day using the
ABioactDCVCBW34 dose metric, from Table 5-37, is divided by the average male and female internal doses at
0.001 mg/kg/day, (0.0027/0.001), from Table 5-35, to yield a unit risk in internal dose units of 9.3 x 10-2. Note that
the original BMDLs and unit risks from BMD modeling were in internal dose units that were then converted to
applied dose units using the values in Table 5-35, so this calculation reverses that conversion.
This document is a draft for review purposes only and does not constitute Agency policy.
148 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
• The general use of linear extrapolation provides reasonable upper-bound estimates that
are believed to be health-protective (U.S. EPA, 2005c) and also provides consistency
across assessments.
Additional uncertainties arise from the specific dosimetry assumptions, the model
structures and parameter estimates in the PBPK models, the dose-response modeling of data in
the observable range, and the application of the results to potentially sensitive human
populations. As discussed in Section 5.2.1.2.1, one uncertainty in the tissue-specific
dose-metrics used here is whether to scale the rate of metabolism by tissue mass or body weight
to the 3/4 in the absence of specific data on clearance; however, in the cases where this is an issue
(the lung, liver, and kidney), the impact of this choice is relatively modest (less than twofold to
about fourfold). An additional dosimetry assumption inherent in this analysis is that equal
concentrations of the active moiety over a lifetime yield equivalent lifetime risk of cancer across
species, and the extent to which this is true for TCE is unknown. Furthermore, it should be noted
that use of tissue-specific dosimetry inherently presumes site concordance of tumors across
species.
With respect to uncertainties in the estimates of internal dose themselves, a quantitative
analysis of the uncertainty and variability in the PBPK model-predicted dose-metric estimates
and their impacts on cancer risk estimates is presented in Section 5.2.1.4.2. Additional
uncertainties in the PBPK model were discussed in Section 3.5. Furthermore, this assessment
examined a variety of dose-metrics for the different tumor types using PBPK models for rats,
mice, and humans, so the impact of dose-metric selection can be assessed. As discussed in
Section 5.2.1.2.1, there is strong support for the primary dose-metrics selected for kidney, liver,
and, to a lesser extent, lung. For the other tumor sites, there is more uncertainty about
dose-metric selection. The cancer slope factor and unit risk estimates obtained using the
preferred dose-metrics were generally similar (within about threefold) to those derived using
default dosimetry assumptions (e.g., equal risks result from equal cumulative equivalent
exposures or doses), with the exception of the bioactivated DCVC dose-metric for rat kidney
tumors and the metric for the amount of TCE oxidized in the respiratory tract for mouse lung
tumors occurring from oral exposure (see Tables 5-39 and 5-40). The higher risk estimates for
kidney tumors based on the bioactivated DCVC dose-metric are to be expected because
pharmacokinetic data indicate, and the PBPK model predicts, substantially more GSH
conjugation (as a fraction of intake), and hence subsequent bioactivation, in humans relative to
rats. Nonetheless, there is substantial uncertainty in the quantitative extrapolation of GSH
conjugation from rodents to humans due to limitations in the available data. The lower risk
estimates for lung tumors from oral TCE exposure based on the metric for the amount of TCE
This document is a draft for review purposes only and does not constitute Agency policy.
149 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
oxidized in the respiratory tract are because there is a greater first-pass effect in human liver
relative to mouse liver following oral exposure and because the gavage dosing used in rodent
studies leads to a large bolus dose that potentially overwhelms liver metabolism to a greater
extent than a more graded oral exposure. Both of these effects result in relatively more TCE
being available for metabolism in the lung for mice than for humans. In addition, mice have
greater respiratory metabolism relative to humans. However, because oxidative metabolites
produced in the liver may contribute to respiratory tract effects, using respiratory tract
metabolism alone as a dose-metric may underestimate lung tumor risk. The slope factor or unit
risk estimates obtained using the alternative dose-metrics were also generally similar to those
derived using default dosimetry assumptions, with the exception of the metric for the amount of
TCE conjugated with GSH for rat kidney tumors, again because humans have greater GSH
conjugation, and the AUC of TCE in blood for all the tumor types resulting from oral exposure,
again because of first-pass effects.
With respect to uncertainties in the dose-response modeling, the two-step approach of
modeling only in the observable range, as put forth in EPA's cancer assessment guidelines (U.S.
EPA, 2005c), is designed in part to minimize model dependence. The ratios of the BMDs to the
BMDLs give some indication of the statistical uncertainties in the dose-response modeling.
These ratios did not exceed a value of 2.5 for all the primary analyses used in this assessment.
Thus, overall, modeling uncertainties in the observable range are considered to be minimal.
Some additional uncertainty is conveyed by uncertainties in the survival adjustments made to
some of the bioassay data; however, their impact is also believed to be minimal relative to the
uncertainties already discussed (i.e., interspecies and low-dose extrapolations).
Regarding the cancer risks to potentially sensitive human populations or life stages,
pharmacokinetic data on 42 individuals were used in the Bayesian population analysis of the
PBPK model discussed in Section 3.5. The impacts of these data on the predicted population
mean are incorporated in the quantitative uncertainty analyses presented in Section 5.2.1.4.2.
These data do not, however, reflect the full range of metabolic variability in the human
population (they are all from healthy, mostly male, human volunteers) and do not address
specific potentially sensitive subgroups (see Section 4.10). Moreover, there is inadequate
information about disease status, coexposures, and other factors that make humans vary in their
responses to TCE. It will be a challenge for future research to quantify the differential risk
indicated by different risk factors or exposure scenarios.
This document is a draft for review purposes only and does not constitute Agency policy.
150 DRAFT—DO NOT CITE OR QUOTE
-------
5.2.1.1.8. Quantitative uncertainty analysis of physiologically based pharmacokinetic
(PBPK) model-based dose-metrics
This document is a draft for review purposes only and does not constitute Agency policy.
151 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
Bioassay
experimental
paradigm
Rodent
model
parameters
\
fixedl
^distribution
PBPK /
model /
{distribution (combined
"uncertainty and variability)
Rodent
internal
dose
Bioassay
responses
V
fixedl
^distribution
Dose-
'esponse*
\Model/
[distribution
0.001 ppm
in air or
0.001 mg/kg-d
continuous
exposure
Human
model
parameters
V
fixed'.
PBPK
model
[distribution (separate
luncertainty and variability)
Human
internal
dose
"Uncertainty
idistribution of
[population mean
Cancer slope
factor =
BMR/BMD
(per internal
dose unit))
Population
mean human
internal dose
per ppm or
per mg/kg-d
Human cancer
slope factor
(per ppm or
per mg/kg-d)
distribution
The Bayesian analysis of the PBPK model for TCE generates distributions of uncertainty
and variability in the internal dose-metrics than can be readily fed into dose-response analysis.
As shown in Figure 5-6, the overall approach taken for the uncertainty analysis is similar to that
used for the point estimates except that distributions are carried through the analysis rather than
median or expected values. In particular, the PBPK model-based rodent internal doses are
carried through to a distribution of BMDs (which also includes sampling variance from the
number of responding and at risk animals in the bioassay). This distribution of BMDs generates
a distribution of cancer slope factors based on internal dose, which then is combined with the
This document is a draft for review purposes only and does not constitute Agency policy.
152 DRAFT—DO NOT CITE OR QUOTE
-------
1 Figure 5-6. Flow-chart for uncertainty analysis of dose-response analyses of
2 rodent bioassays using PBPK model-based dose-metrics. Square nodes
3 indicate point values, circular nodes indicate distributions, and the inverted
4 triangles indicate a (deterministic) functional relationship.
5
This document is a draft for review purposes only and does not constitute Agency policy.
153 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
(uncertainty) distribution of the human population mean conversion to applied dose or exposure.
The resulting distribution for the human population mean risk per unit dose or exposure accounts
for uncertainty in the PBPK model parameters (rodent and human) and the binomial sampling
error in the bioassays. These distributions can then be compared with the point estimates, based
on median rodent dose-metrics and mean human population dose-metrics, reported in
Tables 5-36 and 5-37. Details of the implementation of this uncertainty analysis, which used the
WinBugs software in conjugation with the R statistical package, are reported in Appendix G.
Overall, as shown in Tables 5-41 and 5-42, the 95% confidence upper bound of the
distributions for the linearly extrapolated risk per unit dose or exposure ranged from one- to
eightfold higher than the point slope factors and unit risks derived using the BMDLs reported in
Tables 5-36 and 5-37. The largest differences, up to 4-fold, for rat kidney tumors and 8-fold for
mouse lung tumors, primarily reflect the substantial uncertainty in the internal dose-metrics for
rat kidney DCVC and GSH conjugation and for mouse lung oxidation (see Section 3.5).
Additionally, despite the differences in the degree of uncertainty due to the PBPK model across
endpoints and dose-metrics, the only case where the choice of the most sensitive bioassay for
each sex/speciescombination would change based on the 95% confidence upper bounds reported
in Tables 5-41 and 5-42 would be for female mouse inhalation bioassays. Even in this case, the
difference between slope factor or unit risk estimate for the most sensitive and next most
sensitive study/endpoint was only twofold.
5.2.2. Dose-Response Analyses: Human Epidemiologic Data
Of the epidemiological studies of TCE and cancer, only two had sufficient
exposure-response information for potential dose-response analysis. The two studies, Charbotel
et al. (2006) and Moore et al. (2010), were both case-control studies of TCE and kidney cancer,
and both had quantitative cumulative exposure estimates for the individual subjects. In the study
by Moore et al. (2010), however, the cumulative exposure estimates were assessed by experts
based on categorical metrics for frequency and intensity of exposure and not continuous
measures. Moore et al. (2010) also used a categorical confidence-of-exposure metric to classify
different jobs because of the potential for exposure misclassification from this approach. While
the detailed approach used by Moore et al. should be fairly reliable for general rankings, the
resulting estimates are not expected to be as quantitatively accurate as those in the Charbotel
et al. (2006) study, which relied on a task-exposure matrix based on decades of measurements
from the Arve Valley workshops (Fevotte et al., 2006; see also Section 4.4 for more discussion
of the exposure assessments). Thus, the Charbotel et al. (2006) study was selected as the sole
basis for the derivation of an inhalation unit risk estimate for kidney cancer (see Section 5.2.2.1).
This document is a draft for review purposes only and does not constitute Agency policy.
154 DRAFT—DO NOT CITE OR QUOTE
-------
Table 5-41. Summary of PBPK model-based uncertainty analysis of unit risk estimates for each
sex/species/bioassay/tumor type (inhalation)
Study
Tumor Type
BMR
Dose-metric
Unit risk estimates (ppm) ')
From
Summary statistics of unit risk distribution
Table
5-36
Mean
5% lower
bound
Median
95% upper
bound
Female mouse
Fukuda
Lung AD + CARCa
0.1
AMetLngBW34
2.6 x 10"3
5.65 x 1(T3
2.34 x 1(T4
1.49 x 1(T3
2.18 x 1(T2
T otOxMetabB W3 4
3.2 x 1(T3
1.88 x 1(T3
3.27 x 1(T4
1.52 x 1(T3
4.59 x 1(T3
AUCCBld
1.8 x 1(T3
1.01 x 1(T3
1.54 x 1(T4
8.36 x 1(T4
2.44 x 1(T3
Henschler
Lymphoma13
0.1
TotMetabBW34
1.0 x 10"2
4.38 x 1(T3
6.06 x 1(T4
3.49 x 1(T3
1.11 x 1(T2
Maltoni
Lung AD + CARCa
0.1
AMetLngBW34
1.8 x 10~3
3.88 x 1(T3
1.48 x 1(T4
1.04 x 1(T3
1.52 x 1(T2
T otOxMetabB W3 4
1.9 x 1(T3
1.10 x 1(T3
3.73 x 1(T4
9.52 x 1(T4
2.32 x 1(T3
AUCCBld
1.0 x 1(T3
5.25 x 1(T4
1.63 x 1(T4
4.64 x 1(T4
1.10 x 1(T3
Liver
0.05
AMetLivlBW34
1.2 x 10"3
6.27 x 1(T4
2.18 x 1(T4
5.39 x 1(T4
1.32 x 1(T3
T otOxMetabB W3 4
1.1 X 10~3
5.98 x 1(T4
1.81 x 1(T4
5.07 x 1(T4
1.31 x 1CT3
Male mouse
Maltoni
Liver
0.1
AMetLivlBW34
2.6 x 10"3
1.35 x 1(T3
4.28 x 1(T4
1.16 x 1(T3
2.93 x icr3
T otOxMetabB W3 4
2.0 x 1(T3
1.23 x 1(T3
4.24 x 1(T4
1.06 x 1(T3
2.60 x icr3
Male rat
Maltoni
Leukemiab
0.05
TotMetabBW34
1.8 x 10"3
9.38 x 1(T4
1.26 x 1(T4
7.86 x 1(T4
2.25 x 1CT3
Kidney AD + CARC
0.01
ABioactDCVCBW34
8.3 x 10"2
9.07 x 1(T2
3.66 x 1(T3
3.64 x 1(T2
3.21 x icr1
AMetGSHBW34
5.1 x 1(T2
3.90 x 1(T2
2.71 x 1(T3
2.20 x 1(T2
1.30 x 1CT1
TotMetabBW34
7.3 x 1(T4
3.94 x 1(T4
8.74 x 1(T5
3.42 x 1(T4
8.74 x icr4
Leydig cellb
0.1
TotMetabBW34
5.5 x 10"3
4.34 x 1(T3
1.99 x 1(T3
3.98 x 1(T3
7.87 x icr3
"WinBUGS dose-response analyses did not adequately converge for the AMetLngBW34 dose-metric using the 3r -order multistage model (used for results in
Table 5-36), but did converge when the 2nd-order model was used. Summary statistics reflect results of 2nd-order model calculations.
bPoor dose-response fits in point estimates for AUCCBld, so not included in uncertainty analysis.
AD = adenoma, CARC = carcinoma.
-------
Table 5-42. Summary of PBPK model-based uncertainty analysis of slope factor estimates for each
sex/species/bioassay/tumor type (oral)
5^
cs
si
3
a
>!
£
o5
st
G\
O
&
o
>!
>!
O
a
a
§•
>!
a
o
rs
o
a
>!
o
o
2;
H
o;
a
"T3
H S.
«£
V
o
c
o
H
W
Study
Tumor type
BMR
Dose-metric
Slope factor estimates (mg/kg/day) ')
From
Summary statistics of slope factor distribution
Table 5-37
or 5-38
Mean
5% lower
bound
Median
95% upper
bound
Female mouse
NCI
Liver CARC
0.1
AMetLivlBW34
7.1 x 10"3
3.26 x 10~3
9.35 x 10~4
2.44 x 10~3
8.35 x 10~3
T otOxMetabB W3 4
5.7 x 10~3
2.63 x 10~3
8.76 x 10~4
2.01 x 10~3
6.60 x 10~3
Lung AD + CARCa
0.1
AMetLngBW34
1.3 x 10~4
1.28 x 10~4
6.73 x 10~6
4.12 x 10~5
4.62 x 10~4
T otOxMetabB W3 4
4.0 x 10~3
1.84 x 10~3
5.29 x 10~4
1.39 x 10~3
4.73 x 10~3
AUCCBld
1.5 x 10~4
7.16 x 10~5
4.40 x 10~6
3.39 x 10~5
2.18 x 10~4
Leukemias + sarcomas
0.1
TotMetabBW34
4.9 x 10"3
1.60 x 10~3
1.42 x 10~4
1.13 x 10~3
4.65 x 10~3
AUCCBld
1.4 x 10~4
6.36 x 10~5
3.10 x 10~6
2.90 x 10~5
1.94 x 10~4
Male mouse
NCI
Liver CARC
0.1
AMetLivlBW34
2.9 x 10"2
1.65 x 10~2
4.70 x 10~3
1.25 x 10~2
4.25 x 10~2
T otOxMetabB W3 4
2.3 x 10~2
1.32 x 10~2
4.41 x 10~3
1.01 x 10~2
3.29 x 10~2
Female rat
NTP (1988)
Leukemia
0.05
TotMetabBW34
2.3 x 10"3
1.89 x 10~3
5.09 x 10~4
1.43 x 10~3
4.69 x 10~3
AUCCBld
1.6 x 10~5
1.56 x 10~5
3.39 x 10~6
1.07 x 10~5
3.98 x 10~5
Male rat
NTP (1990)
Kidney AD + CARCb
0.1
ABioactDCVCBW34
1.2 x 10"1
1.40 x 10"1
5.69 x 10~3
5.24 x 10~2
5.18 x 10"1
AMetGSHBW34
7.6 x 10~2
6.18 x 10~2
4.00 x 10~3
3.27 x 10~2
2.11 x 10"1
TotMetabBW34
3.1 x 10~3
2.49 x 10~3
7.14 x 10~4
1.96 x 10~3
5.96 x 10~3
-------
Table 5-42. Summary of PBPK model-based uncertainty analysis of slope factor estimates for each
sex/species/bioassay/tumor type (oral) (continued)
5^
cs
si
3
a
>!
&
o5
§
>!
o
a
a
HH S"1
0 °
HH ^
'/O >!
fel
f i
1 ^
§i
w CV
2; ^
1—1 s
H S.
W £
P
o
c
o
H
W
Study
Tumor type
BMR
Dose-metric
Slope factor estimates (mg/kg/day) ')
From
Summary statistics of slope factor distribution
Table 5-37
or 5-38
Mean
5% lower
bound
Median
95% upper
bound
NTP (1988)
Marshall
Testicular13
0.1
TotMetabBW34
7.1 x 10"2
6.18 x 1(T2
1.92 x 1(T2
4.89 x 1(T2
1.45 x KT1
AUCCBld
6.0 x 1(T4
5.45 x 1(T4
1.18 x 1(T4
3.70 x 1(T4
1.44 x 1(T3
August
Subcut sarcoma
0.05
TotMetabBW34
2.3 x 10"3
1.65 x 1(T3
4.58 x 1CT4
1.27 x 1(T3
4.04 x 1(T3
AUCCBld
2.0 x KT5
1.35 x 1(T5
1.53 x 1(T6
8.34 x 1(T6
3.73 x 1(T5
Osborne-Mendel
Kidney AD + CARCb
0.1
ABioactDCVCBW34
1.6 x 10"1
1.61 x KT1
5.45 x 1CT3
6.35 x 1(T2
6.02 x KT1
AMetGSHBW34
9.7 x 1(T2
7.47 x 1(T2
3.90 x 1(T3
3.85 x 1(T2
2.54 x KT1
TotMetabBW34
4.3 x KT3
2.73 x 1(T3
5.40 x 1CT4
2.10 x 1(T3
6.89 x 1(T3
aWinBUGS dose-response analyses did not adequately converge for AMetLngBW34 dose-metric using the 3r -order multistage model (used for results in
Table 5-37), but did converge when the 2nd-order model was used. Summary statistics reflect results of 2nd-order model calculations.
bUsing poly-3 adjusted incidences from Table 5-38 (software for WinBUGS-based analyses using the MSW model was not developed).
AD = adenoma, CARC = carcinoma.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Other epidemiological studies were used in Section 5.2.2.2 below to provide information for a
comparison of relative risk (RR) estimates across cancer types. These epidemiologic data were
used to derive an adjusted inhalation unit risk estimate for the combined risk of developing
kidney cancer, non-Hodgkin lymphoma (NHL), or liver cancer. The human PBPK model was
then used to perform route-to-route extrapolation to derive an oral slope factor estimate for the
combined risk of kidney cancer, NHL, or liver cancer (see Section 5.2.2.3).
5.2.2.1.1. Inhalation Unit Risk Estimate for Renal Cell Carcinoma Derived from Charbotel
et al. (2006) Data
The Charbotel et al. (2006) case-control study of 86 incident renal cell carcinoma (RCC)
cases and 316 age-and sex-matched controls, with individual cumulative exposure estimates for
TCE for each subject, provides a sufficient human data set for deriving quantitative cancer risk
estimates for RCC in humans. The study is a high-quality study that used a detailed exposure
assessment (Fevotte et al., 2006) and took numerous potential confounding factors, including
exposure to other chemicals, into account (see Section 4.4). A significant dose-response
relationship was reported for cumulative TCE exposure and RCC (Charbotel et al., 2006).
The derivation of an inhalation unit risk estimate, defined as the plausible upper bound
lifetime risk of cancer from chronic inhalation of TCE per unit of air concentration, for RCC
incidence in the U.S. population, based on results of the Charbotel et al. (2006) case-control
study, is presented in the following subsections.
5.2.2.1.2. Renal cell carcinoma (RCC) results from the Charbotel et al. (2006) study
Charbotel et al. (2006) analyzed their data using conditional logistic regression, matching
on sex and age, and reported results (odds ratios [ORs]) for cumulative TCE exposure categories,
adjusted for tobacco smoking and body mass index (Charbotel et al., 2006, Table 6). The
exposure categories were constructed as tertiles based on the cumulative exposure levels in the
exposed control subjects. The results are summarized in Table 5-43, with mean exposure levels
kindly provided by Dr. Charbotel (personal communication from Barbara Charbotel, University
of Lyon, to Cheryl Scott, U.S. EPA, 11 April 2008).
For additional details and discussion of the Charbotel et al. (2006) study, see Section 4.4
and Appendix B.
This document is a draft for review purposes only and does not constitute Agency policy.
15 8 DRAFT—DO NOT CITE OR QUOTE
-------
5.2.2.1.3. Prediction of lifetime extra risk of renal cell carcinoma (RCC) incidence from
trichloroethylene (TCE) exposure
1 The categorical results summarized in Table 5-43 were used for predicting the extra risk
2 of RCC incidence from continuous environmental exposure to TCE. Extra risk is defined as
3
This document is a draft for review purposes only and does not constitute Agency policy.
159 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Table 5-43. Results from Charbotel et al. (2006) on relationship between
TCE exposure and RCC
Cumulative exposure
category
Mean Cumulative exposure
(ppm x yrs)
Adjusted OR
(95% CI)
Nonexposed
1
Low
62.4
1.62 (0.75, 3.47)
Medium
253.2
1.15 (0.47, 2.77)
High
925.0
2.16(1.02,4.60)
CI = confidence interval.
Extra risk = (Rx - Ro)/( 1 - Ro), (Eq. 5-3)
where Rx is the lifetime risk in the exposed population and Ro is the lifetime risk in an
unexposed population (i.e., the background risk). Because kidney cancer is a rare event, the ORs
in Table 5-43 can be used as estimates of the relative risk ratio, RR = Rx/Ro (Rothman and
Greenland, 1998). A weighted linear regression model was used to model the dose-response data
in Table 5-43 to obtain a slope estimate (regression coefficient) for RR of RCC versus
cumulative exposure, under the commonly employed assumption that exposure was measured
without error. Use of a linear model in the observable range of the data is often a good general
approach for epidemiological data because such data are frequently too limited (i.e., imprecise),
as is the case here, to clearly identify an alternate model (U.S. EPA, 2005c). This linear
dose-response function was then used to calculate lifetime extra risks in an actuarial program
(life-table analysis) that accounts for age-specific rates of death and background disease, under
the common assumption that the RR is independent of age.48 In addition, it is generally assumed
that RR estimates transfer across populations, independent of background disease rates—in this
case, the RR estimates based on the Charbotel et al. (2006) study, which was conducted in
France, are assumed to apply to the U.S. population.49
48This program is an adaptation of the approach previously used by the Committee on the Biological Effects of
Ionizing Radiation (NRC. 1988). The same methodology was also used in U.S. EPA's 1,3-butadiene health risk
assessment (U.S. EPA. 2002a'). A spreadsheet illustrating the extra risk calculation for the derivation of the LEC0i
for RCC incidence is presented in Appendix H.
49 In any event, background kidney cancer rates between the U.S. and France are similar, with estimated
age-adjusted incidence rates of 14.1 per 100,000 in the U.S. (Surveillance, Epidemiology, and End Results:
http://seer.cancer.gov/statfacts/htmp/kidrp.html) and 10.4 per 100,000 in France (European Cancer Observatory:
http ://eu-cancer. iarc .fr/cancer-19-kidney. html,en).
This document is a draft for review purposes only and does not constitute Agency policy.
160 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
For the weighted linear regression, the weights used for the RR estimates were the
inverses of the variances, which were calculated from the confidence intervals. Using this
approach,50 a linear regression coefficient of 0.001205 per ppm x year
(standard error = 0.0008195 per ppm x year) was obtained from the categorical results.
For the life-table analysis, U.S. age-specific all-cause mortality rates for 2004 for both
sexes and all race groups combined (CDC, 2007) were used to specify the all-cause background
mortality rates in the actuarial program. Because the goal is to estimate the unit risk for extra
risk of cancer incidence, not mortality, and because the Charbotel et al. (2006) data are incidence
data, RCC incidence rates were used for the cause-specific background "mortality" rates in the
life-table analysis.51 Surveillance, Epidemiology, and End Results (SEER) 2001-2005
cause-specific background incidence rates for RCC were obtained from the SEER public-use
database.52 SEER collects good-quality cancer incidence data from a variety of geographical
areas in the United States. The incidence data used here are from SEER 17, a registry of
17 states, cities, or regions covering about 26% of the United States population
(http://seer.cancer.gov). The risks were computed up to age 85 years for continuous exposures to
TCE.53 Conversions between occupational TCE exposures and continuous environmental
exposures were made to account for differences in the number of days exposed per year (240 vs.
365 days) and in the amount of air inhaled per day (10 vs. 20 m3; U.S. EPA, 1994b). The
standard error for the regression coefficient from the weighted linear regression calculation
described above was used to compute the 95% upper confidence limit (UCL) for the slope
estimate, and this value was used to derive 95% UCLs for risk estimates (or 95% lower
confidence limits [LCLs] for corresponding exposure estimates), based on a normal
approximation.
Point estimates and one-sided 95% UCLs for the extra risk of RCC incidence associated
with varying levels of environmental exposure to TCE based on linear regression of the
Charbotel et al. (2006) categorical results were determined by the actuarial program; the results
50Equations for this weighted linear regression approach are presented in Rothman (1986) and summarized in Appendix H.
5 INo adjustment was made for using RCC incidence rates rather than mortality rates to represent cause-specific
mortality in the actuarial program because the RCC incidence rates are negligible in comparison to the all-cause
mortality rates. Otherwise, all-cause mortality rates for each age interval would have been adjusted to reflect people
dying of a cause other than RCC or being diagnosed with RCC.
52In accordance with the "SEER Program Coding and Staging Manual 2007"
(http://seer.cancer.gov/manuals/2007/SPCSM 2007 AppendixC p6.pdf). pages C-831 to C-833, RRC was
specified as ICD-0-3 histological types coded 8312, 8260, 8310, 8316-8320, 8510, 8959, and 8255 (mixed types).
53 Rates above age 85 years are not included because cause-specific disease rates are less stable for those ages. Note
that 85 years is not employed here as an average lifespan but, rather, as a cut-off point for the life-table analysis,
which uses actual age-specific mortality rates.
This document is a draft for review purposes only and does not constitute Agency policy.
161 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
are presented in Section 5.2.13. The models based on cumulative exposure yield extra risk
estimates that are fairly linear for exposures up to 1 ppm or so.
Consistent with EPA's Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005c),
the same data and methodology were also used to estimate the exposure level (ECS: "effective
concentration corresponding to an extra risk of x%") and the associated 95% lower confidence
limit of the effective concentration corresponding to an extra risk of 1% (LEC,: [lowest effective
concentration], x = 0.01). A 1% extra risk level is commonly used for the determination of the
POD for epidemiological data. Use of a 1% extra risk level for these data is supported by the
fact that, based on the actuarial program, the risk ratio (i.e., Rx/Ro) for an extra risk of 1% for
RCC incidence is 1.9, which is in the range of the ORs reported by Charbotel et al. (see
Table 5-43). Thus, 1% extra risk was selected for determination of the POD, and, consistent
with the Guidelines for Carcinogen Risk Assessment, the LEC value corresponding to that risk
level was used as the actual POD. For the linear model that was selected, the unit risk is
independent of the benchmark risk level used to determine the POD (at low exposures/risk
levels; see Table 5-44); however, selection of a benchmark risk level is generally useful for
comparisons across models.
As discussed in Section 4.4, there is sufficient evidence to conclude that a mutagenic
MOA is operative for TCE-induced kidney tumors, which supports the use of linear low-dose
extrapolation from the POD. The ECoi, LECoi, and inhalation unit risk estimates for RCC
incidence using the linear cumulative exposure model are presented in Table 5-45. Converting
the units, 5.49 x 10~3 per ppm corresponds to a unit risk of 1.02 x 10"6 per (J,g/m3 for RCC
incidence.
Table 5-44. Extra risk estimates for RCC incidence from various levels of
lifetime exposure to TCE, using linear cumulative exposure model
Exposure concentration (ppm)
MLE of extra risk
95% UCL on extra risk
0.001
2.603 x 10~6
5.514 x 10~6
0.01
2.603 x 10~5
5.514 x 10~5
0.1
2.602 x 10~4
5.512 x 10~4
1.0
2.598 x 10~3
5.496 x 10~3
10.0
2.562 x 10~2
5.333 x 10~2
This document is a draft for review purposes only and does not constitute Agency policy.
162 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
Table 5-45. ECoi, LECoi, and unit risk estimates for RCC incidence, using
linear cumulative exposure model
ECoi (ppm)
LECoi (ppm)
unit risk (per ppm)*
3.87
1.82
5.49 x 10~3
*Unit risk = 0.01/LEC0i.
5.2.2.1.4. Uncertainties in the renal cell carcinoma (RCC) unit risk estimate
The two major sources of uncertainty in quantitative cancer risk estimates are generally
interspecies extrapolation and high-dose to low-dose extrapolation. The unit risk estimate for
RCC incidence derived from the Charbotel et al. (2006) results is not subject to interspecies
uncertainty because it is based on human data. A major uncertainty remains in the extrapolation
from occupational exposures to lower environmental exposures. There was some evidence of a
contribution to increased RCC risk from peak exposures; however, there remained an apparent
dose-response relationship for RCC risk with increasing cumulative exposure without peaks, and
the OR for exposure with peaks compared to exposure without peaks was not significantly
elevated (Charbotel et al., 2006). Although the actual exposure-response relationship at low
exposure levels is unknown, the conclusion that a mutagenic MOA is operative for TCE-induced
kidney tumors supports the linear low-dose extrapolation that was used (U.S. EPA, 2005c).
Another notable source of uncertainty in the cancer unit risk estimate is the dose-response
model used to model the study data to estimate the POD. A weighted linear regression across the
categorical ORs was used to obtain a slope estimate; use of a linear model in the observable
range of the data is often a good general approach for human data because epidemiological data
are frequently too limited (i.e., imprecise) to clearly identify an alternate model (U.S. EPA,
2005c). The Charbotel et al. (2006) study is a relatively small case-control study, with only 86
RCC cases, 37 of which had TCE exposure; thus, the dose-response data upon which to specify a
model are indeed limited.
In accordance with EPA's Guidelines for Carcinogen Risk Assessment, the lower bound
on the ECoi is used as the POD; this acknowledges some of the uncertainty in estimating the
POD from the available dose-response data. In this case, the statistical uncertainty associated
with the ECoi is relatively small, as the ratio between the ECoi and the LECoi is about twofold.
The inhalation unit risk estimate of 5.49 x 10 per ppm presented above, which is calculated
based on a linear extrapolation from the POD (LECoi), is expected to provide an upper bound on
This document is a draft for review purposes only and does not constitute Agency policy.
163 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
the risk of cancer incidence. However, for certain applications, such as benefit-cost analyses,
estimates of "central tendency" for the risk below the POD are desired. Because a linear
dose-response model was used in the observable range of the human data and the POD was
within the low-dose linear range for extra risk as a function of exposure, linear extrapolation
below the LECoi has virtually the same slope as the 95% UCL on the actual (linear)
dose-response model in the low-dose range (i.e., below the POD). This is illustrated in
Table 5-44, where the 95% UCL on extra risk for RCC incidence predicted by the dose-response
model is about 5.51 x 10 per ppm for exposures at or below about 0.1 ppm, which is virtually
_o
equivalent to the unit risk estimate of 5.49 x 10 per ppm derived from the LECoi (see
Table 5-45). The same holds for the central tendency (weighted least squares) estimates of the
extra risk from the (linear) dose-response model (i.e., the dose-response model prediction of
_3
2.60 x 10 per ppm from Table 5-44 is virtually identical to the value of 2.58 x 10 per ppm
obtained from linear extrapolation below the ECoi, i.e., by dividing 0.01 extra risk by the ECoi of
3.87 from Table 5-45). In other words, because the dose-response model that was used to model
the data in the observable range is already low-dose linear near the POD, if one assumes that the
same linear model is valid for the low-dose range, one can use the central tendency (weighted
least squares) estimate from the model to derive a statistical "best estimate" of the slope rather
than relying on an extrapolated risk estimate (0.01/ECoi). [The extrapolated risk estimates are
not generally central tendency estimates in any statistical sense because once risk is extrapolated
below the ECoi using the formulation 0.01/ECoi, it is no longer a function of the original model
which generated the ECois and the LECois.]
An important source of uncertainty in the underlying Charbotel et al. (2006) study is the
retrospective estimation of TCE exposures in the study subjects. This case-control study was
conducted in the Arve Valley in France, a region with a high concentration of workshops
devoted to screw cutting, which involves the use of TCE and other degreasing agents. Since the
1960s, occupational physicians of the region have collected a large quantity of well-documented
measurements, including TCE air concentrations and urinary metabolite levels (Fevotte et al.,
2006). The study investigators conducted a comprehensive exposure assessment to estimate
cumulative TCE exposures for the individual study subjects, using a detailed occupational
questionnaire with a customized task-exposure matrix for the screw-cutting workers and a more
general occupational questionnaire for workers exposed to TCE in other industries (Fevotte et
al., 2006). The exposure assessment even attempted to take dermal exposure from hand-dipping
practices into account by equating it with an equivalent airborne concentration based on
biological monitoring data. Despite the appreciable effort of the investigators, considerable
uncertainty associated with any retrospective exposure assessment is inevitable, and some
exposure misclassification is unavoidable. Such exposure misclassification was most likely for
This document is a draft for review purposes only and does not constitute Agency policy.
164 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
the 19 deceased cases and their matched controls, for which proxy respondents were used, and
for exposures outside the screw-cutting industry (295 of 1,486 identified job periods involved
TCE exposure; 120 of these were not in the screw-cutting industry).
Although the exposure estimates from Moore et al. (2010) were not considered to be as
quantitatively accurate as those of Charbotel et al. (2006), as discussed at the beginning of
Section 5.2.2, it is worth noting, in the context of uncertainty in the exposure assessment, that the
exposure estimates in Moore et al. (2010) are substantially lower than those of Charbotel et al.
(2006) for comparable OR estimates. For example, for all subjects and high-confidence
assessments only, respectively, Moore et al. (2010) report OR estimates of 1.19 and 1.77 for
cumulative exposures < 1.58 ppm x years and 2.02 and 2.23 for cumulative
exposures > 1.58 ppm x years. Charbotel et al. (2006), on the other hand, report OR estimates
for all subjects of 1.62, 1.15, and 2.16 for mean cumulative exposures of 62.4, 253.2, and
925.0 ppm x years, respectively. If the exposure estimates for Charbotel et al. (2006) are
overestimated, as suggested by the exposure estimates from Moore et al. (2010), the slope of the
linear regression model, and hence the unit risk estimate, would be correspondingly
underestimated.
Another noteworthy source of uncertainty in the Charbotel et al. (2006) study is the
possible influence of potential confounding or modifying factors. This study population, with a
high prevalence of metal-working, also had relatively high prevalences of exposure to petroleum
oils, cadmium, petroleum solvents, welding fumes, and asbestos (Fevotte et al., 2006). Other
exposures assessed included other solvents (including other chlorinated solvents), lead, and
ionizing radiation. None of these exposures was found to be significantly associated with RCC
at ap = 0.05 significance level. Cutting fluids and other petroleum oils were associated with
RCC at ap = 0.1 significance level; however, further modeling suggested no association with
RCC when other significant factors were taken into account (Charbotel et al., 2006). Moreover,
a review of other studies suggested that potential confounding from cutting fluids and other
petroleum oils is of minimal concern (see Section 4.4.2.3). Nonetheless, a sensitivity analysis
was conducted using the OR estimates further adjusted for cutting fluids and other petroleum oils
from the unpublished report by Charbotel et al. (2005), and an essentially identical unit risk
estimate of 5.46 x 10 per ppm was obtained.54 In addition, the medical questionnaire included
54 The OR estimates further adjusted for cutting fluids and other petroleum oils were 1.52 (95% CI: 0.66, 3.49),
1.07 (0.39, 2.88), and 1.96 (0.71, 5.37) for the low, medium, and high cumulative exposure groups, respectively
(Charbotel et al.. 2005). For the linear regression model, these OR estimates yielded a shallower slope estimate of
0.0009475 per ppm x year but a larger SE of 0.0009709 per ppm x year. In the lifetable analysis, these latter
estimates in turn yielded a slightly higher EC01 estimate (4.92 ppm versus 3.87 ppm), because of the shallower
slope estimate, but an essentially identical LEC01, because of the larger SE.
This document is a draft for review purposes only and does not constitute Agency policy.
165 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
familial kidney disease and medical history, such as kidney stones, infection, chronic dialysis,
hypertension, and use of antihypertensive drugs, diuretics, and analgesics. Body mass index
(BMI) was also calculated, and lifestyle information such as smoking habits and coffee
consumption was collected. Univariate analyses found high levels of smoking and BMI to be
associated with increased odds of RCC, and these two variables were included in the conditional
logistic regressions. Thus, although impacts of other factors are possible, this study took great
pains to attempt to account for potential confounding or modifying factors.
Some other sources of uncertainty associated with the epidemiological data are the
dose-metric and lag period. As discussed above, there was some evidence of a contribution to
increased RCC risk from peak TCE exposures; however, there appeared to be an independent
effect of cumulative exposure without peaks. Cumulative exposure is considered a good
measure of total exposure because it integrates exposure (levels) over time. If there is a
contributing effect of peak exposures, not already taken into account in the cumulative exposure
metric, the linear slope may be overestimated to some extent. Sometimes cancer data are
modeled with the inclusion of a lag period to discount more recent exposures not likely to have
contributed to the onset of cancer. In an unpublished report (Charbotel et al., 2005), Charbotel
et al. also present the results of a conditional logistic regression with a 10-year lag period, and
these results are very similar to the untagged results reported in their published paper, suggesting
that the lag period might not be an important factor in this study.
Some additional sources of uncertainty are not so much inherent in the exposure-response
modeling or in the epidemiologic data themselves but, rather, arise in the process of obtaining
more general Agency risk estimates from the epidemiologic results. EPA cancer risk estimates
are typically derived to represent an upper bound on increased risk of cancer incidence for all
sites affected by an agent for the general population. From experimental animal studies, this is
accomplished by using tumor incidence data and summing across all the tumor sites that
demonstrate significantly increased incidences, customarily for the most sensitive sex and
species, to attempt to be protective of the general human population. However, in estimating
comparable risks from the Charbotel et al. (2006) epidemiologic data, certain limitations are
encountered. For one thing, these epidemiology data represent a geographically limited (Arve
Valley, France) and likely not very diverse population of working adults. Thus, there is
uncertainty about the applicability of the results to a more diverse general population.
Additionally, the Charbotel et al. (2006) study was a study of RCC only, and so the risk estimate
derived from it does not represent all the tumor sites that may be affected by TCE. The issue of
cancer risk at other sites is addressed in the next section (see Section 5.2.2.2).
This document is a draft for review purposes only and does not constitute Agency policy.
166 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
5.2.2.1.5. Conclusions regarding the renal cell carcinoma (RCC) unit risk estimate
An ECoi of 3.9 ppm was calculated using a life-table analysis and linear modeling of the
categorical conditional logistic regression results for RCC incidence reported in a high-quality
case-control study. Linear low-dose extrapolation from the LECoi yielded a lifetime extra RCC
incidence unit risk estimate of 5.5 x 10~3 per ppm (1.0 x 10~6 per |ig/m3) of continuous TCE
exposure. The assumption of low-dose linearity is supported by the conclusion that a mutagenic
MOA is operative for TCE-induced kidney tumors. The inhalation unit risk estimate is expected
to provide an upper bound on the risk of RCC incidence; however, this is just the risk estimate
for RCC. A risk estimate for total cancer risk to humans would need to include the risk for other
potential TCE-associated cancers.
5.2.2.1.6. Adjustment of the Inhalation Unit Risk Estimate for Multiple Sites
Human data on TCE exposure and cancer risk sufficient for dose-response modeling are
only available for RCC, yet human and rodent data suggest that TCE exposure increases the risk
of cancer at other sites as well. In particular, there is evidence from human (and rodent) studies
for increased risks of NHL and liver cancer (see Section 4.11). Therefore, the inhalation unit
risk estimate derived from human data for RCC incidence was adjusted to account for potential
increased risk of those tumor types. To make this adjustment, a factor accounting for the relative
contributions to the extra risk for cancer incidence from TCE exposure for these three tumor
types combined versus the extra risk for RCC alone was estimated, and this factor was applied to
the unit risk estimate for RCC to obtain a unit risk estimate for the three tumor types combined
(i.e., lifetime extra risk for developing any of the three types of tumor). This estimate is
considered a better estimate of total cancer risk from TCE exposure than the estimate for RCC
alone.
Although only the Charbotel et al. (2006) study was found adequate for direct estimation
of inhalation unit risks, the available epidemiologic data provide sufficient information for
estimating the relative potency of TCE across tumor sites. In particular, the relative
contributions to extra risk (for cancer incidence) were calculated from two different data sets to
derive the adjustment factor for adjusting the unit risk estimate for RCC to a unit risk estimate
for the three types of cancers (RCC, NHL, and liver) combined. The first calculation is based on
the results of the meta-analyses of human epidemiologic data for the three tumor types (see
Appendix C); the second calculation is based on the results of the Raaschou-Nielsen et al. (2003)
study, the largest single human epidemiologic study by far with RR estimates for all three tumor
types. The approach for each calculation was to use the RR estimates and estimates of the
This document is a draft for review purposes only and does not constitute Agency policy.
167 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
lifetime background risk in an unexposed population, Ro, to calculate the lifetime risk in the
exposed population, Rx, where Rx = RR x Ro, for each tumor type. Then, the extra risk from
TCE exposure for each tumor type could be calculated using the equation in Section 5.2.2.1.2.
Finally, the extra risks were summed across the three tumor types and the ratio of the sum of the
extra risks to the extra risk for RCC was derived. For the first calculation, the summary relative
risk estimates (RRm's) from the meta-analyses for NHL, kidney cancer, and liver (and biliary)
cancer were used as the RR estimates. For the second calculation, the SIR estimates from the
Raaschou-Nielsen et al. (2003) study were used. For both calculations, Ro for RCC was taken
from the life-table analysis described in Section 5.2.2.1.2 and presented in Appendix H, which
estimated a lifetime risk for RCC incidence up to age 85 years. For Ro values for the other
two sites, SEER statistics for the lifetime risk of developing cancer were used
(http://seer.cancer.gov/statfacts/html/nhl.html and
http:// seer, cancer, gov/ statfacts/html/livibd.htmQ.
In both cases, an underlying assumption in deriving the relative potencies is that the
relative values of the age-specific background incidence risks for the person-years from the
epidemiologic studies for each tumor type approximate the relative values of the lifetime
background incidence risks for those tumor types. In other words, at least on a proportional
basis, the lifetime background incidence risks (for the United States population) for each site
approximate the age-specific background incidence risks for the study populations. A further
assumption is that the lifetime risk of RCC up to 85 years is an adequate approximation to the
full lifetime risk, which is what was used for the other two tumor types. The first calculation,
based on the results of the meta-analyses for the three tumor types, has the advantage of being
based on a large data set, incorporating data from many different studies. However, this
calculation relies on a number of additional assumptions. First, it is assumed that the RRm's
from the meta-analyses, which are based on different groups of studies, reflect similar overall
TCE exposures, i.e., that the overall TCE exposures are similar across the different groups of
studies that went into the different meta-analyses for the three tumor types. Second, it is
assumed that the RRm's, which incorporate RR estimates for both mortality and incidence,
represent good estimates for cancer incidence risk from TCE exposure. In addition, it is assumed
that the RRm for kidney cancer, for which RCC estimates from individual studies were used
when available, is a good estimate for the overall RR for RCC and that the RRm estimate for
NHL, for which different studies used different classification schemes, is a good estimate for the
overall RR for NHL. The second calculation, based on the results of the Raaschou-Nielsen et al.
(2003) study, the largest single study with RR estimates for all three tumor types, has the
advantage of having RR estimates that are directly comparable. In addition, the
This document is a draft for review purposes only and does not constitute Agency policy.
168 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Raaschou-Nielsen et al. study provided data for the precise tumor types of interest for the
calculation, i.e., RCC, NHL, and liver (and biliary) cancer.
The input data and results of the calculations are presented in Table 5-46. The value for
the ratio of the sum of the extra risks to the extra risk for RCC alone was 3.28 in calculation #1
and 4.36 in calculation #2, which together suggest that 4 is a reasonable factor to use to adjust
the inhalation unit risk estimate based on RCC for multiple sites to obtain a total cancer unit risk
estimate.55 Using this factor to adjust the unit risk estimate based on RCCs entails the further
fundamental assumption that the dose-response relationships for the other two tumor types (NHL
and liver cancer) are similarly linear, i.e., that the relative potencies are roughly maintained at
lower exposure levels. This assumption is consistent with EPA's Guidelines for Carcinogen
Risk Assessment (U.S. EPA, 2005c), which recommends low-dose linear extrapolation in the
absence of sufficient evidence to support a nonlinear MOA.
Applying the factor of 4 to the
lifetime extra RCC incidence unit
risk estimate of 5.49 x 10 per ppm
(1.0 x 10~6 per |ig/m3) of continuous
TCE exposure yields a cancer unit
_2
risk estimate of 2.2 x 10 per ppm
(4.1 x io~6 per |ig/m3). Table 5-46
also presents calculations for just
kidney and NHL extra risks
combined, because the strongest
human evidence is for those two
tumor types. For those two tumor
types, the calculations support a
factor of 3.56 Applying this factor to
the RCC unit risk estimate yields an
_2
estimate of 1.6 x 10 per ppm,
which results in the same estimate as
for the three tumor types combined
when finally rounded to one
55 Both the geometric and arithmetic means of the two values for the ratio are 3.8, which rounds to 4, in keeping
with the imprecise nature of the adjustment factor. The factor of 4 is within 25% of either calculated ratio.
56 The geometric and mean of the two values for the ratio, 2.62 and 3.29, is 2.96, and the arithmetic mean is 2.94,
which both round to 3, in keeping with the imprecise nature of the adjustment factor. The factor of 3 is within 15%
of either calculated ratio.
This document is a draft for review purposes only and does not constitute Agency policy.
169 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
_2
significant figure, i.e., 2 x 10 per
ppm (or 3 x 10~6 per |ig/m3, which is
still similar to the three-tumor-type
estimate in those units).
In addition to the uncertainties in the underlying RCC estimate, there are uncertainties
related to the assumptions inherent in these calculations for adjusting to multiple sites, as
detailed above. Nonetheless, the fact that the calculations based on two different data sets
yielded comparable values for the adjustment factor (both within 25% of the selected factor of 4)
provides more robust support for the use of the factor of 4. Additional uncertainties pertain to
the weight of evidence supporting the association of TCE exposure with increased risk of cancer
for the three cancer types. As discussed in Section 4.11.2, it was found that the weight of
evidence for kidney cancer was sufficient to classify TCE as "carcinogenic to humans." It was
also concluded that there was strong evidence that TCE causes NHL as well, although the
evidence for liver cancer was more limited. In addition, the rodent studies demonstrate clear
This document is a draft for review purposes only and does not constitute Agency policy.
170 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 5-46. Relative contributions to extra risk for cancer incidence from
2 TCE exposure for multiple tumor types
3
RR
Ro
Rx
Extra risk
Ratio to
kidney value
Calculation #1: using RR estimates from the meta-analyses
Kidney (RCC)
1.27
0.0107
0.01359
0.002920
1
NHL
1.23
0.0202
0.02485
0.004742
1.62
Liver (and biliary)
cancer
1.29
0.0066
0.008514
0.001927
0.66
sum
0.009589
3.28
Kidney + NHL only
sum
0.007662
2.62
Calculation #2: using RR estimates from Rasschou-Nielsen et al. (2003)
Kidney (RCC)
1.20
0.0107
0.01284
0.002163
1
NHL
1.24
0.0202
0.02505
0.004948
2.29
Liver (and biliary)
cancer
1.35
0.0066
0.008910
0.002325
1.07
sum
0.009436
4.36
Kidney + NHL only
sum
0.007111
3.29
4
5
6 evidence of multisite carcinogenicity, with tumor types including those for which associations
7 with TCE exposure are observed in human studies, i.e., liver and kidney cancers and NHLs.
8 Overall, the evidence was found to be sufficiently persuasive to support the use of the adjustment
9 factor of 4 based on these three cancer types, resulting in a cancer inhalation unit risk estimate of
10 2.2 x 1CT2 per ppm (4.1 x icf6 per |ig/m3). Alternatively, if one were to use the factor based
11 only on the two cancer types with the strongest human evidence, the cancer inhalation unit risk
12 estimate would be only slightly reduced (25%).
13
5.2.2.1.7. Route-to-Route Extrapolation Using Physiologically Based Pharmacokinetic
(PBPK) Model
14 Route-to-route extrapolation of the inhalation unit risk estimate was performed using the
15 PBPK model described in Section 3.5. The (partial) unit risk estimates for NHL and liver cancer
16 were derived as for the total cancer inhalation unit risk estimate in Section 5.2.2.2 above, except
This document is a draft for review purposes only and does not constitute Agency policy.
171 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
that the ratios of extra risk for the individual tumor types relative to kidney cancer were used as
adjustment factors rather than the ratio of the sum. As presented in Table 5-46, for NHL, the
ratios from the two different calculations were 1.62 and 2.29, so a factor of 2 was used; for liver
cancer, the ratios were 0.66 and 1.07, so a factor of 1 was used. (With the ratio of one for kidney
cancer itself, the combined adjustment factor is 4, reproducing the factor of 4 used to estimate
the total cancer unit risk from the multiple sites in Section 5.2.2.2.)
Because different internal dose-metrics are preferred for each target tissue site, a separate
route-to-route extrapolation was performed for each site-specific unit risk estimate calculated in
Sections 5.2.2.1 and 5.2.2.2. As shown in Figure 5-7, the approach taken to apply the human
PBPK model in the low-dose range where external and internal doses are linearly related to
derive a conversion that is the ratio of internal dose per mg/kg/day to internal dose per ppm. The
expected value of the population mean for this conversion factor (in ppm per mg/kg/day) was
used to extrapolate each inhalation unit risk in units of risk per ppm to an oral slope factor in
units of risk per mg/kg/day. Note that this conversion is the mean of the ratio of internal dose
predictions, and is not the same as taking the ratio of the mean of internal dose predictions in
Table 5-35.57
Table 5-47 shows the results of this
route-to-route extrapolation for the
"primary" and "alternative"
dose-metrics. For reference,
route-to-route extrapolation based on
total intake (i.e., ventilation rate x air
concentration = oral dose x BW)
using the parameters in the PBPK
model would yield an expected
population average conversion of
0.95 ppm per mg/kg/day. For
TotMetabBW34,
TotOxMetabBW34, and
AMetLivlBW34, the conversion is
2.0-2.8 ppm per mg/kg/day, greater
than that based on intake. This is
57For route-to-route extrapolation based on dose-response analysis performed on internal dose, as is the case for
rodent bioassays, it would be appropriate to use the values in Table 5-35 to first "unconvert" the unit risk based on
one route, and then recovert to a unit risk based on the other route.
This document is a draft for review purposes only and does not constitute Agency policy.
172 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
because of the greater metabolic first
pass in the liver, which leads to a
higher percentage of intake being
metabolized via oral exposure
relative to inhalation exposure for
the same intake. Conversely, for the
AUC in blood, the conversion is 0.14
ppm per mg/kg/day, less than that
based on intake—the greater first
pass in the liver means lower blood
levels of parent compound via oral
exposure relative to inhalation for
the same intake. The conversion for
the primary dose-metric for the
kidney, ABioactDCVCBW34, is 1.7
ppm per mg/kg/day, less than that for
total, oxidative, or liver oxidative
metabolism. This is because the
majority of metabolism in first pass
through the liver is via oxidation,
whereas with inhalation exposure,
more parent compound reaches the
kidney, in which metabolism is via
GSH conjugation.
When one sums the oral slope factor estimates based on the primary (preferred)
dose-metrics for the three individual tumor types shown in Table 5-47, the resulting total cancer
2
oral slope factor estimate is 4.64 x 10 per mg/kg/day. In the case of the oral
route-extrapolated results, the ratio of the risk estimate for the three tumor types combined to the
risk estimate for kidney cancer alone is 5.0. This value differs from the factor of 4 used for the
This document is a draft for review purposes only and does not constitute Agency policy.
173 DRAFT—DO NOT CITE OR QUOTE
-------
0.001 ppm
in air or
0.001 mg/kg-d
continuous
exposure
Site-specific
human
cancer
unit risk
per ppm
istribution
Human
model
parameters
PBPK
model
[internal dos
per mg/kg/d]/
[internal dose,
per ppm]
.distribution (separate
ncertainty and variability)
fixed\
population
^mean
^rriean
Expected
site-specific
human
cancer unit risk
per mg/kg-d
This document is a draft for review purposes only and does not constitute Agency policy.
174 DRAFT—DO NOT CITE OR QUOTE
-------
1 Figure 5-7. Flow-chart for route-to-route extrapolation of human
2 site-specific cancer inhalation unit risks to oral slope factors. Square nodes
3 indicate point values, circle nodes indicate distributions, and the inverted
4 triangle indicates a (deterministic) functional relationship.
5
This document is a draft for review purposes only and does not constitute Agency policy.
175 DRAFT—DO NOT CITE OR QUOTE
-------
1 Table 5-47. Route-to-route extrapolation of site-specific inhalation unit risks
2 to oral slope factors
3
Kidney
NHL
Liver
Inhalation unit risk
(risk per ppm)
5.49 x 10~3
1.10 x 10~2
5.49 x 10~3
Primary dose-metric
ABioactDCVCBW34a
TotMetabBW34
AMetLivlBW34
ppm per mg/kg/dayb
1.70
1.97
2.82
Oral slope factor
(risk per mg/kg/day)
9.33 x 10~3
2.16 x 10~2
1.55 x 10~2
Alternative dose-metric
TotMetabBW34
AUCCBld
T otOxMetabB W 3 4
ppm per mg/kg/dayb
1.97
0.137
2.04
Oral slope factor
(risk per mg/kg/day)
1.08 x 10~2
1.50 x 10~3
1.12 x 10~2
4
5 "The AMetGSHBW34 dose-metric gives the same route-to-route conversion because there is no route dependence in
6 the pathway between GSH conjugation and DCVC bioactivation.
7 bAverage of expected population mean of males and females. Male and female estimates differed by <1% for
8 ABioactDCVCBW34; TotMetabBW34, AMetLivlBW34, and TotOxMetabBW34, and <15% for AUCCBld.
9 Uncertainty on the population mean route-to-route conversion, expressed as the ratio between the 97.5% quantile
10 the 2.5% quantile, is about 2.6-fold for ABioactDCVCBW34, 1.5-fold for TotMetabBW34, AMetLivlBW34, and
11 TotOxMetabBW34, and about 3.4-fold for AUCCBld.
12
13
14 total cancer inhalation unit risk estimate because of the different dose-metrics used for the
15 different tumor types when the route-to-route extrapolation is performed. If only the kidney
16 cancer and NHL results, for which the evidence is strongest, were combined, the resulting total
_2
17 cancer oral slope factor estimate would be 3.09 x 10 per mg/kg/day, and the ratio of this risk
18 estimate to that for kidney cancer alone would be 3.3.
19 If one were to use some of the risk estimates based on alternative dose-metrics in
20 Table 5-40, the total cancer risk estimate would vary depending on for which tumor type(s) an
21 alternative metric was used. The most extreme difference would occur when the alternative
22 metric is used for NHL and liver tumors; in that case, the resulting total cancer oral slope factor
_2
23 estimate would be 2.20 x 10 per mg/kg/day, and the ratio of this risk estimate to that for kidney
24 cancer alone (based on the primary dose-metric of ABioactDCVCBW34) would be 2.4.
25 The uncertainties in these conversions are relatively modest. As discussed in the note to
26 Table 5-47, the 95% confidence range for the route-to-route conversions at its greatest spans
27 3.4-fold. The greatest uncertainty is in the selection of the dose-metric for NHL, since the use of
28 the alternative dose-metric of AUCCBld yields a converted oral slope factor that is 14-fold lower
This document is a draft for review purposes only and does not constitute Agency policy.
176 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
than that using the primary dose-metric of TotMetabBW34. However, for the other two tumor
sites, the range of conversions is tighter, and lies within threefold of the conversion based solely
on intake.
5.2.3. Summary of Unit Risk Estimates
5.2.3.1.1. Inhalation Unit Risk Estimate
The inhalation unit risk for TCE is defined as a plausible upper bound lifetime extra risk
of cancer from chronic inhalation of TCE per unit of air concentration. The preferred estimate of
the inhalation unit risk for TCE is 2.20 x 10~2 per ppm (2 x 10~2 per ppm [4 x 10~6 per jig/m3]
rounded to one significant figure), based on human kidney cancer risks reported by Charbotel
et al. (2006) and adjusted for potential risk for tumors at multiple sites. This estimate is based on
good-quality human data, thus avoiding the uncertainties inherent in interspecies extrapolation.
This value is supported by inhalation unit risk estimates from multiple rodent bioassays,
the most sensitive of which range from 1 x 10~2 to 2 x 10-1 per ppm [2 x 10~6 to
£ -2
3 x 10 per jig/m ]. From the inhalation bioassays selected for analysis in Section 5.2.1.1, and
using the preferred PBPK model-based dose-metrics, the inhalation unit risk estimate for the
—2 —5 3
most sensitive sex/species is 8 x 10 per ppm [2 x 10 per |ig/m ], based on kidney adenomas
and carcinomas reported by Maltoni et al. (1986) for male Sprague-Dawley rats. Leukemias and
Leydig cell tumors were also increased in these rats, and, although a combined analysis for these
tumor types that incorporated the different site-specific preferred dose-metrics was not
_2
performed, the result of such an analysis is expected to be similar, about 9x10 per ppm
_C T
[2x10 per |ig/m ]. The next most sensitive sex/species from the inhalation bioassays is the
female mouse, for which lymphomas were reported by Henschler et al. (1980); these data yield a
unit risk estimate of 1.0 x 10 2 per ppm [2 x 10 6 per |ig/m3]. In addition, the 90% confidence
intervals reported in Table 5-41 for male rat kidney tumors from Maltoni et al. (1986) and female
mouse lymphomas from Henschler et al. (1980), derived from the quantitative analysis of PBPK
_2
model uncertainty, both included the estimate based on human data of 2 x 10 per ppm.
Furthermore, PBPK model-based route-to-route extrapolation of the results for the most sensitive
sex/species from the oral bioassays, kidney tumors in male Osborne-Mendel rats and testicular
tumors in Marshall rats (NTP, 1988), leads to inhalation unit risk estimates of 2 x 10 1 per ppm
[3 x 10 5 per |ig/m3] and 4 x 10 2 per ppm [8 x 10 6 per |ig/m3], respectively, with the preferred
estimate based on human data falling within the route-to-route extrapolation of the 90%
This document is a draft for review purposes only and does not constitute Agency policy.
177 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
confidence intervals reported in Table 5-42.58 Finally, for all these estimates, the ratios of
BMDs to the BMDLs did not exceed a value of three, indicating that the uncertainties in the
dose-response modeling for determining the POD in the observable range are small.
Although there are uncertainties in these various estimates, as discussed in
Sections 5.2.1.4, 5.2.2.1.3, and 5.2.2.2, confidence in the proposed inhalation unit risk estimate
of 2 x 10 2 per ppm [4 x 10 6 per |ig/m3], based on human kidney cancer risks reported by
Charbotel et al. (2006) and adjusted for potential risk for tumors at multiple sites (as discussed in
Section 5.2.2.2), is further increased by the similarity of this estimate to estimates based on
multiple rodent data sets.
5.2.3.1.2. Oral Slope Factor Estimate
The oral slope factor for TCE is defined as a plausible upper bound lifetime extra risk of
cancer from chronic ingestion of TCE per mg/kg/day oral dose. The preferred estimate of the
_2 2
oral slope factor is 4.64 x 10 per mg/kg/day (5 x 10 per mg/kg/day rounded to
one significant figure), resulting from PBPK model-based route-to-route extrapolation of the
inhalation unit risk estimate based on the human kidney cancer risks reported in Charbotel et al.
(2006) and adjusted for potential risk for tumors at multiple sites. This estimate is based on
good-quality human data, thus avoiding uncertainties inherent in interspecies extrapolation. In
addition, uncertainty in the PBPK model-based route-to-route extrapolation is relatively low
(Chiu. 2006; Chiu and White, 2006). In this particular case, extrapolation using different
dose-metrics yielded expected population mean risks within about a twofold range, and, for any
particular dose-metric, the 95% confidence interval for the extrapolated population mean risks
for each site spanned a range of no more than about threefold.
This value is supported by oral slope factor estimates from multiple rodent bioassays, the
2 i
most sensitive of which range from 3x10 to 3 x 10 per mg/kg/day. From the oral bioassays
selected for analysis in Section 5.2.1.1, and using the preferred PBPK model-based dose-metrics,
58For oral-to-inhalation extrapolation of NTP (19881 male rat kidney tumors, the unit risk estimate of 2.5 x 10-1 per
mg/kg/day using the ABioactDCVCBW34 dose metric, from Table 5-37, is divided by the average male and female
internal doses at 0.001 mg/kg/day, (0.00504/0.001), and then multiplied by the average male and female internal
doses at 0.001 ppm, (0.00324/0.001), both from Table 5-35, to yield a unit risk of 1.6 x 10-1 [3.0 x 10-5 per
|ig/m31. For oral-to-inhalation extrapolation of NTP (1988) male rat testicular tumors, the unit risk estimate of
7.1 x 10-2 per mg/kg/day using the TotMetabBW34 dose metric, from Table 5-37, is divided by the male internal
dose at 0.001 mg/kg/day, (0.0192/0.001), and then multiplied by the male internal doses at 0.001 ppm,
(0.0118/0.001), bothfrom Table 5-35, to yield aunit risk of 4.4 x 10-2 [8.1 x 10-6 per |ig/m31,
This document is a draft for review purposes only and does not constitute Agency policy.
178 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
the oral slope factor estimate for the most sensitive sex/species is 3 x 10 1 per mg/kg/day, based
on kidney tumors in male Osborne-Mendel rats (NTP, 1988). The oral slope factor estimate for
_2
testicular tumors in male Marshall rats (NTP, 1988) is somewhat lower at 7 x 10 per
mg/kg/day. The next most sensitive sex/species result from the oral studies is for male mouse
_2
liver tumors (NCI, 1976), with an oral slope factor estimate of 3 x 10 per mg/kg/day. In
addition, the 90% confidence intervals reported in Table 5-42 for male Osborne-Mendel rat
kidney tumors (NTP, 1988), male F344 rat kidney tumors (NTP, 1990), and male Marshall rat
testicular tumors (NTP, 1988), derived from the quantitative analysis of PBPK model
_2
uncertainty, all included the estimate based on human data of 5 x 10 per mg/kg/day, while the
upper 95% confidence bound for male mouse liver tumors from NCI (1976) was slightly below
_2
this value at 4 x 10 per mg/kg/day. Furthermore, PBPK model-based route-to-route
extrapolation of the most sensitive endpoint from the inhalation bioassays, male rat kidney
tumors from Maltoni et al. (1986), leads to an oral slope factor estimate of 1 x 10 1 per
mg/kg/day, with the preferred estimate based on human data falling within the route-to-route
extrapolation of the 90% confidence interval reported in Table 5-41.59 Finally, for all these
estimates, the ratios of BMDs to the BMDLs did not exceed a value of three, indicating that the
uncertainties in the dose-response modeling for determining the POD in the observable range are
small.
Although there are uncertainties in these various estimates, as discussed in
Sections 5.2.1.4, 5.2.2.1.3, 5.2.2.2, and 5.2.2.3, confidence in the proposed oral slope factor
_2
estimate of 5 x 10 per mg/kg/day, resulting from PBPK model-based route-to-route
extrapolation of the inhalation unit risk estimate based on the human kidney cancer risks reported
in Charbotel et al. (2006) and adjusted for potential risk for tumors at multiple sites (as discussed
in Section 5.2.2.2), is further increased by the similarity of this estimate to estimates based on
multiple rodent data sets.
5.2.3.1.3. Application of Age-Dependent Adjustment Factors
When there is sufficient weight of evidence to conclude that a carcinogen operates
through a mutagenic MO A, and in the absence of chemical-specific data on age-specific
59For the Maltoni et al. (19861 male rat kidney tumors, the unit risk estimate of 8.3 x 10-2 per ppm using the
ABioactDCVCBW34 dose metric, from Table 5-36, is divided by the average male and female internal doses at
0.001 ppm, (0.00324/0.001) and then multiplied by the average male and female internal doses at 0.001 mg/kg/day,
(0.00504/0.001), both from Table 5-35, to yield a unit risk of 1.3 x 10-1 per mg/kg/day.
This document is a draft for review purposes only and does not constitute Agency policy.
179 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
susceptibility, EPA's Supplemental Guidance for Assessing Susceptibility from Early-Life
Exposure to Carcinogens (U.S. EPA, 2005d) advises that increased early-life susceptibility be
assumed and recommends that default age-dependent adjustment factors (ADAFs) be applied to
adjust for this potential increased susceptibility from early-life exposure. As discussed in
Section 4.4, there is sufficient evidence to conclude that a mutagenic MOA is operative for
TCE-induced kidney tumors. In addition, as described in Section 4.10, TCE-specific data are
inadequate for quantification of early-life susceptibility to TCE carcinogenicity. Therefore, as
recommended in the Supplemental Guidance, the default ADAFs are applied.
See the Supplemental Guidance for detailed information on the general application of
these adjustment factors. In brief, the Supplemental Guidance establishes ADAFs for
three specific age groups. The current ADAFs and their age groupings are 10 for <2 years,
three for 2 to <16 years, and one for 16 years and above (U.S. EPA, 2005d). For risk
assessments based on specific exposure assessments, the 10-fold and threefold adjustments to the
slope factor or unit risk estimates are to be combined with age-specific exposure estimates when
estimating cancer risks from early-life (<16-years-of-age) exposure. Currently, due to lack of
appropriate data, no ADAFs are used for other life-stages, such as the elderly. However, the
ADAFs and their age groups may be revised over time. The most current information on the
application of ADAFs for cancer risk assessment can be found at
www, epa. gov/ cancerguidelines.
In the case of TCE, the inhalation unit risk and oral slope factor estimates reflect lifetime
risk for cancer at multiple sites, and a mutagenic MOA has been established for one of these
sites, the kidney. The following subsections illustrate how one might apply the default ADAFs
to the kidney-cancer component of the inhalation unit risk and oral slope factor estimates for
TCE. These are sample calculations, and individual risk assessors should use exposure-related
parameters (e.g., age-specific water ingestion rates) that are appropriate for their particular risk
assessment applications.
In addition to the uncertainties discussed above for the inhalation and oral total cancer
unit risk or slope factor estimates, there are uncertainties in the application of ADAFs to adjust
for potential increased early-life susceptibility. For one thing, the adjustment is made only for
the kidney-cancer component of total cancer risk because that is the tumor type for which the
weight of evidence was sufficient to conclude that TCE-induced carcinogenesis operates through
a mutagenic MOA. However, it may be that TCE operates through a mutagenic MOA for other
tumor types as well or that it operates through other MO As that might also convey increased
early-life susceptibility. Additionally, the ADAFs are general default factors, and it is uncertain
This document is a draft for review purposes only and does not constitute Agency policy.
180 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
to what extent they reflect increased early-life susceptibility for exposure to TCE, if increased
early-life susceptibility occurs.
Furthermore, the assumption of increased early-life susceptibility, invoked by the finding
of a mutagenic MOA for kidney cancer, is in contradiction to the assumption that RR is
independent of age that was used to derive the unit risk estimates in the life-table analysis. In
some other assessments faced with a similar situation, a small modification has been made to the
derivation of the unit risk estimate to avoid the contradictory assumptions (by calculating an
adult-exposure-only unit risk estimate for the application of ADAFs). This has the effect of
slightly reducing the unit risk estimate to which the ADAFs are applied. Because there are
multiple cancer types for TCE but the finding of a mutagenic MOA applies to only one of them,
and because under these circumstances application of the ADAFs already has a minimal impact
on the total risk for most exposure scenarios, as discussed with respect to the examples in
Sections 5.2.3.3.1 and 5.2.3.3.2 below, no attempt was made to modify the kidney cancer unit
risk estimate for this assessment. Such a modification would have substantially increased the
complexity of the calculations, which are already more elaborate than the standard ADAF
applications, without having much quantitative impact on the final risk estimates.
5.2.3.1.4. Example application of age-dependent adjustment factors (ADAFs) for inhalation
exposures.
A calculation template for application of the ADAFs is provided in Table 5-48. In the
"3
example provided, it is assumed that an individual is exposed to 1 |ig/m in air from birth
through age 70 years. Using the template, risk estimates for different exposure scenarios can be
obtained by changing the exposure concentrations (including possibly zero for some age groups).
The steps in the calculation are as follows:
(1) Separate the kidney cancer contribution from the NHL + liver cancer contribution to the
inhalation unit risk estimate. From Section 5.2.2.1.4, the kidney lifetime unit risk is
1.0 x 10 6 per |ig/m3 in air. Subtracting this from the total lifetime unit risk of 4.1 x 10 6
per |ig/m from Section 5.2.2.2 results in the estimated contribution of NHL + liver
cancer being 3.1 x 10~6 per |ig/m3.
(2) Assign a lifetime unit risk estimate for each age group. The template shows the
recommended age groupings from U.S. EPA (2005b) in Column A, along with the age
group duration (Column D), and the fraction of lifetime each age group represents
(Column E; used as a duration adjustment). For each age group, the (unadjusted) lifetime
This document is a draft for review purposes only and does not constitute Agency policy.
181 DRAFT—DO NOT CITE OR QUOTE
-------
1 unit risk estimates for kidney cancer, total cancer, and NHL + liver cancer are shown in
2 columns Column F, I, and J, respectively.
3 (3) For each age group, the kidney cancer inhalation unit risk estimate (Column F) is
"3
4 multiplied by the risk per |ig/m equivalence (Column B), the exposure concentration
5 (Column C), the duration adjustment (Column E), and the ADAF (Column G), to obtain
6 the partial risk from exposure during those ages (Column H). For inhalation exposures, a
7 "risk per |ig/m3 equivalence" of 1 is assumed across age groups (i.e., equivalent risk from
8 equivalent exposure levels in air, independent of body size), as shown in Column B. In
9 this calculation, a unit lifetime exposure of 1 |ig/m3 is assumed, as shown in Column C.
10
This document is a draft for review purposes only and does not constitute Agency policy.
182 DRAFT—DO NOT CITE OR QUOTE
-------
Table 5-48. Sample calculation for total lifetime cancer risk based on the kidney unit risk estimate, potential
risk for NHL and liver cancer, and potential increased early-life susceptibility, assuming a constant lifetime
exposure to 1 jig/m3 of TCE in air
5^
cs
si
3
a
>!
&
o
o
fS"
sj
00 "3
O
>!
Grj
O
a
a
a
§•
>!
o
H §
1 ^
81
o
H
O
HH
H
W
O
V
o
c
o
H
W
Grq
TO
a
"T3
O
t
Col A
Col B
Col C
Col D
Col E
Col F
Col G
Col H
Coll
Col J
Col K
Col L
Exposure scenario parameters
Dose-response assessment calculations
Units:
(Hg/m3)
year
-
(Hg/mT1
-
(Hg/m3)-1
(Hg/m3)-1
Age group
Risk per
|ig/m3 air
equivalence
Exposure
concentration
Age
group
duration
Duration
adjustment
(Col
D/Sum(Col
D))
Kidney cancer
unadjusted
lifetime unit risk
(p 5-137
[5.2.2.1.4])
Default
ADAF
Kidney cancer
ADAF-adjusted
partial risk (Col B
x Col C x Col E x
Col F x Col G)
Kidney
cancer+NHL+
liver cancer
unadjusted
lifetime unit risk
(p 5-139
[5.2.2.2])
NHL+ liver
cancer
lifetime unit
risk (Col I -
Col F)
NHL and liver
cancer partial
risk (Col B x
Col C x Col E x
Col J)
Total partial
risk (Col H +
Col K)
Birth to <1 month
1
1.000
0.083
0.0012
1.0E-06
10
1.2E-08
4.1E-06
3.1E-06
3.7E-09
1.6E-08
1 to <3 months
1
1.000
0.167
0.0024
1.0E-06
10
2.4E-08
4.1E-06
3.1E-06
7.4E-09
3.1E-08
3 to <6 months
1
1.000
0.250
0.0036
1.0E-06
10
3.6E-08
4.1E-06
3.1E-06
1.1E-08
4.7E-08
6 to <12 months
1
1.000
0.500
0.0071
1.0E-06
10
7.1E-08
4.1E-06
3.1E-06
2.2E-08
9.4E-08
1 to <2 years
1
1.000
1.000
0.0143
1.0E-06
10
1.4E-07
4.1E-06
3.1E-06
4.4E-08
1.9E-07
2 to <3 years
1
1.000
1.000
0.0143
1.0E-06
3
4.3E-08
4.1E-06
3.1E-06
4.4E-08
8.7E-08
3 to <6 years
1
1.000
3.000
0.0429
1.0E-06
3
1.3E-07
4.1E-06
3.1E-06
1.3E-07
2.6E-07
6 to <11 years
1
1.000
5.000
0.0714
1.0E-06
3
2.1E-07
4.1E-06
3.1E-06
2.2E-07
4.4E-07
11 to <16 years
1
1.000
5.000
0.0714
1.0E-06
3
2.1E-07
4.1E-06
3.1E-06
2.2E-07
4.4E-07
16 to <21
1
1.000
5.000
0.0714
1.0E-06
1
7.1E-08
4.1E-06
3.1E-06
2.2E-07
2.9E-07
21-70
1
1.000
49.000
0.7000
1.0E-06
1
7.0E-07
4.1E-06
3.1E-06
2.2E-06
2.9E-06
Total exposure
duration:
70
Total unit risk:
4.8E-06
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
(4) For each age group, the NHL + liver cancer unit risk estimate (Column J) is multiplied by
-3
the risk per |ig/m equivalence (Column B), the exposure concentration (Column C), and
the duration adjustment (Column E), to obtain the partial risk from exposure during those
ages (Column K).
(5) For each age group, the ADAF-adjusted partial risk for kidney cancer (Column H) is
added to the partial risk for NHL + liver cancer (Column K), resulting in the total partial
risk (Column L).
(6) The age-group-specific partial risks are added together to obtain the estimated total
lifetime risk (bottom of Column L).
From the example calculation, based on continuous exposure to 1 |ig/m from birth to age 70, the
estimated total lifetime risk is 4.8 x 10 6, which corresponds to a lifetime unit risk estimate of
4.8 x 10 6 per |ig/m3. The risk-specific air concentrations at risk levels of 10 6, 10 5, and 10 4
"3
are 0.21, 2.1, and 21 |ig/m , respectively.
This total cancer unit risk estimate of 4.8 x 10 6 per (J,g/m3 (2.6 x 10 2 per ppm), adjusted
for potential increased early-life susceptibility, is only minimally (17.5%) increased over the
unadjusted total cancer unit risk estimate because the kidney cancer risk estimate that gets
adjusted for potential increased early-life susceptibility is only part of the total cancer risk
estimate. Thus, foregoing the ADAF adjustment in the case of full lifetime calculations will not
seriously impact the resulting risk estimate. For less-than-lifetime exposure calculations, the
impact of applying the ADAFs will increase as the proportion of time at older ages decreases.
The maximum impact will be when exposure is for only the first 2 years of life, in which case the
"3
partial lifetime total cancer risk estimate for exposure to 1 (J,g/m adjusted for potential increased
early-life susceptibility is 10 x (1 (j,g/m3) x (1.0 x 10~6 per (J,g/m3) x (2 / 70) for the kidney cancer
risk + (1 (J,g/m3) x (3.1 x icf6 per (J,g/m3) x (2 / 70) for the NHL and liver cancer, or 3.7 x 10~7,
which is over three times greater than the unadjusted partial lifetime total cancer risk estimate for
exposure to 1 (J,g/m3 of (1 (J-g/m3) x (4.1 x icf6 per (J,g/m3) x (2 / 70), or 1.2 x 10~7.
5.2.3.1.5. Example application of age-dependent adjustment factors (ADAFs) for oral
drinking water exposures
For oral exposures, the calculation of risk estimates adjusted for potential increased
early-life susceptibility is complicated by the fact that for a constant exposure level, e.g., a
constant concentration of TCE in drinking water, doses will vary by age because of different
age-specific uptake rates, e.g., drinking water consumption rates. Different EPA Program or
This document is a draft for review purposes only and does not constitute Agency policy.
184 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Regional Offices may have different default age-specific uptake rates that they use for risk
assessments for specific exposure scenarios, and the calculations presented below are merely to
illustrate the general approach to applying ADAFs for oral TCE exposures, using exposure to
1 [j,g/L of TCE in drinking water from birth through age 70 years as an example. Using the
template, risk estimates for different exposure scenarios can be obtained by changing the intake
rates and exposure concentrations (including possibly zero for some age groups). The steps in
the calculation, illustrated in the template in Table 5-49, are as follows:
(1) Separate the kidney cancer contribution from the NHL + liver cancer contribution to the
oral slope factor estimate. From Section 5.2.2.3, the kidney lifetime oral slope factor is
9.3 x 10 3 per mg/kg/day. Subtracting this from the total lifetime oral slope factor of
_2
4.6 x 10 per mg/kg/day from Section 5.2.2.3 results in an estimated contribution from
NHL + liver cancer of 3.7 x 10 2 per mg/kg/day.
(2) Assign a lifetime oral slope factor estimate for each age group. The template shows the
recommended age groupings from U.S. EPA (2005b) in Column A, along with the age
group duration (Column D), and the fraction of lifetime each age group represents
(Column E; used as a duration adjustment). For each age group, the (unadjusted) lifetime
oral slope factor estimates for kidney cancer, total cancer, and NHL + liver cancer are
shown in columns Column F, I, and J, respectively.
(3) For each age group, the kidney cancer oral slope factor estimate (Column F) is multiplied
by the drinking water ingestion rate (Column B), the exposure concentration (Column C),
the duration adjustment (Column E), and the ADAF (Column G), to obtain the partial risk
from exposure during those ages (Column H). Age-specific water ingestion rates in
L/kg/day, taken from EPA's Child-Specific Exposure Factors Handbook (U.S. EPA,
2008b), are shown in Column B.60 In this calculation, a lifetime unit exposure of 1 |ig/L
is assumed, as shown in Column C.
(4) For each age group, the NHL + liver cancer oral slope factor estimate (Column J) is
multiplied by the drinking water ingestion rate (Column B), the exposure concentration
60 Values for the 90th percentile were taken from Table 3-19 of U.S. EPA (2008c) (consumers-only estimates of
combined direct and indirect water ingestion from community water). The 90th percentile was based on the policy
in the U. S. EPA Office of Water for determining risk through direct and indirect consumption of drinking water.
Community water was used in the illustration because U.S. EPA only regulates community water sources and not
private wells and cisterns or bottled water. Data for "consumers only" (i.e., excluding individuals who did not
ingest community water) were used because formula-fed infants (as opposed to breast-fed infants, who consume
very little community water), children, and young adolescents are often the population of concern with respect to
water consumption. For the 16-21 and 21+ age groups, the standard default rate for adults was used (i.e., 2 L/day +
70 kg, or 0.029 L/kg/day) (U.S. EPA. 1997b. page 3-1). which is identical to the 90th percentile for the 18 to <21
age group.
This document is a draft for review purposes only and does not constitute Agency policy.
185 DRAFT—DO NOT CITE OR QUOTE
-------
1 (Column C), and the duration adjustment (Column E), to obtain the partial risk from
2 exposure during those ages (Column K).
3
This document is a draft for review purposes only and does not constitute Agency policy.
186 DRAFT—DO NOT CITE OR QUOTE
-------
Table 5-49. Sample calculation for total lifetime cancer risk based on the kidney cancer slope factor estimate,
potential risk for NHL and liver cancer, and potential increased early-life susceptibility, assuming a constant
lifetime exposure to 1 jig/L of TCE in drinking water
3^
cs
sj
3
a
>!
ca
*¦
si
00 "3
O
Grj
>!
o
a
a
§•
>!
o
H §
1 ^
81
o
H
O
HH
H
W
O
V
o
c
o
H
W
o
Grq
to
a
"T3
O
t
Col A
Col B
Col C
Col D
Col E
Col F
Col G
Col H
Coll
Col J
Col K
Col L
Exposure scenario parameters
Dose-response assessment calculations
Units:
L water/kg/d
mg/L water
year
-
(mg/kg/day)"1
-
-
(mg/kg/day)"1
(mg/kg/day)"1
-
-
Age group
ingestion
rate
Exposure
concentration
Age
group
duration
Duration
adjustment
(Col
D/Sum(Col
D))
Kidney cancer
unadjusted
lifetime slope
factor (p 5 144
[Table 5-40])
Default
ADAF
Kidney cancer
ADAF-adjusted
partial risk (Col B
x Col C x Col E x
Col F x Col G)
Kidney
cancer+NHL+
liver cancer
unadjusted
lifetime slope
factor (p 5 143
[5.2.2.3])
NHL+ liver
cancer
lifetime slope
factor (Col I -
Col F)
NHL and liver
cancer partial
risk (Col B x
Col C x Col E x
Col J)
Total partial
risk (Col H +
Col K)
Birth to <1 month
0.238
0.001
0.083
0.0012
9.3E-03
10
2.6E-08
4.6E-02
3.7E-02
1.0E-08
3.7E-08
1 to <3 months
0.228
0.001
0.167
0.0024
9.3E-03
10
5.0E-08
4.6E-02
3.7E-02
2.0E-08
7.0E-08
3 to <6 months
0.148
0.001
0.250
0.0036
9.3E-03
10
4.9E-08
4.6E-02
3.7E-02
1.9E-08
6.9E-08
6 to <12 months
0.112
0.001
0.500
0.0071
9.3E-03
10
7.4E-08
4.6E-02
3.7E-02
2.9E-08
1.0E-07
1 to <2 years
0.056
0.001
1.000
0.0143
9.3E-03
10
7.4E-08
4.6E-02
3.7E-02
2.9E-08
1.0E-07
2 to <3 years
0.052
0.001
1.000
0.0143
9.3E-03
3
2.1E-08
4.6E-02
3.7E-02
2.7E-08
4.8E-08
3 to <6 years
0.049
0.001
3.000
0.0429
9.3E-03
3
5.9E-08
4.6E-02
3.7E-02
7.7E-08
1.4E-07
6 to <11 years
0.035
0.001
5.000
0.0714
9.3E-03
3
7.0E-08
4.6E-02
3.7E-02
9.2E-08
1.6E-07
11 to <16 years
0.026
0.001
5.000
0.0714
9.3E-03
3
5.2E-08
4.6E-02
3.7E-02
6.8E-08
1.2E-07
16 to <21
0.029
0.001
5.000
0.0714
9.3E-03
1
1.9E-08
4.6E-02
3.7E-02
7.6E-08
9.5E-08
21-70
0.029
0.001
49.000
0.7000
9.3E-03
1
1.9E-07
4.6E-02
3.7E-02
7.5E-07
9.3E-07
Total exposure
duration:
70
Total unit risk:
1.9E-06
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
(5) For each age group, the ADAF-adjusted partial risk for kidney cancer (Column H) is
added to the partial risk for NHL + liver cancer (Column K), resulting in the total partial
risk (Column L).
(6) The age-group-specific partial risks are added together to obtain the estimated total
lifetime risk (bottom of Column L).
Because the TCE intake is not constant across age groups, one does not calculate a
lifetime unit risk estimate in terms of risk per mg/kg/day adjusted for potential increased
early-life susceptibility. One could calculate a unit risk estimate for TCE in drinking water in
terms of [j,g/L from the result in Table 5-49, but this is dependent on the water ingestion rates
used. Based on the example calculation assuming continuous exposure to 1 [j,g/L of TCE in
drinking water from birth to age 70 years and using the drinking water intake rates shown,
estimated total lifetime risk is 1.9 x 10 6, which corresponds to a lifetime drinking water unit risk
estimate of 1.9 x 10 6 per |ig/L. The corresponding risk-specific drinking water concentrations
at risk levels of 10~6, 10~5, and 10~4 are 0.53, 5.3, and 53 |ig/L, respectively. For different
exposure and intake parameters, the risk-specific drinking water concentrations would need to be
recalculated.
As with the adjusted inhalation risk estimate in Section 5.2.3.3.1, the lifetime total cancer
risk estimate of 1.9 x 10 6 calculated for lifetime exposure to 1 [j,g/L of TCE in drinking water
adjusted for potential increased early-life susceptibility is only minimally (25%) increased over
the unadjusted total cancer unit risk estimate. (This calculation is not shown, but if one omits the
ADAFs for each of the age groups in Table 5-49, the resulting total lifetime risk estimate is
1.5 x 10 6.) Unlike with inhalation exposure under the assumption of ppm equivalence, the oral
intake rates are higher in the potentially more susceptible younger age groups. This would tend
to yield a larger relative impact of adjusting for potential increased early-life susceptibility for
oral risk estimates compared to inhalation risk estimates. In the case of TCE, however, this
impact is partially offset by the lesser proportion of the total oral cancer risk that is accounted for
by the kidney cancer risk, which is the component of total risk that is being adjusted for potential
increased early-life susceptibility, based on the primary dose-metrics (1/5 vs. 1/4 for inhalation).
Thus, as with lifetime inhalation risk, foregoing the ADAF adjustment in the case of full lifetime
calculations will not seriously impact the resulting risk estimate. For less-than-lifetime exposure
calculations, the impact of applying the ADAFs will increase as the proportion of time at older
ages decreases. The maximum impact will be when exposure is for only the first 2 years of life,
in which case the partial lifetime total cancer risk estimate for exposure to 1 [j,g/L adjusted for
_n
potential increased early-life susceptibility is 3.8 x 10 (adding partial risks from Table 5-49 for
This document is a draft for review purposes only and does not constitute Agency policy.
188 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
the appropriate ages groups), which is almost three times greater than the unadjusted partial
lifetime total cancer risk estimate for exposure to 1 [j,g/L of 5 x (0.001 mg/L) x (0.103 L/kg/day)
x (9.33 x 10 3 per mg/kg/day) x (2/70), or 1.4 x 10 1, where 5 is the factor for the multiple
cancer types for oral exposure, 0.103 L/kg/day is the time-weighted ingestion rate for the
1st two years of life using the rates in Table 5-49, 9.33 x io~3 per mg/kg/day is the unadjusted
oral slope factor estimate for kidney cancer, and 2/70 is the duration adjustment.
5.3. KEY RESEARCH NEEDS FOR TCE DOSE-RESPONSE ANALYSES
For noncancer dose-response assessment, key research that would substantially improve
the accuracy or utility of TCE noncancer risk estimates includes:
• Research to obtain toxicokinetic data to better quantify the amount of bioactivation of
DCVC to toxic moiety(ies) in rats and humans, including data on human variability in
DCVC bioactivation.
• Research to obtain mechanistic data that would identify the active moiety(ies) for
TCE-induced immunological effects and developmental cardiac defects. As a
corollary, data on human variability pharmacokinetics of the active moiety after TCE
exposure would also be informative.
• Research to obtain mechanistic data that would quantitatively inform the
pharmacodynamic factors that would make individuals more or less susceptible to
kidney, immunological, and developmental cardiac defects induced by TCE.
• Research to obtain TCE dose-response data on kidney effects, immunological effects,
and developmental cardiac defects at a larger number of doses at and below the
current LOAELs, so as to better describe the dose-response shape at low effect levels.
Ideally, studies would be based on human epidemiologic data with good quantitative
exposure assessment. Studies in laboratory animals would need to address the
limitations in the currently available studies. For example, studies of cardiac defects
would need to address limitations of the Johnson et al. (2003) study described in
Section 4.8.3.3.2.
• Development of a probabilistic approach to noncancer dose-response analysis that
would enable calculation of a risk-specific dose for noncancer effects, while capturing
uncertainty and variability quantitatively.
This document is a draft for review purposes only and does not constitute Agency policy.
189 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
1
For cancer dose-response assessment, key research that would substantially improve the
accuracy or utility of TCE cancer risk estimates includes:
• Research to obtain toxicokinetic data to better quantify the amount of bioactivation of
DCVC to toxic moiety(ies) in humans, including data on human variability in DCVC
bioactivation.
• Research to obtain mechanistic data that would identify the active moiety(ies) for
TCE-induced liver tumors and NHL. As a corollary, data on human variability
pharmacokinetics of the active moiety after TCE exposure would also be informative.
• Research to obtain mechanistic data that would quantitatively inform the
pharmacodynamic factors that would make individuals more or less susceptible to
kidney tumors, liver tumors, and NHL induced by TCE. This includes data on
life-stage-specific susceptibility that would replace the default ADAFs for kidney
tumors and the assumption of no life-stage-specific susceptibility for liver tumors and
NHL.
• Research to obtain human epidemiologic dose-response data on TCE-induced kidney
tumors, liver tumors, and NHL with good quantitative exposure assessment.
• Research to obtain additional human epidemiologic data on TCE exposure and other
tumors, so as to better estimate the total risk of cancer from TCE exposure.
• Development of a probabilistic approach to cancer dose-response analysis that would
enable calculation of a differential susceptibility to carcinogenic effects, while
capturing uncertainty and variability quantitatively
This document is a draft for review purposes only and does not constitute Agency policy.
190 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
6. MAJOR CONCLUSIONS IN THE CHARACTERIZATION OF
HAZARD AND DOSE RESPONSE
6.1. HUMAN HAZARD POTENTIAL
This section summarizes the human hazard potential for trichloroethylene (TCE). For extensive
discussions and references, see Section 2 for Exposure, Section 3 for toxicokinetics and
physiologically based pharmacokinetic (PBPK) modeling, and Sections 4.1-4.9 for the
epidemiologic and experimental studies of TCE noncancer and cancer toxicity. Section 4.10
summarizes information on susceptibility, and Section 4.11 provides a more detailed summary
and references for noncancer toxicity and carcinogenicity.
6.1.1. Exposure (see Section 2)
TCE is a volatile compound with moderate water solubility. Most TCE produced today is used
for metal degreasing. The highest environmental releases are to the air. Ambient air monitoring
"3
data suggest that mean levels have remained fairly constant since 1999 at about 0.3 (j,g/m
(0.06 ppb). As discussed in Section 2, in 2006, ambient air monitors (n = 258) had annual means
3 3
ranging from 0.03-7.73 (J,g/m with a median of 0.13 and an overall average of 0.23 (J,g/m .
Indoor levels are commonly three or more times higher than outdoor levels due to releases from
building materials and consumer products. Vapor intrusion is a likely significant source in
situations where residences are located near soils or groundwater with high contamination levels
and sparse indoor air sampling had detected TCE levels ranging from 1-140 (J,g/m . TCE is
among the most common groundwater contaminants and the one present in the highest
concentration in a summary of ground water analyses reported in 1982. The median level of
TCE in groundwater, based on a large survey by the U.S. Geological Survey for 1985-2001, is
0.15 (J,g/L. It has also been detected in a wide variety of foods in the 1-100 (J,g/kg range. None
of the environmental sampling has been done using statistically based national surveys.
However, a substantial amount of air and groundwater data have been collected allowing
reasonably well supported estimates of typical daily intakes by the general population:
inhalation—13 (j,g/day and water ingestion—0.2 (j,g/day. The limited food data suggest an intake
of about 5 (J,g/day, but this must be considered preliminary. Higher exposures have occurred to
various occupational groups, particularly with vapor degreasing that has the highest potential for
exposure because vapors can escape into the work place. For example, past studies of aircraft
"3
workers have shown short term peak exposures in the hundreds of ppm (>500,000 (J,g/m ) and
"3
long term exposures in the low tens of ppm (>50,000 (j,g/m ). Occupational exposures have
This document is a draft for review purposes only and does not constitute Agency policy.
1 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
likely decreased in recent years due to better release controls, improvements in worker
protection, and substituting other solvents for TCE.
Exposure to a variety of TCE related compounds, which include metabolites of TCE and other
parent compounds that produce similar metabolites, can alter or enhance TCE metabolism and
toxicity by generating higher internal metabolite concentrations than would result from TCE
exposure by itself. Available estimates suggest that exposures to most of these TCE-related
compounds are comparable to or greater than TCE itself.
6.1.2. Toxicokinetics and Physiologically Based Pharmacokinetic Modeling (see Section 3
and Appendix A)
TCE is a lipophilic compound that readily crosses biological membranes. Exposures may occur
via the oral, dermal, and inhalation route, with evidence for systemic availability from each
route. TCE can also be transferred transplacental^ and through breast milk ingestion. TCE is
rapidly and nearly completely absorbed from the gut following oral administration, and animal
studies indicate that exposure vehicle may impact the time course of absorption: oily vehicles
may delay absorption whereas aqueous vehicles result in a more rapid increase in blood
concentrations. See Section 3.1 for additional discussion of TCE absorption.
Following absorption to the systemic circulation, TCE distributes from blood to solid tissues by
each organ's solubility. This process is mainly determined by the blood:tissue partition
coefficients, which are largely determined by tissue lipid content. Adipose partitioning is high,
so adipose tissue may serve as a reservoir for TCE, and accumulation into adipose tissue may
prolong internal exposures. TCE attains high concentrations relative to blood in the brain,
kidney, and liver-all of which are important target organs of toxicity. TCE is cleared via
metabolism mainly in three organs: the kidney, liver, and lungs. See Section 3.2 for additional
discussion of TCE distribution.
The metabolism of TCE is an important determinant of its toxicity. Metabolites are generally
thought to be responsible for toxicity-especially for the liver and kidney. Initially, TCE may be
oxidized via cytochrome P450 (CYP) isoforms or conjugated with glutathione (GSH) by
glutathione S-transferase enzymes. While CYP2E1 is generally accepted to be the CYP isoform
most responsible for TCE oxidation, others forms may also contribute. There are conflicting
data as to which glutathione-S-transferase (GST) isoforms are responsible for TCE conjugation,
with one rat study indicating alpha-class GSTs and another rat study indicating mu and pi-class
GST. The balance between oxidative and conjugative metabolites generally favors the oxidative
pathway, especially at lower concentrations, and inhibition of CYP-dependent oxidation in vitro
This document is a draft for review purposes only and does not constitute Agency policy.
2 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
increases glutathione conjugation in renal preparations. However, different investigators have
reported considerably different rates for TCE conjugation in human liver and kidney cell
fractions, perhaps due to different analytical methods. The inferred flux through the GSH
pathway differs by more than four orders of magnitude across data sets. While the available data
are consistent with the higher values being overestimates, the degree of overestimation is
unclear, and differing results may be attributable to true interindividual variation. Overall, there
remains significant uncertainty in the quantitative estimation of TCE GSH conjugation. See
Section 3.3 for additional discussion of TCE metabolism.
Once absorbed, TCE is excreted primarily either in breath as unchanged TCE or carbon dioxide
[CO2], or in urine as metabolites. Minor pathways of elimination include excretion of
metabolites in saliva, sweat, and feces. Following oral administration or upon cessation of
inhalation exposure, exhalation of unmetabolized TCE is a major elimination pathway. Initially,
elimination of TCE upon cessation of inhalation exposure demonstrates a steep concentration-
time profile: TCE is rapidly eliminated in the minutes and hours postexposure, and then the rate
of elimination via exhalation decreases. Following oral or inhalation exposure, urinary
elimination of parent TCE is minimal, with urinary elimination of the metabolites trichloroacetic
acid and trichloroethanol accounting for the bulk of the absorbed dose of TCE. See Section 3.4
for additional discussion of TCE excretion.
As part of this assessment, a comprehensive Bayesian PBPK model-based analysis of the
population toxicokinetics of TCE and its metabolites was developed in mice, rats, and humans
(also reported in Chiu et al., 2009). This analysis considered a wider range of physiological,
chemical, in vitro, and in vivo data than any previously published analysis of TCE. The
toxicokinetics of the "population average," its population variability, and their uncertainties are
characterized and estimates of experimental variability and uncertainty are included in this
analysis. The experimental database included separate sets for model calibration and evaluation
for rats and humans; fewer data were available in mice, and were all used for model calibration.
The total combination of these approaches and PBPK analysis substantially supports the model
predictions. In addition, the approach employed yields an accurate characterization of the
uncertainty in metabolic pathways for which available data were sparse or relatively indirect,
such as GSH conjugation and respiratory tract metabolism. Key conclusions from the model
predictions include (1) as expected, TCE is substantially metabolized, primarily by oxidation at
doses below saturation; (2) GSH conjugation and subsequent bioactivation in humans appears to
be 10- to 100-fold greater than previously estimated; and (3) mice had the greatest rate of
respiratory tract oxidative metabolism compared to rats and humans. However, there are
uncertainties as to the accuracy of the analytical method used for some of the available in vivo
This document is a draft for review purposes only and does not constitute Agency policy.
3 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
data on GSH conjugation. Because these data are highly influential, the PBPK modeling results
for the flux of GSH conjugation should be interpreted with caution. Thus, there is lower
confidence in the accuracy of GSH conjugation predictions as compared to other dose-metrics,
such as those related to the parent compound, total metabolism, or oxidative metabolites. The
predictions of the PBPK model are subsequently used in noncancer and cancer dose-response
analyses for inter and intraspecies extrapolation of toxicokinetics (see Section 6.2, below). See
Section 3.5 and Appendix A for additional discussion of and details about PBPK modeling of
TCE and metabolites.
6.1.3. Noncancer Toxicity
This section summarizes the weight of evidence for TCE noncancer toxicity. Based on the
available human epidemiologic data and experimental and mechanistic studies, it is concluded
that TCE poses a potential human health hazard for noncancer toxicity to the central nervous
system, the kidney, the liver, the immune system, the male reproductive system, and the
developing fetus. The evidence is more limited for TCE toxicity to the respiratory tract and
female reproductive system. The conclusions pertaining to specific endpoints within these
tissues and systems are summarized below.
6.1.3.1.1. Neurological Effects (see Sections 4.3 and 4.11.1.1 and Appendix D)
Both human and animal studies have associated TCE exposure with effects on several
neurological domains. Multiple epidemiologic studies in different populations have reported
abnormalities in trigeminal nerve function in association with TCE exposure. Two small studies
did not report an association between TCE exposure and trigeminal nerve function. However,
statistical power was limited, exposure misclassification was possible, and, in one case, methods
for assessing trigeminal nerve function were not available. As a result, these studies do not
provide substantial evidence against a causal relationship between TCE exposure and trigeminal
nerve impairment. Laboratory animal studies have also demonstrated TCE-induced changes in
the morphology of the trigeminal nerve following short-term exposures in rats. However, one
study reported no significant changes in trigeminal somatosensory evoked potential in rats
exposed to TCE for 13 weeks. See Section 4.3.1 for additional discussion of studies of
alterations in nerve conduction and trigeminal nerve effects. Human chamber, occupational, and
geographic based/drinking water studies have consistently reported subjective symptoms such as
headaches, dizziness, and nausea which are suggestive of vestibular system impairments. One
This document is a draft for review purposes only and does not constitute Agency policy.
4 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
study reported changes in nystagmus threshold (a measure of vestibular system function)
following an acute TCE exposure. There are only a few laboratory animal studies relevant to
this neurological domain, with reports of changes in nystagmus, balance, and handling reactivity.
See Section 4.3.3 for additional discussion of TCE effects on vestibular function. Fewer and
more limited epidemiologic studies are suggestive of TCE exposure being associated with
delayed motor function, and changes in auditory, visual, and cognitive function or performance
(see Sections 4.3.2, 4.3.4, 4.3.5, and 4.3.6). Acute and subchronic animal studies show
disruption of the auditory system, changes in visual evoked responses to patterns or flash
stimulus, and neurochemical and molecular changes. Animal studies suggest that while the
effects on the auditory system lead to permanent function impairments and histopathology,
effects on the visual system may be reversible with termination of exposure. Additional acute
studies reported structural or functional changes in hippocampus, such as decreased myelination
or decreased excitability of hippocampal CA1 neurons, although the relationship of these effects
to overall cognitive function is not established (see Section 4.3.9). An association between TCE
exposure and sleep changes has also been demonstrated in rats (see Section 4.3.7). Some
evidence exists for motor-related changes in rats/mice exposed acutely/subchronically to TCE,
but these effects have not been reported consistently across all studies (see Section 4.3.6).
Gestational exposure to TCE in humans has been reported to be associated with
neurodevelopmental abnormalities including neural tube defects, encephalopathy, impaired
cognition, aggressive behavior, and speech and hearing impairment. Developmental
neurotoxicological changes have also been observed in animals including aggressive behaviors
following an in utero exposure to TCE and a suggestion of impaired cognition as noted by
decreased myelination in the CA1 hippocampal region of the brain. See Section 4.3.8 for
additional discussion of developmental neurological effects of TCE. Therefore, overall, the
strongest neurological evidence of human toxicological hazard is for changes in trigeminal nerve
function or morphology and impairment of vestibular function, based on both human and
experimental studies, while fewer and more limited evidence exists for delayed motor function,
changes in auditory, visual, and cognitive function or performance, and neurodevelopmental
outcomes.
6.1.3.1.2. Kidney Effects (see Sections 4.4.1, 4.4.4, 4.4.6, and 4.11.1.2)
Kidney toxicity has also been associated with TCE exposure in both human and animal
studies. There are few human data pertaining to TCE-related noncancer kidney toxicity;
however, several available studies reported elevated excretion of urinary proteins, considered
This document is a draft for review purposes only and does not constitute Agency policy.
5 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
nonspecific markers of nephrotoxicity, among TCE-exposed subjects compared to unexposed
controls. While some of these studies include subjects previously diagnosed with kidney cancer,
other studies report similar results in subjects that are disease free. Some additional support for
TCE nephrotoxicity in humans is provided by two studies of end-stage renal disease; a study
reporting a greater incidence of end-stage renal disease in TCE-exposed workers as compared to
unexposed controls and a second study reporting a greater risk for progression from IgA or
membranous nephropathy glomerulonephritis to end-stage renal disease and TCE-exposure. See
Section 4.4.1 for additional discussion of human data on the noncancer kidney effects of TCE.
Laboratory animal and in vitro data provide additional support for TCE nephrotoxicity. TCE
causes renal toxicity in the form of cytomegaly and karyomegaly of the renal tubules in male and
female rats and mice following either oral or inhalation exposure. In rats, the pathology of TCE-
induced nephrotoxicity appears distinct from age-related nephropathy. Increased kidney weights
have also been reported in some rodent studies. See Section 4.4.4 for additional discussion of
laboratory animal data on the noncancer kidney effects of TCE. Further studies with TCE
metabolites have demonstrated a potential role for dichlorovinyl cysteine (DCVC),
trichloroethanol, and trichloroacetic acid (TCA) in TCE-induced nephrotoxicity. Of these,
available data suggest that DCVC induced renal effects are most similar to those of TCE and that
DCVC is formed in sufficient amounts following TCE exposure to account for these effects.
TCE or DCVC have also been shown to be cytotoxic to primary cultures of rat and human renal
tubular cells. See Section 4.4.6 for additional discussion on the role of metabolism in the
noncancer kidney effects of TCE. Overall, multiple lines of evidence support the conclusion that
TCE causes nephrotoxicity in the form of tubular toxicity, mediated predominantly through the
TCE GSH conjugation product DCVC.
6.1.3.1.3. Liver Effects (see Sections 4.5.1, 4.5.3, 4.5.4, 4.5.6, and 4.11.1.3, and Appendix E)
Liver toxicity has also been associated with TCE exposure in both human and animal
studies. Although there are few human studies on liver toxicity and TCE exposure, several
available studies have reported TCE exposure to be associated with significant changes in serum
liver function tests, widely used in clinical settings in part to identify patients with liver disease,
or changes in plasma or serum bile acids. Additional, more limited human evidence for TCE
induced liver toxicity includes reports suggesting an association between TCE exposure and liver
disorders, and case reports of liver toxicity including hepatitis accompanying immune-related
generalized skin diseases, jaundice, hepatomegaly, hepatosplenomegaly, and liver failure in
TCE-exposed workers. Cohort studies examining cirrhosis mortality and either TCE exposure or
This document is a draft for review purposes only and does not constitute Agency policy.
6 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
solvent exposure are generally null, but these studies cannot rule out an association with TCE
because of their use of death certificates where there is a high degree (up to 50%) of
underreporting. Overall, while some evidence exists of liver toxicity as assessed from liver
function tests, the data are inadequate for making conclusions regarding causality. See
Section 4.5.1 for additional discussion of human data on the noncancer liver effects of TCE. In
rats and mice, TCE exposure causes hepatomegaly without concurrent cytotoxicity. Like
humans, laboratory animals exposed to TCE have been observed to have increased serum bile
acids, although the toxicological importance of this effect is unclear. Other effects in the rodent
liver include small transient increases in DNA synthesis, cytomegaly in the form of "swollen" or
enlarged hepatocytes, increased nuclear size probably reflecting polyploidization, and
proliferation of peroxisomes. Available data also suggest that TCE does not induce substantial
cytotoxicity, necrosis, or regenerative hyperplasia, as only isolated, focal necroses and mild to
moderate changes in serum and liver enzyme toxicity markers having been reported. These
effects are consistently observed across rodent species and strains, although the degree of
response at a given mg/kg-day dose appears to be highly variable across strains, with mice on
average appearing to be more sensitive. See Sections 4.5.3 and 4.5.4 for additional discussion of
laboratory animal data on the noncancer liver effects of TCE. While it is likely that oxidative
metabolism is necessary for TCE-induced effects in the liver, the specific metabolite or
metabolites responsible is less clear. However, the available data are strongly inconsistent with
TCA being the sole or predominant active moiety for TCE-induced liver effects, particularly
with respect to hepatomegaly. See Section 4.5.6 for additional discussion on the role of
metabolism in the noncancer liver effects of TCE. Overall, TCE, likely through its oxidative
metabolites, clearly leads to liver toxicity in laboratory animals, with mice appearing to be more
sensitive than other laboratory animal species, but there is only limited epidemiologic evidence
of hepatotoxicity being associated with TCE exposure.
6.1.3.1.4. Immunological Effects (see Sections 4.6.1.1, 4.6.2, and 4.11.1.4)
Effects related the immune system have also been associated with TCE exposure in both
human and animal studies. A relationship between systemic autoimmune diseases, such as
scleroderma, and occupational exposure to TCE has been reported in several recent studies, and a
meta-analysis of scleroderma studies resulted in a statistically significant combined odds ratio for
any exposure in men (odds ratio [OR]: 2.5, 95% confidence interval [CI]: 1.1, 5.4), with a lower
relative risk seen in women (OR: 1.2, 95% CI: 0.58, 2.6). The human data at this time do not
allow a determination of whether the difference in effect estimates between men and women
This document is a draft for review purposes only and does not constitute Agency policy.
1 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
reflects the relatively low background risk of scleroderma in men, gender-related differences in
exposure prevalence or in the reliability of exposure assessment, a gender-related difference in
susceptibility to the effects of TCE, or chance. Additional human evidence for the
immunological effects of TCE includes studies reporting TCE-associated changes in levels of
inflammatory cytokines in occupationally-exposed workers and infants exposed via indoor air at
air concentrations typical of such exposure scenarios (see Section 6.1.1, above); a large number
of case reports (mentioned above) of a severe hypersensitivity skin disorder, distinct from
contact dermatitis and often accompanied by hepatitis; and a reported association between
increased history of infections and exposure to TCE contaminated drinking water. See
Section 4.6.1.1 for additional discussion of human data on the immunological effects of TCE.
Immunotoxicity has also been reported in experimental rodent studies of TCE. Numerous
studies have demonstrated accelerated autoimmune responses in autoimmune-prone mice,
including changes in cytokine levels similar to those reported in human studies, with more severe
effects, including autoimmune hepatitis, inflammatory skin lesions, and alopecia, manifesting at
longer exposure periods. Immunotoxic effects have been also reported in B6C3F1 mice, which
do not have a known particular susceptibility to autoimmune disease. Developmental
immunotoxicity in the form of hypersensitivity responses have been reported in TCE-treated
guinea pigs and mice via drinking water pre- and postnatally. Evidence of localized
immunosuppression has also been reported in mice and rats. See Section 4.6.2 for additional
discussion of laboratory animal data on the immunological effects of TCE. Overall, the human
and animal studies of TCE and immune-related effects provide strong evidence for a role of TCE
in autoimmune disease and in a specific type of generalized hypersensitivity syndrome, while
there are less data pertaining to immunosuppressive effects.
6.1.3.1.5. Respiratory Tract Effects (see Sections 4.7.1.1, 4.7.2.1, 4.7.3, and 4.11.1.5)
The very few human data on TCE and pulmonary toxicity are too limited for drawing
conclusions (see Section 4.7.1.1), but laboratory studies in mice and rats have shown toxicity in
the bronchial epithelium, primarily in Clara cells, following acute exposures to TCE (see
Section 4.7.2.1). A few studies of longer duration have reported more generalized toxicity, such
as pulmonary fibrosis in mice and pulmonary vasculitis in rats. However, respiratory tract
effects were not reported in other longer-term studies. Acute pulmonary toxicity appears to be
dependent on oxidative metabolism, although the particular active moiety is not known. While
earlier studies implicated chloral produced in situ by CYP enzymes in respiratory tract tissue in
toxicity, the evidence is inconsistent and several other possibilities are viable. Although humans
This document is a draft for review purposes only and does not constitute Agency policy.
8 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
appear to have lower overall capacity for enzymatic oxidation in the lung relative to mice, CYP
enzymes do reside in human respiratory tract tissue, suggesting that, qualitatively, the respiratory
tract toxicity observed in rodents is biologically plausible in humans. See Section 4.7.3 for
additional discussion of the role of metabolism in the noncancer respiratory tract toxicity of
TCE. Therefore, overall, data are suggestive of TCE causing respiratory tract toxicity, based
primarily on short-term studies in mice and rats, with available human data too few and limited
to add to the weight of evidence for pulmonary toxicity.
6.1.3.1.6. Reproductive Effects (see Sections 4.8.1 and 4.11.1.6)
A number of human and laboratory animal studies suggest that TCE exposure has the
potential for male reproductive toxicity, with a more limited number of studies examining female
reproductive toxicity. Human studies have reported TCE exposure to be associated (in all but
one case statistically-significantly) with increased sperm density and decreased sperm quality,
altered sexual drive or function, or altered serum endocrine levels. Measures of male fertility,
however, were either not reported or reported to be unchanged with TCE exposure, though the
statistical power of the available studies is quite limited. Epidemiologic studies have identified
possible associations of TCE exposure with effects on female fertility and with menstrual cycle
disturbances, but these data are fewer than those available for male reproductive toxicity. See
Section 4.8.1.1 for additional discussion of human data on the reproductive effects of TCE.
Evidence of similar effects, particularly for male reproductive toxicity, is provided by several
laboratory animal studies that reported effects on sperm, libido/copulatory behavior, and serum
hormone levels, although some studies that assessed sperm measures did not report treatment-
related alterations. Additional adverse effects on male reproduction have also been reported,
including histopathological lesions in the testes or epididymides and altered in vitro sperm-
oocyte binding or in vivo fertilization due to TCE or metabolites. While reduced fertility in
rodents was only observed in one study, this is not surprising given the redundancy and
efficiency of rodent reproductive capabilities. In addition, although the reduced fertility
observed in the rodent study was originally attributed to systemic toxicity, the database as a
whole suggests that TCE does induce reproductive toxicity independent of systemic effects.
Fewer data are available in rodents on female reproductive toxicity. While in vitro oocyte
fertilizability has been reported to be reduced as a result of TCE exposure in rats, a number of
other laboratory animal studies did not report adverse effects on female reproductive function.
See Section 4.8.1.2 for additional discussion of laboratory animal data on the reproductive
effects of TCE. Very limited data are available to elucidate the mode of action (MOA) for these
This document is a draft for review purposes only and does not constitute Agency policy.
9 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
effects, though some aspects of a putative MOA (e.g., perturbations in testosterone biosynthesis)
appear to have some commonalities between humans and animals (see Section 4.8.1.3.2).
Together, the human and laboratory animal data support the conclusion that TCE exposure poses
a potential hazard to the male reproductive system, but are more limited with regard to the
potential hazard to the female reproductive system.
6.1.3.1.7. Developmental Effects (see Sections 4.8.3 and 4.11.1.7)
The relationship between TCE exposure (direct or parental) and developmental toxicity
has been investigated in a number of epidemiologic and laboratory animal studies. Postnatal
developmental outcomes examined include developmental neurotoxicity (addressed above with
neurotoxicity), developmental immunotoxicity (addressed above with immunotoxicity), and
childhood cancers. Prenatal effects examined include death (spontaneous abortion, perinatal
death, pre- or postimplantation loss, resorptions), decreased growth (low birth weight, small for
gestational age, intrauterine growth restriction, decreased postnatal growth), and congenital
malformations, in particular cardiac defects. Some epidemiological studies have reported
associations between parental exposure to TCE and spontaneous abortion or perinatal death, and
decreased birth weight or small for gestational age, although other studies reported mixed or null
findings. While comprising both occupational and environmental exposures, these studies are
overall not highly informative due to the small numbers of cases and limited exposure
characterization or to the fact that exposures were to a mixture of solvents. See Section 4.8.3.1
for additional discussion of human data on the developmental effects of TCE. However,
multiple well conducted studies in rats and mice show analogous effects of TCE exposure: pre-
or postimplantation losses, increased resorptions, perinatal death, and decreased birth weight.
Interestingly, the rat studies reporting these effects used Fischer 344 or Wistar rats, while several
other studies, all of which used Sprague-Dawley rats, reported no increased risk in these
developmental measures, suggesting a strain difference in susceptibility. See Section 4.8.3.2 for
additional discussion of laboratory animal data on the developmental effects of TCE. Therefore,
overall, based on weakly suggestive epidemiologic data and fairly consistent laboratory animal
data, it can be concluded that TCE exposure poses a potential hazard for prenatal losses and
decreased growth or birth weight of offspring.
With respect to congenital malformations, epidemiology and experimental animal studies
of TCE have reported increases in total birth defects, central nervous system defects, oral cleft
defects, eye/ear defects, kidney/urinary tract disorders, musculoskeletal birth anomalies,
lung/respiratory tract disorders, skeletal defects, and cardiac defects. Human occupational cohort
This document is a draft for review purposes only and does not constitute Agency policy.
10 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
studies, while not consistently reporting positive results, are generally limited by the small
number of observed or expected cases of birth defects. While only one of the epidemiological
studies specifically reported observations of eye anomalies, studies in rats have identified
increases in the incidence of fetal eye defects following oral exposures during the period of
organogenesis with TCE or its oxidative metabolites dichloroacetic acid (DCA) and TCA. The
epidemiological studies, while individually limited, as a whole show relatively consistent
elevations, some of which were statistically significant, in the incidence of cardiac defects in
TCE-exposed populations compared to reference groups. In laboratory animal models, avian
studies were the first to identify adverse effects of TCE exposure on cardiac development, and
the initial findings have been confirmed multiple times. Additionally, administration of TCE and
its metabolites TCA and DCA in maternal drinking water during gestation has been reported to
induce cardiac malformations in rat fetuses. It is notable that a number of other studies, several
of which were well-conducted, did not report induction of cardiac defects in rats, mice, or rabbits
in which TCE was administered by inhalation or gavage. However, many of these studies used a
traditional free-hand section technique on fixed fetal specimens, and a fresh dissection technique
that can enhance detection of anomalies was used in the positive studies by Dawson et al. (1993)
and Johnson et al. (2003)(2005). Nonetheless, two studies that used the same or similar fresh
dissection technique did not report cardiac anomalies. Differences in other aspects of
experimental design may have been contributing factors to the differences in observed response.
In addition, mechanistic studies, such as the treatment-related alterations in endothelial cushion
development observed in avian in ovo and in vitro studies, provide a plausible mechanistic basis
for defects in septal and valvular morphogenesis observed in rodents, and consequently support
the plausibility of cardiac defects induced by TCE in humans. Therefore, while the studies by
Dawson et al. (1993) and Johnson et al. (2003)(2005) have significant limitations, including the
lack of clear dose-response relationship for the incidence of any specific cardiac anomaly and the
pooling of data collected over an extended period, there is insufficient reason to dismiss their
findings. See Section 4.8.3.3.2 for additional discussion of the conclusions with respect to
TCE-induced cardiac malformations. Therefore, overall, based on weakly suggestive, but
overall consistent, epidemiologic data, in combination with evidence from experimental animal
and mechanistic studies, it can be concluded that TCE exposure poses a potential hazard for
congenital malformations, including cardiac defects, in offspring.
This document is a draft for review purposes only and does not constitute Agency policy.
11 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
6.1.4. Carcinogenicity (see Sections 4.1, 4.2, 4.4.2, 4.4.5, 4.4.7, 4.5.2, 4.5.5, 4.5.6, 4.5.7,
4.6.1.2, 4.6.2.4, 4.7.1.2, 4.7.2.2, 4.7.4, 4.8.2, 4.9, and 4.11.2, and Appendices B and C)
Following EPA (2005 c) Guidelines for Carcinogen Risk Assessment, based on the
available data as of 2010, TCE is characterized as "carcinogenic to humans" by all routes of
exposure. This conclusion is based on convincing evidence of a causal association between TCE
exposure in humans and kidney cancer. The consistency of increased kidney cancer relative risk
estimates across a large number of independent studies of different designs and populations from
different countries and industries provides compelling evidence given the difficulty, a priori, in
detecting effects in epidemiologic studies when the relative risks are modest, the cancers are
relatively rare, and therefore, individual studies have limited statistical power. This strong
consistency of the epidemiologic data on TCE and kidney cancer argues against chance, bias,
and confounding as explanations for the elevated kidney cancer risks. In addition, statistically
significant exposure-response trends are observed in high-quality studies. These studies were
conducted in populations with high TCE exposure intensity or had the ability to identify
TCE-exposed subjects with high confidence. These studies addressed important potential
confounders and biases, further supporting the observed associations with kidney cancer as
causal. See Section 4.4.2 for additional discussion of the human epidemiologic data on TCE
exposure and kidney cancer. In a meta-analysis of 15 studies with high exposure potential, a
statistically significant summary relative risk estimate was observed for overall TCE exposure
(RRm: 1.27 [95% CI: 1.13, 1.43]). The summary relative risk estimate was greater for the
highest TCE exposure groups (RRm: 1.58 [95% CI: 1.28, 1.96]; n = 13 studies). Meta-analyses
investigating the influence of individual studies and the sensitivity of the results to alternate
relative risk estimate selections found the summary relative risk estimates to be highly robust.
Furthermore, there was no indication of publication bias or significant heterogeneity across the
15 studies. It would require a substantial amount of negative data from informative studies (i.e.,
studies having a high likelihood of TCE exposure in individual study subjects and which meet, to
a sufficient degree, the standards of epidemiologic design and analysis in a systematic review) to
contradict this observed association. See Section 4.4.2.5 and Appendix C for additional
discussion of the kidney cancer meta-analysis.
The human evidence of carcinogenicity from epidemiologic studies of TCE exposure is
compelling for non-Hodgkin lymphoma (NHL) but less convincing than for kidney cancer. High
quality studies generally reported excess relative risk estimates, with statistically significant
increases in three studies with overall TCE exposure, and a statistically significant increase in the
high TCE exposure group and statistically significant trend in a fourth study (see
Section 4.6.1.2). The consistency of the association between TCE exposure and NHL is further
supported by the results of meta-analyses (see Section 4.6.1.2.2 and Appendix C). A statistically
This document is a draft for review purposes only and does not constitute Agency policy.
12 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
significant summary relative risk estimate was observed for overall TCE exposure (RRm: 1.23
[95% CI: 1.07, 1.42]; n=\l studies), and, as with kidney cancer, the summary relative risk
estimate was greater for the highest TCE exposure groups (RRm: 1.43 [95% CI: 1.13, 1.82]; n =
13 studies) than for overall TCE exposure. Sensitivity analyses indicated that these results and
their statistical significance were not overly influenced by any single study or choice of
individual (study-specific) risk estimates, and in all of the influence and sensitivity analyses, the
RRm estimate was statistically significantly increased. Some heterogeneity was observed,
particularly between cohort and case-control studies, but it was not statistically significant. In
addition, there was some evidence of potential publication bias. Thus, while the evidence is
strong for NHL, issues of study heterogeneity, potential publication bias, and weaker exposure-
response results contribute greater uncertainty.
The evidence is more limited for liver and biliary tract cancer mainly because only cohort studies
are available and most of these studies have small numbers of cases due the comparative rarity of
liver and biliary tract cancer. While most high quality studies reported excess relative risk
estimates, they were generally based on small numbers of cases or deaths, with the result of wide
confidence intervals on the estimates. The low number of liver cancer cases in the available
studies made assessing exposure-response relationships difficult. See Section 4.5.2 for
additional discussion of the human epidemiologic data on TCE exposure and liver cancer. A
consistency of the association between TCE exposure and liver cancer is supported by the results
of meta-analyses (see Section 4.5.2 and Appendix C). These meta-analyses found a statistically
significant increased summary relative risk estimate for liver and biliary tract cancer of 1.29
(95%) CI: 1.07, 1.56; n = 9 studies) with overall TCE exposure; but the meta-analyses using only
the highest exposure groups yielded a lower, and nonstatistically significant, summary estimate
for primary liver cancer (1.28 [95%> CI: 0.93, 1.77], n = 8 studies). Although there was no
evidence of heterogeneity or publication bias and the summary estimates were fairly insensitive
to the use of alternative relative risk estimates, the statistical significance of the summary
estimates depends heavily on the one large study by Raaschou-Nielsen et al. (2003). There were
fewer adequate studies with high exposure potential available for meta-analysis of liver cancer
(9 versus 17 for NHL and 15 for kidney), leading to lower statistical power, even with pooling.
Thus, while there is epidemiologic evidence of an association between TCE exposure and liver
cancer, the much more limited database, both in terms of number of available studies and
number of cases upon which the studies are based, contributes to greater uncertainty as compared
to the evidence for kidney cancer or lymphoma.
In addition to the body of evidence pertaining to kidney cancer, NHL, and liver cancer, the
available epidemiologic studies also provide more limited evidence of an association between
This document is a draft for review purposes only and does not constitute Agency policy.
13 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
TCE exposure and other types of cancer, including bladder, esophageal, prostate, cervical, breast,
and childhood leukemia. Differences between these sets of data and the data for kidney cancer,
NHL, and liver cancer are observations from fewer numbers of studies, a mixed pattern of
observed risk estimates, and the general absence of exposure-response data from the studies
using a quantitative TCE-specific exposure measure.
There are several other lines of supporting evidence for TCE carcinogenicity in humans by all
routes of exposure. First, multiple chronic bioassays in rats and mice have reported increased
incidences of tumors with TCE treatment via inhalation and oral gavage, including tumors in the
kidney, liver, and lymphoid tissues-target tissues of TCE carcinogenicity also seen in
epidemiological studies. Of particular note is the site-concordant finding of low, but biologically
and sometimes statistically significant, increases in the incidence of kidney tumors in multiple
strains of rats treated with TCE by either inhalation or corn oil gavage (see Section 4.4.5). The
increased incidences were only detected at the highest tested doses, and were greater in male
than female rats; although, notably, pooled incidences in females from five rat strains tested by
NTP (1988, 1990) resulted in a statistically significant trend. Although these studies have shown
limited increases in kidney tumors, and several individual studies have a number of limitations,
given the rarity of these tumors as assessed by historical controls and the repeatability of this
result across studies and strains, these are considered biologically significant. Therefore, while
individual studies provide only suggestive evidence of renal carcinogenicity, the database as a
whole supports the conclusion that TCE is a kidney carcinogen in rats, with males being more
sensitive than females. No other tested laboratory species (i.e., mice and hamsters) have
exhibited increased kidney tumors, with no adequate explanation for these species differences
(particularly with mice, which have been extensively tested). With respect to the liver, TCE and
its oxidative metabolites chloral hydrate (CH), TCA, and DCA are clearly carcinogenic in mice,
with strain and sex differences in potency that appear to parallel, qualitatively, differences in
background tumor incidence. Data in other laboratory animal species are limited; thus, except
for DCA which is carcinogenic in rats, inadequate evidence exists to evaluate the
hepatocarcinogenicity of these compounds in rats or hamsters. However, to the extent that there
is hepatocarcinogenic potential in rats, TCE is clearly less potent in the strains tested in this
species than in B6C3F1 and Swiss mice. See Section 4.5.5 for additional discussion of
laboratory animal data on TCE-induced liver tumors. Additionally, there is more limited
evidence for TCE-induced lymphatic cancers in rats and mice, lung tumors in mice, and
testicular tumors in rats. With respect to the lymphatic cancers, two studies in mice reported
increased incidences of lymphomas in females of two different strains, and two studies in rats
reported leukemias in males of one strain and females of another. However, these tumors had
This document is a draft for review purposes only and does not constitute Agency policy.
14 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
relatively modest increases in incidence with treatment, and were not reported to be increased in
other studies. See Section 4.6.2.4 for additional discussion of laboratory animal data on
TCE-induced lymphatic tumors. With respect to lung tumors, rodent bioassays have
demonstrated a statistically significant increase in pulmonary tumors in mice following chronic
inhalation exposure to TCE, and nonstatistically significant increases in mice exposed orally; but
pulmonary tumors were not reported in other species tested (i.e., rats and hamsters) (see
Section 4.7.2.2). Finally, increased testicular (interstitial or Leydig cell) tumors have been
observed in multiple studies of rats exposed by inhalation and gavage, although in some cases
high (> 75%) control rates of testicular tumors in rats limited the ability to detect a treatment
effect. See Section 4.8.2.2 for additional discussion of laboratory animal data on TCE-induced
tumors of the reproductive system. Overall, TCE is clearly carcinogenic in rats and mice. The
apparent lack of site concordance across laboratory animal studies may be due to limitations in
design or conduct in a number of rat bioassays and/or genuine interspecies differences in
qualitative or quantitative sensitivity (i.e., potency). Nonetheless, these studies have shown
carcinogenic effects across different strains, sexes, and routes of exposure, and site-concordance
is not necessarily expected for carcinogens. Of greater import is the finding that there is site-
concordance between the main cancers observed in TCE-exposed humans and those observed in
rodent studies—in particular, cancers of the kidney, liver, and lymphoid tissues.
A second line of supporting evidence for TCE carcinogenicity in humans consists of
toxicokinetic data indicating that TCE is well absorbed by all routes of exposure, and that TCE
absorption, distribution, metabolism, and excretion are qualitatively similar in humans and
rodents. As summarized above, there is evidence that TCE is systemically available, distributes
to organs and tissues, and undergoes systemic metabolism from all routes of exposure.
Therefore, although the strongest evidence from epidemiologic studies largely involves
inhalation exposures, the evidence supports TCE carcinogenicity being applicable to all routes of
exposure. In addition, there is no evidence of major qualitative differences across species in
TCE absorption, distribution, metabolism, and excretion. Extensive in vivo and in vitro data
show that mice, rats, and humans all metabolize TCE via two primary pathways: oxidation by
CYPs and conjugation with glutathione via GSTs. Several metabolites and excretion products
from both pathways have been detected in blood and urine from exposed humans as well as from
at least one rodent species. In addition, the subsequent distribution, metabolism, and excretion of
TCE metabolites are qualitatively similar among species. Therefore, humans possess the
metabolic pathways that produce the TCE metabolites thought to be involved in the induction of
rat kidney and mouse liver tumors, and internal target tissues of both humans and rodents
experience a similar mix of TCE and metabolites. See Sections 3.1-3.4 for additional discussion
This document is a draft for review purposes only and does not constitute Agency policy.
15 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
of TCE toxicokinetics. Quantitative interspecies differences in toxicokinetics do exist, and are
addressed through PBPK modeling (see Section 3.5 and Appendix A). Importantly, these
quantitative differences affect only interspecies extrapolations of carcinogenic potency, and do
not affect inferences as to the carcinogenic hazard for TCE.
Finally, available mechanistic data do not suggest a lack of human carcinogenic hazard from
TCE exposure. In particular, these data do not suggest qualitative differences between humans
and test animals that would preclude any of the hypothesized key events in the carcinogenic
MOA in rodents from occurring in humans. For the kidney, the predominance of positive
genotoxicity data in the database of available studies of TCE metabolites derived from GSH
conjugation (in particular DCVC), together with toxicokinetic data consistent with their systemic
delivery to and in situ formation in the kidney, supports the conclusion that a mutagenic MOA is
operative in TCE-induced kidney tumors. While supporting the biological plausibility of this
hypothesized MOA, available data on the von Hippel-Lindau (VHL) gene in humans or
transgenic animals do not conclusively elucidate the role of VHL mutation in TCE-induced renal
carcinogenesis. Cytotoxicity and compensatory cell proliferation, similarly presumed to be
mediated through metabolites formed after GSH-conjugation of TCE, have also been suggested
to play a role in the MOA for renal carcinogenesis, as high incidences of nephrotoxicity have
been observed in animals at doses that induce kidney tumors. Human studies have reported
markers for nephrotoxicity at current occupational exposures, although data are lacking at lower
exposures. Nephrotoxicity is observed in both mice and rats, in some cases with nearly
100% incidence in all dose groups, but kidney tumors are only observed at low incidences in rats
at the highest tested doses. Therefore, nephrotoxicity alone appears to be insufficient, or at least
not rate-limiting, for rodent renal carcinogenesis, since maximal levels of toxicity are reached
before the onset of tumors. In addition, nephrotoxicity has not been shown to be necessary for
kidney tumor induction by TCE in rodents. In particular, there is a lack of experimental support
for causal links, such as compensatory cellular proliferation or clonal expansion of initiated cells,
between nephrotoxicity and kidney tumors induced by TCE. Furthermore, it is not clear if
nephrotoxicity is one of several key events in a MOA, if it is a marker for an "upstream" key
event (such as oxidative stress) that may contribute independently to both nephrotoxicity and
renal carcinogenesis, or if it is incidental to kidney tumor induction. Moreover, while
toxicokinetic differences in the GSH conjugation pathway along with their uncertainty are
addressed through PBPK modeling, no data suggest that any of the proposed key events for
TCE-induced kidney tumors in rats are precluded in humans. See Section 4.4.7 for additional
discussion of the MOA for TCE-induced kidney tumors. Therefore, TCE-induced rat kidney
This document is a draft for review purposes only and does not constitute Agency policy.
16 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
tumors provide additional support for the convincing human evidence of TCE-induced kidney
cancer, with mechanistic data supportive of a mutagenic MOA.
With respect to other tumor sites, data are insufficient to conclude that any of the other
hypothesized MO As are operant. In the liver, a mutagenic MOA mediated by CH, which has
evidence for genotoxic effects, or some other oxidative metabolite of TCE cannot be ruled out,
but data are insufficient to conclude it is operant. A second MOA hypothesis for TCE-induced
liver tumors involves activation of the peroxisome proliferator activated receptor alpha (PPARa)
receptor. Clearly, in vivo administration of TCE leads to activation of PPARa in rodents and
likely does so in humans as well. However, the evidence as a whole does not support the view
that PPARa is the sole operant MOA mediating TCE hepatocarcinogenesis. Rather, there is
evidential support for multiple TCE metabolites and multiple toxicity pathways contributing to
TCE-induced liver tumors. Furthermore, recent experiments have demonstrated that PPARa
activation and the sequence of key events in the hypothesized MOA are not sufficient to induce
hepatocarcinogenesis (Yang et al., 2007). Moreover, the demonstration that the PPARa agonist
di(2-ethylhexyl) phthalate induces tumors in PPARa-null mice supports the view that the events
comprising the hypothesized PPARa activation MOA are not necessary for liver tumor induction
in mice by this PPARa agonist (Ito et al., 2007). See Section 4.5.7 for additional discussion of
the MOA for TCE-induced liver tumors. For mouse lung tumors, as with the liver, a mutagenic
MOA involving CH has also been hypothesized, but there are insufficient data to conclude that it
is operant. A second MOA hypothesis for mouse lung tumors has been posited involving other
effects of oxidative metabolites including cytotoxicity and regenerative cell proliferation, but
experimental support remains limited, with no data on proposed key events in experiments of
duration two weeks or longer. See Section 4.7.4 for additional discussion of the MOA for
TCE-induced lung tumors. A MOA subsequent to in situ oxidative metabolism, whether
involving mutagenicity, cytotoxicity, or other key events, may also be relevant to other tissues
where TCE would undergo CYP metabolism. For instance, CYP2E1, oxidative metabolites, and
protein adducts have been reported in the testes of rats exposed to TCE, and, in some rat
bioassays, TCE exposure increased the incidence of rat testicular tumors. However, inadequate
data exist to adequately define a MOA hypothesis for this tumor site (see Section 4.8.2.3 for
additional discussion of the MOA for TCE-induced testicular tumors).
6.1.5. Susceptibility (see Sections 4.10 and 4.11.3)
There is some evidence that certain populations may be more susceptible to exposure to TCE.
Factors affecting susceptibility examined include lifestage, gender, genetic polymorphisms,
This document is a draft for review purposes only and does not constitute Agency policy.
17 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
race/ethnicity, preexisting health status, and lifestyle factors and nutrition status. Factors that
affect early lifestage susceptibility include exposures such as transplacental transfer and breast
milk ingestion, early lifestage-specific toxicokinetics, and differential outcomes in early
lifestages such as developmental cardiac defects (see Section 4.10.1). Because the weight of
evidence supports a mutagenic MOA being operative for TCE carcinogenicity in the kidney (see
Section 4.4.7), and there is an absence of chemical-specific data to evaluate differences in
carcinogenic susceptibility, early-life susceptibility should be assumed and the age-dependent
adjustment factors (ADAFs) should be applied, in accordance with the Supplemental Guidance
(see summary below in Section 6.2.2.5). Fewer data are available on later lifestages, although
there is suggestive evidence to indicate that older adults may experience increased adverse
effects than younger adults due to greater tissue distribution of TCE. In general, more studies
specifically designed to evaluate effects in early and later lifestages are needed in order to more
fully characterize potential life stage-related TCE toxicity. Gender-specific (see
Section 4.10.2.1) differences also exist in toxicokinetics (e.g., cardiac outputs, percent body fat,
expression of metabolizing enzymes) and susceptibility to toxic endpoints (e.g., gender-specific
effects on the reproductive system, gender differences in baseline risks to endpoints such as
scleroderma or liver cancer). Genetic variation (see Section 4.10.2.2) likely has an effect on the
toxicokinetics of TCE. Increased CYP2E1 activity and GST polymorphisms may influence
susceptibility of TCE due to effects on production of toxic metabolites or may play a role in
variability in toxic response. Differences in genetic polymorphisms related to the metabolism of
TCE have also been observed among various race/ethnic groups (see Section 4.10.2.3).
Preexisting diminished health status (see Section 4.10.2.4) may alter the response to TCE
exposure. Individuals with increased body mass may have an altered toxicokinetic response due
to the increased uptake of TCE into fat. Other conditions that may alter the response to TCE
exposure include diabetes and hypertension, and lifestyle and nutrition factors (see
Section 4.10.2.5) such as alcohol consumption, tobacco smoking, nutritional status, physical
activity, and socioeconomic status. Alcohol intake has been associated with inhibition of TCE
metabolism in both humans and experimental animals. In addition, such conditions have been
associated with increased baseline risks for health effects also associated with TCE, such as
kidney cancer and liver cancer. However, the interaction between TCE and known risk factors
for human diseases is not known, and further evaluation of the effects due to these factors is
needed.
In sum, there is some evidence that certain populations may be more susceptible to exposure to
TCE. Factors affecting susceptibility examined include lifestage, gender, genetic
polymorphisms, race/ethnicity, preexisting health status, and lifestyle factors and nutrition status.
This document is a draft for review purposes only and does not constitute Agency policy.
18 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
However, except in the case of toxicokinetic variability characterized using the PBPK model
described in Section 3.5, there are inadequate chemical-specific data to quantify the degree of
differential susceptibility due to such factors.
6.2. DOSE-RESPONSE ASSESSMENT
This section summarizes the major conclusions of the dose-response analysis for TCE noncancer
effects and carcinogenicity, with more detailed discussions in Section 5.
6.2.1. Noncancer Effects (see Section 5.1)
6.2.1.1.1. Background and Methods
As summarized above, based on the available human epidemiologic data and experimental and
mechanistic studies, it is concluded that TCE poses a potential human health hazard for
noncancer toxicity to the central nervous system, the kidney, the liver, the immune system, the
male reproductive system, and the developing fetus. The evidence is more limited for TCE
toxicity to the respiratory tract and female reproductive system.
Dose-response analysis for a noncancer endpoint generally involves two steps: (1) the
determination of a point of departure (POD) derived from a benchmark dose (BMD),6l a
no-observed-adverse-effect level (NOAEL), or a lowest-observed-adverse-effect level (LOAEL);
and (2) adjustment of the POD by endpoint/study-specific "uncertainty factors" (UFs),
accounting for adjustments and uncertainties in the extrapolation from the study conditions to
conditions of human exposure.
Because of the large number of noncancer health effects associated with TCE exposure and the
large number of studies reporting on these effects, in contrast to toxicological reviews for
chemicals with smaller databases of studies, a formal, quantitative screening process (see
Section 5.1) was used to reduce the number of endpoints and studies to those that would best
inform the selection of the critical effects for the inhalation reference concentration (RfC) and
oral reference dose (RfD).62 As described in Section 5.1, for all studies described in Section 4
61 More precisely, it is the benchmark dose lower bound (BMDL), i.e., the (one-sided) 95% lower confidence
bound on the dose corresponding to the benchmark response (BMR) for the effect, that is used as the POD.
62 In EPA noncancer health assessments, the RfC [RfD] is an estimate (with uncertainty spanning perhaps an order
of magnitude) of a continuous inhalation [daily oral] exposure to the human population (including sensitive
subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime. It can be derived
This document is a draft for review purposes only and does not constitute Agency policy.
19 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
which report adverse noncancer health effects and provided quantitative dose-response data,
PODs on the basis of applied dose, adjusted by endpoint/study-specific UFs, were used to
develop candidate RfCs (cRfCs) and candidate RfDs (cRfDs) intended to be protective for each
endpoint individually. Candidate critical effects-those with the lowest cRfCs and cRfDs taking
into account the confidence in each estimate-were selected within each of the following health
effect domains: (1) neurological, (2) systemic/organ system; (3) immunological;
(4) reproductive; and (5) developmental. For each of these candidate critical effects, the PBPK
model developed in Section 3.5 was used for interspecies, intraspecies, and route-to-route
extrapolation on the basis of internal dose to develop PBPK model-based PODs. Plausible
internal dose-metrics were selected based on what is understood about the role of different TCE
metabolites in toxicity and the MOA for toxicity. These PODs were then adjusted by
endpoint/study-specific UFs, taking into account the use of the PBPK model, to develop PBPK
model-based candidate RfCs (p-cRfCs) and candidate RfDs (p-cRfDs). The most sensitive
cRfCs, p-cRfCs, cRfDs, and p-cRfDs were then evaluated, taking into account the confidence in
each estimate, to arrive at overall candidate RfCs and RfDs for each health effect type. Then, the
RfC and RfD for TCE were selected so as to be protective of the most sensitive effects. In
contrast to the approach used in most assessments, in which the RfC and RfD are each based on
a single critical effect, the final RfC and RfD for TCE were based on multiple critical effects that
resulted in very similar candidate RfC and RfD values at the low end of the full range of values.
This approach was taken here because it provides robust estimates of the RfC and RfD and
because it highlights the multiple effects that are all yielding very similar candidate values.
6.2.1.1.2. Uncertainties and Application of Uncertainty Factors (UFs) (see Section 5.1.1 and
5.1.4)
An underlying assumption in deriving reference values for noncancer effects is that the dose-
response relationship for these effects has a threshold. Thus, a fundamental uncertainty is the
validity of that assumption. For some effects, in particular effects on very sensitive processes
(e.g., developmental processes) or effects for which there is a nontrivial background level and
even small exposures may contribute to background disease processes in more susceptible
people, a practical threshold (i.e., a threshold within the range of environmental exposure levels
of regulatory concern) may not exist.
from a NOAEL, LOAEL, or benchmark concentration [dose], with uncertainty factors generally applied to reflect
limitations of the data used.
This document is a draft for review purposes only and does not constitute Agency policy.
20 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Nonetheless, under the assumption of a threshold, the desired exposure level to have as a
reference value is the maximum level at which there is no appreciable risk for an adverse effect
in sensitive subgroups (of humans). However, because it is not possible to know what this level
is, "uncertainty factors" are used to attempt to address quantitatively various aspects, depending
on the data set, of qualitative uncertainty.
First there is uncertainty about the "point of departure" for the application of UFs. Conceptually,
the POD should represent the maximum exposure level at which there is no appreciable risk for
an adverse effect in the study population under study conditions (i.e., the threshold in the dose-
response relationship). Then, the application of the relevant UFs is intended to convey that
exposure level to the corresponding exposure level for sensitive human subgroups exposed
continuously for a lifetime. In fact, it is again not possible to know that exposure level even for a
laboratory study because of experimental limitations (e.g., the power to detect an effect, dose
spacing, measurement errors, etc.), and crude approximations like the NOAEL or a BMDL are
used. If a LOAEL is used as the POD, the LOAEL-to-NOAEL UF is applied as an adjustment
factor to better approximate the desired exposure level (threshold), although the necessary extent
of adjustment is unknown. The standard value for the LOAEL-to-NOAEL UF is 10, although
sometimes a value of 3 is used if the effect is considered minimally adverse at the response level
observed at the LOAEL or even one if the effect is an early marker for an adverse effect. For
one POD in this assessment, a value of 30 was used for the LOAEL-to-NOAEL UF because the
incidence rate for the adverse effect was > 90% at the LOAEL.
If a BMDL is used as the POD, there are uncertainties regarding the appropriate dose-response
model to apply to the data, but these should be minimal if the modeling is in the observable
range of the data. There are also uncertainties about what BMR to use to best approximate the
desired exposure level (threshold, see above). For continuous endpoints, in particular, it is often
difficult to identify the level of change that constitutes the "cut-point" for an adverse effect.
Sometimes, to better approximate the desired exposure level, a BMR somewhat below the
observable range of the data is selected. In such cases, the model uncertainty is increased, but
this is a trade-off to reduce the uncertainty about the POD not being a good approximation for
the desired exposure level.
For each of these types of PODs, there are additional uncertainties pertaining to adjustments to
the administered exposures (doses). Typically, administered exposures (doses) are converted to
equivalent continuous exposures (daily doses) over the study exposure period under the
assumption that the effects are related to concentration x time, independent of the daily (or
weekly) exposure regimen (i.e., a daily exposure of 6 hours to 4 ppm is considered equivalent to
24 hours of exposure to 1 ppm). However, the validity of this assumption is generally unknown,
This document is a draft for review purposes only and does not constitute Agency policy.
21 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
and, if there are dose-rate effects, the assumption of concentration times time (C x t) equivalence
would tend to bias the POD downwards. Where there is evidence that administered exposure
better correlates to the effect than equivalent continuous exposure averaged over the study
exposure period (e.g., visual effects), administered exposure was not adjusted. For the PBPK
analyses in this assessment, the actual administered exposures are taken into account in the
PBPK modeling, and equivalent daily values (averaged over the study exposure period) for the
dose-metrics are obtained (see above, Section 5.1.3.2). Additional uncertainties about the
PBPK-based estimates include uncertainties about the appropriate dose-metric for each effect,
although for some effects there was better information about relevant dose-metrics than for
others, and uncertainties in the PBPK model predictions for the dose-metrics in humans,
particularly for GSH conjugation (see Section 5.1.3.1).
There is also uncertainty about the other UFs. The human variability UF is to some extent an
adjustment factor because for more sensitive people, the dose-response relationship shifts to
lower exposures. But there is uncertainty about the extent of the adjustment required, i.e., about
the distribution of human susceptibility. Therefore, in the absence of data on a susceptible
population(s) or on the distribution of susceptibility in the general population, an UF of 10 is
generally used, which breaks down (approximately) to a factor of 3 for pharmacokinetic
variability and a factor of 3 for pharmacodynamic variability. This standard value was used for
all the PODs based on applied dose in this assessment with the exception of the PODs for a few
immunological effects that were based on data from a sensitive (autoimmune-prone) mouse
strain. For those PODs, an UF of 3 (reflecting pharmacokinetics only) was used for human
variability. The PBPK analyses in this assessment attempt to account for the pharmacokinetic
portion of human variability using human data on pharmacokinetic variability. For PBPK
model-based candidate reference values, the pharmacokinetic component of this UF was omitted.
A quantitative uncertainty analysis of the PBPK derived dose-metrics used in the assessment is
presented in Section 5.1.4.2 in Section 5. There is still uncertainty regarding the susceptible
subgroups for TCE exposure and the extent of pharmacodynamic variability.
If the data used to determine a particular POD are from laboratory animals, an interspecies
extrapolation UF is used. This UF is also to some extent an adjustment factor for the expected
scaling for toxicologically equivalent doses across species (i.e., according to body weight to the
3/4 power for oral exposures). However, there is also uncertainty about the true extent of
interspecies differences for specific noncancer effects from specific chemical exposures. For
oral exposures, the standard value for the interspecies UF is 10, which can be viewed as breaking
down (approximately) to a factor of 3 for the "adjustment" (nominally pharmacokinetics) and a
factor of 3 for the "uncertainty" (nominally pharmacodynamics). For inhalation exposures for
This document is a draft for review purposes only and does not constitute Agency policy.
22 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
systemic toxicants such as TCE, no adjustment across species is generally assumed for fixed air
concentrations (ppm equivalence), and the standard value for the interspecies UF is 3 reflects
"uncertainty" (nominally pharmacodynamics only). The PBPK analyses in this assessment
attempt to account for the "adjustment" portion of interspecies extrapolation using rodent
pharmacokinetic data to estimate internal doses for various dose-metrics. Equal doses of these
dose-metrics, appropriately scaled, are then assumed to convey equivalent risk across species.
For PBPK model-based candidate reference values, the "adjustment" component of this UF was
omitted. With respect to the "uncertainty" component, quantitative uncertainty analyses of the
PBPK-derived dose-metrics used in the assessment are presented in Section 5.1.4.2 in Section 5.
However, these only address the pharmacokinetic uncertainties in a particular dose-metric, and
there is still uncertainty regarding the true dose-metrics. Nor do the PBPK analyses address the
uncertainty in either cross-species pharmacodynamic differences (i.e., about the assumption that
equal doses of the appropriate dose-metric convey equivalent risk across species for a particular
endpoint from a specific chemical exposure) or in cross-species pharmacokinetic differences not
accounted for by the PBPK model dose-metrics (e.g., departures from the assumed interspecies
scaling of clearance of the active moiety, in the cases where only its production is estimated). A
value of 3 is typically used for the "uncertainty" about cross-species differences, and this
generally represents true uncertainty because it is usually unknown, even after adjustments have
been made to account for the expected interspecies differences, whether humans have more or
less susceptibility, and to what degree, than the laboratory species in question.
RfCs and RfDs apply to lifetime exposure, but sometimes the best (or only) available data come
from less-than-lifetime studies. Lifetime exposure can induce effects that may not be apparent or
as large in magnitude in a shorter study; consequently, a dose that elicits a specific level of
response from a lifetime exposure may be less than the dose eliciting the same level of response
from a shorter exposure period. If the effect becomes more severe with increasing exposure,
then chronic exposure would shift the dose-response relationship to lower exposures, although
the true extent of the shift is unknown. PODs based on subchronic exposure data are generally
divided by a subchronic-to-chronic UF, which has a standard value of 10. If there is evidence
suggesting that exposure for longer time periods does not increase the magnitude of an effect, a
lower value of 3 or 1 might be used. For some reproductive and developmental effects, chronic
exposure is that which covers a specific window of exposure that is relevant for eliciting the
effect, and subchronic exposure would correspond to an exposure that is notably less than the full
window of exposure.
Sometimes a database UF is also applied to address limitations or uncertainties in the database.
The overall database for TCE is quite extensive, with studies for many different types of effects,
This document is a draft for review purposes only and does not constitute Agency policy.
23 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
including 2-generation reproductive studies, as well as neurological and immunological studies.
In addition, there were sufficient data to develop a reliable PBPK model to estimate route-to-
route extrapolated doses for some candidate critical effects for which data were only available
for one route of exposure. Thus, there is a high degree of confidence that the TCE database was
sufficient to identify some sensitive endpoints, and no database UF was used in this assessment.
6.2.1.1.3. Candidate Critical Effects and Reference Values (see Sections 5.1.2 and 5.1.3)
A large number of endpoints and studies were considered within each health effect
domain. Section 5 contains a comprehensive discussion of all endpoints/studies which were
considered for developing candidate reference values (cRfCs, cRfDs, p-cRfCs, and p-cRfDs),
their PODs, and the UFs applied. The summary below reviews the selection of candidate critical
effects for each health effect domain, the confidence in the reference values, the selection of
PBPK model-based dose-metrics, and the impact of PBPK modeling on the candidate reference
values.
6.2.1.1.4. Neurological effects
Candidate reference values were developed for several neurological domains for which there was
evidence of hazard (see Tables 5-2 and 5-13). There is higher confidence in the candidate
reference values for trigeminal nerve, auditory, or psychomotor effects, but the available data
suggest that the more sensitive indicators of TCE neurotoxicity are changes in wakefulness,
regeneration of the sciatic nerve, demyelination in the hippocampus and degeneration of
dopaminergic neurons. Therefore, these more sensitive effects are considered the candidate
critical effects for neurotoxicity, albeit with more uncertainty in the corresponding candidate
reference values. Of these more sensitive effects, there is greater confidence in the changes in
wakefulness reported by Arito et al. (1994). In addition, trigeminal nerve effects are considered
a candidate critical effect because this is the only type of neurological effect for which human
data are available, and the POD for this effect is similar to that from the most sensitive rodent
study (Arito et al., 1994, for changes in wakefulness). Between the two human studies of
trigeminal nerve effects, Ruijten et al. (1991) is preferred for deriving noncancer reference
values because its exposure characterization is considered more reliable.
Because of the lack of specific data as to the metabolites involved and the MOA for the
candidate critical neurologic effects, PBPK model predictions of total metabolism (scaled by
body weight to the 3/4 power) were selected as the preferred dose-metric based on the general
This document is a draft for review purposes only and does not constitute Agency policy.
24 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
observation that TCE toxicity is associated with metabolism. The area-under-the-curve (AUC)
of TCE in blood was used as an alternative dose-metric. With these dose-metrics, the candidate
reference values derived using the PBPK model were only modestly (-threefold or less) different
than those derived on the basis of applied dose.
6.2.1.1.5. Kidney effects
Candidate reference values were developed for histopathological and weight changes in the
kidney (see Tables 5-4 and 5-14), and these are considered to be candidate critical effects for
several reasons. First, they appear to be the most sensitive indicators of toxicity that are
available for the kidney. In addition, as discussed in Sections 3.3 and 3.5, both in vitro and in
vivo pharmacokinetic data indicate substantially more production of GSH-conjugates thought to
mediate TCE kidney effects in humans relative to rats and mice. Several studies are considered
reliable for developing candidate reference values for these endpoints. For histopathological
changes, these were the only available inhalation study (the rat study of Maltoni et al., 1986), the
NTP (1988) study in rats, and the National Cancer Institute (NCI, 1976) study in mice. For
kidney weight changes, both available studies (Kjellstrand et al., 1983a; Woolhiser et al., 2006)
were chosen as candidate critical studies.
Due to the substantial evidence supporting the role of GSH conjugation metabolites in TCE-
induced nephrotoxicity, the preferred PBPK model dose-metrics for kidney effects were the
amount of DCVC bioactivated in the kidney for rat studies and the amount of GSH conjugation
(both scaled by body weight to the % power) for mouse studies (inadequate toxicokinetic data are
available in mice for predicting the amount of DCVC bioactivation). With these dose-metrics,
the candidate reference values derived using the PBPK model were 300- to 400-fold lower than
those derived on the basis of applied dose. As discussed above and in Section 3, this is due to
the available in vivo and in vitro data supporting not only substantially more GSH conjugation in
humans than in rodents, but also substantial interindividual toxicokinetic variability. Overall,
there is high confidence in the nephrotoxic hazard from TCE exposure and in the appropriateness
of the dose-metrics discussed above; however, there is substantial uncertainty in the
extrapolation of GSH conjugation from rodents to humans due to limitations in the available data
(see Section 3.3.3.2).
This document is a draft for review purposes only and does not constitute Agency policy.
25 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
6.2.1.1.6. Liver effects
Hepatomegaly appears to be the most sensitive indicator of toxicity that is available for the liver
and is therefore, considered a candidate critical effect. Several studies are considered reliable for
developing high confidence candidate reference values for this endpoint. Since they all indicated
similar sensitivity but represented different species and/or routes of exposure, they were all
considered candidate critical studies (see Tables 5-4 and 5-14).
Due to the substantial evidence supporting the role of oxidative metabolism in TCE-induced
hepatomegaly (and evidence against TCA being the sole mediator of TCE-induced hepatomegaly
(Evans et al., 2009)), the preferred PBPK model dose-metric for liver effects was the amount of
hepatic oxidative metabolism (scaled by body weight to the 3/4 power). Total (hepatic and
extrahepatic) oxidative metabolism (scaled by body weight to the 3/4 power) was used as an
alternative dose-metric. With these dose-metrics, the candidate reference values derived using
the PBPK model were only modestly (-threefold or less) different than those derived on the
basis of applied dose.
6.2.1.1.7. Immunological effects
There is high qualitative confidence for TCE immunotoxicity and moderate confidence in the
candidate reference values that can be derived from the available studies (see Tables 5-6
and 5-16). Decreased thymus weight reported at relatively low exposures in nonautoimmune-
prone mice is a clear indicator of immunotoxicity (Keil et al., 2009), and is therefore, considered
a candidate critical effect. A number of studies have also reported changes in markers of
immunotoxicity at relatively low exposures. Among markers for autoimmune effects, the more
sensitive measures of autoimmune changes in liver and spleen (Kaneko et al., 2000) and
increased anti-dsDNA and anti-ssDNA antibodies (early markers for systemic lupus
erythematosus) (Keil et al., 2009) are considered the candidate critical effects. For markers of
immunosuppression, the more sensitive measures of decreased PFC response (Woolhiser et al.,
2006), decreased stem cell bone marrow recolonization, and decreased cell-mediated response to
sRBC (both from Sanders et al., 1982b) are considered the candidate critical effects.
Developmental immunological effects are discussed below as part of the summary of
developmental effects (see Section 6.2.1.3.6).
Because of the lack of specific data as to the metabolites involved and the MOA for the
candidate critical immunologic effects, PBPK model predictions of total metabolism (scaled by
body weight to the % power) was selected as the preferred dose-metric based on the general
observation that TCE toxicity is associated with metabolism. The AUC of TCE in blood was
This document is a draft for review purposes only and does not constitute Agency policy.
26 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
used as an alternative dose-metric. With these dose-metrics, the candidate reference values
derived using the PBPK model were, with one exception, only modestly (-threefold or less)
different than those derived on the basis of applied dose. For the Woolhiser et al. (2006)
decreased PFC response, with the alternative dose-metric of AUC of TCE in blood, BMD
modeling based on internal doses changed the candidate reference value by 17-fold higher than
the cRfC based on applied dose. However, the dose-response model fit for this effect using this
metric was substantially worse than the fit using the preferred metric of total oxidative
metabolism, with which the change in candidate reference value was only 1.3-fold.
6.2.1.1.8. Reproductive effects
While there is high qualitative confidence in the male reproductive hazard posed by TCE, there
is lower confidence in the reference values that can be derived from the available studies of these
effects (see Tables 5-8 and 5-17). Relatively high PODs are derived from several studies
reporting less sensitive endpoints (George et al., 1985, 1986; Land et al., 1981), and
correspondingly higher cRfCs and cRfDs suggest that they are not likely to be critical effects.
The studies reporting more sensitive endpoints also tend to have greater uncertainty. For the
human study by Chia et al. (1996), there are uncertainties in the characterization of exposure and
the adversity of the effect measured in the study. For the Kumar et al. (2001b; 2000a; 2000b),
Forkert et al. (2002) and Kan et al. (2007) studies, the severity of the sperm and testes effects
appears to be continuing to increase with duration even at the end of the study, so it is plausible
that a lower exposure for a longer duration may elicit similar effects. For the DuTeaux et al.
(2004a) study, there is also duration- and low-dose extrapolation uncertainty due to the short
duration of the study in comparison to the time period for sperm development as well as the lack
of a NOAEL at the tested doses. Overall, even though there are limitations in the quantitative
assessment, there remains sufficient evidence to consider these to be candidate critical effects.
There is moderate confidence both in the hazard and the candidate reference values for
reproductive effects other than male reproductive effects. While there are multiple studies
suggesting decreased maternal body weight with TCE exposure, this systemic change may not be
indicative of more sensitive reproductive effects. None of the estimates developed from other
reproductive effects is particularly uncertain or unreliable. Therefore, delayed parturition
(Narotsky et al., 1995) and decreased mating (George et al., 1986), which yielded the lowest
cRfDs, were considered candidate critical effects. These effects were also included so that
candidate critical reproductive effects from oral studies would not include only that reported by
DuTeaux et al. (2004a), from which deriving the cRfD entailed a higher degree of uncertainty.
This document is a draft for review purposes only and does not constitute Agency policy.
27 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Because of the general lack of specific data as to the metabolites involved and the MOA for the
candidate critical reproductive effects, PBPK model predictions of total metabolism (scaled by
body weight to the 3/4 power) was selected as the preferred dose-metric based on the general
observation that TCE toxicity is associated with metabolism. The AUC of TCE in blood was
used as an alternative dose-metric. The only exception to this was for the DuTeaux et al. (2004a)
study, which suggested that local oxidative metabolism of TCE in the male reproductive tract
was involved in the effects reported. Therefore, in this case, AUC of TCE in blood was
considered the preferred dose-metric, while total oxidative metabolism (scaled by body weight to
the 3/4 power) was considered the alternative metric. With these dose-metrics, the candidate
reference values derived using the PBPK model were only modestly (~3.5-fold or less) different
than those derived on the basis of applied dose.
6.2.1.1.9. Developmental effects
There is moderate-to-high confidence both in the hazard and the candidate reference values for
developmental effects of TCE (see Tables 5-10 and 5-18). It is also noteworthy that the PODs
for the more sensitive developmental effects were similar to or, in most cases, lower than the
PODs for the more sensitive reproductive effects, suggesting that developmental effects are not a
result of paternal or maternal toxicity. Among inhalation studies, candidate reference values
were only developed for effects in rats reported in Healy et al. (1982), of resorptions, decreased
fetal weight, and delayed skeletal ossification. These were all considered candidate critical
developmental effects. Because resorptions were also reported in oral studies, the most sensitive
(rat) oral study for this effect (and most reliable for dose-response analysis) of Narotsky et al.
(1995) was also selected as a candidate critical study. The confidence in the oral studies and
candidate reference values developed for more sensitive endpoints is more moderate, but still
sufficient for consideration as candidate critical effects. The most sensitive endpoints by far are
the increased fetal heart malformations in rats reported by Johnson et al. (2003) and the
developmental immunotoxicity in mice reported by Peden-Adams et al. (2006), and these are
both considered candidate critical effects. Neurodevelopmental effects are a distinct type among
developmental effects. Thus, the next most sensitive endpoints of decreased rearing
postexposure in mice (Fredriksson et al., 1993), increased exploration postexposure in rats
(Taylor et al., 1985) and decreased myelination in the hippocampus of rats (Isaacson and Taylor,
1989) are also considered candidate critical effects.
Because of the general lack of specific data as to the metabolites involved and the MOA for the
candidate critical reproductive effects, PBPK model predictions of total metabolism (scaled by
This document is a draft for review purposes only and does not constitute Agency policy.
28 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
body weight to the 3/4 power) was selected as the preferred dose-metric based on the general
observation that TCE toxicity is associated with metabolism. The AUC of TCE in blood was
used as an alternative dose-metric. The only exception to this was for the Johnson et al. (2003)
study, which suggested that oxidative metabolites were involved in the effects reported based on
similar effects being reported from TCA and DCA exposure. Therefore, in this case, total
oxidative metabolism (scaled by body weight to the 3/4 power) was considered the preferred dose-
metric, while AUC of TCE in blood was considered the alternative metric. With these dose-
metrics, the candidate reference values derived using the PBPK model were, with one exception,
only modestly (-threefold or less) different than those derived on the basis of applied dose. For
resorptions reported by Narotsky et al. (1995), BMD modeling based on internal doses changed
the candidate reference value by seven to eightfold larger than the corresponding cRfD based on
applied dose. However, there is substantial uncertainty in the low-dose curvature of the
dose-response curve for modeling both with applied and internal dose, so the BMD remains
somewhat uncertain for this endpoint/study. Finally, for two studies (Isaacson and Taylor, 1989;
Peden-Adams et al., 2006), PBPK modeling of internal doses was not performed due to the
inability to model the complicated exposure pattern (in utero, followed by lactational transfer,
followed by drinking water postweaning).
6.2.1.1.10. Summary of most sensitive candidate reference values
As shown in Section 5.1.3 and 5.1.5, the most sensitive candidate reference values are for the
developmental effect of heart malformations in rats (candidate RfC of 0.0004 ppm and candidate
RfD of 0.0005 mg/kg-day), developmental immunotoxicity in mice exposed pre and postnatally
(candidate RfD of 0.0004 mg/kg-day), immunological effects in mice (lowest candidate RfCs of
0.0003-0.003 ppm and lowest candidate RfDs of 0.0005-0.005 mg/kg-day), and kidney effects
in rats and mice (candidate RfCs of 0.0006-0.002 ppm and candidate RfDs of
0.0003-0.001 mg/kg-day). The most sensitive candidate reference values also generally have
low composite uncertainty factors (with the exception of some mouse immunological and kidney
effects), so they are expected to be reflective of the most sensitive effects as well. Thus, the
most sensitive candidate references values for multiple effects span about an order of magnitude
for both inhalation (0.0003-0.003 ppm [0.002-0.02 mg/m3]) and oral (0.0004-0.005 mg/kg-day)
exposures. The most sensitive candidate references values for neurological and reproductive
effects are about an order of magnitude higher (lowest candidate RfCs of 0.007-0.02 ppm
[0.04-0.1 mg/m3] and lowest candidate RfDs of 0.009-0.02 mg/kg-day). Lastly, the liver effects
This document is a draft for review purposes only and does not constitute Agency policy.
29 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
have candidate reference values that are another two orders of magnitude higher (candidate RfCs
of 1-2 ppm [6-10 mg/m3] and candidate RfDs of 0.9-2 mg/kg-day).
6.2.1.1.11. Noncancer Reference Values (see Section 5.1.5)
Reference concentration
The goal is to select an overall RfC that is well supported by the available data (i.e., without
excessive uncertainty given the extensive database) and protective for all the candidate critical
effects, recognizing that individual candidate RfC values are by nature somewhat imprecise. As
discussed in Section 5.1 in Section 5, the lowest candidate RfC values within each health effect
category span a 3,000-fold range from 0.0003-0.9 ppm (see Table 5-26). One approach to
selecting a RfC would be to select the lowest calculated value of 0.0003 ppm for decreased
thymus weight in mice. However, three candidate RfCs (cRfCs and p-cRfCs) are in the
relatively narrow range of 0.0003-0.0006 ppm at the low end of the overall range (see
Table 5-24). Given the somewhat imprecise nature of the individual candidate RfC values, and
the fact that multiple effects/studies lead to similar candidate RfC values, the approach taken in
this assessment is to select a RfC supported by multiple effects/studies. The advantages of this
approach, which is only possible when there is a relatively large database of studies/effects and
when multiple candidate values happen to fall within a narrow range at the low end of the overall
range, are that it leads to a more robust RfC (less sensitive to limitations of individual studies)
and that it provides the important characterization that the RfC exposure level is similar for
multiple noncancer effects rather than being based on a sole explicit critical effect.
Therefore, two critical and one supporting studies/effects were chosen as the basis of the RfC for
TCE noncancer effects (see Tables 5-28 and 5-29). These lowest candidate RfCs, ranging from
0.0003-0.0006 ppm for developmental, kidney, and immunologic effects, are values derived
from route-to-route extrapolation using the PBPK model. The lowest candidate RfC estimate
from an inhalation study is 0.001 ppm for kidney effects, which is higher than the route-to-route
extrapolated candidate RfC estimate from the most sensitive oral study. For all of the candidate
RfCs, the PBPK model was used for inter and intraspecies extrapolation, based on the preferred
dose-metric for each endpoint. There is moderate-to-high confidence in the lowest candidate
RfC for immunological effects (see Section 5.1.2.5), and moderate confidence in the lowest
candidate RfC for developmental effects (heart malformations) (see Section 5.1.2.8); these are
considered the critical effects for deriving the RfC. For kidney effects (toxic nephropathy), there
is high confidence in the nephrotoxic hazard from TCE exposure and in the appropriateness of
This document is a draft for review purposes only and does not constitute Agency policy.
3 0 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
the selected dose-metric; however, as discussed in Section 3.3.3.2, there remains substantial
uncertainty in the extrapolation of GSH conjugation from rodents to humans due to limitations in
the available data, and thus toxic nephropathy is considered a supporting effect.
"3
As a whole, the estimates support a RfC of 0.0004 ppm (0.4 ppb or 2 (j,g/m ). This estimate
essentially reflects the midpoint between the similar candidate RfC estimates for the two critical
effects (0.00033 ppm for decreased thymus weight in mice and 0.00037 ppm for heart
malformations in rats), rounded to one significant figure. This estimate is also within a factor of
2 of the candidate RfC estimate of 0.00006 ppm for the supporting effect of toxic nephropathy in
rats. Thus, this assessment does not rely on a single estimate alone; rather, each estimate is
supported by estimates of similar magnitude from other effects. In other words, there is robust
support for an RfC of 0.0004 ppm provided by estimates for multiple effects from multiple
studies. The estimates are based on PBPK model-based estimates of internal dose for
interspecies, intraspecies, and route-to-route extrapolation, and there is sufficient confidence in
the PBPK model and support from mechanistic data for one of the dose-metrics (total oxidative
metabolism for the heart malformations). There is high confidence that bioactivation of DCVC
and total GSH metabolism would be appropriate dose-metrics for toxic nephropathy, but there is
substantial uncertainty in the PBPK model predictions for these dose-metrics in humans (see
Section 5.1.3.1). Note that there is some human evidence of developmental heart defects from
TCE exposure in community studies (see Section 4.8.3.1.1) and of kidney toxicity in TCE-
exposed workers (see Section 4.4.1).
-3
In summary, the RfC is 0.0004 ppm (0.4 ppb or 2 (J,g/m ) based on route-to-route extrapolated
results from oral studies for the critical effects of heart malformations (rats) and immunotoxicity
(mice). This RfC value is further supported by route-to-route extrapolated results from an oral
study of toxic nephropathy (rats).
6.2.1.1.12. Reference dose
As with the RfC determination above, the goal is to select an overall RfD that is well supported
by the available data (i.e., without excessive uncertainty given the extensive database) and
protective for all the candidate critical effects, recognizing that individual candidate RfD values
are by nature somewhat imprecise. As discussed in Section 5.1 in Section 5, the lowest
candidate RfD values (cRfDs and p-cRfDs) within each health effect category span a nearly
3,000-fold range from 0.0003-0.8 mg/kg-day (see Table 5-26). However, multiple candidate
RfDs are in the relatively narrow range of 0.0003-0.0008 mg/kg-day at the low end of the
overall range. Given the somewhat imprecise nature of the individual candidate RfD values, and
This document is a draft for review purposes only and does not constitute Agency policy.
31 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
the fact that multiple effects/studies lead to similar candidate RfD values, the approach taken in
this assessment is to select a RfD supported by multiple effects/studies. The advantages of this
approach, which is only possible when there is a relatively large database of studies/effects and
when multiple candidate values happen to fall within a narrow range at the low end of the overall
range, are that it leads to a more robust RfD (less sensitive to limitations of individual studies)
and that it provides the important characterization that the RfD exposure level is similar for
multiple noncancer effects rather than being based on a sole explicit critical effect.
Therefore, three critical and two supporting studies/effects were chosen as the basis of the RfD
for TCE noncancer effects (see Tables 5-30 and 5-31). All but one of the lowest candidate RfD
values—0.0008 mg/kg-day for increased kidney weight in rats, 0.0005 mg/kg-day for both heart
malformations in rats and decreased thymus weights in mice, and 0.0003 mg/kg-day for
increased toxic nephropathy in rats—are derived using the PBPK model for inter and
intraspecies extrapolation, based on the preferred dose-metric for each endpoint, and the latter
value is derived also using the PBPK model for route-to-route extrapolation from an inhalation
study. The other of these lowest candidate RfDs—0.0004 mg/kg-day for developmental
immunotoxicity (decreased PFC response and increased delayed-type hypersensitivity) in
mice—is based on applied dose. There is moderate-to-high confidence in the candidate RfDs for
decreased thymus weights (see Section 5.1.2.5) and developmental immunological effects, and
moderate confidence in that for heart malformations (see Section 5.1.2.8); these are considered
the critical effects for deriving the RfC. For kidney effects, there is high confidence in the
nephrotoxic hazard from TCE exposure and in the appropriateness of the selected dose-metric;
however, as discussed in Section 3.3.3.2, there remains substantial uncertainty in the
extrapolation of GSH conjugation from rodents to humans due to limitations in the available
data, and thus these effects are considered supporting effects.
As a whole, the estimates support a RfD of 0.0005 mg/kg-day. This estimate is within 20% of
the estimates for the critical effects—0.0004 mg/kg-day for developmental immunotoxicity
(decreased PFC and increased delayed-type hypersensitivity) in mice and 0.0005 mg/kg-day for
both heart malformations in rats and decreased thymus weights in mice. This estimate is also
within approximately a factor of 2 of the supporting effect estimates of 0.0003 mg/kg-day for
toxic nephropathy in rats and 0.0008 mg/kg-day for increased kidney weight in rats. Thus, this
assessment does not rely on any single estimate alone; rather, each estimate is supported by
estimates of similar magnitude from other effects. In other words, there is strong, robust support
for an RfD of 0.0005 mg/kg-day provided by the concordance of estimates derived from multiple
effects from multiple studies. The estimates for kidney effects, thymus effects, and
developmental heart malformations are based on PBPK model-based estimates of internal dose
This document is a draft for review purposes only and does not constitute Agency policy.
32 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
for interspecies and intraspecies extrapolation, and there is sufficient confidence in the PBPK
model and support from mechanistic data for one of the dose-metrics (total oxidative metabolism
for the heart malformations). There is high confidence that bioactivation of DCVC would be an
appropriate dose-metric for toxic nephropathy, but there is substantial uncertainty in the PBPK
model predictions for this dose-metric in humans (see Section 5.1.3.1). Note that there is some
human evidence of developmental heart defects from TCE exposure in community studies (see
Section 4.8.3.1.1) and of kidney toxicity in TCE-exposed workers (see Section 4.4.1).
In summary, the RfD is 0.0005 mg/kg-day based on the critical effects of heart malformations
(rats), adult immunological effects (mice), and developmental immunotoxicity (mice), and toxic
nephropathy (rats), all from oral studies. This RfD value is further supported by results from an
oral study for the effect of toxic nephropathy (rats) and route-to-route extrapolated results from
an inhalation study for the effect of increased kidney weight (rats).
6.2.2. Cancer (see Section 5.2)
6.2.2.1.1. Background and Methods (rodent: see Section 5.2.1.1; human: see Section 5.2.2.1)
As summarized above, following EPA (2005c) Guidelines for Carcinogen Risk Assessment, TCE
is characterized as "carcinogenic to humans" by all routes of exposure, based on convincing
evidence of a causal association between TCE exposure in humans and kidney cancer, but there
is also human evidence of TCE carcinogenicity in the liver and lymphoid tissues. This
conclusion is further supported by rodent bioassay data indicating carcinogenicity of TCE in rats
and mice at tumor sites that include those identified in human epidemiologic studies. Therefore,
both human epidemiologic studies as well as rodent bioassays were considered for deriving
PODs for dose-response assessment of cancer endpoints. For PODs derived from rodent
bioassays, default dosimetry procedures were applied to convert applied rodent doses to human
equivalent doses. Essentially, for inhalation exposures, "ppm equivalence" across species was
assumed. For oral doses, 3/4-power body-weight scaling was used, with a default average human
body weight of 70 kg. In addition to applied doses, several internal dose-metrics estimated using
a PBPK model for TCE and its metabolites were used in the dose-response modeling for each
tumor type. In general, an attempt was made to use tissue-specific dose-metrics representing
particular pathways or metabolites identified from available data as having a likely role in the
induction of a tissue-specific cancer. Where insufficient information was available to establish
particular metabolites or pathways of likely relevance to a tissue-specific cancer, more general
This document is a draft for review purposes only and does not constitute Agency policy.
3 3 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
"upstream" metrics had to be used. In addition, the selection of dose-metrics was limited to
metrics that could be adequately estimated by the PBPK model.
Regarding low-dose extrapolation, a key consideration in determining what extrapolation
approach to use is the MOA(s). However, MOA data are lacking or limited for each of the
cancer responses associated with TCE exposure, with the exception of the kidney tumors. For
the kidney tumors, the weight of the available evidence supports the conclusion that a mutagenic
MOA is operative; this MOA supports linear low-dose extrapolation. For the other TCE-induced
tumors, the MOA(s) is unknown. When the MOA(s) cannot be clearly defined, EPA generally
uses a linear approach to estimate low-dose risk (2005c), based on the following general
principles:
• A chemical's carcinogenic effects may act additively to ongoing biological processes,
given that diverse human populations are already exposed to other agents and have
substantial background incidences of various cancers.
• A broadening of the dose-response curve (i.e., less rapid fall-off of response with
decreasing dose) in diverse human populations and, accordingly, a greater potential for
risks from low-dose exposures (Lutz et al., 2005; Zeise et al., 1987) is expected for two
reasons: First, even if there is a "threshold" concentration for effects at the cellular level,
that threshold is expected to differ across individuals. Second, greater variability in
response to exposures would be anticipated in heterogeneous populations than in inbred
laboratory species under controlled conditions (due to, e.g., genetic variability, disease
status, age, nutrition, and smoking status).
• The general use of linear extrapolation provides reasonable upper-bound estimates that
are believed to be health-protective (U.S. EPA, 2005c) and also provides consistency
across assessments.
6.2.2.1.2. Inhalation Unit Risk Estimate (rodent: see Section 5.2.1.3; human: see
Section 5.2.2.1 and 5.2.2.2)
The inhalation unit risk for TCE is defined as a plausible upper bound lifetime extra risk of
cancer from chronic inhalation of TCE per unit of air concentration. The inhalation unit risk for
TCE is 2.20 x 10~2 per ppm (2 x 10~2 per ppm [4 x 10~6 per jig/m3] rounded to one significant
figure), based on human kidney cancer risks reported by Charbotel et al. (2006) and adjusted for
potential risk for tumors at multiple sites. This estimate is based on good-quality human data,
thus avoiding the uncertainties inherent in interspecies extrapolation. The Charbotel et al. (2006)
case-control study of 86 incident renal cell carcinoma (RCC) cases and 316 age- and sex-
matched controls, with individual cumulative exposure estimates for TCE inhalation for each
This document is a draft for review purposes only and does not constitute Agency policy.
34 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
subject, provides a sufficient human data set for deriving quantitative cancer risk estimates for
RCC in humans. The study is a high-quality study which used a detailed exposure assessment
(Fevotte et al., 2006) and took numerous potential confounding factors, including exposure to
other chemicals, into account. A significant dose-response relationship was reported for
cumulative TCE exposure and RCC (Charbotel et al., 2006). Human data on TCE exposure and
cancer risk sufficient for dose-response modeling are only available for RCC, yet human and
rodent data suggest that TCE exposure increases the risk of cancer at other sites as well. In
particular, there is evidence from human (and rodent) studies for increased risks of lymphoma
and liver cancer. Therefore, the inhalation unit risk estimate derived from human data for RCC
incidence was adjusted to account for potential increased risk of those tumor types. To make this
adjustment, a factor accounting for the relative contributions to the extra risk for cancer
incidence from TCE exposure for these three tumor types combined versus the extra risk for
RCC alone was estimated, and this factor was applied to the unit risk estimate for RCC to obtain
a unit risk estimate for the three tumor types combined (i.e., lifetime extra risk for developing
any of the three types of tumor). This estimate is considered a better estimate of total cancer risk
from TCE exposure than the estimate for RCC alone. Although only the Charbotel et al. (2006)
study was found adequate for direct estimation of inhalation unit risks, the available
epidemiologic data provide sufficient information for estimating the relative potency of TCE
across tumor sites. In particular, the relative contributions to extra risk (for cancer incidence)
were calculated from two different data sets to derive the adjustment factor for adjusting the unit
risk estimate for RCC to a unit risk estimate for the three types of cancers (RCC, lymphoma, and
liver) combined. The first calculation is based on the results of the meta-analyses of human
epidemiologic data for the three tumor types; the second calculation is based on the results of the
Raaschou-Nielsen et al. (2003) study, the largest single human epidemiologic study by far with
RR estimates for all three tumor types. These calculations support an adjustment factor of.
The inhalation unit risk based on human epidemiologic data is supported by inhalation unit risk
2
estimates from multiple rodent bioassays, the most sensitive of which range from 1x10 to 2 x
10_1 per ppm [2 x 10~6 to 3 x 10~5 per |ig/m3| From the inhalation bioassays selected for
analysis in Section 5.2.1.1, and using the preferred PBPK model-based dose-metrics, the
_2 _c
inhalation unit risk estimate for the most sensitive sex/species is 8 x 10 per ppm [2 x 10 per
"3
[j,g/m ], based on kidney adenomas and carcinomas reported by Maltoni et al. (1986) for male
Sprague-Dawley rats. Leukemias and Leydig cell tumors were also increased in these rats, and,
although a combined analysis for these tumor types which incorporated the different site-specific
preferred dose-metrics was not performed, the result of such an analysis is expected to be
—2 -5 3
similar, about 9x10 per ppm [2 x 10 per (J,g/m ]. The next most sensitive sex/species from
This document is a draft for review purposes only and does not constitute Agency policy.
3 5 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
the inhalation bioassays is the female mouse, for which lymphomas were reported by Henschler
et al. (1980); these data yield a unit risk estimate of 1.0 x 10~2 per ppm [2 x 10~6 per (J,g/m3]. In
addition, the 90% confidence intervals (i.e., 5 to 95% bounds) reported in Table 5-41 for male rat
kidney tumors from Maltoni et al. (1986) and female mouse lymphomas from Henschler et al.
(1980), derived from the quantitative analysis of PBPK model uncertainty, both included the
_2
estimate based on human data of 2 x 10 per ppm. Furthermore, PBPK model-based route-to-
route extrapolation of the results for the most sensitive sex/species from the oral bioassays,
kidney tumors in male Osborne-Mendel rats and testicular tumors in Marshall rats (NTP, 1988),
_ 1 _c o _2
leads to inhalation unit risk estimates of 2 x 10 per ppm [3x10 per (J,g/m ] and 4x10 per
ppm [8 x 10 6 per [j,g/m3], respectively, with the preferred estimate based on human data falling
within the route-to-route extrapolation of the 90% confidence intervals reported in Table 5-42.
Finally, for all these estimates, the ratios of BMDs to the BMDLs did not exceed a value of 3,
indicating that the uncertainties in the dose-response modeling for determining the POD in the
observable range are small.
Although there are uncertainties in these various estimates, confidence in the proposed inhalation
unit risk estimate of 2 x 10~2 per ppm [4 x 10~6 per [j,g/m3], based on human kidney cancer risks
reported by Charbotel et al. (2006) and adjusted for potential risk for tumors at multiple sites (as
summarized above in Section 6.1.4), is further increased by the similarity of this estimate to
estimates based on multiple rodent data sets. Application of the ADAFs for the kidney cancer
risks, due to the weight of evidence supporting a mutagenic MOA for this endpoint, is
summarized below in Section 6.2.2.5.
6.2.2.1.3. Oral Slope Factor Estimate (rodent: see Section 5.2.1.3; human: see Section
5.2.2.3)
The oral slope factor for TCE is defined as a plausible upper bound lifetime extra risk of cancer
_2
from chronic ingestion of TCE per mg/kg-day oral dose. The oral slope factor is 4.64 x 10 per
2
mg/kg-day (5 x 10 per mg/kg-day rounded to one significant figure), resulting from PBPK
model-based route-to-route extrapolation of the inhalation unit risk estimate based on the human
kidney cancer risks reported in Charbotel et al. (2006) and adjusted for potential risk for tumors
at multiple sites. This estimate is based on good-quality human data, thus avoiding uncertainties
inherent in interspecies extrapolation. In addition, uncertainty in the PBPK model-based route-
to-route extrapolation is relatively low (Chiu, 2006; Chiu and White, 2006). In this particular
case, extrapolation using different dose-metrics yielded expected population mean risks within
about a twofold range, and, for any particular dose-metric, the 95% confidence interval for the
This document is a draft for review purposes only and does not constitute Agency policy.
3 6 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
extrapolated population mean risks for each site spanned a range of no more than about
threefold.
This value is supported by oral slope factor estimates from multiple rodent bioassays, the most
2 i
sensitive of which range from 3x10 to 3 x 10 per mg/kg-day. From the oral bioassays
selected for analysis in Section 5.2.1.1, and using the preferred PBPK model-based dose-metrics,
the oral slope factor estimate for the most sensitive sex/species is 3 x 10 1 per mg/kg-day, based
on kidney tumors in male Osborne-Mendel rats (NTP, 1988). The oral slope factor estimate for
_2
testicular tumors in male Marshall rats (NTP, 1988) is somewhat lower at 7 x 10 per mg/kg-
day. The next most sensitive sex/species result from the oral studies is for male mouse liver
_2
tumors (NCI, 1976), with an oral slope factor estimate of 3 x 10 per mg/kg-day. In addition,
the 90% confidence intervals reported in Table 5-42 for male Osborne-Mendel rat kidney tumors
(NTP, 1988), male F344 rat kidney tumors (NTP, 1990), and male Marshall rat testicular tumors
(NTP, 1988), derived from the quantitative analysis of PBPK model uncertainty, all included the
_2
estimate based on human data of 5 x 10 per mg/kg-day, while the upper 95% confidence bound
_2
for male mouse liver tumors from NCI (1976) was slightly below this value at 4 x 10 per
mg/kg-day. Furthermore, PBPK model-based route-to-route extrapolation of the most sensitive
endpoint from the inhalation bioassays, male rat kidney tumors from Maltoni et al. (1986), leads
to an oral slope factor estimate of 1 x 10 1 per mg/kg-day, with the preferred estimate based on
human data falling within the route-to-route extrapolation of the 90% confidence interval
reported in Table 5-41. Finally, for all these estimates, the ratios of BMDs to the BMDLs did
not exceed a value of three, indicating that the uncertainties in the dose-response modeling for
determining the POD in the observable range are small.
Although there are uncertainties in these various estimates, confidence in the proposed oral slope
_2
factor estimate of 5 x 10 per mg/kg-day, resulting from PBPK model-based route-to-route
extrapolation of the inhalation unit risk estimate based on the human kidney cancer risks reported
in Charbotel et al. (2006) and adjusted for potential risk for tumors at multiple sites (as
summarized above), is further increased by the similarity of this estimate to estimates based on
multiple rodent data sets. Application of the ADAFs for the kidney cancer risks, due to the
weight of evidence supporting a mutagenic MOA for this endpoint, is summarized below in
Section 6.2.2.5.
This document is a draft for review purposes only and does not constitute Agency policy.
3 7 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
6.2.2.1.4. Uncertainties in Cancer Dose-Response Assessment
6.2.2.1.5. Uncertainties in estimates based on human epidemiologic data (see
Section 5.2.2.1.3)
All risk assessments involve uncertainty, as study data are extrapolated to make inferences about
potential effects in humans from environmental exposure. The values for the slope factor and
unit risk estimates are based on good quality human data, which avoids interspecies
extrapolation, one of the major sources of uncertainty in quantitative cancer risk estimates.
A remaining major uncertainty in the unit risk estimate for RCC incidence derived from
the Charbotel et al. (2006) study is the extrapolation from occupational exposures to lower
environmental exposures. There was some evidence of a contribution to increased RCC risk
from peak exposures; however, there remained an apparent dose-response relationship for RCC
risk with increasing cumulative exposure without peaks, and the OR for exposure with peaks
compared to exposure without peaks was not significantly elevated (Charbotel et al., 2006).
Although the actual exposure-response relationship at low exposure levels is unknown, the
conclusion that a mutagenic MOA is operative for TCE-induced kidney tumors supports the
linear low-dose extrapolation that was used (U.S. EPA, 2005c). Additional support for use of
linear extrapolation is discussed above in Section 6.2.2.1.
Another source of uncertainty is the dose-response model used to model the study data to
estimate the POD. A weighted linear regression across the categorical ORs was used to obtain a
slope estimate; use of a linear model in the observable range of the data is often a good general
approach for human data because epidemiological data are frequently too limited (the Charbotel
et al. [(2006)] study had 86 RCC cases, 37 of which had TCE exposure) to clearly identify an
alternate model (U.S. EPA, 2005c). The ratio of the maximum likelihood estimate of the
effective concentration for a 1% response (ECoi) to the LECoi, which gives some indication of
the statistical uncertainties in the dose-response modeling, was about a factor of 2.
A further source of uncertainty is the retrospective estimation of TCE exposures in the Charbotel
et al. (2006) study. This case-control study was conducted in the Arve Valley in France, a region
with a high concentration of screw cutting workshops using TCE and other degreasing agents.
Since the 1960s, occupational physicians of the region have collected a large quantity of well-
documented measurements, including TCE air concentrations and urinary metabolite levels
(Fevotte et al., 2006). The study investigators conducted a comprehensive exposure assessment
to estimate cumulative TCE exposures for the individual study subjects, using a detailed
occupational questionnaire with a customized task-exposure matrix for the screw-cutting workers
and a more general occupational questionnaire for workers exposed to TCE in other industries
This document is a draft for review purposes only and does not constitute Agency policy.
3 8 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
(Fevotte et al., 2006). The exposure assessment also attempted to take dermal exposure from
hand-dipping practices into account by equating it with an equivalent airborne concentration
based on biological monitoring data. Despite the appreciable effort of the investigators,
considerable uncertainty associated with any retrospective exposure assessment is inevitable, and
some exposure misclassification is unavoidable. Such exposure misclassification was most
likely for the 19 deceased cases and their matched controls, for which proxy respondents were
used, and for exposures outside the screw-cutting industry. The exposure estimates from the
RCC study of Moore et al. (2010) were not considered to be as quantitatively accurate as those of
Charbotel et al. (2006) and so were not used for derivation of a unit risk estimate (see Section
5.2.2); nonetheless, it should be noted that these exposure estimates are substantially lower than
those of Charbotel et al. (2006) for comparable OR estimates. If the exposure estimates for
Charbotel et al. (2006) are overestimated, as suggested by the exposure estimates from Moore et
al. (2010), the slope of the linear regression model, and hence the unit risk estimate, would be
correspondingly underestimated.
Another source of uncertainty in the Charbotel et al. (2006) study is the possible influence of
potential confounding or modifying factors. This study population, with a high prevalence of
metal-working, also had relatively high prevalences of exposure to petroleum oils, cadmium,
petroleum solvents, welding fumes, and asbestos (Fevotte et al., 2006). Other exposures
assessed included other solvents (including other chlorinated solvents), lead, and ionizing
radiation. None of these exposures was found to be significantly associated with RCC at a
p = 0.05 significance level. Cutting fluids and other petroleum oils were associated with RCC at
ap = 0.1 significance level; however, further modeling suggested no association with RCC when
other significant factors were taken into account (Charbotel et al., 2006). Moreover, a review of
other studies suggested that potential confounding from cutting fluids and other petroleum oils is
of minimal concern (see Section 4.4.2.3). Nonetheless, a sensitivity analysis was conducted
using the OR estimates further adjusted for cutting fluids and other petroleum oils from the
unpublished report by Charbotel et al. (2005), and an essentially identical unit risk estimate of
5.46 x 10 per ppm was obtained. In addition, the medical questionnaire included familial
kidney disease and medical history, such as kidney stones, infection, chronic dialysis,
hypertension, and use of antihypertensive drugs, diuretics, and analgesics. Body mass index
(BMI) was also calculated, and lifestyle information such as smoking habits and coffee
consumption was collected. Univariate analyses found high levels of smoking and BMI to be
associated with increased odds of RCC, and these two variables were included in the conditional
logistic regressions. Thus, although impacts of other factors are possible, this study took great
pains to attempt to account for potential confounding or modifying factors.
This document is a draft for review purposes only and does not constitute Agency policy.
3 9 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Some other sources of uncertainty associated with the epidemiological data are the dose-metric
and lag period. As discussed above, there was some evidence of a contribution to increased RCC
risk from peak TCE exposures; however, there appeared to be an independent effect of
cumulative exposure without peaks. Cumulative exposure is considered a good measure of total
exposure because it integrates exposure (levels) over time. If there is a contributing effect of
peak exposures, not already taken into account in the cumulative exposure metric, the linear
slope may be overestimated to some extent. Sometimes cancer data are modeled with the
inclusion of a lag period to discount more recent exposures not likely to have contributed to the
onset of cancer. In an unpublished report, Charbotel et al. (2005) also present the results of a
conditional logistic regression with a 10-year lag period, and these results are very similar to the
untagged results reported in their published paper, suggesting that the lag period might not be an
important factor in this study.
Some additional sources of uncertainty are not so much inherent in the exposure-response
modeling or in the epidemiologic data themselves but, rather, arise in the process of obtaining
more general Agency risk estimates from the epidemiologic results. EPA cancer risk estimates
are typically derived to represent an upper bound on increased risk of cancer incidence for all
sites affected by an agent for the general population. From experimental animal studies, this is
accomplished by using tumor incidence data and summing across all the tumor sites that
demonstrate significantly increased incidences, customarily for the most sensitive sex and
species, to attempt to be protective of the general human population. However, in estimating
comparable risks from the Charbotel et al. (2006) epidemiologic data, certain limitations are
encountered. For one thing, these epidemiology data represent a geographically limited (Arve
Valley, France) and likely not very diverse population of working adults. Thus, there is
uncertainty about the applicability of the results to a more diverse general population.
Additionally, the Charbotel et al. (2006) study was a study of RCC only, and so the risk estimate
derived from it does not represent all the tumor sites that may be affected by TCE. This
uncertainty was addressed by adjusting the RCC estimate to multiple sites, but there are also
uncertainties related to the assumptions inherent in the calculations for this adjustment. As
discussed in Section 5.2.2.2, adequate quantitative dose-response data were only available for
one cancer type in humans, so other human data were used to adjust the estimate derived for
RCC to include risk for other cancers with substantial human evidence of hazard (lymphoma and
liver cancer). The relative contributions to extra risk (for cancer incidence) were calculated from
two different data sets to derive an adjustment factor. The first calculation is based on the results
of the meta-analyses for the three tumor types; the second calculation is based on the results of
the Raaschou-Nielsen et al. (2003) study, the largest single study by far with RR estimates for all
This document is a draft for review purposes only and does not constitute Agency policy.
40 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
three tumor types. The fact that the calculations based on two different data sets yielded
comparable values for the adjustment factor (both within 25% of the selected factor of 4)
provides more robust support for the use of the factor of 4. Additional uncertainties pertain to
the weight of evidence supporting the association of TCE exposure with increased risk of cancer
for the three cancer types. As discussed in Section 4.11.2, it is concluded that the weight of
evidence for kidney cancer is sufficient to classify TCE as "carcinogenic to humans." It is also
concluded that there is strong evidence that TCE causes lymphoma as well, although the
evidence for liver cancer was more limited. In addition, the rodent studies demonstrate clear
evidence of multisite carcinogenicity, with tumor types including those for which associations
with TCE exposure are observed in human studies, i.e., liver and kidney cancers and lymphomas.
Overall, the evidence is sufficiently persuasive to support the use of the adjustment factor of 4
based on these three cancer types. Alternatively, if one were to use the factor based only on the
two cancer types with the strongest human evidence, the cancer inhalation unit risk estimate
would be only slightly reduced (25%).
Finally, the value for the oral slope factor estimate was based on route-to-route extrapolation of
the inhalation unit risk based on human data using predictions from the PBPK model. Because
different internal dose-metrics are preferred for each target tissue site, a separate route-to-route
extrapolation was performed for each site-specific slope factor estimate. As discussed above,
uncertainty in the PBPK model-based route-to-route extrapolation is relatively low (Chiu, 2006;
Chiu and White, 2006). In this particular case, extrapolation using different dose-metrics yielded
expected population mean risks within about a twofold range, and, for any particular dose-
metric, the 95% confidence interval for the extrapolated population mean risks for each site
spanned a range of no more than about threefold.
6.2.2.1.6. Uncertainties in estimates based on rodent bioassays (see Section 5.2.1.4)
With respect to rodent-based cancer risk estimates, the cancer risk is typically estimated from the
total cancer burden from all sites that demonstrate an increased tumor incidence for the most
sensitive experimental species and sex. It is expected that this approach is protective of the
human population, which is more diverse but is exposed to lower exposure levels. In the case of
TCE, the impact of selection of the bioassay is limited, since, as discussed in Sections 5.2.1.3
and 5.2.3, estimates based on the two or three most sensitive bioassays are within an order of
magnitude of each other, and are consistent across routes of exposure when extrapolated using
the PBPK model.
This document is a draft for review purposes only and does not constitute Agency policy.
41 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Another source of uncertainty in the TCE rodent-based cancer risk estimates is interspecies
extrapolation. Several plausible PBPK model-based dose-metrics were used for extrapolation of
toxicokinetics, but the cancer slope factor and unit risk estimates obtained using the preferred
dose-metrics were generally similar (within about threefold) to those derived using default
dosimetry assumptions, with the exception of the bioactivated DCVC dose-metric for rat kidney
tumors and the metric for the amount of TCE oxidized in the respiratory tract for mouse lung
tumors occurring from oral exposure. However, there is greater biological support for these
selected dose-metrics. The uncertainty in the PBPK model predictions themselves was analyzed
quantitatively through an analysis of the impact of parameter uncertainties in the PBPK model.
The 95% lower bounds on the BMD including parameter uncertainties in the PBPK model were
no more than fourfold lower than those based on central estimates of the PBPK model
predictions. The greatest uncertainty was for slope factors and unit risks derived from rat kidney
tumors, primarily reflecting the substantial uncertainty in the rat internal dose and in the
extrapolation of GSH conjugation from rodents to humans.
Regarding low-dose extrapolation, a key consideration in determining what extrapolation
approach to use is the MOA(s). However, MOA data are lacking or limited for each of the
cancer responses associated with TCE exposure, with the exception of the kidney tumors. For
the kidney tumors, the weight of the available evidence supports the conclusion that a mutagenic
MOA is operative; this MOA supports linear low-dose extrapolation. For the other TCE-induced
tumors, the data either support a complex MOA or are inadequate to specify the key events and
MO As involved. When the MOA(s) cannot be clearly defined, EPA generally uses a linear
approach to estimate low-dose risk (U.S. EPA, 2005c), based on the general principles discussed
above.
With respect to uncertainties in the dose-response modeling, the two-step approach of modeling
only in the observable range, as put forth in EPA's Guidelines for Carcinogen Risk Assessment
(U.S. EPA, 2005c), is designed in part to minimize model dependence. The ratios of the BMDs
to the BMDLs, which give some indication of the statistical uncertainties in the dose-response
modeling, did not exceed a value of 2.5 for all the primary analyses used in this assessment.
Thus, overall, modeling uncertainties in the observable range are considered to be minimal.
Some additional uncertainty is conveyed by uncertainties in the survival adjustments made to
some of the bioassay data; however, a comparison of the results of two different survival
adjustment methods suggest that their impact is minimal relative to the uncertainties already
discussed.
This document is a draft for review purposes only and does not constitute Agency policy.
42 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
6.2.2.1.7. Application of Age-Dependent Adjustment Factors (see Section 5.2.3.3)
When there is sufficient weight of evidence to conclude that a carcinogen operates through a
mutagenic MO A, and in the absence of chemical-specific data on age-specific susceptibility,
EPA's Supplemental Guidance for Assessing Susceptibility from Early-Life Exposure to
Carcinogens (U.S. EPA, 2005c) recommends the application of default ADAFs to adjust for
potential increased susceptibility from early-life exposure. See the Supplemental Guidance for
detailed information on the general application of these adjustment factors. In brief, the
Supplemental Guidance establishes ADAFs for three specific age groups. The current ADAFs
and their age groupings are 10 for <2 years, 3 for 2 to <16 years, and 1 for 16 years and above
(U.S. EPA, 2005c). For risk assessments based on specific exposure assessments, the 10-fold
and threefold adjustments to the slope factor or unit risk estimates are to be combined with age-
specific exposure estimates when estimating cancer risks from early-life (<16 years age)
exposure.
In the case of TCE, the inhalation unit risk and oral slope factor estimates reflect lifetime risk for
cancer at multiple sites, and a mutagenic MOA has been established for one of these sites, the
kidney. In addition, as discussed in Section 4.10, inadequate TCE-specific data exists to quantify
early-life susceptibility to TCE carcinogenicity; therefore, as recommended in the Supplemental
Guidance, the default ADAFs are used. As illustrated in the example calculations in
Sections 5.2.3.3.1 and 5.2.3.3.2, application of the default ADAFs to the kidney cancer
inhalation unit risk and oral slope factor estimates for TCE is likely to have minimal impact on
the total cancer risk except when exposure is primarily during early life.
In addition to the uncertainties discussed above for the inhalation and oral total cancer unit risk
and slope factor estimates, there are uncertainties in the application of ADAFs to adjust for
potential increased early-life susceptibility. The adjustment is made only for the kidney cancer
component of total cancer risk because that is the tumor type for which the weight of evidence
was sufficient to conclude that TCE-induced carcinogenesis operates through a mutagenic MOA.
However, it may be that TCE operates through a mutagenic MOA for other tumor types as well
or that it operates through other MO As that might also convey increased early-life susceptibility.
Additionally, the ADAFs from the 2005 Supplemental Guidance are not specific to TCE, and it
is uncertain to what extent they reflect increased early-life susceptibility to kidney cancer from
exposure to TCE, if increased early-life susceptibility occurs.
This document is a draft for review purposes only and does not constitute Agency policy.
43 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
6.3. OVERALL CHARACTERIZATION OF TCE HAZARD AND DOSE RESPONSE
There is substantial potential for human exposure to TCE, as it has a widespread presence
in ambient air, indoor air, soil, and groundwater. At the same time, humans are likely to be
exposed to a variety of compounds that are either metabolites of TCE or which have common
metabolites or targets of toxicity. Once exposed, humans, as well as laboratory animal species,
rapidly absorb TCE, which is then distributed to tissues via systemic circulation, extensively
metabolized, and then excreted primarily in breath as unchanged TCE or CO2, or in urine as
metabolites.
Based on the available human epidemiologic data and experimental and mechanistic
studies, it is concluded that TCE poses a potential human health hazard for noncancer toxicity to
the central nervous system, the kidney, the liver, the immune system, the male reproductive
system, and the developing fetus. The evidence is more limited for TCE toxicity to the
respiratory tract and female reproductive system. Following EPA (2005c) Guidelines for
Carcinogen Risk Assessment, TCE is characterized as "carcinogenic to humans" by all routes of
exposure. This conclusion is based on convincing evidence of a causal association between TCE
exposure in humans and kidney cancer. The human evidence of carcinogenicity from
epidemiologic studies of TCE exposure is compelling for lymphoma, but less convincing than
for kidney cancer, and more limited for liver and biliary tract cancer. Less human evidence is
found for an association between TCE exposure and other types of cancer, including bladder,
esophageal, prostate, cervical, breast, and childhood leukemia, breast. Further support for the
characterization of TCE as "carcinogenic to humans" by all routes of exposure is derived from
positive results in multiple rodent cancer bioassays in rats and mice of both sexes, similar
toxicokinetics between rodents and humans, mechanistic data supporting a mutagenic MOA for
kidney tumors, and the lack of mechanistic data supporting the conclusion that any of the
MOA(s) for TCE-induced rodent tumors are irrelevant to humans.
As TCE toxicity and carcinogenicity are generally associated with TCE metabolism,
susceptibility to TCE health effects may be modulated by factors affecting toxicokinetics,
including lifestage, gender, genetic polymorphisms, race/ethnicity, preexisting health status,
lifestyle, and nutrition status. In addition, while some of these factors are known risk factors for
effects associated with TCE exposure, it is not known how TCE interacts with known risk factors
for human diseases.
For noncancer effects, the most sensitive types of effects, based either on human
equivalent concentrations/doses or on candidate RfCs/RfDs, appear to be developmental, kidney,
and immunological (adult and developmental) effects. The neurological and reproductive effects
appear to be about an order of magnitude less sensitive, with liver effects another two orders of
This document is a draft for review purposes only and does not constitute Agency policy.
44 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
"3
magnitude less sensitive. The RfC of 0.0004 ppm (0.4 ppb or 2 (J,g/m ) is based on route-to-
route extrapolated results from oral studies for the critical effects of heart malformations (rats)
and immunotoxicity (mice). This RfC value is further supported by route-to-route extrapolated
results from an oral study of toxic nephropathy (rats). Similarly, the RfD for noncancer effects
of 0.0005 mg/kg-day is based on the critical effects of heart malformations (rats), adult
immunological effects (mice), and developmental immunotoxicity (mice), all from oral studies.
This RfD value is further supported by results from an oral study for the effect of toxic
nephropathy (rats) and route-to-route extrapolated results from an inhalation study for the effect
of increased kidney weight (rats). There is high confidence in these noncancer reference values,
as they are supported by moderate- to high-confidence estimates for multiple effects from
multiple studies.
For cancer, the inhalation unit risk is 2 x 10~2 per ppm [4 x 10~6 per jig/m3], based on
human kidney cancer risks reported by Charbotel et al. (2006) and adjusted, using human
epidemiologic data, for potential risk for tumors at multiple sites. The oral slope factor for
2
cancer is 5 x 10 per mg/kg-day, resulting from PBPK model-based route-to-route extrapolation
of the inhalation unit risk estimate based on the human kidney cancer risks reported in Charbotel
et al. (2006) and adjusted, using human epidemiologic data, for potential risk for tumors at
multiple sites. There is high confidence in these unit risks for cancer, as they are based on good
quality human data, as well as being similar to unit risk estimates based on multiple rodent
bioassays. There is both sufficient weight of evidence to conclude that TCE operates through a
mutagenic MOA for kidney tumors and a lack of TCE-specific quantitative data on early-life
susceptibility. Generally, the application of age-dependent adjustment factors (ADAFs) is
recommended when assessing cancer risks for a carcinogen with a mutagenic MOA. However,
because the ADAF adjustment applies only to the kidney cancer component of the total risk
estimate, it is likely to have a minimal impact on the total cancer risk except when exposures are
primarily during early life.
This document is a draft for review purposes only and does not constitute Agency policy.
45 DRAFT—DO NOT CITE OR QUOTE
-------
This document is a draft for review purposes only and does not constitute Agency policy.
46 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
DRAFT- DO NOT QUOTE OR CITE
7. REFERENCES
Abbas, R. and Fisher, J. W. (1997). A physiologically based pharmacokinetic model for
trichloroethylene and its metabolites, chloral hydrate, trichloroacetate, dichloroacetate,
trichloroethanol, and trichloroethanol glucuronide in B6C3F1 mice. Toxicol Appl
Pharmacol, 147, 15-30. http://dx.doi.org/10.1006/taap.1997.8190
Abbas, R. R.; Seckel, C. S.; Kidney, J. K.; Fisher, J. W. (1996). Pharmacokinetic analysis of
chloral hydrate and its metabolism in B6C3F1 mice. Drug Metab Dispos, 24, 1340-1346.
http://www.ncbi.nlm.nih.gov/pubmed/8971140
Abbas, R.; Seckel, C. S.; MacMahon, K. L.; Fisher, J. W. (1997). Determination of kinetic rate
constants for chloral hydrate, trichloroethanol, trichloroacetic acid and dichloroacetic
acid: A physiologically based modeling approach. Toxicologist, 36, 32-33.
Abbate, C.; Giorgianni, C.; Munao, F.; Brecciaroli, R. (1993). Neurotoxicity induced by
exposure to toluene: an electrophysiologic study. Int Arch Occup Environ Health, 64,
389-392.
Abraham, D.; Patel, P.; Cooper, A. (1995). Isolation from rat kidney of a cytosolic high
molecular weight cysteine-S-conjugate beta-lyase with activity toward leukotriene E4. J
Biol Chem, 270, 180-188. http://www.ncbi.nlm.nih.gov/pubmed/7814371
Abraham, D.; Thomas, R.; Cooper, A. (1995). Glutamine transaminase K is not a major cysteine
S-conjugate beta-lyase of rat kidney mitochondria: evidence that a high-molecular weight
enzyme fulfills this role. Mol Pharmacol, 48, 855-860.
http://www.ncbi.nlm.nih.gov/pubmed/7476916
Adachi, H.; Kubota, I.; Okamura, N.; Iwata, H.; Tsujimoto, M.; Nakazato, H., . . . Noguchi, T.
(1989). Purification and characterization of human microsomal dipeptidase. J Biochem,
105, 957-961. http://www.ncbi.nlm.nih.gov/pubmed/2768222
Adami, H. O.; Chow, W. H.; Nyren, O.; Berne, C.; Linet, M. S.; Ekbom, A., . . . Fraumeni, J. F.
(1996). Excess risk of primary liver cancer in patients with diabetes mellitus. J Natl
Cancer Inst, 88, 1472-1477. http://www.ncbi.nlm.nih.gov/pubmed/8841022
Adams, C.; Keil, D.; Meyers, K.; Eudaly, A.; Smythe, J.; Eudaly, J., . . . Peden-Adams, M. M.
(2003). Lifetime exposure to trichloroethylene (TCE) modulates immune function.
Toxicologist, 72, 375.
Adams, L. A. and Lindor, K. D. (2007). Nonalcoholic fatty liver disease. Ann Epidemiol, 17,
863-869. http://dx.doi.Org/10.1016/j.annepidem.2007.05.013
Adgate, J. L.; Church, T. R.; Ryan, A. D.; Ramachandran, G.; Fredrickson, A. L.; Stock, T. H., . .
. Sexton, K. (2004). Outdoor, indoor, and personal exposure to VOCs in children.
Environ Health Perspect, 112, 1386-1392. http://dx.doi.org/10.1289/ehp.7107
Adgate, J.; Eberly, L.; Stroebel, C.; Pellizzari, E.; Sexton, K. (2004). Personal, indoor, and
outdoor VOC exposures in a probability sample of children. J Expo Anal Environ
Epidemiol, 14, S4-S13. http://dx.doi.org/10.1038/sj.jea.7500353
ADHS. (Arizona Department of Health Services). (1990). Incidence study of childhood cancer in
Maricopa County: 1965-1986. Phoenix, AZ: Author.
ADHS. (Arizona Department of Health Services). (1995). Update of the incidence of childhood
cancers and testicular cancer in Southwest Tuscon: 1987-1991. Phoenix, AZ: Author.
1
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Adler, I. D. (1993). Synopsis of the in vivo results obtained with the 10 known or suspected
aneugens tested in the CEC collaborative study. DNA Repair, 287, 131-137.
http://dx.doi. org/10.1016/0027-5107(93)90152-6
Adli, M. and Baldwin, A. S. (2006). IKK-i/IKKepsilon controls constitutive, cancer cell-
associated NF-kappaB activity via regulation of Ser-536 p65/RelA phosphorylation. J
Biol Chem, 281, 26976-26984. http://dx.doi.org/10.1074/jbc.M603133200
Aickin, M. (2004). Bayes without priors. J Clin Epidemiol, 57, 4-13.
http://dx.doi.org/10.1016/S0895-4356(03)00251-8
Aickin, M.; Chapin, C. A.; Flood, T. J.; Englender, S. J.; Caldwell, G. G. (1992). Assessment of
the spatial occurrence of childhood leukaemia mortality using standardized rate ratios
with a simple linear Poisson model. Int J Epidemiol, 21, 649-655.
http://www.ncbi.nlm.nih.gov/pubmed/1521967
Aikata, H.; Takaishi, H.; Kawakami, Y.; Takahashi, S.; Kitamoto, M.; Nakanishi, T., . . . Ide, T.
(2000). Telomere reduction in human liver tissues with age and chronic inflammation.
Exp Cell Res, 256, 578-582. http://dx.doi.org/10.1006/excr.2000.4862
Akira, S. and Takeda, K. (2004). Toll-like receptor signalling. Nat Rev Immunol, 4, 499-511.
http://dx.doi.org/10.1038/nril391
Akiyama, T. E.; Nicol, C. J.; Fievet, C.; Staels, B.; Ward, J. M.; Auwerx, J., . . . Peters, J. M.
(2001). Peroxisome proliferator-activated receptor-alpha regulates lipid homeostasis, but
is not associated with obesity: studies with congenic mouse lines. J Biol Chem, 276,
39088-39093. http://dx.doi.org/10.1074/jbc.M107073200
Albanese, R.; Banks, H.; Evans, M.; Potter, L. (2002). Physiologically based pharmacokinetic
models for the transport of trichloroethylene in adipose tissue. Bull Math Biol, 64, 97-
131. http://dx.doi.org/10.1006/bulm.2001.0268
Albee, R.; Nitschke, K.; Mattsson, J.; Stebbins, K. (1997). Dichloroacetylene: effects on the rat
trigeminal nerve somatosensory evoked potential. Neurotoxicol Teratol, 19, 27-37.
http://www.ncbi.nlm.nih.gov/pubmed/9088008
Albee, R.; Spencer, P.; Johnson, K.; Bradley, G.; Marable, B.; Wilmer, J.; Mattsson, J. (2006).
Lack of trigeminal nerve toxicity in rats exposed to trichloroethylene vapor for 13 weeks.
Int J Toxicol, 25, 531-540. http://dx.doi.org/10.1080/10915810600972678
Alberati-Giani, D.; Malherbe, P.; Kohler, C.; Lang, G.; Kiefer, V.; Lahm, H. W.; Cesura, A. M.
(1995). Cloning and characterization of a soluble kynurenine aminotransferase from rat
brain: Identity with kidney cysteine conjugate [beta]-lyase. JNeurochem, 64, 1448-1455.
http://dx.doi.Org/10.1046/j.1471-4159.1995.64041448.x
Albers, J. W.; Wald, J. J.; Werner, R. A.; Franzblau, A.; Berent, S. (1999). Absence of
polyneuropathy among workers previously diagnosed with solvent-induced toxic
encephalopathy. J Occup Environ Med, 41, 500-509.
http://www.ncbi.nlm.nih.gov/pubmed/10390702
Albert, D.; Albert, A.; Vernace, M.; Sebastian, J.; Hsia, E. (2005). Analysis of a cluster of cases
of Wegener granulomatosis. J Clin Rheumatol, 11, 188-193.
http://www.ncbi.nlm.nih.gov/pubmed/16357755
Albertini, S. (1990). Analysis of nine known or suspected spindle poisons for mitotic
chromosome malsegregation using Saccharomyces cerevisiae D61.M. Mutagenesis, 5,
453-459. http://www.ncbi.nlm.nih.gov/pubmed/2263203
2
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Alexander, D. D.; Kelsh, M. A.; Mink, P. J.; Mandel, J. H.; Basu, R.; Weingart, M. (2007). A
meta-analysis of occupational trichloroethylene exposure and liver cancer. Int Arch
Occup Environ Health, 81, 127-143. http://dx.doi.org/10.1007/s00420-007-0201-4
Alexander, D. D.; Mink, P. J.; Adami, H. O.; Chang, E. T.; Cole, P.; Mandel, J. S.; Trichopoulos,
D. (2007). The non-Hodgkin lymphomas: a review of the epidemiologic literature. Int J
Cancer, 120 Suppl 12, 1-39. http://dx.doi.org/10.1002/ijc.22719
Alexander, D. D.; Mink, P. J.; Mandel, J. H.; Kelsh, M. A. (2006). A meta-analysis of
occupational trichloroethylene exposure and multiple myeloma or leukaemia. Occup Med
(Lond), 56, 485-493. http://dx.doi.org/10.1093/occmed/kql083
Alimov, A.; Kost-Alimova, M.; Liu, J.; Li, C.; Bergerheim, U.; Imreh, S., . . . Zabarovsky, E. R.
(2000). Combined LOH/CGH analysis proves the existence of interstitial 3p deletions in
renal cell carcinoma. Oncogene, 19, 1392-1399. http://dx.doi.org/10.1038/sj.onc.1203449
Allen, B. C. and Fisher, J. W. (1993). Pharmacokinetic modeling of trichloroethylene and
trichloroacetic acid in humans. Risk Anal, 13, 71-86.
http://www.ncbi.nlm.nih.gov/pubmed/8451462
Allen, B. C.; Shipp, A. M.; Crump, K. S.; Kilian, B.; Hogg, M. L.; Tudor, J.; Keller, B. (1987).
Investigation of cancer risk assessment methods (3 volumes plus summary report).
Washington, D.C.: U.S. Environmental Protection Agency.
Allen, D.; Pearse, G.; Haseman, J.; Maronpot, R. (2004). Prediction of rodent carcinogenesis: an
evaluation of prechronic liver lesions as forecasters of liver tumors in NTP
carcinogenicity studies. Toxicol Pathol, 32, 393-401.
http://dx.doi.org/10.1080/01926230490440934
Allen, J. W.; Collins, B. W.; Evansky, P. A. (1994). Spermatid micronucleus analyses of
trichloroethylene and chloral hydrate effects in mice. Mutat Res Lett, 323, 81-88.
http://dx.doi. org/10.1016/0165-7992(94)90049-3
Altmann, L.; Welge, P.; Mensing, T.; Lilienthal, H.; Voss, B.; Wilhelm, M. (2002). Chronic
exposure to trichloroethylene affects neuronal plasticity in rat hippocampal slices.
Environ Toxicol Pharmacol, 12, 157-167.
Altuntas, T.; Park, S.; Kharasch, E. (2004). Sulfoxidation of cysteine and mercapturic acid
conjugates of the sevoflurane degradation product fluoromethyl-2,2-difluoro-l-
(trifluoromethyl)vinyl ether (compound A). Chem Res Toxicol, 17, 435-445.
http://dx.doi.org/10.1021/tx034254k
Amacher, D. E. and Zelljadt, I. (1983). The morphological transformation of Syrian hamster
embryo cells by chemicals reportedly nonmutagenic to Salmonella typhimurium.
Carcinogenesis, 4, 291-296.
Amet, Y.; Berthou, F.; Fournier, G.; Dreano, Y.; Bardou, L.; Cledes, J.; Menez, J. F. (1997).
Cytochrome P450 4A and 2E1 expression in human kidney microsomes. Biochem
Pharmacol, 53, 765-771. http://dx.doi.org/10.1016/S0006-2952(96)00821-0
Ammini, C.; Fernandez-Canon, J.; Shroads, A.; Cornett, R.; Cheung, J.; James, M., . . .
Stacpoole, P. (2003). Pharmacologic or genetic ablation of maleylacetoacetate isomerase
increases levels of toxic tyrosine catabolites in rodents. Biochem Pharmacol, 66, 2029-
2038. http://www.ncbi.nlm.nih.gov/pubmed/14599561
Andelman, J. B. (1985). Human exposures to volatile halogenated organic chemicals in indoor
and outdoor air. Environ Health Perspect, 62, 313-318.
http://www.ncbi.nlm.nih.gov/pubmed/4085436
3
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Anders, M. W. and Jakobson, I. (1985). Biotransformation of halogenated solvents. Scand J
Work Environ Health, 11, 23-32. http://www.ncbi.nlm.nih.gov/pubmed/3906866
Andersen, M. E.; 3rd, C. H.; Gargas, M. L.; Smith, F. A.; Reitz, R. H. (1987). Physiologically
based pharmacokinetics and the risk assessment process for methylene chloride. Toxicol
Appl Pharmacol, 87, 185-205. http://dx.doi.org/10.1016/0041-008X(87)90281-X
Andersen, M. E.; Gargas, M. L.; 3rd, C. H.; Severyn, K. M. (1987). Quantitative evaluation of
the metabolic interactions between trichloroethylene and 1,1-dichloroethylene in vivo
using gas uptake methods. Toxicol Appl Pharmacol, 89, 149-157.
http://www.ncbi.nlm.nih.gov/pubmed/3603553
Anderson, B. J.; McKee, A. D.; Holford, N. H. (1997). Size, myths and the clinical
pharmacokinetics of analgesia in paediatric patients. Clin Pharmacokinet, 33, 313-327.
http://www.ncbi.nlm.nih.gov/pubmed/9391745
Anderson, S. P.; Howroyd, P.; Liu, J.; Qian, X.; Bahnemann, R.; Swanson, C., . . . Corton, J. C.
(2004). The transcriptional response to a peroxisome proliferator-activated receptor alpha
agonist includes increased expression of proteome maintenance genes. J Biol Chem, 279,
52390-52398. http://dx.doi.org/10.1074/jbc.M409347200
Andersson, L.; Haraldsson, B.; Johansson, C.; Barregard, L. (2008). Methodological issues on
the use of urinary alpha- 1-microglobuline in epidemiological studies. Nephrol Dial
Transplant, 23, 1252-1256. http://dx.doi.org/10.1093/ndt/gfm729
Anna, C. Ft.; Maronpot, R. R.; Pereira, M. A.; Foley, J. F.; Malarkey, D. E.; Anderson, M. W.
(1994). ras proto-oncogene activation in dichloroacetic acid-, trichloroethylene- and
tetrachloroethylene-induced liver tumors inB6C3Fl mice. Carcinogenesis, 15, 2255-
2261. http://dx.doi.org/10.1093/carcin/15.10.2255
Anti, M.; Marra, G.; Rapaccini, G. L.; Rumi, C.; Bussa, S.; Fadda, G., . . . Pompili, M. (1994).
DNA ploidy pattern in human chronic liver diseases and hepatic nodular lesions. Flow
cytometric analysis on echo-guided needle liver biopsy. Cancer, 73, 281-288.
http://www.ncbi.nlm.nih.gov/pubmed/8293389
Anttila, A.; Pukkala, E.; Sallmen, M.; Hernberg, S.; Hemminki, K. (1995). Cancer incidence
among Finnish workers exposed to halogenated hydrocarbons. J Occup Environ Med, 37,
797-806. http://www.ncbi.nlm.nih.gov/pubmed/7552463
Antti-Poika, M. (1982). Overall prognosis of patients with diagnosed chronic organic solvent
intoxication. Int Arch Occup Environ Health, 51, 127-138.
http://dx.doi.org/10.1007/BF00378157
Anway, M. D.; Cupp, A. S.; Uzumcu, M.; Skinner, M. K. (2005). Epigenetic transgenerational
actions of endocrine disruptors and male fertility. Science, 308, 1466-1469.
http://dx.doi.org/10.1126/science.1108190
Aranyi, C.; O'Shea, W. J.; Graham, J. A.; Miller, F. J. (1986). The effects of inhalation of
organic chemical air contaminants on murine lung host defenses. Fundam Appl Toxicol,
6, 713-720. http://dx.doi.org/10.1016/0272-0590(86)90184-3
Aratani, J.; Suzuki, H.; Hashimoto, K. (1993). Measurement of vibratory perception threshold
(VPT) in workers exposed to organic solvents. Environ Res, 61, 357-361.
http://dx.doi.org/10.1006/enrs.1993.1080
Arif, A. A. and Shah, S. M. (2007). Association between personal exposure to volatile organic
compounds and asthma among US adult population. Int Arch Occup Environ Health, 80,
711-719. http://dx.doi.org/10.1007/s00420-007-0183-2
4
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Arito, H.; Takahashi, M.; Ishikawa, T. (1994). Effect of subchronic inhalation exposure to low-
level trichloroethylene on heart rate and wakefulness-sleep in freely moving rats. Sangyo
Eiseigaku Zasshi, 36, 1-8. http://www.ncbi.nlm.nih.gov/pubmed/8126932
Aronson, K.; Siemiatycki, J.; Dewar, R.; Gerin, M. (1996). Occupational risk factors for prostate
cancer: Results from a case-control study in Montreal, Quebec, Canada. Am J Epidemiol,
143, 363-373. http://www.ncbi.nlm.nih.gov/pubmed/8633620
Arp, E. W.; Wolf, P. H.; Checkoway, H. (1983). Lymphocytic leukemia and exposures to
benzene and other solvents in the rubber industry. J Occup Med, 25, 598-602.
http://www.ncbi.nlm.nih.gov/pubmed/6886869
Aryal, B. K.; Khuder, S. A.; Schaub, E. A. (2001). Meta-analysis of systemic sclerosis and
exposure to solvents. Am J Ind Med, 40, 271-274. http://dx.doi.org/10.1002/ajim.1098
Asal, N. R.; Geyer, J. R.; Risser, D. R.; Lee, E. T.; Kadamani, S.; Cherng, N. (1988). Risk
factors in renal cell carcinoma. II. Medical history, occupation, multivariate analysis, and
conclusions. Cancer Detect Prev, 13, 263-279.
http://www.ncbi.nlm.nih.gov/pubmed/3266567
Asal, N. R.; Risser, D. R.; Kadamani, S.; Geyer, J. R.; Lee, E. T.; Cherng, N. (1988). Risk
factors in renal cell carcinoma: I. Methodology, demographics, tobacco, beverage use,
and obesity. Cancer Detect Prev, 11, 359-377.
http://www.ncbi.nlm.nih.gov/pubmed/3390857
Ashby, J.; Brady, A.; Elcombe, C. R.; Elliott, B. M.; Ishmael, J.; Odum, J., . . . Purchase, I. F.
(1994). Mechanistically-based human hazard assessment of peroxisome proliferator-
induced hepatocarcinogenesis. Hum Exp Toxicol, 13, S1-S117.
http://dx.doi.org/10.1177/096032719401300201
Ashley, D. L.; Bonin, M. A.; Cardinali, F. L.; McCraw, J. M.; Wooten, J. V. (1994). Blood
concentrations of volatile organic compounds in a nonoccupationally exposed US
population and in groups with suspected exposure. Clin Chem, 40, 1401-1404.
http://www.ncbi.nlm.nih.gov/pubmed/8013127
Astrand, I. (1975). Uptake of solvents in the blood and tissues of man. A review. Scand J Work
Environ Health, 1, 199-218. http://www.ncbi.nlm.nih.gov/pubmed/779027
Astrand, I. and Ovrum, P. (1976). Exposure to trichloroethylene I Uptake and distribution in
man. Scand J Work Environ Health, 4, 199-211.
http://www.ncbi.nlm.nih.gov/pubmed/1019594
ATSDR. (1990). Toxicological profile for cis-, trans-1,2-dichloroethene.
ATSDR. (1996). Toxicological profile for 1,2-dichloroethene. Retrieved from
http://www.atsdr.cdc.gov/ToxProfiles/tp.asp?id=464&tid=82.
ATSDR. (1997). Toxicological profile for tetrachloroethylene. Retrieved from
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=265&tid=48.
ATSDR. (1997). Toxicological profile for trichloroethylene. Retrieved from
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=265&tid=48.
ATSDR. (1998). Volatile organic compounds in drinking water and adverse pregnancy
outcomes: United States Marine Corps Base, Camp Lejeune, North Carolina. Retrieved
from http://www.atsdr.cdc.gov/hs/lejeune/.
ATSDR. (2003). Impact of trichloroethylene exposure on oral motor, speech, and hearing in
children. Retrieved from file://L:\Lab\NrCEA\TCE Documents\TCE PDFs4\ATSDR2003
TCE_OralMotor_Speech_Hearing_Children FinalReport.pdf.
5
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
ATSDR. (2003). Progress report: Survey of specific childhood cancers and birth defects among
children whose mothers were pregnant while living at U.S. Marine Corps Base Camp
Lejeune, North Carolina, 1968-1985. Retrieved from
http://www.atsdr.cdc.gov/sites/lejeune/lejeunehealthsurvey.pdf.
ATSDR. (Agency for Toxic Substances and Disease Registry). (1995). Toxicological profile for
1,1,1-trichloroethane. Atlanta, GA: U.S. Department of Health and Humans Services.
ATSDR. (Agency for Toxic Substances and Disease Registry). (1996). Biennial report to
congress (1991 and 1992). Atlanta, GA: U.S. Department of Health and Human
Services.
ATSDR. (Agency for Toxic Substances and Disease Registry). (1997). Toxicological profile for
trichloroethylene (update). (PB98-101165). Atlanta, GA: U.S. Department of Health and
Human Services, Public Health Service, Agency for Toxic Substances and Disease
Registry. Retrieved from http://www.atsdr.cdc.gov/toxprofiles/tpl9-p.pdf.
ATSDR. (Agency for Toxic Substances and Disease Registry). (1998). Health consultation:
Evaluation of health impacts from potential past exposure to tetrachloroethylene in the
Natick public water supply. Atlanta, GA: Agency for Toxic Substances & Disease
Registry. Retrieved from http://www.atsdr.cdc.gov/hac/pha/pha.asp?docid=1270&pg=0.
ATSDR. (Agency for Toxic Substances and Disease Registry). (2000). Public Health
Assessment for Tucson International Airport Area (a/k/a El Vado Residential Properties)
Tucson, Pima County, Arizona, Region 9. CERCLIS No. AZD980737530. Office of
Environmental Health, Environmental Health Consultation Services: U.S. Department of
Health and Humans Services.
ATSDR. (Agency for Toxic Substances and Disease Registry). (2001). Final Report: Evaluation
of priority health conditions in a community with historical contamination by
trichloroethylene. Atlanta, GA: U.S. Department of Health and Human Services.
ATSDR. (Agency for Toxic Substances and Disease Registry). (2003). Health consultation.
View-master factory supply well (a.k.a. Mattel Portland Operations) Beaverton,
Washington County, Oregon. Atlanta, GA: U.S. Department of Health and Human
Services. Retrieved from http://www.atsdr.cdc.gov/hac/pha/pha.asp?docid=293&pg=0.
ATSDR. (Agency for Toxic Substances and Disease Registry). (2004). Feasibility investigation
of worker exposure to trichloroethylene at the view-master factory in Beaverton, Oregon.
Atlanta, GA: Author. Retrieved from
http://www.ntis.gov/search/product.aspx?ABBR=PB2005102173.
ATSDR. (Agency for Toxic Substances and Disease Registry). (2004). Interaction profile: 1,1,1-
trichloroethane, 1,1-dichloroethane, trichloroethylene, and tetrachloroethylene. Atlanta,
GA: Author. Retrieved from http://www.atsdr.cdc.gov/interactionprofiles/ip02.html.
ATSDR. (Agency for Toxic Substances and Disease Registry). (2006). Health Consultation:
Public health implications of exposures to low-level volatile organic compounds in public
drinking water: Endicott are investigation Broome County, New York. Atlanta, GA: U.S.
Department of Health and Human Services.
ATSDR. (Agency for Toxic Substances and Disease Registry). (2006). Health statistics review:
Cancer and birth outcome analysis: Endicott area investigation: Endicott area, town of
Union, Broome County, New York. Atlanta, GA: U.S. Department of Health and
Humans Services. Retrieved from
6
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
http://www.atsdr.cdc.gov/HAC/pha/EndicottAreaInvestigation/EndicottHealthStatsRevie
wHC052606.pdf.
ATSDR. (Agency for Toxic Substances and Disease Registry). (2006). Toxicological profile for
1,1,1 trichloroethane. Atlanta, GA: Author. Retrieved from
http://www.atsdr.cdc.gov/toxpro2.html.
ATSDR. (Agency for Toxic Substances and Disease Registry). (2007). Analyses of groundwater
flow, contaminant fate and transport and distribution of drinking water at Tarawa Terrace
and vicinity, U.S. Marine Corps Base Camp Lejeune, North Carolina: Historical
reconstruction and present-day conditions. Atlanta, GA: U.S. Department of Health and
Human Services, Agency for Toxic Substances and Disease Registry. Retrieved from
http ://www.atsdr. cdc.gov/ sites/lej eune/tarawaterrace.html.
ATSDR. (Agency for Toxic Substances and Disease Registry). (2008). Health consultation.
Cancer and birth outcome analysis, Endicott area, Town of Union, Broome County, New
York. Health statistics review follow-up. Atlanta, GA: U.S. Department of Health and
Humans Services. Retrieved from
http://www.atsdr.cdc.gOv/hac/pha//EndicottAreaInvestigationFollowUp/EndicottAreaHC
051508.pdf.
ATSDR. (Agency for Toxic Substances and Disease Registry). (2009). Study on birth defects
and childhood cancers. Atlanta, GA: U.S. Department of Health and Human Services,
Public Health Services. Retrieved from
http://www.atsdr.cdc.gov/sites/lejeune/update.html.
Attallah, A. M.; Tabll, A. A.; Salem, S. F.; El-Sadany, M.; Ibrahim, T. A.; Osman, S.; El-
Dosoky, I. M. (1999). DNA ploidy of liver biopsies from patients with liver cirrhosis and
hepatocellular carcinoma: a flow cytometric analysis. Cancer Lett, 142, 65-69.
http://www.ncbi.nlm.nih.gov/pubmed/10424782
Auperin, A.; Benhamou, S.; Ory-Paoletti, C.; Flamant, R. (1994). Occupational risk factors for
renal cell carcinoma: A case-control study. Occup Environ Med, 51, 426-428.
http://dx.doi.Org/10.l 136/oem. 51.6.426
Austin, E. W.; Parrish, J. M.; Kinder, D. H.; Bull, R. J. (1996). Lipid peroxidation and formation
of 8-hydroxydeoxyguanosine from acute doses of halogenated acetic acids. Fundam Appl
Toxicol, 31, 77-82. http://dx.doi.org/10.1006/faat.1996.0078
Austin, H.; Delzell, E.; Grufferman, S.; Levine, R.; Morrison, A. S.; Stolley, P. D.; Cole, P.
(1987). Case-control study of hepatocellular carcinoma, occupation, and chemical
exposures. J Occup Med, 29, 665-669. http://www.ncbi.nlm.nih.gov/pubmed/2821204
Auwerx, J.; Schoonjans, K.; Fruchart, J. C.; Staels, B. (1996). Transcriptional control of
triglyceride metabolism: fibrates and fatty acids change the expression of the LPL and
apo C-III genes by activating the nuclear receptor PPAR. Atherosclerosis, 124 Suppl,
S29-S37. http://www.ncbi.nlm.nih.gov/pubmed/8831913
Axelson, O.; Selden, A.; Andersson, K.; Hogstedt, C. (1994). Updated and expanded Swedish
cohort study on trichloroethylene and cancer risk. J Occup Med, 36, 556-562.
http://www.ncbi.nlm.nih.gov/pubmed/8027881
Baden, J. M.; Kelley, M.; Mazze, R. I.; Simmon, V. F. (1979). Mutagenicity of inhalation
anaesthetics: trichloroethylene, divinyl ether, nitrous oxide and cyclopropane. Br J
Anaesth, 51,417-421.
7
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Badger, T.; Ronis, M.; Seitz, H.; Albano, E.; Ingelman-Sundberg, M.; Lieber, C. (2003). Alcohol
metabolism: role in toxicity and carcinogenesis. Alcohol Clin Exp Res, 27, 336-347.
http://dx.doi.org/10.1097/01.ALC.0000052583.87673.37
Bai, C. L.; Canfield, P. J.; Stacey, N. H. (1992). Effects of hexachloro-l,3-butadiene and 1,1,2,2-
tetrachloroethylene on individual serum bile acids. Toxicol Ind Health, 8, 191-203.
http://www.ncbi.nlm.nih.gov/pubmed/1502697
Bai, C.; Canfield, P. J.; Stacey, N. H. (1992). Individual serum bile acids as early indicators of
carbon tetrachloride- and chloroform-induced liver injury. Toxicology, 75, 221-234.
Bai, C.-L. and Stacey, N. H. (1993). Mechanism of trichloroethylene-induced elevation of
individual serum bile acids: II in vitro and in vivo interference by trichloroethylene with
bile acid transport in isolated rat hepatocytes. Toxicol Appl Pharmacol, 121, 296-302.
http://dx.doi.org/10.1006/taap.1993.1157
Bailly, M.; Bain, C.; Favrot, M. C.; Ozturk, M. (1995). Somatic mutations of von Hippel-Lindau
(VHL) tumor-suppressor gene in European kidney cancers. Int J Cancer, 63, 660-664.
http://www.ncbi.nlm.nih.gov/pubmed/7591282
Baker, V. A.; Harries, H. M.; Waring, J. F.; Duggan, C. M.; Ni, H. A.; Jolly, R. A., . . . Rockett,
J. C. (2004). Clofibrate-induced gene expression changes in rat liver: a cross-laboratory
analysis using membrane cDNA arrays. Environ Health Perspect, 112, 428-438.
http://www.ncbi.nlm.nih.gov/pubmed/15033592
Bakke, B.; Stewart, P.; Waters, M. (2007). Uses of and exposure to trichloroethylene in U.S.
industry: a systematic literature review. J Occup Environ Hyg, 4, 375-390.
http://dx.doi.org/10.1080/15459620701301763
Balkwill, F. and Coussens, L. M. (2004). Cancer: an inflammatory link. Nature, 431, 405-406.
http://dx.doi.org/10.1038/431405a
Balkwill, F.; Charles, K. A.; Mantovani, A. (2005). Smoldering and polarized inflammation in
the initiation and promotion of malignant disease. Cancer Cell, 7, 211-217.
http://dx.doi.Org/10.1016/j.ccr.2005.02.013
Ballestar, E. and Esteller, M. (2002). The impact of chromatin in human cancer: linking DNA
methylation to gene silencing. Carcinogenesis, 23, 1103-1109.
http://www.ncbi.nlm.nih.gov/pubmed/12117766
Balluz, L.; Philen, R.; Ortega, L.; Rosales, C.; Brock, J.; Barr, D.; Kieszak, S. (2001).
Investigation of systemic lupus erythematosus in Nogales, Arizona. Am J Epidemiol,
154, 1029-1036. http://www.ncbi.nlm.nih.gov/pubmed/11724719
Band, P. R.; Le, N. D.; Fang, R.; Deschamps, M.; Gallagher, R. P.; Yang, P. (2000).
Identification of occupational cancer risks in British Columbia: A population-based case-
control study of 995 incident breast cancer cases by menopausal status, controlling for
confounding factors. J Occup Environ Med, 42, 284-310.
http://www.ncbi.nlm.nih.gov/pubmed/10738708
Baneijee, S. and Van Duuren, B. (1978). Covalent binding of the carcinogen trichloroethylene to
hepatic microsomal proteins and to exogenous DNA in vitro. Cancer Res, 38, 776-780.
http://www.ncbi.nlm.nih.gov/pubmed/626981
Banks, P. M. (1992). Changes in diagnosis of non-Hodgkin's lymphomas overtime. Cancer Res,
52, 5453s-5455s. http://www.ncbi.nlm.nih.gov/pubmed/1394153
Bannasch, P. (1996). Pathogenesis of hepatocellular carcinoma: sequential cellular, molecular,
and metabolic changes. 14, 161-197. http://www.ncbi.nlm.nih.gov/pubmed/9055578
8
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Bannasch, P. (2001). Hormonal and hormone-like effects eliciting hepatocarcinogenesis. Folia
Histochem Cytobiol, 39 Suppl 2, 28-29. http://www.ncbi.nlm.nih.gov/pubmed/11820616
Bannasch, P.; Haertel, T.; Su, Q. (2003). Significance of hepatic preneoplasia in risk
identification and early detection of neoplasia. Toxicol Pathol, 31, 134-139.
http://dx.doi.org/10.1080/01926230390173923
Bannasch, P.; Nehrbass, D.; Kopp-Schneider, A. (2001). Significance of hepatic preneoplasia for
cancer chemoprevention. IARC Sci Publ, 154, 223-240.
http://www.ncbi.nlm.nih.gov/pubmed/11220662
Barak, Y.; Sadovsky, Y.; Shalom-Barak, T. (2008). PPAR Signaling in Placental Development
and Function. PPAR Research, 2008, 142082. http://dx.doi.org/10.1155/2008/142082
Baraona, E.; Shoichet, L.; Navder, K.; Lieber, C. (2002). Mediation by nitric oxide of the
stimulatory effects of ethanol on blood flow. Life Sci, 70, 2987-2995.
http://www.ncbi.nlm.nih.gov/pubmed/12138012
Baraona, E.; Zeballos, G.; Shoichet, L.; Mak, K.; Lieber, C. (2002). Ethanol consumption
increases nitric oxide production in rats, and its peroxynitrite-mediated toxicity is
attenuated by polyenylphosphatidylcholine. Alcohol Clin Exp Res, 26, 883-889.
http://www.ncbi.nlm.nih.gov/pubmed/12068258
Bardodej, Z. and Vyskocil, J. (1956). The problem of trichloroethylene in occupational
medicine: trichloroethylene metabolism and its effect on the nervous system evaluated as
a means of hygienic control. Arch Environ Occup Health, 13, 581-592.
Barrass, N. C.; Price, R. J.; Lake, B. G.; Orton, T. C. (1993). Comparison of the acute and
chronic mitogenic effects of the peroxisome proliferators methylclofenapate and clofibric
acid in rat liver. Carcinogenesis, 14, 1451-1456.
http://www.ncbi.nlm.nih.gov/pubmed/8330363
Barret, L.; Arsac, P.; Vincent, M.; Faure, J.; Garrel, S.; Reymond, F. (1982). Evoked trigeminal
nerve potential in chronic trichloroethylene intoxication. Clin Toxicol, 19, 419-423.
http://www.ncbi.nlm.nih.gov/pubmed/7143527
Barret, L.; Faure, J.; Guilland, B.; Chomat, D.; Didier, B.; Debru, J. (1984). Trichloroethylene
occupational exposure: elements for better prevention. Int Arch Occup Environ Health,
53, 283-289. http://www.ncbi.nlm.nih.gov/pubmed/6715052
Barret, L.; Garrel, S.; Danel, V.; Debru, J. L. (1987). Chronic trichloroethylene intoxication: a
new approach by trigeminal-evoked potentials? Arch Environ Occup Health, 42, 297-
302. http://www.ncbi.nlm.nih.gov/pubmed/3452297
Barret, L.; Torch, S.; Leray, C.; Sarlieve, L.; Saxod, R. (1992). Morphometric and biochemical
studies in trigeminal nerve of rat after trichloroethylene or dichloroacetylene oral
administration. Neurotoxicology, 13, 601-614.
http://www.ncbi.nlm.nih.gov/pubmed/1282229
Barret, L.; Torch, S.; Usson, Y.; Gonthier, B.; Saxod, R. (1991). A morphometric evaluation of
the effects of trichloroethylene and dichloroacetylene on the rat mental nerve Preliminary
results. Neurosci Lett, 131, 141-144. http://www.ncbi.nlm.nih.gov/pubmed/1762688
Barter, Z. E.; Bayliss, M. K.; Beaune, P. H.; Boobis, A. R.; Carlile, D. J.; Edwards, R. J., . . .
Rostami-Hodjegan, A. (2007). Scaling factors for the extrapolation of in vivo metabolic
drug clearance from in vitro data: reaching a consensus on values of human microsomal
protein and hepatocellularity per gram of liver. Curr Drug Metab, 8, 33-45.
9
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Barton, H. A. and Clewell, H. J., III. (2000). Evaluating noncancer effects of trichloroethylene:
Dosimetry, mode of action, and risk assessment. Environ Health Perspect, 108, 323-334.
http://www.ncbi.nlm.nih.gov/pubmed/10807562
Barton, H. A. and Das, S. (1996). Alternatives for a risk assessment on chronic noncancer effects
from oral exposure to trichloroethylene. Regul Toxicol Pharmacol, 24, 269-285.
http://dx.doi.org/10.1006/rtph.1996.0140
Barton, H. A.; Chiu, W. A.; Setzer, R. W.; Andersen, M. E.; Bailer, A. J.; Bois, F. Y., . . . Tan,
Y. M. (2007). Characterizing uncertainty and variability in physiologically-based
pharmacokinetic (PBPK) models: State of the science and needs for research and
implementation. Toxicol Sci, 99, 395-402. http://dx.doi.org/10.1093/toxsci/kfml00
Barton, H. A.; Creech, J. R.; Godin, C. S.; Randall, G. M.; Seckel, C. S. (1995). Chloroethylene
mixtures: pharmacokinetic modeling and in vitro metabolism of vinyl chloride,
trichloroethylene, and trans-1,2-dichloroethylene in rat. Toxicol Appl Pharmacol, 130,
237-247. http://dx.doi.org/10.1006/taap.1995.1029
Barton, H. A.; Flemming, C. D.; Lipscomb, J. C. (1996). Evaluating human variability in
chemical risk assessment: hazard identification and dose-response assessment for
noncancer oral toxicity of trichloroethylene. Toxicology, 111, 271-287.
http://www.ncbi.nlm.nih.gov/pubmed/8711742
Barton, H.; Bull, R.; Schultz, I.; Andersen, M. (1999). Dichloroacetate (DCA) dosimetry:
interpreting DCA-induced liver cancer dose response and the potential for DCA to
contribute to trichloroethylene-induced liver cancer. Toxicol Lett, 106, 9-21.
http://www.ncbi.nlm.nih.gov/pubmed/10378446
Bartonicek, V. (1962). Metabolism and excretion of trichloroethylene after inhalation by human
subjects. Occup Environ Med, 19, 134-141.
http://www.ncbi.nlm.nih.gov/pubmed/13865497
Bartonicek, V. and Teisinger, J. (1962). Effect of tetraethyl thiuram disulphide (disulfiram) on
metabolism of trichloroethylene in man. Occup Environ Med, 19, 216-221.
http://dx.doi.Org/10.l 136/oem. 19.3.216
Bartosiewicz, M.; Penn, S.; Buckpitt, A. (2001). Applications of gene arrays in environmental
toxicology: fingerprints of gene regulation associated with cadmium chloride,
benzo(a)pyrene, and trichloroethylene. Environ Health Perspect, 109, 71-74.
http://www.ncbi.nlm.nih.gOv/pubmed/l 1171528
Bartsch, H.; Malaveille, C.; Barbin, A.; Planche, G. (1979). Mutagenic and alkylating
metabolites of halo-ethylenes, chlorobutadienes and dichlorobutenes produced by rodent
or human liver tissues: Evidence for oxirane formation by P450-linked microsomal
mono-oxygenases. Arch Toxicol, 41, 249-277. http://dx.doi.org/10.1007/BF00296896
Beaudreuil, S.; Lasfargues, G.; Laueriere, L.; El Ghoul, Z.; Fourquet, F.; Longuet, C., . . .
Buchler, M. (2005). Occupational exposure in ANCA-positive patients: A case-control
study. Kidney Int, 67, 1961-1966. http://dx.doi.Org/10.llll/j.1523-1755.2005.00295.x
Beaumont, J. J. and Breslow, N. E. (1981). Power considerations in epidemiologic studies of
vinyl chloride workers. Am J Epidemiol, 114, 725-734.
http://www.ncbi.nlm.nih.gov/pubmed/7030065
Becker, N.; Deeg, E.; Meters, A. (2004). Population-based study of lymphoma in Germany:
rationale, study design and first results. Leuk Res, 28, 713-724.
http://dx.doi.Org/10.1016/j.leukres.2003.l 1.010
10
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Beckstead, M. J.; Weiner, J. L.; Eger, E. I.; Gong, D. H.; Mihic, S. J. (2000). Glycine and
[gamma]-aminobutyric acid(A) receptor function is enhanced by inhaled drugs of abuse.
Mol Pharmacol, 57, 1199-1205. http://www.ncbi.nlm.nih.gov/pubmed/10825391
Beekman, J. M.; Boess, F.; Hildebrand, H.; Kalkuhl, A.; Suter, L. (2006). Gene expression
analysis of the hepatotoxicant methapyrilene in primary rat hepatocytes: an
interlaboratory study. Environ Health Perspect, 114, 92-99.
http://www.ncbi.nlm.nih.gov/pubmed/16393664
Begg, C. B. and Mazumdar, M. (1994). Operating characteristics of a rank correlation test for
publication bias. Biometrics, 50, 1088-1101.
http://www.ncbi.nlm.nih.gov/pubmed/7786990
Beland, F. (1999). NTP technical report on the toxicity and metabolism studies of chloral hydrate
(CAS No. 302-17-0). Administered by gavage to F344/N rats and B6C3F1 mice. U.S.
Dept. of Health and Human Services, National Institutes of Health: Rockville, MD.
Benedetti, M. S.; Whomsley, R.; Canning, M. (2007). Drug metabolism in the paediatric
population and in the elderly. Drug Discov Today, 12, 599-610.
http://dx.doi.Org/10.1016/j.drudis.2007.06.011
Benichou, J.; Chow, W. H.; McLaughlin, J. K.; Mandel, J. S.; Fraumeni, J. F. (1998). Population
attributable risk of renal cell cancer in Minnesota. Am J Epidemiol, 148, 424-430.
http://www.ncbi.nlm.nih.gov/pubmed/9737554
Benson, R. (2004). [Correspondence from Robert Benson (U. S. EPA) to Dr. P. D. Johnson
(University of Arizona)].
Beppu, K. (1968). Transmission of the anesthetic agents through the placenta in painless delivery
and their effects on newborn infants. Keio J Med, 17, 81-107.
http://www.ncbi.nlm.nih.gov/pubmed/5727834
Berger, J. and Daxenbichler, G. (2002). DNA methylation of nuclear receptor genes—possible
role in malignancy. J Steroid Biochem Mol Biol, 80, 1-11.
http://www.ncbi.nlm.nih.gov/pubmed/11867259
Berger, T. and Horner, C. M. (2003). In vivo exposure of female rats to toxicants may affect
oocyte quality. Reprod Toxicol, 17, 273-281. http://dx.doi.org/10.1016/S0890-
6238(03)00009-1
Bergeron, A.; Jorquera, R.; Tanguay, R. (2003). [Hereditary tyrosinemia: an endoplasmic
reticulum stress disorder?]. Med Sci (Paris), 19, 976-980.
http://dx.doi.org/10.1051/medsci/20031910976
Bergman, K. (1983). Interactions of trichloroethylene with DNA in vitro and with RNA and
DNA of various mouse tissues in vivo. Arch Toxicol, 54, 181-193.
http://www.ncbi.nlm.nih.gov/pubmed/6197950
Berman, E.; Schlicht, M.; Moser, V. C.; MacPhail, R. C. (1995). A multidisciplinary approach to
toxicological screening: I. Systemic toxicity. J Toxicol Environ Health, 45, 127-143.
http://dx.doi.org/10.1080/15287399509531986
Bernad, P. G.; Newell, S.; Spyker, D. A. (1987). Neurotoxicity and behavior abnormalities in a
cohort chronically exposed to trichloroethylene. Vet Hum Toxicol, 29, 475.
Bernauer, U.; Birner, G.; Dekant, W.; Henschler, D. (1996). Biotransformation of
trichloroethene: dose-dependent excretion of 2,2,2-trichloro-metabolites and mercapturic
acids in rats and humans after inhalation. Arch Toxicol, 70, 338-346.
http://www.ncbi.nlm.nih.gov/pubmed/8975632
11
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Bernillon, P. and Bois, F. Y. (2000). Statistical issues in toxicokinetic modeling: a bayesian
perspective. Environ Health Perspect, 108, 883-893.
http://www.ncbi.nlm.nih.gov/pubmed/11035998
Beroud, C.; Joly, D.; Gallou, C.; Staroz, F.; Orfanelli, M. T.; Junien, C. (1998). Software and
database for the analysis of mutations in the VHL gene. Nucleic Acids Res, 26, 256-258.
http://www.ncbi.nlm.nih.gov/pubmed/9399847
Besson, H.; Brennan, P.; Becker, N; Meters, A.; De Sanjose, S.; Font, R., . . . Boffetta, P.
(2006). Tobacco smoking, alcohol drinking and non-Hodgkin's lymphoma: A European
multicenter case-control study (Epilymph). Int J Cancer, 119, 901-908.
http://dx.doi.org/10.1002/ijc.21913
Best, D. H. and Coleman, W. B. (2007). Treatment with 2-AAF blocks the small hepatocyte-like
progenitor cell response in retrorsine-exposed rats. J Hepatol, 46, 1055-1063.
http://dx.doi.Org/10.1016/j.jhep.2007.01.040
Beyer, R. P.; Fry, R. C.; Lasarev, M. R.; McConnachie, L. A.; Meira, L. B.; Palmer, V. S., . . .
Consortium, M. o. t. T. R. (2007). Multicenter study of acetaminophen hepatotoxicity
reveals the importance of biological endpoints in genomic analyses. Toxicol Sci, 99, 326-
337. http ://dx. doi. org/10.1093/toxsci/kfm 150
Bhunya, S. P. and Behera, B. C. (1987). Relative genotoxicity of trichloroacetic acid (TCA) as
revealed by different cytogenetic assays: Bone marrow chromosome aberration,
micronucleus and sperm-head abnormality in the mouse. Mutat Res Genet Toxicol
Environ Mutagen, 188, 215-221. http://dx.doi.org/10.1016/0165-1218(87)90092-9
Bhunya, S. P. and Jena, G. B. (1996). The evaluation of clastogenic potential of trichloroacetic
acid (TCA) in chick in vivo test system. Mutat Res Genet Toxicol Environ Mutagen, 367,
253-259. http://dx.doi.org/10.1016/S0165-1218(96)90085-3
Bignami, M.; Conti, G.; Conti, L.; Crebelli, R.; Misuraca, F.; Puglia, A. M., . . . Carere, A.
(1980). Mutagenicity of halogenated aliphatic hydrocarbons in Salmonella typhimurium,
Streptomyces coelicolor and Aspergillus nidulans. Chem Biol Interact, 30, 9-23.
http ://dx. doi. org/10.1016/0009-2797(80)90110-6
Binaschi, S. and Cantu, L. (1983). Vestibular Function In Solvent Exposure: Clinical Criteria.
46, 205-210.
Birner, G.; Bernauer, U.; Werner, M.; Dekant, W. (1997). Biotransformation, excretion and
nephrotoxicity of haloalkene-derived cysteine S-conjugates. Arch Toxicol, 72, 1-8.
http://www.ncbi.nlm.nih.gov/pubmed/9458184
Birner, G.; Richling, C.; Henschler, D.; Anders, M. W.; Dekant, W. (1994). Metabolism of
tetrachloroethene in rats: Identification of N[epsilon]-(dichloroacetyl)-L-lysine and
N[epsilon]-(trichloroacetyl)-L-lysine as protein adducts. Chem Res Toxicol, 7, 724-732.
http://dx.doi.org/10.1021/tx00042a003
Birner, G.; Vamvakas, S.; Dekant, W.; Henschler, D. (1993). Nephrotoxic and genotoxic N-
acetyl-S-dichlorovinyl-L-cysteine is a urinary metabolite after occupational 1,1,2-
trichloroethene exposure in humans: implications for the risk of trichloroethene exposure.
Environ Health Perspect, 99, 281-284. http://www.ncbi.nlm.nih.gov/pubmed/8319644
Birner, G.; Werner, M.; Rosner, E.; Mehler, C.; Dekant, W. (1998). Biotransformation,
excretion, and nephrotoxicity of the hexachlorobutadiene metabolite (E)-N-acetyl-S-
(1,2,3,4, 4-pentachlorobutadienyl)-L-cysteine sulfoxide. Chem Res Toxicol, 11, 750-757.
http://dx.doi.org/10.1021/tx970216n
12
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Bjornsson, H. T.; Sigurdsson, M. I.; Fallin, M. D.; Irizarry, R. A.; Aspelund, T.; Cui, H., . . .
Feinberg, A. P. (2008). Intra-individual change over time in DNA methylation with
familial clustering. JAMA, 299, 2877-2883. http://dx.doi.org/10.1001/jama.299.24.2877
Blackburn, A.; Coggan, M.; Tzeng, H.; Lantum, H.; Polekhina, G.; Parker, M., . . . Board, P.
(2001). GSTZld: a new allele of glutathione transferase zeta and maleylacetoacetate
isomerase. Pharmacogenetics, 11, 671-678.
http://www.ncbi.nlm.nih.gov/pubmed/11692075
Blackburn, A.; Tzeng, H.; Anders, M.; Board, P. (2000). Discovery of a functional
polymorphism in human glutathione transferase zeta by expressed sequence tag database
analysis. Pharmacogenetics, 10, 49-57. http://www.ncbi.nlm.nih.gov/pubmed/10739172
Blackshaw, A. W. (1970). Histochemical localization of testicular enzyme. In A. D. Johnson, W.
R. Gomes & N. L. VanDemark (Eds.), The testis (pp. 73-173). New York, NY: Academic
Press.
Blain, L.; Lachapelle, P.; Molotchnikoff, S. (1994). Electroretinal responses are modified by
chronic exposure to trichloroethylene. Neurotoxicology, 15, 627-631.
http://www.ncbi.nlm.nih.gov/pubmed/7854598
Blair, A.; Haas, T.; Prosser, R.; Morrissette, M.; Blackman, K.; Grauman, D., . . . Moran, F.
(1989). Mortality among United States Coast Guard marine inspectors. Arch Environ
Health, 44, 150 - 156.
Blair, A.; Hartge, P.; Stewart, P. A.; McAdams, M.; Lubin, J. (1998). Mortality and cancer
incidence of aircraft maintenance workers exposed to trichloroethylene and other organic
solvents and chemicals: Extended follow-up. Occup Environ Med, 55, 161-171.
http://dx.doi.Org/10.l 136/oem.55.3.161
Blair, A.; Linos, A.; Stewart, P. A.; Burmeister, L. F.; Gibson, R.; Everett, G., . . . Cantor, K. P.
(1993). Evaluation of risks for non-Hodgkin's lymphoma by occupation and industry
exposures from a case-control study. Am J Ind Med, 23, 301-312.
http://dx.doi.org/10.1002/ajim.4700230207
Blake, J. E.; Compton, K. V.; Schmidt, W.; Orrego, H. (1988). Accuracy of death certificates in
the diagnosis of alcoholic liver cirrhosis. Alcohol Clin Exp Res, 12, 168-172.
http://www.ncbi.nlm.nih.gov/pubmed/3279851
Blake, M. J.; Castro, L.; Leeder, J. S.; Kearns, G. L. (2005). Ontogeny of drug metabolizing
enzymes in the neonate. Semin Fetal Neonatal Med, 10, 123-138.
http://dx.doi.Org/10.1016/j.siny.2004.l 1.001
Blatter, B. M.; Roeleveld, N.; Zielhuis, G. A.; Verbeek, A. L. (1997). Assessment of
occupational exposure in a population based case-control study: comparing postal
questionnaires with personal interviews. Occup Environ Med, 54, 54-59.
http://www.ncbi.nlm.nih.gov/pubmed/9072035
Blechacz, B. R. and Gores, G. J. (2008). Cholangiocarcinoma. Clin Liver Dis, 12, 131-150, ix.
http://dx.doi.Org/10.1016/j.cld.2007.l 1.003
Bloemen, L. and Tomenson, J. (1995). Increased incidence of renal cell tumours in a cohort of
cardboard workers exposed to trichloroethylene. Arch Toxicol, 70, 129-133.
http://www.ncbi.nlm.nih.gov/pubmed/8773187
Bloemen, L.; Monster, A.; Kezic, S.; Commandeur, J.; Veulemans, H.; Vermeulen, N.; Wilmer,
J. (2001). Study on the cytochrome P-450- and glutathione-dependent biotransformation
13
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
of trichloroethylene in humans. Int Arch Occup Environ Health, 74, 102-108.
http://www.ncbi.nlm.nih.gov/pubmed/11317702
Blossom, S. J. and Doss, J. C. (2007). Trichloroethylene alters central and peripheral immune
function in autoimmune-prone MRL(+/+) mice following continuous developmental and
early life exposure. J Immunotoxicol, 4, 129-141.
http://dx.doi.org/10.1080/15476910701337035
Blossom, S. J. and Gilbert, K. M. (2006). Exposure to a metabolite of the environmental
toxicant, trichloroethylene, attenuates CD4+ T cell activation-induced cell death by
metalloproteinase-dependent FasL shedding. Toxicol Sci, 92, 103-114.
http://dx.doi.org/10.1093/toxsci/klj212
Blossom, S. J.; Doss, J. C.; Gilbert, K. M. (2007). Chronic exposure to a trichloroethylene
metabolite in autoimmune-prone MRL+/+ mice promotes immune modulation and
alopecia. Toxicol Sci, 95, 401-411. http://dx.doi.org/10.1093/toxsci/kfll49
Blossom, S. J.; Doss, J. C.; Hennings, L. J.; Jernigan, S.; Melnyk, S.; James, S. J. (2008).
Developmental exposure to trichloroethylene promotes CD4+ T cell differentiation and
hyperactivity in association with oxidative stress and neurobehavioral deficits in MRL+/+
mice. Toxicol Appl Pharmacol, 231, 344-353.
http://dx.doi.Org/10.1016/j.taap.2008.05.009
Blossom, S. J.; Pumford, N. R.; Gilbert, K. M. (2004). Activation and attenuation of apoptosis of
CD4+ T cells following in vivo exposure to two common environmental toxicants,
trichloroacetaldehyde hydrate and trichloroacetic acid. J Autoimmun, 23, 211-220.
http://dx.doi.Org/10.1016/j.jaut.2004.06.007
Board, P.; Chelvanayagam, G.; Jermiin, L.; Tetlow, N.; Tzeng, H.; Anders, M.; Blackburn, A.
(2001). Identification of novel glutathione transferases and polymorphic variants by
expressed sequence tag database analysis. Drug Metab Dispos, 29, 544-547.
http://www.ncbi.nlm.nih.gov/pubmed/11259348
Boerrigter, M. E. (2004). Mutagenicity of the peroxisome proliferators clofibrate, Wyeth 14,643
and di-2-ethylhexyl phthalate in the lacZ plasmid-based transgenic mouse mutation
assay. Journal of Carcinogenesis, 3, 7. http://dx.doi.org/10.1186/1477-3163-3-7
Boffetta, P. and de Vocht, F. (2007). Occupation and the risk of non-Hodgkin lymphoma. Cancer
Epidemiol Biomarkers Prev, 16, 369-372. http://dx.doi.org/10.1158/1055-9965.EPI-06-
1055
Bogen, K. T. (1990). Uncertainty in Environmental Health Risk Assessment. New York: Garlend
Publishing Company.
Bogen, K. T.; BW Jr, C.; Machicao, L. K. (1992). Dermal absorption of dilute aqueous
chloroform, trichloroethylene, and tetrachloroethylene in hairless guinea pigs. Toxicol
Sci, 18, 30-39. http://dx.doi.org/10.1016/0272-0590(92)90192-K
Boice, J. D.; Cohen, S. S.; Mumma, M. T.; Dupree Ellis, E.; Eckerman, K. F.; Leggett, R. W., . .
. Henderson, B. E. (2006). Mortality among radiation workers at Rocketdyne (Atomics
International), 1948-1999. RadiatRes, 166, 98-115. http://dx.doi.Org/10.1667/RR3582.l
Boice, J. D.; Marano, D. E.; Cohen, S. S.; Mumma, M. T.; Blot, W. J.; Brill, A. B., . . .
McLaughlin, J. K. (2006). Mortality among Rocketdyne workers who tested rocket
engines, 1948-1999. J Occup Environ Med, 48, 1070-1092.
http://dx.doi.org/10.1097/01.jom.0000240661.33413.b5
14
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Boice, J.; Marano, D.; Fryzek, J.; Sadler, C.; McLaughlin, J. (1999). Mortality among aircraft
manufacturing workers. Occup Environ Med, 56, 581-597.
http://dx.doi.Org/10.l 136/oem. 56.9.581
Bois, F. (2000). Statistical analysis of Clewell et al. PBPK model of trichloroethylene kinetics.
Environ Health Perspect, 108, 307-316. http://www.ncbi.nlm.nih.gov/pubmed/10807560
Bois, F. (2000). Statistical analysis of Fisher et al. PBPK model of trichloroethylene kinetics.
Environ Health Perspect, 108 Suppl 2, 275-282.
http://www.ncbi.nlm.nih.gov/pubmed/10807558
Bolm-Audorff, U.; Ahrens, W.; Jockel, K.-H.; Molik, B.; Greiser, E.; Timm, J. (1988).
Esperience with supplementary questionnaires in a lung cancer case-referent study. In D.
Henon & M. Goldberg (Eds.), Proceedings of the workshop on methodology of
assessment of occupational exposures in the context of epidemiological detection of
cancer risks (pp. 117-168). Luxembourg
Bolt, H. M. and Filser, J. G. (1977). Irreversible binding of chlorinated ethylenes to
macromolecules. Environ Health Perspect, 21, 107-112.
http://www.ncbi.nlm.nih.gov/pubmed/612434
Bolt, H.; Lammert, M.; Selinski, S.; Briining, T. (2004). Urinary alphal-microglobulin excretion
as biomarker of renal toxicity in trichloroethylene-exposed persons. Int Arch Occup
Environ Health, 77, 186-190. http://dx.doi.org/10.1007/s00420-003-0500-3
Bombail, V.; Moggs, J. G.; Orphanides, G. (2004). Perturbation of epigenetic status by toxicants.
Toxicol Lett, 149, 51-58. http://dx.doi.Org/10.1016/j.toxlet.2004.01.003
Bonatti, S.; Cavalieri, Z.; Viaggi, S.; Abbondandolo, A. (1992). The analysis of 10 potential
spindle poisons for their ability to induce CREST-positive micronuclei in human diploid
fibroblasts. Mutagenesis, 7, 111-114. http://www.ncbi.nlm.nih.gov/pubmed/1579065
Boorman, G. A.; Dellarco, V.; Dunnick, J. K.; Chapin, R. E.; Hunter, S.; Hauchman, F., . . . Sills,
R. C. (1999). Drinking water disinfection byproducts: review and approach to toxicity
evaluation. Environ Health Perspect, 107, 207-217.
http://www.ncbi.nlm.nih.gov/pubmed/10229719
Borzio, M.; Fargion, S.; Borzio, F.; Fracanzani, A. L.; Croce, A. M.; Stroffolini, T., . . . Roncalli,
M. (2003). Impact of large regenerative, low grade and high grade dysplastic nodules in
hepatocellular carcinoma development. J Hepatol, 39, 208-214.
http://www.ncbi.nlm.nih.gov/pubmed/12873817
Boublik, T.; Fried, V.; Hala, E. (1984). The vapour pressures of pure substances: selected values
of the temperature dependence of the vapour pressures of some pure substances in the
normal and low pressure region. Amsterdam, The Netherlands: Elsevier.
Bove, F. J. (1996). Public drinking water contamination and birthweight, prematurity, fetal
deaths, and birth defects. Toxicol Ind Health, 12, 255-266.
http://www.ncbi.nlm.nih.gov/pubmed/8794538
Bove, F. J.; Fulcomer, M. C.; Klotz, J. B.; Esmart, J.; Dufficy, E. M.; Savrin, J. E. (1995). Public
drinking water contamination and birth outcomes. Am J Epidemiol, 141, 850-862.
Bovenzi, M.; Barbone, F.; Pisa, F.; Betta, A.; Romeo, L.; Tonello, A., . . . Caramaschi, P. (2004).
A case-control study of occupational exposures and systemic sclerosis. Int Arch Occup
Environ Health, 77, 10-16. http://dx.doi.org/10.1007/s00420-003-0462-5
Bovo, R.; Aimoni, C.; Martini, A. (2006). Immune-mediated inner ear disease. Acta Otolaryngol,
126, 1012-1021. http://dx.doi.org/10.1080/00016480600606723
15
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
Bowler, R. M.; Mergler, D.; Huel, G.; Harrison, R.; Cone, J. (1991). Neuropsychological
impairment among former microelectronics workers. Neurotoxicology, 12, 87-103.
http://www.ncbi.nlm.nih.gov/pubmed/2014071
Box, G. E. P.; Hunter, W. G.; Hunter, J. S. (1978). Statistics for experimenters: an introduction
to design, data analysis, and model building. New York, NY: Wiley.
Boyault, S.; Rickman, D. S.; de Reynies, A.; Balabaud, C.; Rebouissou, S.; Jeannot, E., . . .
Zucman-Rossi, J. (2007). Transcriptome classification of HCC is related to gene
alterations and to new therapeutic targets. Hepatology, 45, 42-52.
http://dx.doi.org/10.1002/hep.21467
Boyer, A.; Finch, W.; Runyan, R. (2000). Trichloroethylene inhibits development of embryonic
heart valve precursors in vitro. Toxicol Sci, 53, 109-117.
http://www.ncbi.nlm.nih.gov/pubmed/10653528
Boyes, W. K.; Bercegeay, M.; Krantz, T.; Evans, M.; Benignus, V.; Simmons, J. E. (2005).
Momentary brain concentration of trichloroethylene predicts the effects on rat visual
function. Toxicol Sci, 87, 187-196. http://dx.doi.org/10.1093/toxsci/kfi242
Boyes, W. K.; Bushnell, P. J.; Crofton, K. M.; Evans, M.; Simmons, J. E. (2000). Neurotoxic and
pharmacokinetic responses to trichloroethylene as a function of exposure scenario.
Environ Health Perspect, 108, 317-322. http://www.ncbi.nlm.nih.gov/pubmed/10807561
Boyes, W.; Bercegeay, M.; Ali, J.; Krantz, T.; McGee, J.; Evans, M., . . . Simmons, J. (2003).
Dose-based duration adjustments for the effects of inhaled trichloroethylene on rat visual
function. Toxicol Sci, 76, 121-130. http://dx.doi.org/10.1093/toxsci/kfg213
Boyle, C. A. and Brann, E. A. (1992). Proxy respondents and the validity of occupational and
other exposure data. The Selected Cancers Cooperative Study Group. Am J Epidemiol,
136, 712-721. http://www.ncbi.nlm.nih.gov/pubmed/1442737
Bradford, B. U.; Kono, H.; Isayama, F.; Kosyk, O.; Wheeler, M. D.; Akiyama, T. E., . . . Rusyn,
I. (2005). Cytochrome P450 CYP2E1, but not nicotinamide adenine dinucleotide
phosphate oxidase, is required for ethanol-induced oxidative DNA damage in rodent
liver. Hepatology, 41, 336-344. http://dx.doi.org/10.1002/hep.20532
Braissant, O. and Wahli, W. (1998). Differential expression of peroxisome proliferator-activated
receptor-alpha, -beta, and -gamma during rat embryonic development. Endocrinology,
139, 2748-2754. http://www.ncbi.nlm.nih.gov/pubmed/9607781
Branda, M. and Wands, J. R. (2006). Signal transduction cascades and hepatitis B and C related
hepatocellular carcinoma. Hepatology, 43, 891-902. http://dx.doi.org/10.1002/hep.21196
Brauch, H.; Weirich, G.; Hornauer, M.; Storkel, S.; Wohl, T.; Briining, T. (1999).
Trichloroethylene exposure and specific somatic mutations in patients with renal cell
carcinoma. J Natl Cancer Inst, 91, 854-861.
http://www.ncbi.nlm.nih.gov/pubmed/10340905
Brauch, H.; Weirich, G.; Klein, B.; Rabstein, S.; Bolt, H.; Briining, T. (2004). VHL mutations in
renal cell cancer: does occupational exposure to trichloroethylene make a difference.
Toxicol Lett, 151, 301-310. http://dx.doi.Org/10.1016/j.toxlet.2003.12.074
Brautbar, N. and Williams, J. (2002). Industrial solvents and liver toxicity: risk assessment, risk
factors and mechanisms. Int J Hyg Environ Health, 205, 479-491.
http://www.ncbi.nlm.nih.gov/pubmed/12455270
16
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Brenner, H. and Gefeller, O. (1993). Use of the positive predictive value to correct for disease
misclassification in epidemiologic studies. Am J Epidemiol, 138, 1007-1015.
http://www.ncbi.nlm.nih.gov/pubmed/8256775
Breslow, N. E. and Day, N. E. (1980). Statistical methods in cancer research: Volume 1: The
analysis of case-control studies (Vol. 32). Lyon, France: International Agency for
Research on Cancer.
Breslow, N. E. and Day, N. E. (1987). Statistical methods in cancer research, volume 2 - the
design and analysis of cohort studies. Oxford, UK: World Health Organization,
International Agency for Research on Cancer.
Bringmann, G.; God, R.; Feineis, D.; Wesemann, W.; Riederer, P.; Rausch, W., . . . Sontag, K.
(1995). The TaClo concept: l-trichloromethyl-l,2,3,4-tetrahydro-beta-carboline (TaClo),
a new toxin for dopaminergic neurons. J Neural Transm Suppl, 46, 235-244.
http://www.ncbi.nlm.nih.gov/pubmed/8821060
Briving, C.; Jacobson, I.; Hamberger, A.; Kjellstrand, P.; Haglid, K. G.; Rosengren, L. E. (1986).
Chronic effects of perchloroethylene and trichloroethylene on the gerbil brain amino
acids and glutathione. Neurotoxicology, 7, 101-108.
http://www.ncbi.nlm.nih.gov/pubmed/2872637
Brodsky, W. Y. and Uryvaeva, I. V. (1977). Cell polyploidy: its relation to tissue growth and
function. IntRev Cytol, 50, 275-332. http://www.ncbi.nlm.nih.gov/pubmed/332657
Bronley-DeLancey, A.; McMillan, D.; McMillan, J.; Jollow, D.; Mohr, L.; Hoel, D. (2006).
Application of cryopreserved human hepatocytes in trichloroethylene risk assessment:
relative disposition of chloral hydrate to trichloroacetate and trichloroethanol. Environ
Health Perspect, 114, 1237-1242. http://www.ncbi.nlm.nih.gov/pubmed/16882532
Bronzetti, G.; Corsi, C.; Nieri, R. (1980). [Use of the D7 strain of S. cerevisiae in the
determination of environmental risk. 1. Genetic effects of trichloroethylene]. Boll Soc Ital
Biol Sper, 56, 1315-1321. http://www.ncbi.nlm.nih.gov/pubmed/7006634
Bronzetti, G.; Galli, A.; Corsi, C.; Cundari, E.; Del Carratore, R.; Nieri, R.; Paolini, M. (1984).
Genetic and biochemical investigation on chloral hydrate in vitro and in vivo. Mutat Res,
141, 19-22. http://www.ncbi.nlm.nih.gov/pubmed/6384771
Bross, G.; DiFranceisco, D.; Desmond, M. E. (1983). The effects of low dosages of
trichloroethylene on chick development. Toxicology, 28, 283-294.
http://www.ncbi.nlm.nih.gov/pubmed/6648978
Brown, H. S.; Bishop, D. R.; Rowan, C. A. (1984). The role of skin absorption as a route of
exposure for volatile organic compounds (VOCs) in drinking water. Am J Public Health,
74, 479-484. http://www.ncbi.nlm.nih.gov/pubmed/6711723
Brown, R. P.; Delp, M. D.; Lindstedt, S. L.; Rhomberg, L. R.; Beliles, R. P. (1997).
Physiological parameter values for physiologically based pharmacokinetic models.
Toxicol Ind Health, 13, 407-484. http://dx.doi.org/10.1177/074823379701300401
Brownson, R. C. (1988). A case-control study of renal cell carcinoma in relation to occupation,
smoking, and alcohol consumption. Arch Environ Health, 43, 238-241.
http://www.ncbi.nlm.nih.gov/pubmed/3382249
Briining, T.; Lammert, M.; Kempkes, M.; Thier, R.; Golka, K.; Bolt, H. (1997). Influence of
polymorphisms of GSTM1 and GSTT1 for risk of renal cell cancer in workers with long-
term high occupational exposure to trichloroethene. Arch Toxicol, 71, 596-599.
http://www.ncbi.nlm.nih.gov/pubmed/9285043
17
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
Briining, T.; Mann, H.; Melzer, H.; Sundberg, A.; Bolt, H. (1999). Pathological excretion
patterns of urinary proteins in renal cell cancer patients exposed to trichloroethylene.
Occup Med (Lond), 49, 299-305. http://www.ncbi.nlm.nih.gov/pubmed/10628055
Briining, T.; Pesch, B.; Wiesenhiitter, B.; Rabstein, S.; Lammert, M.; Baumiiller, A.; Bolt, H.
(2003). Renal cell cancer risk and occupational exposure to trichloroethylene: results of a
consecutive case-control study in Arnsberg, Germany. Am J Ind Med, 43, 274-285.
http://dx.doi.org/10.1002/ajim.10185
Briining, T.; Sundberg, A.; Birner, G.; Lammert, M.; Bolt, H.; Appelkvist, E., . . . Dallner, G.
(1999). Glutathione transferase alpha as a marker for tubular damage after
trichloroethylene exposure. Arch Toxicol, 73, 246-254.
http://www.ncbi.nlm.nih.gov/pubmed/10463390
Briining, T.; Vamvakas, S.; Makropoulos, V.; Birner, G. (1998). Acute intoxication with
trichloroethene: clinical symptoms, toxicokinetics, metabolism, and development of
biochemical parameters for renal damage. Toxicol Sci, 41, 157-165.
http://dx.doi.org/10.1006/toxs.1997.2401
Briining, T.; Weirich, G.; Hornauer, M.; Hofler, H.; Brauch, H. (1997). Renal cell carcinomas in
trichloroethene (TRI) exposed persons are associated with somatic mutations in the von
Hippel-Lindau (VHL) tumour suppressor gene. Arch Toxicol, 71, 332-335.
http://www.ncbi.nlm.nih.gov/pubmed/9137812
Brzezinski, M. R.; Boutelet-Bochan, H.; Person, R. E.; Fantel, A. G.; Juchau, M. R. (1999).
Catalytic activity and quantitation of cytochrome P-450 2E1 in prenatal human brain. J
Pharmacol Exp Ther, 289, 1648-1653. http://www.ncbi.nlm.nih.gov/pubmed/10336564
Buben, J. A. and O'Flaherty, E. J. (1985). Delineation of the role of metabolism in the
hepatotoxicity of trichloroethylene and perchloroethylene: A dose-effect study. Toxicol
Appl Pharmacol, 78, 105-122. http://www.ncbi.nlm.nih.gov/pubmed/2994252
Buckpitt, A.; Chang, A.-M.; Weir, A.; Van Winkle, L.; Duan, X.; Philpot, R.; Plopper, C. (1995).
Relationship of cytochrome P450 activity to Clara cell cytotoxicity IV Metabolism of
naphthalene and naphthalene oxide in microdissected airways from mice, rats, and
hamsters. Mol Pharmacol, 47, 74-81. http://www.ncbi.nlm.nih.gov/pubmed/7838135
Budavari, S., (Ed.). (1996). The Merck Index: An Encyclopedia of Chemicals, Drugs, and
Biologicals (12th ed.). Rahway, NJ: Merck & Co., Inc.
Bull, R. J. (2000). Mode of action of liver tumor induction by trichloroethylene and its
metabolites, trichloroacetate and dichloroacetate. Environ Health Perspect, 108, 241-259.
http://www.ncbi.nlm.nih.gov/pubmed/10807555
Bull, R. J. (2004). Trichloroethylene and liver tumors in mice. Symposium on New Scientific
Research Related to Health Effects of TCE. Washington, DC.
Bull, R. J.; Orner, G. A.; Cheng, R. S.; Stillwell, L.; Stauber, A. J.; Sasser, L. B., . . . Thrall, B.
D. (2002). Contribution of dichloroacetate and trichloroacetate to liver tumor induction in
mice by trichloroethylene. Toxicol Appl Pharmacol, 182, 55-65.
http://dx.doi.org/10.1006/taap.2002.9427
Bull, R. J.; Sanchez, I. M.; Nelson, M. A.; Larson, J. L.; Lansing, A. J. (1990). Liver tumor
induction in B6C3F1 mice by dichloroacetate and trichloroacetate. Toxicology, 63, 341-
359. http://dx.doi.org/10.1016/0300-483X(90)90195-M
18
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Bull, R. J.; Sasser, L. B.; Lei, X. C. (2004). Interactions in the tumor-promoting activity of
carbon tetrachloride, trichloroacetate, and dichloroacetate in the liver of male B6C3F1
mice. Toxicology, 199, 169-183. http://dx.doi.Org/10.1016/j.tox.2004.02.018
Bunce, N. J. and Schneider, U. A. (1994). Chemical lifetimes of chlorinated aliphatic priority
pollutants in the Canadian troposphere. J Photochem Photobiol A, 81, 93 - 101.
Burg, J. and Gist, G. (1997). The potential impact on women from environmental exposures. J
Worn Health Gend Base Med, 6, 159-161.
http://www.ncbi.nlm.nih.gov/pubmed/9140851
Burg, J. and Gist, G. (1999). Health effects of environmental contaminant exposure: an intrafile
comparison of the Trichloroethylene Subregistry. Arch Environ Health, 54, 231-241.
http://www.ncbi.nlm.nih.gov/pubmed/10433181
Burg, J. R.; Gist, G.; Allred, S. L.; Rudtke, T. M.; Pallos, L. L.; Cusack, C. D. (1995). The
national exposure registry - morbidity analyses of noncancer outcomes from the
trichloroethylene subregistry baseline data. 4, 237 - 257.
Burmaster, D. E. (1982). The new pollution: groundwater contamination. Environment, 24, 6-
13;33-36.
Bushnell, P. (1997). Concentration-time relationships for the effects of inhaled trichloroethylene
on signal detection behavior in rats. Fundam Appl Toxicol, 36, 30-38.
http://www.ncbi.nlm.nih.gov/pubmed/9073464
Bushnell, P. and Oshiro, W. (2000). Behavioral components of tolerance to repeated inhalation
of trichloroethylene (TCE) in rats. Neurotoxicol Teratol, 22, 221-229.
http://www.ncbi.nlm.nih.gov/pubmed/10758351
Butler, T. C. (1949). Metabolic transformations of trichloroethylene. J Pharmacol Exp Ther, 97,
84-92.
Buxton, P. and Hayward, M. (1967). Polyneuritis cranialis associated with industrial
trichloroethylene poisoning. J Neurol Neurosurg Psychiatry, 30, 511-518.
http://www.ncbi.nlm.nih.gov/pubmed/4170824
Byers, V. S.; Levin, A. S.; Ozonoff, D. M.; Baldwin, R. W. (1988). Association between clinical
symptoms and lymphocyte abnormalities in a population with chronic domestic exposure
to industrial solvent-contaminated domestic water supply and a high incidence of
leukaemia. Cancer Immunol Immunother, 27, 77-81.
http://www.ncbi.nlm.nih.gov/pubmed/3260823
Cai, H. and Guengerich, F. (2001). Reaction of trichloroethylene and trichloroethylene oxide
with cytochrome P450 enzymes: Inactivation and sites of modification. Chem Res
Toxicol, 14, 451-458. http://www.ncbi.nlm.nih.gov/pubmed/11304134
Cai, H. and Guengerich, F. (2001). Reaction of trichloroethylene oxide with proteins and dna:
instability of adducts and modulation of functions. Chem Res Toxicol, 14, 54-61.
http://www.ncbi.nlm.nih.gOv/pubmed/l 1170508
Cai, H. and Guengerich, F. P. (1999). Mechanisms of aqueous decomposition of
trichloroethylene oxide. 121, 11656-11663.
Cai, P.; Konig, R.; Boor, P. J.; Kondraganti, S.; Kaphalia, B. S.; Khan, M. F.; Ansari, G. A. S.
(2008). Chronic exposure to trichloroethene causes early onset of SLE-like disease in
female MRL +/+ mice. Toxicol Appl Pharmacol, 228, 68-75.
http://dx.doi.Org/10.1016/j.taap.2007.ll.031
19
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
Cai, P.; Konig, R.; Khan, M. F.; Kaphalia, B. S.; Ansari, G. A. (2007). Differential immune
responses to albumin adducts of reactive intermediates of trichloroethene in MRL+/+
mice. Toxicol Appl Pharmacol, 220, 278-283.
http://dx.doi.Org/10.1016/j.taap.2007.01.020
Cai, P.; Konig, R.; Khan, M. F.; Qiu, S.; Kaphalia, B. S.; Ansari, G. A. (2006). Autoimmune
response in MRL+/+ mice following treatment with dichloroacetyl chloride or
dichloroacetic anhydride. Toxicol Appl Pharmacol, 216, 248-255.
http://dx.doi.Org/10.1016/j.taap.2006.05.010
Cakmak, A.; Ekici, A.; Ekici, M.; Arslan, M.; Iteginli, A.; Kurtipek, E.; Kara, T. (2004).
Respiratory findings in gun factory workers exposed to solvents. Respir Med, 98, 52-56.
http://www.ncbi.nlm.nih.gov/pubmed/14959814
Caldwell, J. and Keshava, N. (2006). Key issues in the modes of action and effects of
trichloroethylene metabolites for liver and kidney tumorigenesis. Environ Health
Perspect, 114, 1457-1463. http://www.ncbi.nlm.nih.gov/pubmed/16966105
Caldwell, J. C.; Keshava, N.; Evans, M. V. (2008). Difficulty of mode of action determination
for trichloroethylene: An example of complex interactions of metabolites and other
chemical exposures. Environ Mol Mutagen, 49, 142-154.
http://dx.doi.org/10.1002/em.20350
Caldwell, P. T.; Manziello, A.; Howard, J.; Palbykin, B.; Runyan, R. B.; Selmin, O. (2010).
Gene expression profiling in the fetal cardiac tissue after folate and low-dose
trichloroethylene exposure. Birth Defects Res A Clin Mol Teratol, 88, 111-127.
http://dx.doi.org/10.1002/bdra.20631
Caldwell, P. T.; Thorne, P. A.; Johnson, P. D.; Boitano, S.; Runyan, R. B.; Selmin, O. (2008).
Trichloroethylene disrupts cardiac gene expression and calcium homeostasis in rat
myocytes. Toxicol Sci, 104, 135-143. http://dx.doi.org/10.1093/toxsci/kfn078
Call en, D. F.; Wolf, C. R.; Philpot, R. M. (1980). Cytochrome P-450 mediated genetic activity
and cytotoxicity of seven halogenated aliphatic hydrocarbons in Saccharomyces
cerevisiae. Mutat Res Genet Toxicol Environ Mutagen, 77, 55-63.
http://dx.doi.org/10.1016/0165-1218(80)90120-2
Calvisi, D. F.; Factor, V. M.; Ladu, S.; Conner, E. A.; Thorgeirsson, S. S. (2004). Disruption of
beta-catenin pathway or genomic instability define two distinct categories of liver cancer
in transgenic mice. Gastroenterology, 126, 1374-1386.
http://www.ncbi.nlm.nih.gov/pubmed/15131798
Calvisi, D. F.; Ladu, S.; Conner, E. A.; Factor, V. M.; Thorgeirsson, S. S. (2004). Disregulation
of E-cadherin in transgenic mouse models of liver cancer. Lab Invest, 84, 1137-1147.
http://dx.doi.org/10.1038/labinvest.3700147
Calvisi, D. F.; Ladu, S.; Gorden, A.; Farina, M.; Conner, E. A.; Lee, J. S., . . . Thorgeirsson, S. S.
(2006). Ubiquitous activation of Ras and Jak/Stat pathways in human HCC.
Gastroenterology, 130, 1117-1128. http://dx.doi.Org/10.1053/j.gastro.2006.01.006
Calvisi, D. F.; Ladu, S.; Gorden, A.; Farina, M.; Lee, J. S.; Conner, E. A., . . . Thorgeirsson, S. S.
(2007). Mechanistic and prognostic significance of aberrant methylation in the molecular
pathogenesis of human hepatocellular carcinoma. J Clin Invest, 117, 2713-2722.
http://dx.doi.org/10.1172/JCI31457
20
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Campo, P.; Blachere, V.; Payan, J. P.; Cossec, B.; Ducos, P. (2006). Toluene toxicokineties and
metabolism parameters in the rat and guinea pig. Environ Toxicol Pharmacol, 21, 276-
282.
Campo, P.; Loquet, G.; Blachere, V.; Roure, M. (1999). Toluene and styrene intoxication route
in the rat cochlea. Neurotoxicol Teratol, 21, 427-434.
http://www.ncbi.nlm.nih.gov/pubmed/10440486
Cao, K.; Hollenbach, J.; Shi, X.; Shi, W.; Chopek, M.; Fernandez-Vina, M. A. (2001). Analysis
of the frequencies of HLA-A, B, and C alleles and haplotypes in the five major ethnic
groups of the United States reveals high levels of diversity in these loci and contrasting
distribution patterns in these populations. Hum Immunol, 62, 1009-1030.
http://www.ncbi.nlm.nih.gov/pubmed/11543903
Cappaert, N. L. M.; Klis, S. F. L.; Baretta, A. B.; Muijser, H.; Smoorenburg, G. F. (2000). Ethyl
benzene-induced ototoxicity in rats: A dose-dependent mid-frequency hearing loss. J
Assoc Res Otolaryngol, 1, 292-299. http://dx.doi.org/10.1007/sl01620010050
Cappaert, N. L. M.; Klis, S. F. L.; Muijser, H.; de Groot, J. C. M. J.; Kulig, B. M.; Smoorenburg,
G. F. (1999). The ototoxic effects of ethyl benzene in rats. Hear Res, 137, 91-102.
http://dx.doi.org/10.1016/S0378-5955(99)00141-0
Carney, E. W.; Thorsrud, B. A.; Dugard, P. H.; Zablotny, C. L. (2006). Developmental toxicity
studies in Crl:CD (SD) rats following inhalation exposure to trichloroethylene and
perchloroethylene. Birth Defects Res B Dev Reprod Toxicol, 77, 405-412.
http ://dx. doi. org/10.1002/bdrb .20091
Carpenter, S. P.; Lasker, J. M.; Raucy, J. L. (1996). Expression, induction, and catalytic activity
of the ethanol-inducible cytochrome P450 (CYP2E1) in human fetal liver and
hepatocytes. Mol Pharmacol, 49, 260-268.
http://www.ncbi.nlm.nih.gov/pubmed/8632758
Carrieri, M.; Magosso, D.; Piccoli, P.; Zanetti, E.; Trevisan, A.; Bartolucci, G. B. (2007). Acute,
nonfatal intoxication with trichloroethylene. Arch Toxicol, 81, 529-532.
http://dx.doi.org/10.1007/s00204-007-0180-y
Carter, J. H.; Carter, H. W.; DeAngelo, A. B. (1995). Biochemical, pathologic and morphometric
alterations induced in male B6C3F1 mouse liver by short-term exposure to dichloroacetic
acid. Toxicol Lett, 81, 55-71. http://dx.doi.org/10.1016/0378-4274(95)03409-9
Carter, J. H.; Carter, H. W.; Deddens, J. A.; Hurst, B. M.; George, M. H.; DeAngelo, A. B.
(2003). A 2-year dose-response study of lesion sequences during hepatocellular
carcinogenesis in the male B6C3F(1) mouse given the drinking water chemical
dichloroacetic acid. Environ Health Perspect, 111, 53-64.
http://dx.doi.org/10.1289/ehp.5442
CDC. (Centers for Disease Control and Prevention). (2007). National Vital Statistics Reports:
Deaths: Final data for 2004. Hyattsville, MD. Retrieved from
http://www.cdc.gov/nchs/products/nvsr.htm.
CDHS. (1986). Organic chemical contamination of large public water systems in California.
Chakrabarti, S. K. and Tuchweber, B. (1988). Studies of acute nephrotoxic potential of
trichloroethylene in Fischer 344 rats. J Toxicol Environ Health, 23, 147-158.
http://www.ncbi.nlm.nih.gov/pubmed/3343692
CHALUPA, B.; SYNKOVA, J.; SEVCIK, M. (1960). The assessment of
electroencephalographic changes and memory disturbances in acute intoxications with
21
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
industrial poisons. Br J Ind Med, 17, 238-241.
http://www.ncbi.nlm.nih.gov/pubmed/13692202
Chang, L. W.; Daniel, F. B.; DeAngelo, A. B. (1992). Analysis of DNA strand breaks induced in
rodent liver in vivo, hepatocytes in primary culture, and a human cell line by chlorinated
acetic acids and chlorinated acetaldehydes. Environ Mol Mutagen, 20, 277-288.
http://dx.doi.org/10.1002/em.2850200406
Chang, Y.; Tai, C.; Lin, R.; Yang, S.; Chen, C.; Shih, T.; Liou, S. (2003). A proportionate cancer
morbidity ratio study of workers exposed to chlorinated organic solvents in Taiwan. Ind
Health, 41, 77-87. http://www.ncbi.nlm.nih.gov/pubmed/12725467
Chang, Y.; Tai, C.; Yang, S.; Chen, C.; Shih, T.; Lin, R.; Liou, S. (2003). A cohort mortality
study of workers exposed to chlorinated organic solvents in Taiwan. Ann Epidemiol, 13,
652-660. http://dx.doi.Org/l0.1016/S1047-2797(03)00038-3
Chang, Y.; Tai, C.; Yang, S.; Lin, R.; Sung, F.; Shih, T.; Liou, S. (2005). Cancer incidence
among workers potentially exposed to chlorinated solvents in an electronics factory. J
Occup Health, 47, 171-180. http://www.ncbi.nlm.nih.gov/pubmed/15824483
Channel, S. R.; Latendresse, J. R.; Kidney, J. K.; Grabau, J. H.; Lane, J. W.; Steel-Goodwin, L.;
Gothaus, M. C. (1998). A subchronic exposure to trichloroethylene causes lipid
peroxidation and hepatocellular proliferation in male B6C3F1 mouse liver. Toxicol Sci,
43, 145-154. http://dx.doi.Org/10.1093/toxsci/43.2.145
Chapin, R. E. and Sloane, R. A. (1997). Reproductive assessment by continuous breeding:
evolving study design and summaries of ninety studies. Environ Health Perspect, 105
Suppl 1, 199-205. http://www.ncbi.nlm.nih.gov/pubmed/9114287
Chapin, R. E.; Norton, R. M.; Popp, J. A.; Bus, J. S. (1982). The effects of 2,5-hexanedione on
reproductive hormones and testicular enzyme activities in the F-344 rat. Toxicol Appl
Pharmacol, 62, 262-272. http://www.ncbi.nlm.nih.gov/pubmed/6120587
Charbonneau, M.; Couture, J.; Plaa, G. L. (1991). Inhalation versus oral administration of
acetone: effect of the vehicle on the potentiation of CC14-induced liver injury. Toxicol
Lett, 57, 47-54.
Charbotel, B.; Fevotte, J.; Hours, M.; Martin, J. L.; Bergeret, A. (2005). Case-control study on
renal cell cancer and occupational trichloroethylene exposure in the Arve Valley
(France). Lyon, France: Institut Universitaire de Medecine du Travail.
Charbotel, B.; Fevotte, J.; Hours, M.; Martin, J. L.; Bergeret, A. (2006). Case-control study on
renal cell cancer and occupational exposure to trichloroethylene. Part II: Epidemiological
aspects. Ann Occup Hyg, 50, 777-787. http://dx.doi.org/10.1093/annhyg/mel039
Charbotel, B.; Fevotte, J.; Martin, J.-L.; Bergeret, A. (2009). Cancer du rein et expositions au
trichloroethylene : les valeurs limites d'exposition professionnelle fran9aises en vigueur
sont-elles adaptees ? Rev Epidemiol Sante Publique, 57, 41-47.
http://dx.doi.Org/10.1016/j.respe.2008.09.008
Charbotel, B.; Gad, S.; Caiola, D.; Beroud, C.; Fevotte, J.; Bergeret, A., . . . Richard, S. (2007).
Trichloroethylene exposure and somatic mutations of the VHL gene in patients with
Renal Cell Carcinoma. J Occup Med Toxicol, 2, 13. http://dx.doi.org/10.1186/1745-
6673-2-13
Charles, J. M. and Leeming, N. M. (1998). Chronic dietary toxicity/oncogenicity studies on 2,4-
dichlorophenoxybutyric acid in rodents. Toxicol Sci, 46, 21-30.
http://dx.doi.org/10.1006/toxs.1998.2502
22
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Checkoway, H.; Pearce, N.; Crawford-Brown, D. J. (1989). Research methods in occupational
epidemiology. New York: Oxford University Press.
Chen, C. W. (2000). Biologically based dose-response model for liver tumors induced by
trichloroethylene. Environ Health Perspect, 108 Suppl 2, 335-342.
http://www.ncbi.nlm.nih.gov/pubmed/10807563
Chen, C.; Hennig, G.; Whiteley, H.; Manautou, J. (2002). Protection against acetaminophen
hepatotoxicity by clofibrate pretreatment: role of catalase induction. J Biochem Mol
Toxicol, 16, 227-234. http://dx.doi.org/10.1002/jbt.10043
Chen, X.; Cheung, S. T.; So, S.; Fan, S. T.; Barry, C.; Higgins, J., . . . Brown, P. O. (2002). Gene
expression patterns in human liver cancers. Mol Biol Cell, 13, 1929-1939.
http://dx.doi.org/10.1091/mbc.02-02-0023.
Chen, Y.; Cai, J.; Anders, M.; Stevens, J.; Jones, D. (2001). Role of mitochondrial dysfunction in
S-(l,2-dichlorovinyl)-l-cysteine-induced apoptosis. Toxicol Appl Pharmacol, 170, 172-
180. http://dx.doi.org/10.1006/taap.2000.9107
Chen, Z. Y.; White, C. C.; He, C. Y.; Liu, Y. F.; Eaton, D. L. (1995). Zonal differences in DNA
synthesis activity and cytochrome P450 gene expression in livers of male F344 rats
treated with five nongenotoxic carcinogens. J Environ Pathol Toxicol Oncol, 14, 83-99.
http://www.ncbi.nlm.nih.gov/pubmed/9372837
Cherrie, J.; Kromhout, H.; Semple, S. (2001). The importance of reliable exposure estimates in
deciding whether trichloroethylene can cause kidney cancer. J Cancer Res Clin Oncol,
127, 400-404. http://www.ncbi.nlm.nih.gov/pubmed/11414201
Cheung, C.; Akiyama, T. E.; Ward, J. M.; Nicol, C. J.; Feigenbaum, L.; Vinson, C.; Gonzalez, F.
J. (2004). Diminished hepatocellular proliferation in mice humanized for the nuclear
receptor peroxisome proliferator-activated receptor alpha. Cancer Res, 64, 3849-3854.
http://dx.doi.org/10.1158/0008-5472.CAN-04-0322
Chia, S. E.; Goh, V. H.; Ong, C. N. (1997). Endocrine profiles of male workers with exposure to
trichloroethylene. Am J Ind Med, 32, 217-222. http://dx.doi.org/10.1002/(SICI)1097-
0274(199709)32:3<217: :AID-AJIM6>3.0.CO;2-U
Chia, S.-E.; Ong, C.-N.; Tsakok, M. F.; Ho, A. (1996). Semen parameters in workers exposed to
trichloroethylene. Reprod Toxicol, 10, 295-299. http://dx.doi.org/10.1016/0890-
6238(96)00058-5
Chieco, P.; Moslen, M. T.; Reynolds, E. S. (1981). Effect of administrative vehicle on oral 1,1-
dichloroethylene toxicity. Toxicol Appl Pharmacol, 57, 146-155.
Chiesa, R.; Donato, F.; Tagger, A.; Favret, M.; Ribero, M. L.; Nardi, G., . . . Callea, F. (2000).
Etiology of hepatocellular carcinoma in Italian patients with and without cirrhosis.
Cancer Epidemiol Biomarkers Prev, 9, 213-216.
http://www.ncbi.nlm.nih.gov/pubmed/10698484
Chiu, B. C. and Weisenburger, D. D. (2003). An update of the epidemiology of non-Hodgkin's
lymphoma. 4, 161-168. http://www.ncbi.nlm.nih.gov/pubmed/14715098
Chiu, W. A. (2006). Statistical issues in physiologically based pharmacokinetic modeling
Toxicokinetics And Risk Assessment (pp. 269). New York: Informa Healthcare.
Chiu, W. A. (In Press). Trichloroacetic acid: Updated estimates of its bioavailability and its
contribution to trichloroethylene-induced mouse hepatomegaly.
http://dx.doi.Org/10.1016/j.tox.2011.04.009
23
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Chiu, W. A. and Bois, F. Y. (2007). An Approximate Method for Population Toxicokinetic
Analysis With Aggregated Data. Journal of Agricultural, Biological, and Environmental
Statistics, 12, 346-363. http://dx.doi.org/10.1198/108571107X229340
Chiu, W. A. and White, P. (2004). PBPK Model-Based Route Extrapolation: Exact and
Approximate Solutions for Use in Risk Assessment.
Chiu, W. A. and White, P. (2006). Steady-state solutions to PBPK models and their applications
to risk assessment I: Route-to-route extrapolation of volatile chemicals. Risk Anal, 26,
769-780. http://dx.doi.org/10.1111/j. 1539-6924.2006.00762.X
Chiu, W. A.; Micallef, S.; Monster, A. C.; Bois, F. Y. (2007). Toxicokinetics of inhaled
trichloroethylene and tetrachloroethylene in humans at 1 ppm: empirical results and
comparisons with previous studies. Toxicol Sci, 95, 23-36.
http://dx.doi.org/10.1093/toxsci/kfll29
Chiu, W. A.; Okino, M. S.; Evans, M. V. (2009). Characterizing uncertainty and population
variability in the toxicokinetics of trichloroethylene and metabolites in mice, rats, and
humans using an updated database, physiologically based pharmacokinetic (PBPK)
model, and Bayesian approach. Toxicol Appl Pharmacol, 241, 36-60.
http://dx.doi.Org/10.1016/j.taap.2009.07.032
Chiu, W. A.; Okino, M. S.; Lipscomb, J. C.; Evans, M. V. (2006). Issues in the pharmacokinetics
of trichloroethylene and its metabolites. Environ Health Perspect, 114, 1450-1456.
http://dx.doi.org/10.1289/ehp.8691
Chiu, W.; Caldwell, J.; Keshava, N.; Scott, C. (2006). Key scientific issues in the health risk
assessment of trichloroethylene. Environ Health Perspect, 114, 1445-1449.
http://www.ncbi.nlm.nih.gov/pubmed/16966103
Cho, W. C. (2007). OncomiRs: the discovery and progress of microRNAs in cancers. Mol
Cancer, 6, 60. http://dx.doi.org/10.1186/1476-4598-6-60
Chow, W. H. and Devesa, S. S. (2008). Contemporary epidemiology of renal cell cancer. Cancer
J, 14, 288-301. http://dx.doi.org/10.1097/PPO.0b013e3181867628
Chow, W.; Gridley, G.; McLaughlin, J.; Mandel, J.; Wacholder, S.; Blot, W., . . . Fraumeni, J.
(1994). Protein intake and risk of renal cell cancer. J Natl Cancer Inst, 86, 1131-1139.
http://www.ncbi.nlm.nih.gov/pubmed/8028035
Cicmanec, J.; Condie, L.; Olson, G.; Wang, S. (1991). 90-Day toxicity study of dichloroacetate
in dogs. Fundam Appl Toxicol, 17, 376-389.
http://www.ncbi.nlm.nih.gov/pubmed/1765225
Clapp, R. W. and Hoffman, K. (2008). Cancer mortality in IBM Endicott plant workers, 1969-
2001: an update on a NY production plant. Environ Health, 7, 13.
http://dx.doi.org/10.1186/1476-069X-7-13
Clay, P. (2008). Assessment of the genotoxicity of trichloroethylene and its metabolite, S-(l,2-
dichlorovinyl)-L-cysteine (DCVC), in the comet assay in rat kidney. Mutagenesis, 23,
27-33. http://dx.doi.org/10.1093/mutage/gem034
Clegg, E. D.; Cook, J. C.; Chapin, R. E.; Foster, P. M. D.; Daston, G. P. (1997). Leydig cell
hyperplasia and adenoma formation: Mechanisms and relevance to humans. Reprod
Toxicol, 11, 107-121. http://dx.doi.org/10.1016/S0890-6238(96)00203-l
Clewell, H. J. and Andersen, M. E. (2004). Applying mode-of-action and pharmacokinetic
considerations in contemporary cancer risk assessments: An example with
24
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
trichloroethylene. CritRev Toxicol, 34, 385-445.
http://dx.doi.org/10.1080/10408440490500795
Clewell, H. J., Ill; Gentry, P. R.; Covington, T. R.; Gearhart, J. M. (2000). Development of a
physiologically based pharmacokinetic model of trichloroethylene and its metabolites for
use in risk assessment. Environ Health Perspect, 108, 283-305.
http://www.ncbi.nlm.nih.gov/pubmed/10807559
Clewell, H.; Andersen, M.; Barton, H. (2002). A consistent approach for the application of
pharmacokinetic modeling in cancer and noncancer risk assessment. Environ Health
Perspect, 110, 85-93. http://www.ncbi.nlm.nih.gov/pubmed/11781169
Coberly, S.; Oudiz, D. J.; Overstreet, J. W.; Wiley, L. M. (1992). Effects of maternal exposure to
trichloroethylene (TCE) on cell proliferation in the mouse preimplantation embryo.
Reprod Toxicol, 6, 241-245. http://www.ncbi.nlm.nih.gov/pubmed/1591481
Cocco, P.; t'Mannetje, A.; Fadda, D.; Melis, M.; Becker, N.; de Sanjose, S., . . . Boffetta, P.
(2010). Occupational exposure to solvents and risk of lymphoma subtypes: results from
the Epilymph case-control study. Occup Environ Med, 67, 341-347.
http://dx.doi.Org/10.l 136/oem.2009.046839
Cogger, V. C.; Warren, A.; Fraser, R.; Ngu, M.; McLean, A. J.; Le Couteur, D. G. (2003).
Hepatic sinusoidal pseudocapillarization with aging in the non-human primate. Exp
Gerontol, 38, 1101-1107. http://www.ncbi.nlm.nih.gov/pubmed/14580863
Cohn, P.; Fagliano, J.; Klotz, J. (1994). Assessing human health effects from chemical
contaminants in drinking water. 91, 719-722.
http://www.ncbi.nlm.nih.gov/pubmed/7800286
Cohn, P.; Klotz, J.; Bove, F.; Berkowitz, M.; Fagliano, J. (1994). Drinking water contamination
and the incidence of leukemia and non-Hodgkin's lymphoma. Environ Health Perspect,
102, 556-561. http://www.ncbi.nlm.nih.gov/pubmed/9679115
Cojocel, C.; Beuter, W.; Miiller, W.; Mayer, D. (1989). Lipid peroxidation: A possible
mechanism of trichloroethylene-induced nephrotoxicity. Toxicology, 55, 131-141.
http://dx.doi.org/10.1016/0300-483X(89)90180-7
Coleridge, H. M.; Coleridge, J. C. G.; Luck, J. C.; Norman, J. (1968). The effect of four volatile
anaesthetic agents on the impulse activity of two types of pulmonary receptor. Br J
Anaesth, 40, 484-492.
Collier, J.; Selmin, O.; Johnson, P.; Runyan, R. (2003). Trichloroethylene effects on gene
expression during cardiac development. Birth Defects Res A Clin Mol Teratol, 67, 488-
495. http://dx.doi.org/10.1002/bdra.10073
Colt, J. and Blair, A. (1998). Parental occupational exposures and risk of childhood cancer.
Environ Health Perspect, 106 Suppl 3, 909-925.
http://www.ncbi.nlm.nih.gov/pubmed/9646055
Columbano, A. and Ledda-Columbano, G. M. (2003). Mitogenesis by ligands of nuclear
receptors: an attractive model for the study of the molecular mechanisms implicated in
liver growth. Cell Death Differ, 10 Suppl 1, S19-S21.
http://dx.doi.org/10.1038/sj.cdd.4401113
Columbano, A.; Ledda-Columbano, G. M.; Coni, P. P.; Faa, G.; Liguori, C.; Santa Cruz, G.;
Pani, P. (1985). Occurrence of cell death (apoptosis) during the involution of liver
hyperplasia. Lab Invest, 52, 670-675. http://www.ncbi.nlm.nih.gov/pubmed/4010225
25
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Columbano, A.; Ledda-Columbano, G. M.; Ennas, M. G.; Curto, M.; Chelo, A.; Pani, P. (1990).
Cell proliferation and promotion of rat liver carcinogenesis: Different effect of hepatic
regeneration and mitogen induced hyperplasia on the development of enzyme-altered
foci. Carcinogenesis, 11, 771-776. http://dx.doi.Org/10.1093/carcin/ll.5.771
Columbano, A.; Ledda-Columbano, G. M.; Lee, G.; Rajalakshmi, S.; Sarma, D. S. (1987).
Inability of mitogen-induced liver hyperplasia to support the induction of enzyme-altered
islands induced by liver carcinogens. Cancer Res, 47, 5557-5559.
http://www.ncbi.nlm.nih.gov/pubmed/2889525
Colvin, M.; Myers, J.; Nell, V.; Rees, D.; Cronje, R. (1993). A cross-sectional survey of
neurobehavioral effects of chronic solvent exposure on workers in a paint manufacturing
plant. Environ Res, 63, 122-132. http://dx.doi.org/10.1006/enrs.1993.1134
Commandeur, J. and Vermeulen, N. (1990). Identification of N-acetyl(2,2-dichlorovinyl)- and N-
acetyl(l,2-dichlorovinyl)-L-cysteine as two regioisomeric mercapturic acids of
trichloroethylene in the rat. Chem Res Toxicol, 3, 212-218.
http://www.ncbi.nlm.nih.gov/pubmed/1966701
Coni, P.; Simbula, G.; De Prati, A. C.; Menegazzi, M.; Suzuki, H.; Sarma, D. S. R., . . .
Columbano, A. (1993). Differences in the steady-state levels of c-fos, c-jun and c-myc
messenger RNA during mitogen-induced liver growth and compensatory regeneration.
Hepatology, 17, 1109-1116.
Conner, E. A.; Lemmer, E. R.; Omori, M.; Wirth, P. J.; Factor, V. M.; Thorgeirsson, S. S.
(2000). Dual functions of E2F-1 in a transgenic mouse model of liver carcinogenesis.
Oncogene, 19, 5054-5062. http://dx.doi.org/10.1038/sj.onc.1203885
Conner, E. A.; Lemmer, E. R.; Sanchez, A.; Factor, V. M.; Thorgeirsson, S. S. (2003). E2F1
blocks and c-Myc accelerates hepatic ploidy in transgenic mouse models. Biochem
BiophysRes Commun, 302, 114-120. http://www.ncbi.nlm.nih.gov/pubmed/12593856
Contrera, J. F.; Jacobs, A. C.; DeGeorge, J. J. (1997). Carcinogenicity testing and the evaluation
of regulatory requirements for pharmaceuticals. Regul Toxicol Pharmacol, 25, 130-145.
http://dx.doi.org/10.1006/rtph.1997.1085
Cook, J. C.; Klinefelter, G. R.; Hardisty, J. F.; Sharpe, R. M.; Foster, P. M. D. (1999). Rodent
leydig cell tumorigenesis: A review of the physiology, pathology, mechanisms and
relevance to humans. CritRev Toxicol, 29, 169-261.
Cooper, A. (2004). The role of glutamine transaminase K (GTK) in sulfur and alpha-keto acid
metabolism in the brain, and in the possible bioactivation of neurotoxicants. Neurochem
Int, 44, 557-577. http://dx.doi.Org/10.1016/j.neuint.2003.12.002
Cooper, G. and Stroehla, B. (2003). The epidemiology of autoimmune diseases. Autoimmun
Rev, 2, 119-125. http://www.ncbi.nlm.nih.gov/pubmed/12848952
Cooper, G. S.; Makris, S. L.; Nietert, P. J.; Jinot, J. (2009). Evidence of autoimmune-related
effects of trichloroethylene exposure from studies in mice and humans. Environ Health
Perspect, 117, 696-702. http://dx.doi.org/10.1289/ehp.11782
Cooper, G. S.; Parks, C. G.; Treadwell, E. L.; St Clair, E. W.; Gilkeson, G. S.; Dooley, M. A.
(2004). Occupational risk factors for the development of systemic lupus erythematosus. J
Rheumatol, 31, 1928-1933. http://www.ncbi.nlm.nih.gov/pubmed/15468355
Cooper, G.; Miller, F.; Pandey, J. (1999). The role of genetic factors in autoimmune disease:
implications for environmental research. Environ Health Perspect, 107 Suppl 5, 693-700.
http://www.ncbi.nlm.nih.gov/pubmed/10502533
26
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Coopman, V.; Cordonnier, J.; De Letter, E.; Piette, M. (2003). Tissue distribution of
trichloroethylene in a case of accidental acute intoxication by inhalation. Forensic Sci Int,
134, 115-119. http://www.ncbi.nlm.nih.gov/pubmed/12850404
Copeland, K. T.; Checkoway, H.; McMichael, A. J.; Holbrook, R. H. (1977). Bias due to
misclassification in the estimation of relative risk. Am J Epidemiol, 105, 488-495.
http://www.ncbi.nlm.nih.gov/pubmed/871121
Cornett, R.; James, M.; Henderson, G.; Cheung, J.; Shroads, A.; Stacpoole, P. (1999). Inhibition
of glutathione S-transferase zeta and tyrosine metabolism by dichloroacetate: A potential
unifying mechanism for its altered biotransformation and toxicity. Biochem Biophys Res
Commun, 262, 752-756. http://dx.doi.org/10.1006/bbrc.1999.1287
Corton, J. C. (2008). Evaluation of the role of peroxisome proliferator-activated receptor alpha
(PPARalpha) in mouse liver tumor induction by trichloroethylene and metabolites. Crit
Rev Toxicol, 38, 857-875. http://dx.doi.org/10.1080/10408440802209796
Corton, J. C. and Lapinskas, P. J. (2005). Peroxisome proliferator-activated receptors: Mediators
of phthalate ester-induced effects in the male reproductive tract? Toxicol Sci, 83, 4-17.
http://dx.doi.org/10.1093/toxsci/kfi011
Cosby, N. C. and Dukelow, W. R. (1992). Toxicology of maternally ingested trichloroethylene
(TCE) on embryonal and fetal development in mice and of TCE metabolites on in vitro
fertilization. Toxicol Sci, 19, 268-274. http://www.ncbi.nlm.nih.gov/pubmed/1516784
Costa, A. K. and Ivanetich, K. M. (1984). Chlorinated ethylenes: their metabolism and effect on
DNA repair in rat hepatocytes. Carcinogenesis, 5, 1629-1636.
http://www.ncbi.nlm.nih.gov/pubmed/6499114
Costa, G.; Merletti, F.; Segnan, N. (1989). A mortality cohort study in a north Italian aircraft
factory. Br J Ind Med, 46, 738-743. http://www.ncbi.nlm.nih.gov/pubmed/2818961
Costantini, A. S.; Benvenuti, A.; Vineis, P.; Kriebel, D.; Tumino, R.; Ramazzotti, V., . . . Miligi,
L. (2008). Risk of leukemia and multiple myeloma associated with exposure to benzene
and other organic solvents: Evidence from the Italian Multicenter Case-control study. Am
J Ind Med, 51, 803-811. http://dx.doi.org/10.1002/ajim.20592
Costas, K.; Knorr, R. S.; Condon, S. K. (2002). A case-control study of childhood leukemia in
Woburn, Massachusetts: The relationship between leukemia incidence and exposure to
public drinking water. Sci Total Environ, 300, 23-35. http://dx.doi.org/10.1016/S0048-
9697(02)00169-9
Costet, P.; Legendre, C.; More, J.; Edgar, A.; Galtier, P.; Pineau, T. (1998). Peroxisome
proliferator-activated receptor alpha-isoform deficiency leads to progressive dyslipidemia
with sexually dimorphic obesity and steatosis. J Biol Chem, 273, 29577-29585.
http://www.ncbi.nlm.nih.gov/pubmed/9792666
Counts, J.; Sarmiento, J.; Harbison, M.; Downing, J.; McClain, R.; Goodman, J. (1996). Cell
proliferation and global methylation status changes in mouse liver after phenobarbital
and/or choline-devoid, methionine-deficient diet administration. Carcinogenesis, 17,
1251-1257. http://www.ncbi.nlm.nih.gov/pubmed/8681439
Coyle, Y. M.; Hynan, L. S.; Euhus, D. M.; Minhajuddin, A. T. (2005). An ecological study of the
association of environmental chemicals on breast cancer incidence in Texas. Breast
Cancer Res Treat, 92, 107-114. http://dx.doi.org/10.1007/sl0549-004-8268-z
Crabb, D.; Pinaire, J.; Chou, W.; Sissom, S.; Peters, J.; Harris, R.; Stewart, M. (2001).
Peroxisome proliferator-activated receptors (PPAR) and the mitochondrial aldehyde
27
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
dehydrogenase (ALDH2) promoter in vitro and in vivo. Alcohol Clin Exp Res, 25, 945-
952. http://www.ncbi.nlm.nih.gov/pubmed/11505017
Crebelli, R. and Carere, A. (1989). Genetic toxicology of 1,1,2-trichloroethylene. Mutat Res,
221, 11-37. http://dx.doi.org/10.1016/0165-l 110(89)90043-2
Crebelli, R.; Bignami, M.; Conti, L.; Carere, A. (1982). Mutagenicity of trichloroethylene in
Salmonella typhimurium TA100. Ann 1st Super Sanita, 18, 117-121.
http://www.ncbi.nlm.nih.gov/pubmed/6763484
Crebelli, R.; Conti, G.; Conti, L.; Carere, A. (1985). Mutagenicity of trichloroethylene,
trichloroethanol and chloral hydrate in Aspergillus nidulans. Mutat Res Genet Toxicol
Environ Mutagen, 155, 105-111. http://dx.doi.org/10.1016/0165-1218(85)90126-0
Crebelli, R.; Conti, G.; Conti, L.; Carere, A. (1991). In vitro studies with nine known or
suspected spindle poisons: results in tests for chromosome malsegregation in Aspergillus
nidulans. Mutagenesis, 6, 131-136. http://www.ncbi.nlm.nih.gov/pubmed/2056914
Crofton, K. and Zhao, X. (1997). The ototoxicity of trichloroethylene: extrapolation and
relevance of high-concentration, short-duration animal exposure data. Fundam Appl
Toxicol, 38, 101-106. http://www.ncbi.nlm.nih.gov/pubmed/9268609
Crofton, K. M. and Zhao, X. (1993). Mid-frequency hearing loss in rats following inhalation
exposure to trichloroethylene: evidence from reflex modification audiometry.
Neurotoxicol Teratol, 15, 413-423. http://www.ncbi.nlm.nih.gov/pubmed/8302243
Crofton, K. M.; Lassiter, T. L.; Rebert, C. S. (1994). Solvent-induced ototoxicity in rats: an
atypical selective mid-frequency hearing deficit. Hear Res, 80, 25-30.
http://www.ncbi.nlm.nih.gov/pubmed/7852200
Crump, K.; Allen, B.; Shipp, A. (1989). Choice of dose measure for extrapolating carcinogenic
risk from animals to humans: An empirical investigation of 23 chemicals. Health Phys,
57, 387-393. http://www.ncbi.nlm.nih.gov/pubmed/2606698
Cruzan, G.; Bus, J.; Banton, M.; Gingell, R.; Carlson, G. (2009). Mouse specific lung tumors
from CYP2F2-mediated cytotoxic metabolism: an endpoint/toxic response where data
from multiple chemicals converge to support a mode of action. Regul Toxicol Pharmacol,
55, 205-218. http://dx.doi.Org/10.1016/j.yrtph.2009.07.002
Cummings, B. S. and Lash, L. H. (2000). Metabolism and toxicity of trichloroethylene and S-
(l,2-dichlorovinyl)-L-cysteine in freshly isolated human proximal tubular cells. Toxicol
Sci, 53, 458-466. http://www.ncbi.nlm.nih.gov/pubmed/10696794
Cummings, B. S.; Lasker, J. M.; Lash, L. H. (2000). Expression of glutathione-dependent
enzymes and cytochrome P450s in freshly isolated and primary cultures of proximal
tubular cells from human kidney. J Pharmacol Exp Ther, 293, 677-685.
http://www.ncbi.nlm.nih.gov/pubmed/10773044
Cummings, B. S.; Parker, J. C.; Lash, L. H. (2000). Role of cytochrome P450 and glutathione S-
transferase alpha in the metabolism and cytotoxicity of trichloroethylene in rat kidney.
Biochem Pharmacol, 59, 531-543. http://dx.doi.org/10.1016/S0006-2952(99)00374-3
Cummings, B. S.; Parker, J. C.; Lash, L. H. (2001). Cytochrome p450-dependent metabolism of
trichloroethylene in rat kidney. Toxicol Sci, 60, 11-19.
http://www.ncbi.nlm.nih.gov/pubmed/11222868
Cummings, B. S.; Zangar, R. C.; Novak, R. F.; Lash, L. H. (1999). Cellular distribution of
cytochromes P-450 in the rat kidney. Drug Metab Dispos, 27, 542-548.
28
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Cummings, B.; Zangar, R.; Novak, R.; Lash, L. (2000). Cytotoxicity of trichloroethylene and S-
(1, 2-dichlorovinyl)-L-cysteine in primary cultures of rat renal proximal tubular and distal
tubular cells. Toxicology, 150, 83-98. http://www.ncbi.nlm.nih.gov/pubmed/10996665
Cutler, J.; Parker, G.; Rosen, S.; Prenney, B.; Healey, R.; Caldwell, G. (1986). Childhood
leukemia in Woburn, Massachusetts. Public Health Rep, 101, 201-205.
http://www.ncbi.nlm.nih.gov/pubmed/3083476
Dahlgren, J.; Takhar, H.; Anderson-Mahoney, P.; Kotlerman, J.; Tarr, J.; Warshaw, R. (2007).
Cluster of systemic lupus erythematosus (SLE) associated with an oil field waste site: a
cross sectional study. Environ Health, 6, 8. http://dx.doi.org/10.1186/1476-069X-6-8
Dai, X.; De Souza, A. T.; Dai, H.; Lewis, D. L.; Lee, C. K.; Spencer, A. G., . . . He, Y. D.
(2007). PPARalpha siRNA-treated expression profiles uncover the causal sufficiency
network for compound-induced liver hypertrophy. 3, e30.
http://dx.doi.org/10.1371/journal.pcbi.0030030
Dai, Y.; Leng, S.; Li, L.; Niu, Y.; Huang, H.; Cheng, J.; Zheng, Y. (2004). Genetic
polymorphisms of cytokine genes and risk for trichloroethylene-induced severe
generalized dermatitis: a case-control study. Biomarkers, 9, 470-478.
http://dx.doi.org/10.1080/13547500400026920
Dajee, M.; Lazarov, M.; Zhang, J. Y.; Cai, T.; Green, C. L.; Russell, A. J., . . . Khavari, P. A.
(2003). NF-kappaB blockade and oncogenic Ras trigger invasive human epidermal
neoplasia. Nature, 421, 639-643. http://dx.doi.org/10.1038/nature01283
Dalbey, W. and Bingham, E. (1978). Metabolism of trichloroethylene by the isolated perfused
lung. Toxicol Appl Pharmacol, 43, 267-277.
http://www.ncbi.nlm.nih.gov/pubmed/635916
Dallas, C. E. (1991). Validation and application of pharmacokinetic models for interspecies
extrapolations in toxicity risk assessments of volatile organics.
Dallas, C.; Gallo, J.; Ramanathan, R.; Muralidhara, S.; Bruckner, J. (1991). Physiological
pharmacokinetic modeling of inhaled trichloroethylene in rats. Toxicol Appl Pharmacol,
110, 303-314. http://www.ncbi.nlm.nih.gov/pubmed/1891776
Daniel, F. B.; DeAngelo, A. B.; Stober, J. A.; Olson, G. R.; Page, N. P. (1992).
Hepatocarcinogenicity of chloral hydrate, 2-chloroacetaldehyde, and dichloroacetic acid
in the male B6C3F1 mouse. Fundam Appl Toxicol, 19, 159-168.
http://dx.doi.org/10.1016/0272-0590(92)90147-A
Daniell, W. E.; Claypoole, K. H.; Checkoway, H.; Smith-Weller, T.; Dager, S. R.; Townes, B.
D.; Rosenstock, L. (1999). Neuropsychological function in retired workers with previous
long-term occupational exposure to solvents. Occup Environ Med, 56, 93-105.
http://www.ncbi.nlm.nih.gov/pubmed/10448313
Darnerud, P.; Brandt, I.; Feil, V.; Bakke, J. (1989). Dichlorovinyl cysteine (DCVC) in the mouse
kidney: tissue-binding and toxicity after glutathione depletion and probenecid treatment.
Arch Toxicol, 63, 345-350. http://www.ncbi.nlm.nih.gov/pubmed/2818197
Das, R. M. and Scott, J. E. (1994). Trichloroethylene-induced pneumotoxicity in fetal and
neonatal mice. Toxicol Lett, 73, 227-239. http://www.ncbi.nlm.nih.gov/pubmed/8091431
David, R. M.; Clewell, H. J.; Gentry, P. R.; Covington, T. R.; Morgott, D. A.; Marino, D. J.
(2006). Revised assessment of cancer risk to dichloromethane II. Application of
probabilistic methods to cancer risk determinations. Regul Toxicol Pharmacol, 45, 55-65.
http://dx.doi.Org/10.1016/j.yrtph.2005.12.003
29
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
David, R. M.; Moore, M. R.; Cifone, M. A.; Finney, D. C.; Guest, D. (1999). Chronic
peroxisome proliferation and hepatomegaly associated with the hepatocellular
tumorigenesis of di(2-ethylhexyl)phthalate and the effects of recovery. Toxicol Sci, 50,
195-205. http://www.ncbi.nlm.nih.gov/pubmed/10478855
Davidson, I. W. F. and Beliles, R. P. (1991). Consideration of the target organ toxicity of
trichloroethylene in terms of metabolite toxicity and pharmacokinetics. Drug Metab Rev,
23, 493-599. http://dx.doi.org/10.3109/03602539109029772
Davis, R. R.; Murphy, W. J.; Snawder, J. E.; Striley, C. A.; Henderson, D.; Khan, A.; Krieg, E.
F. (2002). Susceptibility to the ototoxic properties of toluene is species specific. Hear
Res, 166, 24-32. http://www.ncbi.nlm.nih.gov/pubmed/12062755
Davis, S.; Laszlo Pallos, L.; Wu, J.; Sapp, J.; Cusack, C. (2005). ATSDR's trichloroethylene
subregistry methods and results: 1989-2000. Arch Environ Occup Health, 60, 130-139.
http://www.ncbi.nlm.nih.gov/pubmed/17153085
Dawson, B. V.; Johnson, P. D.; Goldberg, S. J.; Ulreich, J. B. (1990). Cardiac teratogenesis of
trichloroethylene and dichloroethylene in a mammallian model. J Am Coll Cardiol, 16,
1304-1309.
Dawson, B.; Johnson, P.; Goldberg, S.; Ulreich, J. (1993). Cardiac teratogenesis of halogenated
hydrocarbon-contaminated drinking water. J Am Coll Cardiol, 21, 1466-1472.
http://www.ncbi.nlm.nih.gov/pubmed/8473657
De Baere, S.; Meyer, E.; Dirinck, I.; Lambert, W.; Piette, M.; Van Peteghem, C.; De Leenheer,
A. (1997). Tissue distribution of trichloroethylene and its metabolites in a forensic case. J
Anal Toxicol, 21, 223-227. http://www.ncbi.nlm.nih.gov/pubmed/9171208
de Jong, P. and Brenner, B. (2004). From secondary to primary prevention of progressive renal
disease: the case for screening for albuminuria. Kidney Int, 66, 2109-2118.
http://dx.doi.Org/10.llll/j.1523-1755.2004.66001.x
De Roos, A. J.; Olshan, A. F.; Teschke, K.; Poole, C.; Savitz, D. A.; Blatt, J., . . . Pollock, B. H.
(2001). Parental occupational exposures to chemicals and incidence of neuroblastoma in
offspring. Am J Epidemiol, 154, 106-114.
http://www.ncbi.nlm.nih.gov/pubmed/11447042
De Roos, A.; Olshan, A.; Teschke, K.; Poole, C.; Savitz, D.; Blatt, J., . . . Pollock, B. (2001).
Parental occupational exposures to chemicals and incidence of neuroblastoma in
offspring. Am J Epidemiol, 154, 106-114. http://dx.doi.Org/10.1093/aje/154.2.106
Deane, M.; Swan, S.; Harris, J.; Epstein, D.; Neutra, R. (1989). Adverse pregnancy outcomes in
relation to water contamination, Santa Clara County, California, 1980-1981. Am J
Epidemiol, 129, 894-904. http://www.ncbi.nlm.nih.gov/pubmed/2705433
DeAngelo, A. B.; Daniel, F. B.; McMillan, L.; Wernsing, P.; Savage, R. E. J. (1989). Species
and strain sensitivity to the induction of peroxisome proliferation by chloroacetic acids.
Toxicol Appl Pharmacol, 101, 285-298. http://dx.doi.org/10.1016/0041-008X(89)90277-
9
DeAngelo, A. B.; Daniel, F. B.; Most, B. M.; Olson, G. R. (1996). The carcinogenicity of
dichloroacetic acid in the male fischer 344 rat. Toxicology, 114, 207-221.
http://dx.doi.org/10.1016/S0300-483X(96)03510-X
DeAngelo, A. B.; Daniel, F. B.; Most, B. M.; Olson, G. R. (1997). Failure of monochloroacetic
acid and trichloroacetic acid administered in the drinking water to produce liver cancer in
30
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
male F344/N rats. J Toxicol Environ Health, 52, 425-445.
http://dx.doi.org/10.1080/00984109708984074
DeAngelo, A. B.; Daniel, F. B.; Wong, D. M.; George, M. H. (2008). The induction of
hepatocellular neoplasia by trichloroacetic acid administered in the drinking water of the
male B6C3F1 mouse. J Toxicol Environ Health A, 71, 1056-1068.
http://dx.doi.org/10.1080/15287390802111952
DeAngelo, A. B.; George, M. H.; House, D. E. (1999). Hepatocarcinogenicity in the male
B6C3F1 mouse following a lifetime exposure to dichloroacetic acid in the drinking
water: Dose-response determination and modes of action. J Toxicol Environ Health A,
58, 485-507. http://www.ncbi.nlm.nih.gov/pubmed/10632141
Deerberg, F. and Miiller-Peddinghaus, R. (1970). [Comparative studies on the occurrence of
spontaneous diseases and spontaneous deaths in conventional and specific germ-free
female NMRI-HAN mice], Z Versuchstierkd, 12, 341-347.
http://www.ncbi.nlm.nih.gov/pubmed/5508977
Deerberg, F.; Pittermann, W.; Rapp, K. (1974). [Lung tumor in the mouse: a progressive tumor].
Vet Pathol, 11, 430-441. http://www.ncbi.nlm.nih.gov/pubmed/4462259
Dees, C. and Travis, C. (1993). The mitogenic potential of trichloroethylene in B6C3F1 mice.
Toxicol Lett, 69, 129-137. http://www.ncbi.nlm.nih.gov/pubmed/8212055
Dees, C. and Travis, C. (1994). Trichloroacetate stimulation of liver DNA synthesis in male and
female mice. Toxicol Lett, 70, 343-355. http://dx.doi.org/10.1016/0378-4274(94)90129-5
Degrassi, F. and Tanzarella, C. (1988). Immunofluorescent staining of kinetochores in
micronuclei: A new assay for the detection of aneuploidy. Mutat Res Genet Toxicol
Environ Mutagen, 203, 339-345. http://dx.doi.org/10.1016/0165-l 161(88)90030-1
Dehon, B.; Humbert, L.; Devisme, L.; Stievenart, M.; Mathieu, D.; Houdret, N.; Lhermitte, M.
(2000). Tetrachloroethylene and trichloroethylene fatality: case report and simple
headspace SPME-capillary gas chromatographic determination in tissues. J Anal Toxicol,
24, 22-26. http://www.ncbi.nlm.nih.gov/pubmed/10654565
Dekant, W. (1990). Bioactivation of halogenated alkenes by glutathione conjugation a
mechanistic explanation for their renal carcinogenicity. J Cancer Res Clin Oncol, 116
(Suppl. Part 2), 1063.
Dekant, W.; Berthold, K.; Vamvakas, S.; Henschler, D.; Anders, M. W. (1988). Thioacylating
intermediates as metabolites of S-(l,2-dichlorovinyl)-L-cysteine and S-(l,2,2-
trichlorovinyl)-L-cysteine formed by cysteine conjugate beta-lyase. Chem Res Toxicol, 1,
175-178. http://dx.doi.org/10.1021/tx00003a008
Dekant, W.; Metzler, M.; Henschler, D. (1984). Novel metabolites of trichloroethylene through
dechlorination reactions in rats, mice and humans. Biochem Pharmacol, 33, 2021-2027.
http://www.ncbi.nlm.nih.gov/pubmed/6743352
Dekant, W.; Metzler, M.; Henschler, D. (1986). Identification of S-l,2-dichlorovinyl-N-acetyl-
cysteine as a urinary metabolite of trichloroethylene: A possible explanation for its
nephrocarcinogenicity in male rats. Biochem Pharmacol, 35, 2455-2458.
http://dx.doi.org/10.1016/0006-2952(86)90039-0
Dekant, W.; Schulz, A.; Metzler, M.; Henschler, D. (1986). Absorption, elimination, and
metabolism of trichlorethylene: a quantitative comparison between rats and mice.
Xenobiotica, 16, 143-152.
31
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Dekant, W.; Vamvakas, S.; Berthold, K.; Schmidt, S.; Wild, D.; Henschler, D. (1986). Bacterial
beta-lyase mediated cleavage and mutagenicity of cysteine conjugates derived from the
nephrocarcinogenic alkenes trichloroethylene, tetrachloroethylene and
hexachlorobutadiene. Chem Biol Interact, 60, 31-45. http://dx.doi.org/10.1016/0009-
2797(86)90015-3
Delahunt, B.; Bethwaite, P. B.; Nacey, J. N. (1995). Occupational risk for renal cell carcinoma.
A case-control study based on the New Zealand Cancer Registry. Br J Urol, 75, 578-582.
http://dx.doi.Org/10.llll/j.1464-410X.1995.tb07410.x
DeMarini, D. M.; Perry, E.; Shelton, M. L. (1994). Dichloroacetic acid and related compounds:
Induction of prophage in E. coli and mutagenicity and mutation spectra in Salmonella
TA100. Mutagenesis, 9, 429-437. http://dx.doi.Org/10.1093/mutage/9.5.429
DerSimonian, R. and Laird, N. (1986). Meta-analysis in clinical trials. Contemp Clin Trials, 7,
177-188. http://www.ncbi.nlm.nih.gov/pubmed/3802833
Dewar, R.; Siemiatyckia, J.; Geacuterin, M. (1994). Loss of statistical power associated with the
use of a job-exposure matrix in occupational case-control studies. Appl Occup Environ
Hyg, 6, 508-515. http://dx.doi.org/10.1080/1047322X.1991.10387920
Ding, R.; Shen, T.; Zhu, Q. X. (2006). [Effects of TCE and PCE on cultured human keratinocyte
lipid peroxidation and protective effect of vitamin E on it], Zhonghua Lao Dong Wei
Sheng Zhi Ye Bing Za Zhi, 24, 662-665. http://www.ncbi.nlm.nih.gov/pubmed/17181943
Diot, E.; Lesire, V.; Guilmot, J.; Metzger, M.; Pilore, R.; Rogier, S., . . . Lasfargues, G. (2002).
Systemic sclerosis and occupational risk factors: a case-control study. Occup Environ
Med, 59, 545-549. http://www.ncbi.nlm.nih.gov/pubmed/12151611
DiRenzo, A. B.; Gandolfi, A. J.; Sipes, I. G. (1982). Microsomal bioactivation and covalent
binding of aliphatic halides to DNA. Toxicol Lett, 11, 243-252.
Dix, K. J.; Kedderis, G. L.; Borghoff, S. J. (1997). Vehicle-dependent oral absorption and target
tissue dosimetry of chloroform in male rats and female mice. Toxicol Lett, 91, 197-209.
http://www.ncbi.nlm.nih.gov/pubmed/9217240
Djouadi, F.; Weinheimer, C. J.; Saffitz, J. E.; Pitchford, C.; Bastin, J.; Gonzalez, F. J.; Kelly, D.
P. (1998). A gender-related defect in lipid metabolism and glucose homeostasis in
peroxisome proliferator- activated receptor alpha- deficient mice. J Clin Invest, 102,
1083-1091. http://dx.doi.org/10.1172/JCI3949
Dobrev, I. D.; Andersen, M. E.; Yang, R. S. (2001). Assessing interaction thresholds for
trichloroethylene in combination with tetrachloroethylene and 1,1,1-trichloroethane using
gas uptake studies and PBPK modeling. Arch Toxicol, 75, 134-144.
http://dx.doi.org/10.1007/s002040100216
Dobrev, I. D.; Andersen, M. E.; Yang, R. S. (2002). In silico toxicology: Simulating interaction
thresholds for human exposure to mixtures of trichloroethylene, tetrachloroethylene, and
1,1,1-trichloroethane. Environ Health Perspect, 110, 1031-1039.
http ://dx. doi. org/123 61929
Dohn, D. R. and Anders, M. W. (1982). Assay of cysteine conjugate beta-lyase activity with S-
(2-benzothiazolyl)cysteine as the substrate. Anal Biochem, 120, 379-386.
http ://dx. doi. org/10.1016/0003 -2697(82)903 61 -X
Dolinoy, D. C.; Weidman, J. R.; Jirtle, R. L. (2007). Epigenetic gene regulation: linking early
developmental environment to adult disease. Reprod Toxicol, 23, 297-307.
http://dx.doi.Org/10.1016/j.reprotox.2006.08.012
32
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Donoghue, A. M.; Dryson, E. W.; Wynn-Williams, G. (1995). Contrast sensitivity in organic-
solvent-induced chronic toxic encephalopathy. J Occup Environ Med, 37, 1357-1363.
http://www.ncbi.nlm.nih.gov/pubmed/8749741
Dorfmueller, M. A.; Henne, S. P.; York, R. G.; Bornschein, R. L.; Manson, J. M. (1979).
Evaluation of teratogenicity and behavioral toxicity with inhalation exposure of maternal
rats to trichloroethylene. Toxicology, 14, 153-166.
http://www.ncbi.nlm.nih.gov/pubmed/538767
Dome, J. L.; Walton, K.; Renwick, A. G. (2005). Human variability in xenobiotic metabolism
and pathway-related uncertainty factors for chemical risk assessment: a review. Food
Chem Toxicol, 43, 203-216. http://dx.doi.Org/10.1016/j.fct.2004.05.011
Dosemeci, M.; Cocco, P.; Chow, W. H. (1999). Gender differences in risk of renal cell
carcinoma and occupational exposures to chlorinated aliphatic hydrocarbons. Am J Ind
Med, 36, 54-59. http://www.ncbi.nlm.nih.gov/pubmed/10361587
Dosemeci, M.; Cocco, P.; Gomez, M.; Stewart, P. A.; Heineman, E. F. (1994). Effects of three
features of a job-exposure matrix on risk estimates. Epidemiology, 5, 124-127.
http://www.ncbi.nlm.nih.gov/pubmed/8117771
Dosemeci, M.; Wacholder, S.; Lubin, J. H. (1990). Does nondifferential misclassification of
exposure always bias a true effect toward the null value? Am J Epidemiol, 132, 746-748.
http://www.ncbi.nlm.nih.gov/pubmed/2403115
D0ssing, M.; Petersen, K. T.; Vyberg, M.; Olsen, J. H. (1997). Liver cancer among employees in
Denmark. Am J Ind Med, 32, 248-254. http://www.ncbi.nlm.nih.gov/pubmed/9219654
Douglas, G.; Gingerich, J.; Soper, L.; Potvin, M.; Bjarnason, S. (1999). Evidence for the lack of
base-change and small-deletion mutation induction by trichloroethylene in lacZ
transgenic mice. Environ Mol Mutagen, 34, 190-194.
http://www.ncbi.nlm.nih.gov/pubmed/10529743
Dow, J. and Green, T. (2000). Trichloroethylene induced vitamin B(12) and folate deficiency
leads to increased formic acid excretion in the rat. Toxicology, 146, 123-136.
http://www.ncbi.nlm.nih.gov/pubmed/10814845
Dowsley, T.; Ulreich, J.; Bolton, J.; Park, S.; Forkert, P. (1996). CYP2E1-dependent
bioactivation of 1,1-dichloroethylene in murine lung: formation of reactive intermediates
and glutathione conjugates. Toxicol Appl Pharmacol, 139, 42-48.
http://dx.doi.org/10.1006/taap.1996.0141
Drake, V. J.; Koprowski, S. L.; Hu, N.; Smith, S. M.; Lough, J. (2006). Cardiogenic effects of
trichloroethylene and trichloroacetic acid following exposure during heart specification
of avian development. Toxicol Sci, 94, 153-162. http://dx.doi.org/10.1093/toxsci/kfl083
Drake, V.; Koprowski, S.; Lough, J.; Hu, N.; Smith, S. (2006). Trichloroethylene exposure
during cardiac valvuloseptal morphogenesis alters cushion formation and cardiac
hemodynamics in the avian embryo. Environ Health Perspect, 114, 842-847.
http://www.ncbi.nlm.nih.gov/pubmed/16759982
Driscoll, T.; Hamdan, H.; Wang, G.; Wright, P.; Stacey, N. (1992). Concentrations of individual
serum or plasma bile acids in workers exposed to chlorinated aliphatic hydrocarbons. Br J
Ind Med, 49, 700-705. http://www.ncbi.nlm.nih.gov/pubmed/1419858
Dryver, E.; Brandt, L.; Kauppinen, T.; Olsson, H. (2004). Occupational exposures and non-
Hodgkin's lymphoma in Southern Sweden. Int J Occup Environ Health, 10, 13-21.
http://www.ncbi.nlm.nih.gov/pubmed/15070021
33
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
D'Souza, R.; Bruckner, J.; Feldman, S. (1985). Oral and intravenous trichloroethylene
pharmacokinetics in the rat. J Toxicol Environ Health, 15, 587-601.
http://www.ncbi.nlm.nih.gov/pubmed/4046066
DT Jr, L. and Calo, J. M. (1981). Distribution coefficients of chlorinated hydrocarbons in dilute
air-water systems for groundwater contamination applications. Journal of Chemical and
Engineering Data, 26, 382-585.
Du, C. L. and Wang, J. D. (1998). Increased morbidity odds ratio of primary liver cancer and
cirrhosis of the liver among vinyl chloride monomer workers. Occup Environ Med, 55,
528-532. http://www.ncbi.nlm.nih.gov/pubmed/9849539
Duffel, M. W. and Jakoby, W. B. (1982). Cysteine S-conjugate N-acetyltransferase from rat
kidney microsomes. Mol Pharmacol, 21, 444-448.
http://www.ncbi.nlm.nih.gov/pubmed/6892478
Dugard, P. (2000). Effects of trichloroethylene (TCE) on an in vitro chick atrioventricular canal
culture. Toxicol Sci, 56, 437-438. http://www.ncbi.nlm.nih.gov/pubmed/10960254
Dumas, S.; Parent, M. E.; Siemiatycki, J.; Brisson, J. (2000). Rectal cancer and occupational risk
factors: a hypothesis-generating, exposure-based case-control study. Int J Cancer, 87,
874-879.
DuTeaux, S. B.; Berger, T.; Hess, R. A.; Sartini, B. L.; Miller, M. G. (2004). Male reproductive
toxicity of trichloroethylene: sperm protein oxidation and decreased fertilizing ability.
Biol Reprod, 70, 1518-1526. http://dx.doi.org/10.1095/biolreprod.103.022210
DuTeaux, S.; Hengel, M.; DeGroot, D.; Jelks, K.; Miller, M. (2003). Evidence for
trichloroethylene bioactivation and adduct formation in the rat epididymis and efferent
ducts. Biol Reprod, 69, 771-779. http://dx.doi.org/10.1095/biolreprod.102.014845
DuTeaux, S.; Newman, J.; Morisseau, C.; Fairbairn, E.; Jelks, K.; Hammock, B.; Miller, M.
(2004). Epoxide hydrolases in the rat epididymis: possible roles in xenobiotic and
endogenous fatty acid metabolism. Toxicol Sci, 78, 187-195.
http://dx.doi.org/10.1093/toxsci/kfh066
Eacho, P. I.; Lanier, T. L.; Brodhecker, C. A. (1991). Hepatocellular DNA synthesis in rats given
peroxisome proliferating agents: Comparison of WY-14,643 to clofibric acid, nafenopin
andLY171883. Carcinogenesis, 12, 1557-1561.
http://www.ncbi.nlm.nih.gov/pubmed/1893515
Eastin, W. C.; Mennear, J. H.; Tennant, R. W.; Stoll, R. E.; Branstetter, D. G.; Bucher, J. R., . . .
Mahler, J. F. (2001). Tg.AC genetically altered mouse: assay working group overview of
available data. Toxicol Pathol, 29 Suppl, 60-80.
http://dx.doi.org/10.1080/019262301753178483
Eastmond, D. A. and Tucker, J. D. (1989). Identification of aneuploidy-inducing agents using
cytokinesis-blocked human lymphocytes and an antikinetochore antibody. Environ Mol
Mutagen, 13, 34-43. http://www.ncbi.nlm.nih.gov/pubmed/2783409
Eastmond, D. A.; Hartwig, A.; Anderson, D.; Anwar, W. A.; Cimino, M. C.; Dobrev, I., . . .
Vickers, C. (2009). Mutagenicity testing for chemical risk assessment: Update of the
WHO/IPCS Harmonized Scheme. Mutagenesis, 24, 341-349.
http://dx.doi.org/10.1093/mutage/gep014
EFSA. (European Food Safety Authority). (2006). Opinion of the scientific panel on food
additives, flavourings, processing aids and materials in contact with food (AFC) on a
34
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
request from the commission related to a new long-term carcinogenicity study on
aspartame. EFSA J, 4, 1-44. http://dx.doi.Org/10.2903/j.efsa.2006.356
Egger, M.; Smith, G. D.; Schneider, M.; Minder, C. (1997). Bias in meta-analysis detected by a
simple, graphical test. Br Med J, 315, 629-634.
EHS. (Environmental Health Strategies). (1997). Final Report. Correspondence from Paul A.
Cammer, Ph.D., Trichloroethylene Issues Group, to Cheryl Siegel Scott, US
Environmental Protection Agency. December 22, 1997. Menlo Park, CA: Author.
Eichenlaub-Ritter, U.; Baart, E.; Yin, H.; Betzendahl, I. (1996). Mechanisms of spontaneous and
chemically-induced aneuploidy in mammalian oogenesis: basis of sex-specific
differences in response to aneugens and the necessity for further tests. Mutat Res, 372,
279-294. http://www.ncbi.nlm.nih.gov/pubmed/9015146
El Ghawabi, S. M.; Mansoor, M. B.; El Gamel, M. S.; El Saharti, A. A.; El Enany, F. F. (1973).
Chronic trichloroethylene exposure. 56, 715-724.
http://www.ncbi.nlm.nih.gov/pubmed/4618847
Elcombe, C. R. (1985). Species differences in carcinogenicity and peroxisome proliferation due
to trichloroethylene: a biochemical human hazard assessment. J Ind Hyg Toxicol, 8, 6-17.
http://www.ncbi.nlm.nih.gov/pubmed/3868383
Elcombe, C. R.; Rose, M. S.; Pratt, I. S. (1985). Biochemical, histological, and ultrastructural
changes in rat and mouse liver following the administration of trichloroethylene: Possible
relevance to species differences in hepatocarcinogenicity. Toxicol Appl Pharmacol, 79,
365-376. http://dx.doi.org/10.1016/0041-008X(85)90135-8
Elfarra, A. A. and Anders, M. W. (1984). Renal processing of glutathione conjugates role in
nephrotoxicity. Biochem Pharmacol, 33, 3729-3732.
http://www.ncbi.nlm.nih.gov/pubmed/6508831
Elfarra, A. A.; Jakobson, I.; Anders, M. W. (1986). Mechanism of S-(l,2-
dichlorovinyl)glutathione-induced nephrotoxicity. Biochem Pharmacol, 35, 283-288.
http://www.ncbi.nlm.nih.gov/pubmed/2867768
Elfarra, A. and Hwang, I. (1990). In vivo metabolites of S-(2-benzothiazolyl)-L-cysteine as
markers of in vivo cysteine conjugate beta-lyase and thiol glucuronosyl transferase
activities. DrugMetab Dispos, 18, 917-922.
http://www.ncbi.nlm.nih.gov/pubmed/1981537
Elfarra, A.; Krause, R.; Last, A.; Lash, L.; Parker, J. (1998). Species- and sex-related differences
in metabolism of trichloroethylene to yield chloral and trichloroethanol in mouse, rat, and
human liver microsomes. Drug Metab Dispos, 26, 779-785.
http://www.ncbi.nlm.nih.gov/pubmed/9698293
Elfarra, A.; Lash, L.; Anders, M. (1987). Alpha-ketoacids stimulate rat renal cysteine conjugate
beta-lyase activity and potentiate the cytotoxicity of S-(l,2-dichlorovinyl)-L-cysteine.
Mol Pharmacol, 31, 208-212. http://www.ncbi.nlm.nih.gov/pubmed/3807895
Elizur, A.; Adair-Kirk, T. L.; Kelley, D. G.; Griffin, G. L.; Demello, D. E.; Senior, R. M. (2008).
Tumor necrosis factor-alpha from macrophages enhances LPS-induced clara cell
expression of keratinocyte-derived chemokine. Am J Respir Cell Mol Biol, 38, 8-15.
http://dx.doi.Org/10.l 165/rcmb.2007-02030C
Elofsson, S. A.; Gamberale, F.; Hindmarsh, T. (1980). Exposure to organic solvents. A cross-
sectional epidemiologic investigation on occupationally exposed car and industrial spray
35
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
painters with special reference to the nervous system. Scand J Work Environ Health, 6,
239 -273.
Elovaara, E.; Hemminki, K.; Vainio, H. (1979). Effects of methylene chloride, trichloroethane,
trichloroethylene, tetrachloroethylene and toluene on the development of chick embryos.
Toxicology, 12, 111-119. http://dx.doi.org/10.1016/0300-483X(79)90037-4
El-Serag, H. B. (2004). Hepatocellular carcinoma: recent trends in the United States.
Gastroenterology, 127, S27-S34. http://www.ncbi.nlm.nih.gov/pubmed/15508094
El-Serag, H. B. (2007). Epidemiology of hepatocellular carcinoma in USA. Hepatology
Research, 37 Suppl 2, S88-S94. http://dx.doi.Org/10.llll/j.1872-034X.2007.00168.x
El-Serag, H. B. and Rudolph, K. L. (2007). Hepatocellular carcinoma: epidemiology and
molecular carcinogenesis. Gastroenterology, 132, 2557-2576.
http://dx.doi.Org/10.1053/j.gastro.2007.04.061
Emmert, B.; Biinger, J.; Keuch, K.; Miiller, M.; Emmert, S.; Hallier, E.; Westphal, G. A. (2006).
Mutagenicity of cytochrome P450 2E1 substrates in the Ames test with the metabolic
competent S. typhimurium strain YG7108pin3ERb5. Toxicology, 228, 66-76.
http://dx.doi.Org/10.1016/j.tox.2006.08.013
Engel, L. S.; Vaughan, T. L.; Gammon, M. D.; Chow, W. H.; Risch, H. A.; Dubrow, R., . . .
Fraumeni, J. F. (2002). Occupation and risk of esophageal and gastric cardia
adenocarcinoma. Am J Ind Med, 42, 11-22. http://dx.doi.org/10.1002/ajim.10077
Epstein, D. L.; Nolen, G. A.; Randall, J. L.; Christ, S. A.; Read, E. J.; Stober, J. A.; Smith, M. K.
(1992). Cardiopathic effects of dichloroacetate in the fetal Long-Evans rat. Teratology,
46, 225-235. http://dx.doi.org/10.1002/tera.1420460306
Ertle, T.; Henschler, D.; Mueller, G.; Spassowski, M. (1972). Metabolism of trichloroethylene in
man I The significance of trichloroethanol in long-term exposure conditions. Arch
Toxicol, 29, 171-188. http://www.ncbi.nlm.nih.gov/pubmed/4642755
Evans, M. V.; Chiu, W. A.; Okino, M. S.; Caldwell, J. C. (2009). Development of an updated
PBPK model for trichloroethylene and metabolites in mice, and its application to discern
the role of oxidative metabolism in TCE-induced hepatomegaly. Toxicol Appl
Pharmacol, 236, 329-340. http://dx.doi.Org/10.1016/j.taap.2009.02.013
Evert, M.; Schneider-Stock, R.; Dombrowski, F. (2003). Apolipoprotein A-IV mRNA
overexpression in early preneoplastic hepatic foci induced by low-number pancreatic islet
transplants in streptozotocin-diabetic rats. Pathol Res Pract, 199, 373-379.
http://www.ncbi.nlm.nih.gov/pubmed/12924437
Eyre, R.; Stevens, D.; Parker, J.; Bull, R. (1995). Acid-labile adducts to protein can be used as
indicators of the cysteine S-conjugate pathway of trichloroethene metabolism. J Toxicol
Environ Health, 46, 443-464. http://www.ncbi.nlm.nih.gov/pubmed/8523471
Eyre, R.; Stevens, D.; Parker, J.; Bull, R. (1995). Renal activation of trichloroethene and S-(l,2-
dichlorovinyl)-L-cysteine and cell proliferative responses in the kidneys of F344 rats and
B6C3F1 mice. J Toxicol Environ Health, 46, 465-481.
http://www.ncbi.nlm.nih.gov/pubmed/8523472
Fagliano, J.; Berry, M.; Bove, F.; Burke, T. (1990). Drinking water contamination and the
incidence of leukemia: An ecologic study. Am J Public Health, 80, 1209-1212.
http://www.ncbi.nlm.nih.gov/pubmed/2400032
Fahrig, R.; Madle, S.; Baumann, H. (1995). Genetic toxicology of trichloroethylene (TCE).
MutatRes, 340, 1-36. http://www.ncbi.nlm.nih.gov/pubmed/8524349
36
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Fairman, M. J.; Scott, B. B.; Toothill, C.; Losowsky, M. S. (1977). Jejunal mucosal gamma
glutamyl transferase activity in coeliac disease. Gut, 18, 484-487.
http://www.ncbi.nlm.nih.gov/pubmed/17564
FalkFilipsson, A. and Victorin, K. (2003). Comparison of available benchmark dose softwares
and models using trichloroethylene as a model substance. Regul Toxicol Pharmacol, 37,
343-355. http://www.ncbi.nlm.nih.gov/pubmed/12758215
Fan, A. (1988). Trichloroethylene: water contamination and health risk assessment. Rev Environ
Contam Toxicol, 101, 55-92. http://www.ncbi.nlm.nih.gov/pubmed/3275994
Fang, H. L.; Strom, S. C.; Cai, H.; Falany, C. N.; Kocarek, T. A.; Runge-Morris, M. (2005).
Regulation of human hepatic hydroxysteroid sulfotransferase gene expression by the
peroxisome proliferator-activated receptor alpha transcription factor. Mol Pharmacol, 67,
1257-1267. http://dx.doi.org/10.1124/mol.104.005389
Farazi, P. A. and Depinho, R. A. (2006). Hepatocellular carcinoma pathogenesis: from genes to
environment. Nat Rev Cancer, 6, 674-687. http://dx.doi.org/10.1038/nrcl934
Fattovich, G.; Stroffolini, T.; Zagni, I.; Donato, F. (2004). Hepatocellular carcinoma in cirrhosis:
incidence and risk factors. Gastroenterology, 127, S35-S50.
http://www.ncbi.nlm.nih.gov/pubmed/15508101
Fechter, L. D.; Gearhart, C.; Fulton, S.; Campbell, J.; Fisher, J.; Na, K., . . . Pouyatos, B. (2007).
Promotion of noise-induced cochlear injury by toluene and ethylbenzene in the rat.
Toxicol Sci, 98, 542-551.
Fechter, L. D.; Liu, Y.; Herr, D. W.; Crofton, K. M. (1998). Trichloroethylene ototoxicity:
Evidence for a cochlear origin. Toxicol Sci, 42, 28-35.
http://dx.doi.org/10.1093/toxsci/42.1.28
Feinberg, A. P.; Ohlsson, R.; Henikoff, S. (2006). The epigenetic progenitor origin of human
cancer. Nat Rev Genet, 7, 21-33. http://dx.doi.org/10.1038/nrgl748
Feitelson, M. A.; Sun, B.; Satiroglu Tufan, N. L.; Liu, J.; Pan, J.; Lian, Z. (2002). Genetic
mechanisms of hepatocarcinogenesis. Oncogene, 21, 2593-2604.
http://dx.doi.org/10.1038/sj.onc.1205434
Feldman, R.; Chirico-Post, J.; Proctor, S. (1988). Blink reflex latency after exposure to
trichloroethylene in well water. Arch Environ Health, 43, 143-148.
http://www.ncbi.nlm.nih.gov/pubmed/3377549
Feldman, R.; Niles, C.; Proctor, S.; Jabre, J. (1992). Blink reflex measurement of effects of
trichloroethylene exposure on the trigeminal nerve. Muscle Nerve, 15, 490-495.
http://dx.doi.org/10.1002/mus.880150413
Ferlay, J.; Bray, F.; Pisani, P.; Parkin, D. M. (2004). GLOBOCAN 2002: Cancer incidence,
mortality and prevalence worldwide.
Fernandez, J. G.; Droz, P. O.; Humbert, B. E.; Caperos, J. R. (1977). Trichloroethylene
exposure: simulation of uptake, excretion, and metabolism using a mathematical model.
Occup Environ Med, 34, 43-55. http://www.ncbi.nlm.nih.gov/pubmed/843463
Ferreira-Gonzalez, A.; DeAngelo, A. B.; Nasim, S.; Garrett, C. T. (1995). Ras oncogene
activation during hepatocarcinogenesis in B6C3F1 male mice by dichloroacetic and
trichloroacetic acids. Carcinogenesis, 16, 495-500.
http://dx.doi.Org/10.1093/carcin/16.3.495
Fevotte, J.; Charbotel, B.; Muller-Beaute, P.; Martin, J. L.; Hours, M.; Bergeret, A. (2006). Case-
control study on renal cell cancer and occupational exposure to trichloroethylene. Part I:
37
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Exposure assessment. Ann Occup Hyg, 50, 765-775.
http://dx.doi.org/10.1093/annhyg/mel040
Figgs, L.; Dosemeci, M.; Blair, A. (1995). United States non-Hodgkin's lymphoma surveillance
by occupation 1984-1989: a twenty-four state death certificate study. Am J Ind Med, 27,
817-835. http://www.ncbi.nlm.nih.gov/pubmed/7645576
Filser, J. G. and Bolt, H. M. (1979). Pharmacokinetics of halogenated ethylenes in rats. Arch
Toxicol, 42, 123-136. http://dx.doi.org/10.1007/BF00316492
Finckh, A.; Cooper, G. S.; Chibnik, L. B.; Costenbader, K. H.; Watts, J.; Pankey, H., . . .
Karlson, E. W. (2006). Occupational silica and solvent exposures and risk of systemic
lupus erythematosus in urban women. Arthritis Rheum, 54, 3648-3654.
http ://dx. doi. org /10.1002/art.22210
Fiserova-Bergerova, V.; Tichy, M.; Di Carlo, F. J. (1984). Effects of biosolubility on pulmonary
uptake and disposition of gases and vapors of lipophilic chemicals. Drug Metab Rev, 15,
1033-1070.
Fisher, J. (2000). Physiologically based pharmacokinetic models for trichloroethylene and its
oxidative metabolites. Environ Health Perspect, 108, 265-273.
http://www.ncbi.nlm.nih.gov/pubmed/10807557
Fisher, J. W. and Allen, B. C. (1993). Evaluating the risk of liver cancer in humans exposed to
trichloroethylene using physiological models. Risk Anal, 13, 87-95.
http://www.ncbi.nlm.nih.gov/pubmed/8451463
Fisher, J. W.; Gargas, M. L.; Allen, B. C.; Andersen, M. E. (1991). Physiologically based
pharmacokinetic modeling with trichloroethylene and its metabolite, trichloroacetic acid,
in the rat and mouse. Toxicol Appl Pharmacol, 109, 183-195.
http://www.ncbi.nlm.nih.gov/pubmed/2068722
Fisher, J. W.; Mahle, D.; Abbas, R. (1998). A human physiologically based pharmacokinetic
model for trichloroethylene and its metabolites, trichloroacetic acid and free
trichloroethanol. Toxicol Appl Pharmacol, 152, 339-359.
http://dx.doi.org/10.1006/taap.1998.8486
Fisher, J. W.; Whittaker, T. A.; Taylor, D. H.; 3rd, C. H.; Andersen, M. E. (1989).
Physiologically based pharmacokinetic modeling of the pregnant rat: a multiroute
exposure model for trichloroethylene and its metabolite, trichloroacetic acid. Toxicol
Appl Pharmacol, 99, 395-414. http://dx.doi.org/10.1016/0041-008X(89)90149-X
Fisher, J. W.; Whittaker, T. A.; Taylor, D. H.; Clewell, H. J., Ill; Andersen, M. E. (1990).
Physiologically based pharmacokinetic modeling of the lactating rat and nursing pup: a
multiroute exposure model for trichloroethylene and its metabolite, trichloroacetic acid.
Toxicol Appl Pharmacol, 102, 497-513. http://www.ncbi.nlm.nih.gov/pubmed/2315918
Fisher, J.; Channel, S.; Eggers, J.; Johnson, P.; MacMahon, K.; Goodyear, C., . . . Graeter, L.
(2001). Trichloroethylene, trichloroacetic acid, and dichloroacetic acid: Do they affect
fetal rat heart development. Int J Toxicol, 20, 257-267.
http://www.ncbi.nlm.nih.gov/pubmed/11766123
Fisher, J.; Mahle, D.; Bankston, L.; Greene, R.; Gearhart, J. (1997). Lactational transfer of
volatile chemicals in breast milk. Am Ind Hyg Assoc J, 58, 425-431.
http://dx.doi.org/10.1080/15428119791012667
Fitzgerald, J. E.; Sanyer, J. L.; Schardein, J. L.; Lake, R. S.; McGuire, E. J.; de la Iglesia, F. A.
(1981). Carcinogen bioassay and mutagenicity studies with the hypolipidemic agent
38
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
gemfibrozil. J Natl Cancer Inst, 67, 1105-1116.
http://www.ncbi.nlm.nih.gov/pubmed/7029098
Fleming-Jones, M. E. and Smith, R. E. (2003). Volatile organic compounds in foods: A five year
study. J AgricFood Chem, 51, 8120-8127.
Flodin, U.; Fredriksson, M.; Persson, B. (1987). Multiple myeloma and engine exhausts, fresh
wood, and creosote: a case-referent study. Am J Ind Med, 12, 519-529.
Flood, T. (1988). Report on mortality in Maricopa County.
Flood, T. J. (1997). Case-referent study of childhood leukemia in Maricopa County, Arizona:
1965-1990. Phoenix, AZ: Arizona Dept. of Health Services, Bureau of Epidemiology
and Disease Control Services, Office of Chronic Disease Epidemiology, Office of
Environmental Health.
Forand, S. P. (2004). Leukaemia incidence among workers in the shoe and boot manufacturing
industry: a case-control study. Environ Health, 3, 7. http://dx.doi.org/10.1186/1476-
069X-3-7
Ford, E.; Rhodes, S.; McDiarmid, M.; Schwartz, S.; Brown, J. (1995). Deaths from acute
exposure to trichloroethylene. J Occup Environ Med, 37, 749-754.
http://www.ncbi.nlm.nih.gov/pubmed/7670923
Forkert, P. (1999). 1,1-Dichloroethylene-induced Clara cell damage is associated with in situ
formation of the reactive epoxide. Immunohistochemical detection of its glutathione
conjugate. Am J Respir Cell Mol Biol, 20, 1310-1318.
http://www.ncbi.nlm.nih.gov/pubmed/10340951
Forkert, P. (1999). In vivo formation and localization of 1,1-dichloroethylene epoxide in murine
liver: identification of its glutathione conjugate 2-S-glutathionyl acetate. J Pharmacol Exp
Ther, 290, 1299-1306. http://www.ncbi.nlm.nih.gov/pubmed/10454507
Forkert, P. G. and Birch, D. W. (1989). Pulmonary toxicity of trichloroethylene in mice: covalent
binding and morphological manifestations. Drug Metab Dispos, 17, 106-113.
http://www.ncbi.nlm.nih.gov/pubmed/2566461
Forkert, P. G. and Forkert, L. (1994). Trichloroethylene induces pulmonary fibrosis in mice. Can
J Physiol Pharmacol, 72, 205-210. http://www.ncbi.nlm.nih.gov/pubmed/8069766
Forkert, P. G.; Sylvestre, P. L.; Poland, J. S. (1985). Lung injury induced by trichloroethylene.
Toxicology, 35, 143-160. http://www.ncbi.nlm.nih.gov/pubmed/4002240
Forkert, P.; Baldwin, R.; Millen, B.; Lash, L.; Putt, D.; Shultz, M.; Collins, K. (2005).
Pulmonary bioactivation of trichloroethylene to chloral hydrate: relative contributions of
CYP2E1, CYP2F, and CYP2B1. Drug Metab Dispos, 33, 1429-1437.
http://dx.doi.Org/10.l 124/dmd. 105.005074
Forkert, P.; Lash, L.; Nadeau, V.; Tardif, R.; Simmonds, A. (2002). Metabolism and toxicity of
trichloroethylene in epididymis and testis. Toxicol Appl Pharmacol, 182, 244-254.
http://www.ncbi.nlm.nih.gov/pubmed/12183104
Forkert, P.; Lash, L.; Tardif, R.; Tanphaichitr, N.; Vandevoort, C.; Moussa, M. (2003).
Identification of trichloroethylene and its metabolites in human seminal fluid of workers
exposed to trichloroethylene. Drug Metab Dispos, 31, 306-311.
http://www.ncbi.nlm.nih.gov/pubmed/12584157
Forkert, P.; Malkinson, A.; Rice, P.; Moussa, M. (1999). Diminished CYP2E1 expression and
formation of 2-S-glutathionyl acetate, a glutathione conjugate derived from 1,1-
39
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
dichloroethylene epoxide, in murine lung tumors. Drug Metab Dispos, 27, 68-73.
http://www.ncbi.nlm.nih.gov/pubmed/9884311
Forkert, P.; Millen, B.; Lash, L.; Putt, D.; Ghanayem, B. (2006). Pulmonary bronchiolar
cytotoxicity and formation of dichloroacetyl lysine protein adducts in mice treated with
trichloroethylene. J Pharmacol Exp Ther, 316, 520-529.
http://dx.doi.org/10.1124/jpet.105.093062
Fort, D. J.; Stover, E. L.; Rayburn, J. R.; Hull, M.; Bantle, J. A. (1993). Evaluation of the
developmental toxicity of trichloroethylene and detoxification metabolites using
Xenopus. Birth Defects Res B Dev Reprod Toxicol, 13, 35-45.
http://www.ncbi.nlm.nih.gov/pubmed/8100651
Fort, D.; Rayburn, J.; DeYoung, D.; Bantle, J. (1991). Assessing the efficacy of an Aroclor 1254-
induced exogenous metabolic activation system for FETAX. Drug Chem Toxicol, 14,
143-160. http://dx.doi.org/10.3109/01480549109017873
Fox, A.; Sullivan, B.; Buffini, J.; Neichin, M.; Nicora, R.; Hoehler, F., . . . Stoltz, R. (1996).
Reduction of serum lactate by sodium dichloroacetate, and human pharmacokinetic-
pharmacodynamic relationships. J Pharmacol Exp Ther, 279, 686-693.
http://www.ncbi.nlm.nih.gov/pubmed/8930172
Fox, A.; Yang, X.; Murli, H.; Lawlor, T.; Cifone, M.; Reno, F. (1996). Absence of mutagenic
effects of sodium dichloroacetate. Fundam Appl Toxicol, 32, 87-95.
http://www.ncbi.nlm.nih.gov/pubmed/8812237
Fox, T. R.; Schumann, A. M.; Watanabe, P. G.; Yano, B. L.; Maher, V. M.; McCormick, J. J.
(1990). Mutational analysis of the H-ras oncogene in spontaneous C57BL/6 x C3H/He
mouse liver tumors and tumors induced with genotoxic and nongenotoxic
hepatocarcinogens. Cancer Res, 50, 4014-4019.
http://www.ncbi.nlm.nih.gOv/pubmed/2191770
Fraga, M. F.; Ballestar, E.; Villar-Garea, A.; Boix-Chornet, M.; Espada, J.; Schotta, G., . . .
Esteller, M. (2005). Loss of acetylation at Lysl6 and trimethylation at Lys20 of histone
H4 is a common hallmark of human cancer. Nat Genet, 37, 391-400.
http://dx.doi.org/10.1038/ngl531
Fraga, M. F.; Herranz, M.; Espada, J.; Ballestar, E.; Paz, M. F.; Ropero, S., . . . Esteller, M.
(2004). A mouse skin multistage carcinogenesis model reflects the aberrant DNA
methylation patterns of human tumors. Cancer Res, 64, 5527-5534.
http://dx.doi.org/10.1158/0008-5472.CAN-03-4061
Franco, A.; Costoya, M. A.; Roca, E. (2007). Estimating risk during showering exposure to
VOCs of workers in a metal-degreasing facility. J Toxicol Environ Health A, 70, 627-
637. http://dx.doi.org/10.1080/15287390600974551
Franco, J. L.; Teixeira, A.; Meotti, F. C.; Ribas, C. M.; Stringari, J.; Garcia Pomblum, S. C., . . .
Farina, M. (2006). Cerebellar thiol status and motor deficit after lactational exposure to
methylmercury. Environ Res, 102, 22-28. http://dx.doi.Org/10.1016/j.envres.2006.02.003
Fredriksson, A.; Danielsson, B. R. G.; Eriksson, P. (1993). Altered behaviour in adult mice
orally exposed to tri- and tetrachloroethylene as neonates. Toxicol Lett, 66, 13-19.
http://dx.doi.org/10.1016/0378-4274(93)90074-8
Fredriksson, M.; Bengtsson, N. O.; Hardell, L.; Axelson, O. (1989). Colon cancer, physical
activity, and occupational exposures. A case-control study. Cancer, 63, 1838-1842.
http://www.ncbi.nlm.nih.gov/pubmed/2702592
40
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Fritschi, L. and Siemiatycki, J. (1996). Lymphoma, myeloma and occupation: results of a case-
control study. Int J Cancer, 67, 498-503. http://dx.doi.org/10.1002/(SICI)1097-
0215(19960807)67:4<498::AID-nC6>3.0.CO;2-N
Fritschi, L. and Siemiatycki, J. (1996). Melanoma and occupation: results of a case-control
study. Occup Environ Med, 53, 168-173. http://www.ncbi.nlm.nih.gov/pubmed/8704857
Fritschi, L.; Nadon, L.; Benke, G.; Lakhani, R.; Latreille, B.; Parent, M. E.; Siemiatycki, J.
(2003). Validation of expert assessment of occupational exposures. Am J Ind Med, 43,
519-522. http://dx.doi.org/10.1002/ajim.10208
Fritschi, L.; Siemiatycki, J.; Richardson, L. (1996). Self-assessed versus expert-assessed
occupational exposures. Am J Epidemiol, 144, 521-527.
http://www.ncbi.nlm.nih.gov/pubmed/8781468
Frayland, L.; Madsen, L.; Vaagenes, H.; Totland, G. K.; Auwerx, J.; Kryvi, H., . . . Berge, R. K.
(1997). Mitochondrion is the principal target for nutritional and pharmacological control
of triglyceride metabolism. J Lipid Res, 38, 1851-1858.
http://www.ncbi.nlm.nih.gov/pubmed/9323594
Fujita, J.; Marino, M. W.; Wada, H.; Jungbluth, A. A.; Mackrell, P. J.; Rivadeneira, D. E., . . .
Daly, J. M. (2001). Effect of TNF gene depletion on liver regeneration after partial
hepatectomy in mice. Surgery, 129, 48-54.
http://www.ncbi.nlm.nih.gOv/pubmed/l 1150033
Fukuda, K.; Takemoto, K.; Tsuruta, H. (1983). Inhalation carcinogenicity of trichloroethylene in
mice and rats. Ind Health, 21, 243-254. http://www.ncbi.nlm.nih.gov/pubmed/6654707
Furge, K. A.; Dykema, K.; Petillo, D.; Westphal, M.; Zhang, Z. F.; Kort, E. J.; Teh, B. T. (2007).
Combining differential expression, chromosomal and pathway analyses for the molecular
characterization of renal cell carcinoma. 1, S21-S27.
http://www.ncbi.nlm.nih.gov/pubmed/18542781
Furnus, C. C.; Ulrich, M. A.; Terreros, M. C.; Dulout, F. N. (1990). The induction of aneuploidy
in cultured Chinese hamster cells by propionaldehyde and chloral hydrate. Mutagenesis,
5, 323-326. http://www.ncbi.nlm.nih.gov/pubmed/2398816
Fiirstenberger, G. and Senn, H. J. (2002). Insulin-like growth factors and cancer. Lancet Oncol,
3, 298-302. http://www.ncbi.nlm.nih.gov/pubmed/12067807
Fuscoe, J. C.; Afshari, A. J.; George, M. H.; DeAngelo, A. B.; Tice, R. R.; Salman, T.; Allen, J.
W. (1996). In vivo genotoxicity of dichloroacetic acid: Evaluation with the mouse
peripheral blood micronucleus assay and the single cell gel assay. Environ Mol Mutagen,
27, 1-9. http://dx.doi.org/10.1002/(SICI)1098-2280(1996)27:1<1::AID-EM1>3.0.CO;2-L
Gaglio, P. J.; Liu, H.; Dash, S.; Cheng, S.; Dunne, B.; Ratterree, M., . . . LaBrecque, D. (2002).
Liver regeneration investigated in a non-human primate model (Macaca mulatta). J
Hepatol, 37, 625-632. http://www.ncbi.nlm.nih.gov/pubmed/12399229
Gagnaire, F.; Marignac, B.; Langlais, C.; Bonnet, P. (2001). Ototoxicity in rats exposed to ortho-
, meta- and para-xylene vapours for 13 weeks. Pharmacol Toxicol, 89, 6-14.
http://www.ncbi.nlm.nih.gov/pubmed/11484912
Gallou, C.; Longuemaux, S.; Delomenie, C.; Mejean, A.; Martin, N.; Martinet, S., . . . Dupret, J.
(2001). Association of GSTT1 non-null andNATl slow/rapid genotypes with von
Hippel-Lindau tumour suppressor gene transversions in sporadic renal cell carcinoma.
Pharmacogenetics, 11, 521-535. http://www.ncbi.nlm.nih.gov/pubmed/11505222
41
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
Galloway, S. M.; Armstrong, M. J.; Reuben, C.; Colman, S.; Brown, B.; Cannon, C., . . . Zeiger,
E. (1987). Chromosome aberrations and sister chromatid exchanges in Chinese hamster
ovary cells: Evaluations of 108 chemicals. Environ Mol Mutagen, 10, 1-175.
http://dx.doi.org/10.1002/em.2850100502
Gama, C. and Meira, J. B. (1978). Occupational acro-osteolysis. J Bone Joint Surg Am, 60, 86-
90.
Gamberale, G.; Annwall, G.; Olson, B. A. (1976). Exposure to trichloroethylene: III
psychological functions. Scand J Work Environ Health, 4, 220-224.
http://www.ncbi.nlm.nih.gov/pubmed/1019596
Gandhi, M.; Aweeka, F.; Greenblatt, R. M.; Blaschke, T. F. (2004). Sex differences in
pharmacokinetics and pharmacodynamics. Annu Rev Pharmacol Toxicol, 44, 499-523.
http://dx.doi.Org/10.l 146/annurev.pharmtox.44.101802.121453
Gandillet, A.; Alexandre, E.; Royer, C.; Cinqualbre, J.; Jaeck, D.; Richert, L. (2003). Hepatocyte
ploidy in regenerating livers after partial hepatectomy, drug-induced necrosis, and
cirrhosis. Eur Surg Res, 35, 148-160. http://dx.doi.org/10.1159/000070044
Garabrant, D. H.; Lacey J. V., J.; Laing, T. J.; Gillespie, B. W.; Mayes, M. D.; Cooper, B. C.;
Schottenfeld, D. (2003). Scleroderma and solvent exposure among women. Am J
Epidemiol, 157, 493-500. http://dx.doi.org/10.1093/aje/kwf223
Garabrant, D.; Held, J.; Langholz, B.; Bernstein, L. (1988). Mortality of aircraft manufacturing
workers in southern California. Am J Ind Med, 13, 683-693.
http://www.ncbi.nlm.nih.gov/pubmed/3389363
Garfinkel, L. and Stellman, S. D. (1988). Smoking and lung cancer in women: findings in a
prospective study. Cancer Res, 48, 6951-6955.
http://www.ncbi.nlm.nih.gov/pubmed/3180103
Gargas, M. L.; Burgess, R. J.; Voisard, D. E.; Cason, G. H.; Andersen, M. E. (1989). Partition
coefficients of low-molecular-weight volatile chemicals in various liquids and tissues.
Toxicol Appl Pharmacol, 98, 87-99. http://www.ncbi.nlm.nih.gov/pubmed/2929023
Gargas, M. L.; Seybold, P. G.; Andersen, M. E. (1988). Modeling the tissue solubilities and
metabolic rate constant (Vmax) of halogenated methanes, ethanes, and ethylenes. Toxicol
Lett, 43, 235-256. http://dx.doi.org/10.1016/0378-4274(88)90031-8
Gart, J. J.; Chu, K. C.; Tarone, R. E. (1979). Statistical issues in interpretation of chronic
bioassay tests for carcinogenicity. J Natl Cancer Inst, 62, 957-974.
http://www.ncbi.nlm.nih.gov/pubmed/285297
Garte, S.; Gaspari, L.; Alexandrie, A. K.; Ambrosone, C.; Autrup, H.; Autrup, J. L., . . . Taioli, E.
(2001). Metabolic gene polymorphism frequencies in control populations. Cancer
Epidemiol Biomarkers Prev, 10, 1239-1248.
http://www.ncbi.nlm.nih.gov/pubmed/11751440
Gash, D.; Rutland, K.; Hudson, N.; Sullivan, P.; Bing, G.; Cass, W., . . . Prince, T. (2008).
Trichloroethylene: Parkinsonism and complex 1 mitochondrial neurotoxicity. Ann
Neurol, 63, 184-192. http://dx.doi.org/10.1002/ana.21288
Ge, R.; Tao, L.; Kramer, P. M.; Cunningham, M. L.; Pereira, M. A. (2002). Effect of peroxisome
proliferators on the methylation and protein level of the c-myc protooncogene in B6C3F1
mice liver. JBiochem Mol Toxicol, 16, 41-47. http://dx.doi.org/10.1002/jbt.10019
42
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Ge, R.; Wang, W.; Kramer, P. M.; Yang, S.; Tao, L.; Pereira, M. A. (2001). Wy-14,643-induced
hypomethylation of the c-myc gene in mouse liver. Toxicol Sci, 62, 28-35.
http://www.ncbi.nlm.nih.gov/pubmed/11399790
Ge, R.; Yang, S.; Kramer, P. M.; Tao, L.; Pereira, M. A. (2001). The effect of dichloroacetic acid
and trichloroacetic acid on DNA methylation and cell proliferation in B6C3F1 mice. J
Biochem Mol Toxicol, 15, 100-106. http://dx.doi.Org/10.1002/jbt.5
Gelderblom, W. C.; Galendo, D.; Abel, S.; Swanevelder, S.; Marasas, W. F.; Wild, C. P. (2001).
Cancer initiation by fumonisin B(l) in rat liver—role of cell proliferation. Cancer Lett,
169, 127-137. http://www.ncbi.nlm.nih.gov/pubmed/11431101
Gelfand, A. E. and Smith, A. F. M. (1990). Sampling based approaches to calculating marginal
densities. J Am Stat Assoc, 85, 398 - 409.
Gelman, A.; Bois, F.; Jiang, J. (1996). Physiological pharmacokinetic analysis using population
modeling and informative prior distributions. J Am Stat Assoc, 91, 1400-1412.
Gelman, A.; Carlin, J. B.; Stern, H. S.; Rubin, D. B. (2004). Bayesian data analysis: Chapman
and Hall.
George, J. D.; Reel, J. R.; Myers, C. B.; Lawton, A. D. (1985). Trichloroethylene: Reproduction
and fertility assessment in CD-I mice when administered in the feed. (RTI-152).
Research Triangle Park, NC: NIHEHS.
George, J. D.; Reel, J. R.; Myers, C. B.; Lawton, A. D. (1986). Trichloroethylene: Reproduction
and fertility assessment in F344 rats when administered in the feed. Retrieved from
file://L:\Lab\NCEA\TCE Documents\TCE PDFs4\ George et al rats 1986 Part l.pdf.
George, J.; Byth, K.; Farrell, G. C. (1995). Age but not gender selectively affects expression of
individual cytochrome P450 proteins in human liver. Biochem Pharmacol, 50, 727-730.
http://www.ncbi.nlm.nih.gov/pubmed/7669077
George, M. H.; Moore, T.; Kilburn, S.; Olson, G. R.; DeAngelo, A. B. (2000). Carcinogenicity
of chloral hydrate administered in drinking water to the male F344/N rat and male
B6C3F1 mouse. Toxicol Pathol, 28, 610-618.
http://www.ncbi.nlm.nih.gov/pubmed/10930049
Ghantous, H.; Danielsson, B. R. G.; Dencker, L.; Gorczak, J.; Vesterberg, O. (1986).
Trichloroacetic acid accumulates in murine amniotic fluid after tri- and
tetrachloroethylene inhalation. Basic Clin Pharmacol Toxicol, 58, 105-114.
http://dx.doi.Org/10.llll/j.1600-0773.1986.tb00078.x
Ghoshal, A. K. and Farber, E. (1984). The induction of liver cancer by dietary deficiency of
choline and methionine without added carcinogens. Carcinogenesis, 5, 1367-1370.
http://www.ncbi.nlm.nih.gov/pubmed/6488458
Giardino, N. J. and Andelman, J. B. (1996). Characterization of the emissions of
trichloroethylene, chloroform, and l,2-dibromo-3-chloropropane in a full-size,
experimental shower. J Expo Sci Environ Epidemiol, 6, 413-423.
Gibson, D.; Aardema, M.; Kerckaert, G.; Carr, G.; Brauninger, R.; LeBoeuf, R. (1995).
Detection of aneuploidy-inducing carcinogens in the Syrian hamster embryo (SHE) cell
transformation assay. MutatRes, 343, 7-24.
http://www.ncbi.nlm.nih.gov/pubmed/7753109
Gilbert, K. M.; Whitlow, A. B.; Pumford, N. R. (2004). Environmental contaminant and
disinfection by-product trichloroacetaldehyde stimulates T cells in vitro. Int
Immunopharmacol, 4, 25-36. http://dx.doi.Org/10.1016/j.intimp.2003.10.001
43
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Gilbert, K.; Griffin, J.; Pumford, N. (1999). Trichloroethylene activates CD4+ T cells: potential
role in an autoimmune response. Drug Metab Rev, 31, 901-916.
http://dx.doi.org/10.1081/DMR-100101945
Gilbert, K.; Pumford, N.; Blossom, S. (2006). Environmental contaminant trichloroethylene
promotes autoimmune disease and inhibits T-cell apoptosis in MRL(+/+) mice. J
Immunotoxicol, 3, 263-267. http://dx.doi.org/10.1080/15476910601023578
Gilkeson, G. S.; Keil, D.; Peden-Adams, M. B. J. D. B. T. I. T. t. s. b. o. r. a. (2004). Immune
effects of trichloroethylene on autoimmune disease in mice. 87 - 98 PUBL- The Medical
University of South Carolina Press.
Gilks, W. R.; Richardson, S.; Spiegelhalter, D. J. (1996). Markov Chain Monte Carlo in Practice.
Boca Raton, FL: Chapman & Hall/CRC.
Giller, S.; Le Curieux, F.; Gauthier, L.; Erb, F.; Marzin, D. (1995). Genotoxicity assay of chloral
hydrate and chloropicrine. MutatRes, 348, 147-152.
http://www.ncbi.nlm.nih.gov/pubmed/8544866
Giller, S.; Le Curieux, F.; Erb, F.; Marzin, D. (1997). Comparative genotoxicity of halogenated
acetic acids found in drinking water. Mutagenesis, 12, 321-328.
http://www.ncbi.nlm.nih.gov/pubmed/9379909
Ginsberg, G.; Hattis, D.; Miller, R.; Sonawane, B. (2004). Pediatric pharmacokinetic data:
Implications for environmental risk assessment for children. Pediatrics, 113, 973-983.
Ginsberg, G.; Hattis, D.; Russ, A.; Sonawane, B. (2005). Pharmacokinetic and pharmacodynamic
factors that can affect sensitivity to neurotoxic sequelae in elderly individuals. Environ
Health Perspect, 113, 1243-1249. http://dx.doi.org/10.1289/ehp.7568
Ginsberg, G.; Hattis, D.; Sonawane, B. (2004). Incorporating pharmacokinetic differences
between children and adults in assessing children's risks to environmental toxicants.
Toxicol Appl Pharmacol, 198, 164-183. http://dx.doi.Org/10.1016/j.taap.2003.10.010
Ginsberg, G.; Hattis, D.; Sonawane, B.; Russ, A.; Banati, P.; Kozlak, M., . . . Goble, R. (2002).
Evaluation of child/adult pharmacokinetic differences from a database derived from the
therapeutic drug literature. Toxicol Sci, 66, 185-200.
http://www.ncbi.nlm.nih.gov/pubmed/11896285
Gochfeld, M. (2007). Framework for gender differences in human and animal toxicology.
Environ Res, 104, 4-21. http://dx.doi.Org/10.1016/j.envres.2005.12.005
Goel, S.; Rao, G.; Pandya, K.; Shanker, R. (1992). Trichloroethylene toxicity in mice: A
biochemical, hematological and pathological assessment. Indian J Exp Biol, 30, 402-406.
http://www.ncbi.nlm.nih.gov/pubmed/1459618
Goeptar, A. R.; Commandeur, J. N. M.; van Ommen, B.; van Bladeren, P. J.; Vermeulen, N. P.
E. (1995). Metabolism and kinetics of trichloroethylene in relation to toxicity and
carcinogenicity Relevance of the mercapturic acid pathway. Chem Res Toxicol, 8, 3-21.
http://www.ncbi.nlm.nih.gov/pubmed/7703363
Goh, V. H.; Chia, S. E.; Ong, C. N. (1998). Effects of chronic exposure to low doses of
trichloroethylene on steroid hormone and insulin levels in normal men. Environ Health
Perspect, 106, 41-44. http://www.ncbi.nlm.nih.gov/pubmed/9417767
Gold, L. S. and E, Z. (Eds.). (1997). Handbook of carcinogenic potency and genotoxicity
databases. Boca Raton: CRC Press.
Gold, L. S.; Stewart, P. A.; Milliken, K.; Purdue, M.; Severson, R.; Seixas, N., . . . De Roos, A.
J. (In Press). The relationship between multiple myeloma and occupational exposure to
44
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
six chlorinated solvents. Occup Environ Med.
http://dx.doi.Org/10.l 136/oem.2009.054809
Goldberg, M.; Parent, M.; Siemiatycki, J.; Desy, M.; Nadon, L.; Richardson, L., . . . Valois, M.
(2001). A case-control study of the relationship between the risk of colon cancer in men
and exposures to occupational agents. Am J Ind Med, 39, 531-546.
http://www.ncbi.nlm.nih.gov/pubmed/11385637
Goldberg, S.; Lebowitz, M.; Graver, E.; Hicks, S. (1990). An association of human congenital
cardiac malformations and drinking water contaminants. J Am Coll Cardiol, 16, 155-164.
http://www.ncbi.nlm.nih.gov/pubmed/2358589
Goldsworthy, T. L. and Popp, J. A. (1987). Chlorinated hydrocarbon-induced peroxisomal
enzyme activity in relation to species and organ carcinogenicity. Toxicol Appl
Pharmacol, 88, 225-233. http://dx.doi.org/10.1016/0041-008X(87)90008-1
Goldsworthy, T. L.; Lyght, O.; Burnett, V. L.; Popp, J. A. (1988). Potential role of [alpha]-
2[mu]-globulin, protein droplet accumulation, and cell replication in the renal
carcinogenicity of rats exposed to trichloroethylene, perchloroethylene, and
pentachloroethane. Toxicol Appl Pharmacol, 96, 367-379.
http://dx.doi. org/10.1016/0041-008X(88)90095-6
Gomez, M.; Cocco, P.; Dosemeci, M.; Stewart, P. (1994). Occupational exposure to chlorinated
aliphatic hydrocarbons: Job exposure matrix. Am J Ind Med, 26, 171-183.
http://www.ncbi.nlm.nih.gov/pubmed/7977394
Gonzalez-Leon, A.; Merdink, J.; Bull, R.; Schultz, I. (1999). Effect of pre-treatment with
dichloroacetic or trichloroacetic acid in drinking water on the pharmacokinetics of a
subsequent challenge dose in B6C3F1 mice. Chem Biol Interact, 123, 239-253.
http://www.ncbi.nlm.nih.gov/pubmed/10654841
Gorla, G. R.; Malhi, H.; Gupta, S. (2001). Polyploidy associated with oxidative injury attenuates
proliferative potential of cells. J Cell Sci, 114, 2943-2951.
http://www.ncbi.nlm.nih.gov/pubmed/11686298
Graham, D. J.; Staffa, J. A.; Shatin, D.; Andrade, S. E.; Schech, S. D.; La Grenade, L., . . . Piatt,
R. (2004). Incidence of hospitalized rhabdomyolysis in patients treated with lipid-
lowering drugs. JAMA, 292, 2585-2590. http://dx.doi.org/10.1001/jama.292.21.2585
Gramantieri, L.; Melchiorri, C.; Chieco, P.; Gaiani, S.; Stecca, B.; Casali, A.; Bolondi, L. (1996).
Alteration of DNA ploidy and cell nuclearity in human hepatocellular carcinoma
associated with HBV infection. J Hepatol, 25, 848-853.
http://www.ncbi.nlm.nih.gov/pubmed/9007712
Grandjean, E.; Munchinger, R.; Turrian, V.; Haas, P. A.; Knoepfel, H.-K.; Rosenmund, H.
(1955). Investigations into the effects of exposure to trichlorethylene in mechanical
engineering. Occup Environ Med, 12, 131-142.
http://www.ncbi.nlm.nih.gov/pubmed/14363593
Grasl-Kraupp, B.; Waldhor, T.; Huber, W.; Schulte-Hermann, R. (1993). Glutathione S-
transferase isoenzyme patterns in different subtypes of enzyme-altered rat liver foci
treated with the peroxisome proliferator nafenopin or with phenobarbital. Carcinogenesis,
14, 2407-2412. http://www.ncbi.nlm.nih.gov/pubmed/8242872
Grawe, J.; Niisse, M.; Adler, I. (1997). Quantitative and qualitative studies of micronucleus
induction in mouse erythrocytes using flow cytometry. I. Measurement of micronucleus
induction in peripheral blood polychromatic erythrocytes by chemicals with known and
45
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
suspected genotoxicity. Mutagenesis, 12, 1-8.
http://www.ncbi.nlm.nih.gov/pubmed/9025090
Green, L. and Lash, T. (1999). Re: "Renal cell cancer correlated with occupational
exposure to trichloroethylene". J Cancer Res Clin Oncol, 125, 430-432.
http://www.ncbi.nlm.nih.gov/pubmed/10394965
Green, T. (2000). Pulmonary toxicity and carcinogenicity of trichloroethylene: species
differences and modes of action. Environ Health Perspect, 108 Suppl 2, 261-264.
http://www.ncbi.nlm.nih.gov/pubmed/10807556
Green, T. and Prout, M. S. (1985). Species differences in response to trichloroethylene: II
Biotransformation in rats and mice. Toxicol Appl Pharmacol, 79, 401-411.
http://www.ncbi.nlm.nih.gov/pubmed/3929429
Green, T.; Dow, J.; Ellis, M. K.; Foster, J. R.; Odum, J. (1997). The role of glutathione
conjugation in the development of kidney tumours in rats exposed to trichloroethylene.
ChemBiol Interact, 105, 99-117. http://www.ncbi.nlm.nih.gov/pubmed/9251723
Green, T.; Dow, J.; Foster, J. (2003). Increased formic acid excretion and the development of
kidney toxicity in rats following chronic dosing with trichloroethanol, a major metabolite
of trichloroethylene. Toxicology, 191, 109-119.
http://www.ncbi.nlm.nih.gov/pubmed/12965114
Green, T.; Dow, J.; Foster, J.; Hext, P. (1998). Formic acid excretion in rats exposed to
trichloroethylene: a possible explanation for renal toxicity in long-term studies.
Toxicology, 127, 39-47. http://www.ncbi.nlm.nih.gov/pubmed/9699792
Green, T.; Dow, J.; Ong, C.; Ng, V.; Ong, H.; Zhuang, Z., . . . Bloemen, L. (2004). Biological
monitoring of kidney function among workers occupationally exposed to
trichloroethylene. Occup Environ Med, 61, 312-317.
http://www.ncbi.nlm.nih.gov/pubmed/15031388
Green, T.; Mainwaring, G. W.; Foster, J. R. (1997). Trichloroethylene-induced mouse lung
tumors: studies of the mode of action and comparisons between species. Fundam Appl
Toxicol, 37, 125-130. http://dx.doi.org/10.1006/faat.1997.2312
Green, T.; Odum, J.; Nash, J. A.; Foster, J. R. (1990). Perchloroethylene-induced rat kidney
tumors: An investigation of the mechanisms involved and their relevance to humans.
Toxicol Appl Pharmacol, 103, 77-89. http://dx.doi.org/10.1016/0041-008X(90)90264-U
Greenberg, M.; Burton, G.; Fisher, J. (1999). Physiologically based pharmacokinetic modeling of
inhaled trichloroethylene and its oxidative metabolites in B6C3F1 mice. Toxicol Appl
Pharmacol, 154, 264-278. http://dx.doi.org/10.1006/taap.1998.8594
Greene, A. K.; Wiener, S.; Puder, M.; Yoshida, A.; Shi, B.; Perez-Atayde, A. R., . . . O'Reilly,
M. S. (2003). Endothelial-directed hepatic regeneration after partial hepatectomy. Ann
Surg, 237, 530-535. http://dx.doi.org/10.1097/01.SLA.0000059986.96051.EA
Greenland, S. (1992). A semi-Bayes approach to the analysis of correlated multiple associations,
with an application to an occupational cancer-mortality study. Stat Med, 11, 219-230.
http://www.ncbi.nlm.nih.gov/pubmed/1579760
Greenland, S.; Salvan, A.; Wegman, D. H.; Hallock, M. F.; Smith, T. J. (1994). A case-control
study of cancer mortality at a transformer-assembly facility. Int Arch Occup Environ
Health, 66, 49-54. http://dx.doi.org/10.1007/BF00386579
46
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Gregersen, P. (1988). Neurotoxic effects of organic solvents in exposed workers: two controlled
follow-up studies after 55 and 106 years. Am J Ind Med, 14, 681-701.
http://www.ncbi.nlm.nih.gov/pubmed/2852897
Greim, H.; Bonse, G.; Radwan, Z.; Reichert, D.; Henschler, D. (1975). Mutagenicity in vitro and
potential carcinogenicity of chlorinated ethylenes as a function of metabolic oxirane
formation. Biochem Pharmacol, 24, 2013-2017. http://dx.doi.org/10.1016/0006-
2952(75)90396-2
Greten, F. R.; Eckmann, L.; Greten, T. F.; Park, J. M.; Li, Z. W.; Egan, L. J., . . . Karin, M.
(2004). IKKbeta links inflammation and tumorigenesis in a mouse model of colitis-
associated cancer. Cell, 118, 285-296. http://dx.doi.Org/10.1016/j.cell.2004.07.013
Griffin, J. M.; Blossom, S. J.; Jackson, S. K.; Gilbert, K. M.; Pumford, N. R. (2000).
Trichloroethylene accelerates an autoimmune response by Thl T cell activation in MRL
+/+ mice. Immunopharmacology, 46, 123-137. http://dx.doi.org/10.1016/S0162-
3109(99)00164-2
Griffin, J. M.; Gilbert, K. M.; Lamps, L. W.; Pumford, N. R. (2000). CD4(+) T-cell activation
and induction of autoimmune hepatitis following trichloroethylene treatment in MRL+/+
mice. Toxicol Sci, 57, 345-352. http://www.ncbi.nlm.nih.gov/pubmed/11006364
Griffin, J. M.; Gilbert, K. M.; Pumford, N. R. (2000). Inhibition of CYP2E1 reverses CD4+ T-
cell alterations in trichloroethylene-treated MRL+/+ mice. Toxicol Sci, 54, 384-389.
http://www.ncbi.nlm.nih.gov/pubmed/10774820
Grosse, Y.; Baan, R.; Secretan-lauby, B.; El ghissassi, F.; Bouvard, V.; Benbrahim-tallaa, L., . . .
Straif, K. (2011). Carcinogenicity of chemicals in industrial and consumer products, food
contaminants and flavourings, and water chlorination byproducts. Lancet Oncol, 12, 328-
329. http://dx.doi.org/10.1016/S1470-2045(l 1)70088-2
Groupwater Vapor Project, Endicott, New York:,. (2006). Summary of findings, working draft.
Cited in ATSDR. Endicott Area, Town of Union, Broome County, New York: Health
Statistics Reviews, Cancer and Birth Outcome Analysis.
Groves, C.; Hayden, P.; Lock, E.; Schnellmann, R. (1993). Differential cellular effects in the
toxicity of haloalkene and haloalkane cysteine conjugates to rabbit renal proximal
tubules. J Biochem Toxicol, 8, 49-56. http://www.ncbi.nlm.nih.gov/pubmed/8492303
Gu, Z. W.; Sele, B.; Chmara, D.; Jalbert, P.; Vincent, M.; Vincent, F., . . . Faure, J. (1981).
[Effects of trichl or ethylene and its metabolites on the rate of sister chromatid exchange.
In vivo and in vitro study on the human lymphocytes], Ann Genet, 24, 105-106.
http://www.ncbi.nlm.nih.gov/pubmed/6977287
Gu, Z. W.; Sele, B.; Jalbert, P.; Vincent, M.; Vin. (1981). Induction d'echanges entre les
chromatides soeurs (SCE) par le trichloroethylene et ses metabolites [Induction of sister
chromatid exchange by trichloroethylene and its metabolites], Toxicol Eur Res, 3, 63-67.
Gualandi, G. (1987). Use of alpha- and beta-tubulin mutants for the study of spontaneous and
induced chromosomal mis-distribution in Aspergillus nidulans. MutatRes, 178, 33-41.
Guehl, D.; Bezard, E.; Dovero, S.; Boraud, T.; Bioulac, B.; Gross, C. (1999). Trichloroethylene
and parkinsonism: A human and experimental observation. Eur J Neurol, 6, 609-611.
http://www.ncbi.nlm.nih.gov/pubmed/10457397
Guengerich, F. P. (2004). Metabolism of trichlorethylene and covalent binding of reaction
products. Proceedings of symposium on new scientific research related to the health
effects of trichlorethylene. Washington, DC.
47
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Guengerich, F. P. and Shimada, T. (1991). Oxidation of toxic and carcinogenic chemicals by
human cytochrome P-450 enzymes. Chem Res Toxicol, 4, 391-407.
http://dx.doi.org/10.1021/tx00022a001
Guengerich, F. P.; Crawford, W. M. J.; Watanabe, P. G. (1979). Activation of vinyl chloride to
covalently bound metabolites: roles of 2-chloroethylene oxide and 2-chloroacetaldehyde.
Biochemistry, 18, 5177-5182.
Guengerich, F. P.; Kim, D.-H.; Iwasaki, M. (1991). Role of human cytochrome P-450 IIE1 in the
oxidation of many low molecular weight cancer suspects. Chem Res Toxicol, 4, 168-179.
http://dx.doi.org/10.1021/tx00020a008
Gun, R. T.; Grygorcewicz, C.; Nettelbeck, T. J. (1978). Choice reaction time in workers using
trichloroethylene. Med J Aust, 1, 535-536. http://www.ncbi.nlm.nih.gov/pubmed/683058
Gupta, S. (2000). Hepatic polyploidy and liver growth control. Semin Cancer Biol, 10, 161-171.
http://dx.doi.org/10.1006/scbi.2000.0317
Gustavsson, P.; Jakobsson, R.; Johansson, H.; Lewin, F.; Norell, S.; Rutkvist, L. E. (1998).
Occupational exposures and squamous cell carcinoma of the oral cavity, pharynx, larynx,
and oesophagus: A case-control study in Sweden. Occup Environ Med, 55, 393-400.
http://www.ncbi.nlm.nih.gov/pubmed/9764099
Guyton, K. Z.; Barone, S.; Brown, R. C.; Euling, S. Y.; Jinot, J.; Makris, S. (2008). Mode of
action frameworks: a critical analysis. J Toxicol Environ Health B Crit Rev, 11, 16-31.
http://dx.doi.org/10.1080/10937400701600321
Guyton, K. Z.; Chiu, W. A.; Bateson, T. F.; Jinot, J.; Scott, C. S.; Brown, R. C.; Caldwell, J. C.
(2009). A reexamination of the PPAR-alpha activation mode of action as a basis for
assessing human cancer risks of environmental contaminants. Environ Health Perspect,
117, 1664-1672. http://dx.doi.org/10.1289/ehp.0900758
Gyorke, S. and Terentyev, D. (2008). Modulation of ryanodine receptor by luminal calcium and
accessory proteins in health and cardiac disease. Cardiovasc Res, 77, 245-255.
http ://dx. doi. org /10.1093/cvr/cvm03 8
Habib, G. M.; Barrios, R.; Shi, Z. Z.; Lieberman, M. W. (1996). Four distinct membrane-bound
dipeptidase RNAs are differentially expressed and show discordant regulation with
gamma-glutamyl transpeptidase. J Biol Chem, 271, 16273-16280.
http://www.ncbi.nlm.nih.gov/pubmed/8663190
Hack, C. E.; Chiu, W. A.; Jay Zhao, Q.; Clewell, H. J. (2006). Bayesian population analysis of a
harmonized physiologically based pharmacokinetic model of trichloroethylene and its
metabolites. Regul Toxicol Pharmacol, 46, 63-83.
http://dx.doi.Org/10.1016/j.yrtph.2006.05.012
Haddad, S.; Charest-Tardif, G.; K, K. (2000). Physiologically based modeling of the maximal
effect of metabolic interactions on the kinetics of components of complex chemical
mixtures. J Toxicol Environ Health A, 61, 209-223.
http://www.ncbi.nlm.nih.gov/pubmed/11036509
Haglid, K. G.; Briving, C.; Hansson, H.-A.; Rosengren, L.; Kjellstrand, P.; Stavron, D., . . .
Wronski, A. (1981). Trichloroethylene: long-lasting changes in the brain after
rehabilitation. Neurotoxicology, 2, 659-673.
http://www.ncbi.nlm.nih.gov/pubmed/7200581
Hakkola, J.; Pasanen, M.; Hukkanen, J.; Pelkonen, O.; Maenpaa, J.; Edwards, R. J., . . . Raunio,
H. (1996). Expression of xenobiotic-metabolizing cytochrome P450 forms in human full-
48
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
term placenta. Biochem Pharmacol, 51, 403-411. http://dx.doi.org/10.1016/0006-
2952(95)02184-1
Hakkola, J.; Pelkonen, O.; Pasanen, M.; Raunio, H. (1998). Xenobiotic-metabolizing cytochrome
P450 enzymes in the human feto-placental unit: Role in intrauterine toxicity. Crit Rev
Toxicol, 28, 35-72. http://dx.doi.org/10.1080/10408449891344173
Hakkola, J.; Raunio, H.; Purkunen, R.; Pelkonen, O.; Saarikoski, S.; Cresteil, T.; Pasanen, M.
(1996). Detection of cytochrome P450 gene expression in human placenta in first
trimester of pregnancy. Biochem Pharmacol, 52, 379-383.
http://dx.doi. org/10.1016/0006-2952(96)00216-X
Hakkola, J.; Tanaka, E.; Pelkonen, O. (1998). Developmental expression of cytochrome P450
enzymes in human liver. Pharmacol Toxicol, 82, 209-217.
http://dx.doi.Org/10.llll/j.1600-0773.1998.tb01427.x
Halmes, N.; Perkins, E.; McMillan, D.; Pumford, N. (1997). Detection of trichloroethylene-
protein adducts in rat liver and plasma. Toxicol Lett, 92, 187-194.
http://www.ncbi.nlm.nih.gov/pubmed/9334829
Hamada, T. and Tanaka, H. (1995). Transfer of methyl chloroform, trichloroethylene and
tetrachloroethylene to milk, tissues and expired air following intraruminal or oral
administration in lactating goats and milk-fed kids. Environ Pollut, 87, 313-318.
http://www.ncbi.nlm.nih.gov/pubmed/15091581
Hamdan, H. and Stacey, N. H. (1993). Mechanism of trichloroethylene-induced elevation of
individual serum bile acids: I correlation of trichloroethylene concentrations to bile acids
in rat serum. Toxicol Appl Pharmacol, 121, 291-295.
http://dx.doi.org/10.1006/taap.1993.1156
Hanahan, D. and Weinberg, R. A. (2000). The hallmarks of cancer. Cell, 100, 57-70.
http://dx.doi.org/10.1016/S0092-8674(00)81683-9
Hansch, C.; Leo, A.; Hoekman, D. (1995). Exploring QSAR: Hydrophobic, electronic, and steric
constants. Washington, DC: American Chemical Society.
Hansen, J.; Raaschou-Nielsen, O.; Christensen, J. M.; Johansen, I.; McLaughlin, J. K.; Lipworth,
L., . . . Olsen, J. H. (2001). Cancer incidence among Danish workers exposed to
trichloroethylene. J Occup Environ Med, 43, 133-139.
http://www.ncbi.nlm.nih.gOv/pubmed/l 1227631
Hardell, L.; Bengtsson, N. O.; Jonsson, U.; Eriksson, S.; Larsson, L. G. (1984). Aetiological
aspects on primary liver cancer with special regard to alcohol, organic solvents and acute
intermittent porphyria - An epidemiological investigation. Br J Cancer, 50, 389-397.
http://www.ncbi.nlm.nih.gov/pubmed/6087869
Hardell, L.; Eriksson, M.; Degerman, A. (1994). Exposure to phenoxyacetic acids,
chlorophenols, or organic solvents in relation to histopathology, stage, and anatomical
localization of non-Hodgkin's lymphoma. Cancer Res, 54, 2386-2389.
http://www.ncbi.nlm.nih.gov/pubmed/8162585
Hardell, L.; Eriksson, M.; Lenner, P.; Lundgren, E. (1981). Malignant lymphoma and exposure
to chemicals, especially organic solvents, chlorophenols and phenoxy acids: A case-
control study. Br J Cancer, 43, 169-176. http://www.ncbi.nlm.nih.gov/pubmed/7470379
Hardin, B. D.; Bond, G. P.; Sikov, M. R.; Andrew, F. D.; Beliles, R. P.; Niemeier, R. W. (1981).
Testing of selected workplace chemicals for teratogenic potential. Scand J Work Environ
Health, 7, 66-75. http://www.ncbi.nlm.nih.gov/pubmed/7330632
49
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Hardin, B. D.; Kelman, B. J.; Brent, R. L. (2004). Trichloroethylene and cardiac malformations.
Environ Health Perspect, 112, A607-608; author reply A608-609.
http://www.ncbi.nlm.nih.gov/pubmed/15289177
Hardin, B.; Kelman, B.; Brent, R. (2005). Trichloroethylene and dichloroethylene: a critical
review of teratogenicity. Birth Defects Res A Clin Mol Teratol, 73, 931-955.
http://dx.doi.org/10.1002/bdra.20192
Harrington, J. M.; Whitby, H.; Gray, C. N.; Reid, F. J.; Aw, T. C.; Waterhouse, J. A. (1989).
Renal disease and occupational exposure to organic solvents: A case referent approach.
Br JInd Med, 46, 643-650. http://dx.doi.Org/10.1136/oem.46.9.643
Harrington-Brock, K.; Doerr, C. L.; Moore, M. M. (1998). Mutagenicity of three disinfection by-
products: Di- and trichloroacetic acid and chloral hydrate in L5178Y(+/-) --3.7.2C mouse
lymphoma cells. Mutat Res Genet Toxicol Environ Mutagen, 413, 265-276.
http://www.ncbi.nlm.nih.gov/pubmed/9651541
Harris, N. L.; Jaffe, E. S.; Diebold, J.; Flandrin, G.; Muller-Hermelink, H. K.; Vardiman, J.
(2000). Lymphoma classification—from controversy to consensus: The R E. A.L. and
WHO Classification of lymphoid neoplasms. Ann Oncol, 11, S3-S10.
http://www.ncbi.nlm.nih.gov/pubmed/10707771
Haseman, J. K.; Huff, J.; Boorman, G. A. (1984). Use of historical control data in
carcinogenicity studies in rodents. Toxicol Pathol, 12, 126-135.
http://dx.doi.org/10.1177/019262338401200203
Hasmall, S. C. and Roberts, R. A. (2000). The nongenotoxic hepatocarcinogens
diethylhexylphthalate and methylclofenapate induce DNA synthesis preferentially in
octoploid rat hepatocytes. Toxicol Pathol, 28, 503-509.
http://dx.doi.org/10.1177/019262330002800401
Hassall, C.; Brendel, K.; Gandolfi, A. (1983). Regulation of a S(trans-l,2-dichlorovinyl)-L-
cysteine-induced renal tubular toxicity by glutathione. J Appl Toxicol, 3, 321-325.
http://www.ncbi.nlm.nih.gov/pubmed/6677655
Hassan, M. M.; Frome, A.; Patt, Y. Z.; El-Serag, H. B. (2002). Rising prevalence of hepatitis C
virus infection among patients recently diagnosed with hepatocellular carcinoma in the
United States. J Clin Gastroenterol, 35, 266-269.
http://www.ncbi.nlm.nih.gov/pubmed/12192205
Hattis, D.; Ginsberg, G.; Sonawane, B.; Smolenski, S.; Russ, A.; Kozlak, M.; Goble, R. (2003).
Differences in pharmacokinetics between children and adults: II. Childrens variability in
drug elimination half-lives and in some parameters needed for physiologically-based
pharmacokinetic modeling. Risk Anal, 23, 117-142. http://dx.doi.org/10.llll/1539-
6924.00295
Haworth, S.; Lawlor, T.; Mortelmans, K.; Speck, W.; Zeiger, E. (1983). Salmonella mutagenicity
test results for 250 chemicals. Environ Mutagen, 5, 3-142.
http://dx.doi.org/10.1002/em.2860050703
Hayden, M. S. and Ghosh, S. (2004). Signaling to NF-kappaB. Genes Dev, 18, 2195-2224.
http://dx.doi.org/10.1101/gad.1228704
Hays, T.; Rusyn, I.; Burns, A. M.; Kennett, M. J.; Ward, J. M.; Gonzalez, F. J.; Peters, J. M.
(2005). Role of peroxisome proliferator-activated receptor-alpha (PPARalpha) in
bezafibrate-induced hepatocarcinogenesis and cholestasis. Carcinogenesis, 26, 219-227.
http://dx.doi.org/10.1093/carcin/bgh285
50
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Healy, T. E. J.; Poole, T. R.; Hopper, A. (1982). Rat fetal development and maternal exposure to
trichloroethylene 100 ppm. Br J Anaesth, 54, 337-341.
http://www.ncbi.nlm.nih.gov/pubmed/7066129
Heath, C. W.; Lally, C. A.; Calle, E. E.; McLaughlin, J. K.; Thun, M. J. (1997). Hypertension,
diuretics, and antihypertensive medications as possible risk factors for renal cell cancer.
Am J Epidemiol, 145, 607-613. http://www.ncbi.nlm.nih.gov/pubmed/9098177
Hegi, M. E.; Fox, T. R.; Belinsky, S. A.; Devereux, T. R.; Anderson, M. W. (1993). Analysis of
activated protooncogenes in B6C3F1 mouse liver tumors induced by ciprofibrate, a
potent peroxisome proliferator. Carcinogenesis, 14, 145-149.
http://www.ncbi.nlm.nih.gov/pubmed/8425263
Hein, M. J.; Waters, M. A.; Ruder, A. M.; Stenzel, M. R.; Blair, A.; Stewart, P. A. (2010).
Statistical modeling of occupational chlorinated solvent exposures for case-control
studies using a literature-based database. Ann Occup Hyg, 54, 459-472.
http://dx.doi.org/10.1093/annhyg/meq027
Heineman, E. F.; Cocco, P.; Gomez, M. R.; Dosemeci, M.; Stewart, P. A.; Hayes, R. B., . . .
Blair, A. (1994). Occupational exposure to chlorinated aliphatic hydrocarbons and risk of
astrocytic brain cancer. Am J Ind Med, 26, 155-169.
http://dx.doi.org/10.1002/ajim.4700260203
Heinemann, K.; Willich, S.; Heinemann, L.; DoMinh, T.; Mohner, M.; Heuchert, G. (2000).
Occupational exposure and liver cancer in women: results of the Multicentre
International Liver Tumour Study (MILTS). Occup Med (Lond), 50, 422-429.
http://www.ncbi.nlm.nih.gov/pubmed/10994245
Hejtmancik, M. R.; Trela, B. A.; Kurtz, P. J.; Persing, R. L.; Ryan, M. J.; Yarrington, J. T.;
Chhabra, R. S. (2002). Comparative gavage subchronic toxicity studies of o-chloroaniline
and m-chloroaniline in F344 rats and B6C3F1 mice. Toxicol Sci, 69, 234-243.
Helliwell, P. J. andHutton, A. M. (1950). Trichlorethylene anesthesia. Anaesthesia, 5, 4-13.
http://dx.doi.Org/10.llll/j.1365-2044.1950.tbl3680.x
Henderson, G.; Yan, Z.; James, M.; Davydova, N.; Stacpoole, P. (1997). Kinetics and
metabolism of chloral hydrate in children: identification of dichloroacetate as a
metabolite. Biochem Biophys Res Commun, 235, 695-698.
http://dx.doi.org/10.1006/bbrc.1997.6868
Hendler, A. H. and Crow, W. L. (1992). Paper No. 92-75.05.
Hendriksen, P. J.; Freidig, A. P.; Jonker, D.; Thissen, U.; Bogaards, J. J.; Mumtaz, M. M., . . .
Stierum, R. H. (2007). Transcriptomics analysis of interactive effects of benzene,
trichloroethylene and methyl mercury within binary and ternary mixtures on the liver and
kidney following subchronic exposure in the rat. Toxicol Appl Pharmacol, 225, 171-188.
http://dx.doi.Org/10.1016/j.taap.2007.08.017
Hennig, U. G.; Rudd, N. L.; Hoar, D. I. (1988). Kinetochore immunofluorescence in
micronuclei: a rapid method for the in situ detection of aneuploidy and chromosome
breakage in human fibroblasts. Mutat Res, 203, 405-414.
http://www.ncbi.nlm.nih.gov/pubmed/3057373
Henning, S. M. and Swendseid, M. E. (1996). The role of folate, choline, and methionine in
carcinogenesis induced by methyl-deficient diets. Adv Exp Med Biol, 399, 143-155.
http://www.ncbi.nlm.nih.gov/pubmed/8937554
51
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Henschler, D.; Eder, E.; Neudecker, T.; Metzler, M. (1977). Carcinogenicity of
trichloroethylene: fact or artifact? Arch Toxicol, 37, 233-236.
Henschler, D.; Elsaesser, H.; Romen, W.; Eder, E. (1984). Carcinogenicity study of
trichloroethylene, with and without epoxide stabilizers, in mice. J Cancer Res Clin Oncol,
107, 149-156. http://www.ncbi.nlm.nih.gov/pubmed/6736101
Henschler, D.; Romen, W.; Elsaesser, H. M.; Reichert, D.; Eder, E.; Radwan, Z. (1980).
Carcinogenicity study of trichloroethylene by longterm inhalation in three animal species.
Arch Toxicol, 43, 237-248.
Henschler, D.; Vamvakas, S.; Lammert, M.; Dekant, W.; Kraus, B.; Thomas, B.; Ulm, K. (1995).
Increased incidence of renal cell tumors in a cohort of cardboard workers exposed to
trichloroethene. Arch Toxicol, 69, 291-299.
http://www.ncbi.nlm.nih.gov/pubmed/7654132
Herbert, V.; Gardner, A.; Colman, N. (1980). Mutagenicity of dichloroacetate, an ingredient of
some formulations of pangamic acid (trade-named "vitamin B15"). Am J
ClinNutr, 33, 1179-1182. http://www.ncbi.nlm.nih.gov/pubmed/6992558
Herman, J.; Umar, A.; Polyak, K.; Graff, J.; Ahuja, N.; Issa, J., . . . Baylin, S. (1998). Incidence
and functional consequences of hMLHl promoter hypermethylation in colorectal
carcinoma. PNAS, 95, 6870-6875. http://www.ncbi.nlm.nih.gov/pubmed/9618505
Hernberg, S.; Kauppinen, T.; Riala, R.; Korkala, M. L.; Asikainen, U. (1988). Increased risk for
primary liver cancer among women exposed to solvents. Scand J Work Environ Health,
14, 356-365. http://www.ncbi.nlm.nih.gov/pubmed/3212412
Hernberg, S.; Korkala, M. L.; Asikainen, U.; Riala, R. (1984). Primary liver cancer and exposure
to solvents. Int Arch Occup Environ Health, 54, 147-153.
http://dx.doi.org/10.1007/BF00378517
Herren-Freund, S. L.; Pereira, M. A.; Khoury, M. D.; Olson, G. (1987). The carcinogenicity of
trichloroethylene and its metabolites, trichloroacetic acid and dichloroacetic acid, in
mouse liver. Toxicol Appl Pharmacol, 90, 183-189. http://dx.doi.org/10.1016/0041-
008X(87)90325-5
Herrera, E.; Samper, E.; Martin-Caballero, J.; Flores, J. M.; Lee, H. W.; Blasco, M. A. (1999).
Disease states associated with telomerase deficiency appear earlier in mice with short
telomeres. EMBO J, 18, 2950-2960. http://dx.doi.org/10.1093/emboj/18.ll.2950
Hers, I.; Zapf-Gilje, R.; Li, L.; Atwater, J. (2001). The use of indoor air measurements to
evaluate intrusion of subsurface VOC vapors into buildings. J Air Waste Manag Assoc,
51, 1318-1331. http://www.ncbi.nlm.nih.gov/pubmed/11575885
Higgins, J. P.; Thompson, S. G.; Deeks, J. J.; Altman, D. G. (2003). Measuring inconsistency in
meta-analyses. Br Med J Int Ed, 327, 557-560.
http://dx.doi.Org/10.l 136/bmj. 327.7414.557
Hill, A. B. (1965). The environment and disease: Association or causation? Proc R Soc Med, 58,
295-300. http://www.ncbi.nlm.nih.gov/pubmed/14283879
Hinchman, C. A. and Ballatori, N. (1990). Glutathione-degrading capacities of liver and kidney
in different species. Biochem Pharmacol, 40, 1131-1135. http://dx.doi.org/10.1016/0006-
2952(90)90503-D
Hines, R. N. and McCarver, D. G. (2002). The ontogeny of human drug-metabolizing enzymes:
Phase I oxidative enzymes. J Pharmacol Exp Ther, 300, 355-360.
http://dx.doi.Org/10.1124/jpet.300.2.355
52
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Hirsch, A. R.; Rankin, K. M.; Panelli, P. P. (1996). Trichloroethylene exposure and headache. 7,
126 - 138.
Hissink, E. M.; Bogaards, J. J. P.; Freidig, A. P.; Commandeur, J. N. M.; Vermeulen, N. P. E.;
Bladeren, P. J. v. (2002). The use of in vitro metabolic parameters and physiologically
based pharmacokinetic (PBPK) modeling to explore the risk assessment of
trichloroethylene. Environ Toxicol Pharmacol, 11, 259 - 271.
Hobara, T.; Kobayashi, H.; Higashihara, E.; Kawamoto, T.; Sakai, T. (1984). Acute effects of
1,1,1-trichloroethane, trichloroethylene, and toluene on the hematologic parameters in
dogs. Arch Environ Contam Toxicol, 13, 589-593.
http://www.ncbi.nlm.nih.gov/pubmed/6486885
Hodgen, G. D. and Sherins, R. J. (1973). Enzymes as markers of testicular growth and
development in the rat. Endocrinology, 93, 985-989.
http://www.ncbi.nlm.nih.gov/pubmed/4728225
Hoel, D. G. and Walburg, H. E. (1972). Statistical analysis of survival experiments. J Natl
Cancer Inst, 49, 361-372. http://www.ncbi.nlm.nih.gov/pubmed/5076821
Hoffman, S.; Mishima, N.; Krug, E. L. (2004). An improved model for evaluating
trichloroethylene and its metabolites as cardiac specific teratogens. In D. Jollow (Ed.),
Trichloroethylene: the scientific basis of risk assessment: The Medical University of
South Carolina Press.
Hoivik, D. J.; Quails, C. W.; Mirabile, R. C.; Cariello, N. F.; Kimbrough, C. L.; Colton, H. M., . .
. Miller, R. T. (2004). Fibrates induce hepatic peroxisome and mitochondrial proliferation
without overt evidence of cellular proliferation and oxidative stress in cynomolgus
monkeys. Carcinogenesis, 25, 1757-1769. http://dx.doi.org/10.1093/carcin/bghl82
Hooper, N. M.; Hesp, R. J.; Tieku, S. (1994). Metabolism of aspartame by human and pig
intestinal microvillar peptidases. Biochem J, 298 Pt 3, 635-639.
http://www.ncbi.nlm.nih.gov/pubmed/8141778
Hoover, R. N. and Strasser, P. H. (1980). Artificial sweeteners and human bladder cancer.
Preliminary results. Lancet, 1, 837-840. http://www.ncbi.nlm.nih.gov/pubmed/6103207
Horesovsky, G.; Ginsler, J.; Everitt, J.; Walker, C. (1994). Increased susceptibility to in vitro
transformation of cells carrying the Eker tumor susceptibility mutation. Carcinogenesis,
15, 2183-2187. http://www.ncbi.nlm.nih.gov/pubmed/7955051
Hornung, R. W.; Pinney, S. M.; Lodwick, J.; Killough, G. G.; Brewer, D. E.; Nasuta, J. (2008).
Estimation of radon exposures to workers at the Fernald Feed Materials Production
Center 1952-1988. J Expo Sci Environ Epidemiol, 18, 512-523.
http://dx.doi.org/10.1038/sj.jes.7500645
Hortelano, S.; Dewez, B.; Genaro, A. M.; Diaz-Guerra, M. J.; Bosca, L. (1995). Nitric oxide is
released in regenerating liver after partial hepatectomy. Hepatology, 21, 776-786.
http://www.ncbi.nlm.nih.gov/pubmed/7533126
Horvath, A. L.; Getzen, F. W.; Maczynska, Z. (1999). IUPAC-NIST Solubility Data Series 67.
Halogenated Ethanes and Ethenes with Water. J Phys Chem Ref Data, 28, 395.
http://dx.doi.org/10.1063/L556039
Howard, P. H.; Boethling, R. S.; Jarvis, W. F.; Meylan, W. M.; Michalenko, E. M. (1991).
Handbook of environmental degradation rates. Chelsea, MI: Lewis Publishers.
Howroyd, P.; Swanson, C.; Dunn, C.; Cattley, R. C.; Corton, J. C. (2004). Decreased longevity
and enhancement of age-dependent lesions in mice lacking the nuclear receptor
53
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
peroxisome proliferator-activated receptor alpha (PPARalpha). Toxicol Pathol, 32, 591-
599. http://www.ncbi.nlm.nih.gov/pubmed/15603543
Hrelia, P.; Maffei, F.; Vigagni, F.; Fimognari, C.; Flori, P.; Stanzani, R.; Forti, G. C. (1994).
Interactive effects between trichloroethylene and pesticides at metabolic and genetic level
in mice. Environ Health Perspect, 102, 31-34.
HSDB. (Hazardous Substances Data Bank). (1996). Trichloroethylene. Bethesda, MD: National
Library of Medicine. Retrieved from http://toxnet.nlm.nih.gov/cgi-
bin/sis/htmlgen?HSDB
HSDB. (Hazardous Substances Data Bank). (2002). Trichloroethylene. Bethesda, MD: National
Library of Medicine. Retrieved from http://toxnet.nlm.nih.gov/cgi-
bin/sis/htmlgen?HSDB
Hsia, C. Y.; Huo, T. I.; Chiang, S. Y.; Lu, M. F.; Sun, C. L.; Wu, J. C., . . . Lee, S. D. (2007).
Evaluation of interleukin-6, interleukin-10 and human hepatocyte growth factor as tumor
markers for hepatocellular carcinoma. Eur J Surg Oncol, 33, 208-212.
http://dx.doi.Org/10.1016/j.ejso.2006.10.036
Hu, C.; Jiang, L.; Geng, C.; Zhang, X.; Cao, J.; Zhong, L. (2008). Possible involvement of
oxidative stress in trichloroethylene-induced genotoxicity in human HepG2 cells. Mutat
Res, 652, 88-94. http://dx.doi.Org/10.1016/j.mrgentox.2008.01.002
Hu, Y.; Hakkola, J.; Oscarson, M.; Ingelman-Sundberg, M. (1999). Structural and functional
characterization of the 5'-flanking region of the rat and human cytochrome P450 2E1
genes: Identification of a polymorphic repeat in the human gene. Biochem Biophys Res
Commun, 263, 286-293. http://dx.doi.org/10.1006/bbrc.1999.1362
Huang, H. L.; Li, L. Y.; Chen, B. J.; Huang, J. X.; Kuang, S. R. (2002). New problems caused by
occupational trichloroethylene exposure. Int J Immunopathol Pharmacol, 15, 30-32.
Hunter, E.; Rogers, E.; Schmid, J.; Richard, A. (1996). Comparative effects of haloacetic acids in
whole embryo culture. Teratology, 54, 57-64. http://dx.doi.org/10.1002/(SICI)1096-
9926(199606)54:2<57::AID-TERAl>3.0.CO;2-l
IARC. (International Agency for Research on Cancer). (1995). Dry cleaning, some chlorinated
solvents and other industrial chemicals: Summary of data reported and evaluation. (IARC
Monographs on the Evaluation of Carcinogenic Risks to Humans, Volume 63). Lyon,
France: International Agency for Research on Cancer, World Health Organization.
IARC. (International Agency for Research on Cancer). (1995). Dry cleaning, some chlorinated
solvents and other industrial chemicals: views and expert opinions of an IARC Working
Group on the Evaluation of Carcinogenic Risks of Humans; February; Lyon, France.
Lyon, France: Author.
IARC. (International Agency for Research on Cancer). (1995). Tetrachloroethylene. Lyon,
France: Author. Retrieved from
http://monographs.iarc.fr/ENG/Monographs/vol63/index.php.
IARC. (International Agency for Research on Cancer). (2004). Tobacco smoke and involuntary
smoking. Lyon, France: Author.
Iavicoli, I.; Marinaccio, A.; Carelli, G. (2005). Effects of occupational trichloroethylene
exposure on cytokine levels in workers. J Occup Environ Med, 47, 453-457.
http://www.ncbi.nlm.nih.gov/pubmed/15891523
ICRP. (International Commission of Radiological Protection). (2003). Basic anatomical and
physiological data for use in radiological protection: reference values. A report of age-
54
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
and gender-related differences in the anatomical and physiological characteristics of
reference individuals. ICRP Publication 89. (89). New York: International Commission
On Radiological Protection.
IEI. (International Epidemiology Institute). (2005). Rocketdyne worker health study: Final
report: Executive summary & appendices. Rockville, MD: Author. Retrieved from
http://www.boeing.com/aboutus/environment/santa_susana/healthstudy.html.
Ikbal, M.; Tastekin, A.; Dogan, H.; Pirim, I.; Ors, R. (2004). The assessment of genotoxic effects
in lymphocyte cultures of infants treated with chloral hydrate. Mutat Res, 564, 159-164.
http ://dx. doi. org /10.1016/j. mrgentox.2004.08.007
Ikeda, M. (1977). Metabolism of trichloroethylene and tetrachloroethylene in human subjects.
Environ Health Perspect, 21, 239-245. http://www.ncbi.nlm.nih.gov/pubmed/612449
Ikeda, M. and Imamura, T. (1973). Biological half-life of trichloroethylene and
tetrachloroethylene in human subjects. Int Arch Occup Environ Health, 31, 209-224.
http://www.ncbi.nlm.nih.gov/pubmed/4593976
Ikeda, M.; Ohtsuji, H.; Imamura, T.; Komoike, Y. (1972). Urinary excretion of total trichloro-
compounds, trichloroethanol, and trichloroacetic acid as a measure of exposure to
trichloroethylene and tetrachloroethylene. Br J Ind Med, 29, 328-333.
http://dx.doi.Org/10.l 136/oem. 29.3.328
Ikeda, M.; Otsuji, H.; Kawai, H.; Kuniyoshi, M. (1971). Excretion kinetics of urinary metabolites
in a patient addicted to richloroethylene. Br J Ind Med, 28, 203-206.
http://www.ncbi.nlm.nih.gov/pubmed/5572691
ILSI. (International Life Sciences Institute). (1992). Similarities and differences between
children and adults: Implications for risk assessment. New York: ILSI Press.
Inoue, O.; Seiji, K.; Kawai, T.; Jin, C.; Liu, Y.-T.; Chen, Z., . . . Ikeda, M. (1989). Relationship
between vapor exposure and urinary metabolite excretion among workers exposed to
trichloroethylene. Am J Ind Med, 15, 103-110.
http://www.ncbi.nlm.nih.gov/pubmed/2784624
IOM. (Institute of Medicine). (2003). Gulf war and health. Washington, DC: National
Academies Press.
Isaacs, K.; Evans, M.; Harris, T. (2004). Visualization-based analysis for a mixed-inhibition
binary PBPK model: determination of inhibition mechanism. J Pharmacokinet
Pharmacodyn, 31, 215-242. http://www.ncbi.nlm.nih.gov/pubmed/15518245
Isaacson, L. and Taylor, D. (1989). Maternal exposure to 1,1,2-trichloroethylene affects myelin
in the hippocampal formation of the developing rat. Brain Res, 488, 403-407.
http://www.ncbi.nlm.nih.gov/pubmed/2743138
Isaacson, L. G.; Spohler, S. A.; Taylor, D. H. (1990). Trichloroethylene affects learning and
decreases myelin in the rat hippocampus. Neurotoxicol Teratol, 12, 375-381.
http://www.ncbi.nlm.nih.gov/pubmed/2392097
Isacson, P.; Bean, J. A.; Splinter, R.; Olson, D. B.; Kohler, J. (1985). Drinking water and cancer
incidence in Iowa. III. Association of cancer with indices of contamination. Am J
Epidemiol, 121, 856-869. http://www.ncbi.nlm.nih.gov/pubmed/4014178
Ito, Y.; Yamanoshita, O.; Asaeda, N.; Tagawa, Y.; Lee, C. H.; Aoyama, T., . . . Nakajima, T.
(2007). Di(2-ethylhexyl)phthalate induces hepatic tumorigenesis through a peroxisome
proliferator-activated receptor alpha-independent pathway. J Occup Health, 49, 172-182.
http://www.ncbi.nlm.nih.gov/pubmed/17575397
55
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Iyer, R.; Frink, E.; Ebert, T.; Anders, M. (1998). Cysteine conjugate beta-lyase-dependent
metabolism of compound A (2-[fluoromethoxy]-l,l,3,3,3-pentafluoro-l-propene) in
human subjects anesthetized with sevoflurane and in rats given compound A.
Anesthesiology, 88, 611-618. http://www.ncbi.nlm.nih.gov/pubmed/9523802
Jacob, S.; Hery, M.; Protois, J. C.; Rossert, J.; Stengel, B. (2007). New insight into solvent-
related end-stage renal disease: Occupations, products and types of solvents at risk.
Occup Environ Med, 64, 843-848. http://dx.doi.org/10.1136/oem.2007.032482
Jaffe, D.; Gandolfi, A.; Nagle, R. (1984). Chronic toxicity of S-(trans-l,2-dichlorovinyl)-L-
cysteine in mice. J Appl Toxicol, 4, 315-319.
http://www.ncbi.nlm.nih.gov/pubmed/6520320
Jaffe, D.; Hassall, C.; Gandolfi, A.; Brendel, K. (1985). Production of DNA single strand breaks
in rabbit renal tissue after exposure to 1,2-dichlorovinylcysteine. Toxicology, 35, 25-33.
http://www.ncbi.nlm.nih.gov/pubmed/4002236
Jaffe, E. S.; N.L., H.; Stein, H.; Vardiman, J. W. (2001). IARC WHO classification of tumours,
no 3: Pathology and genetics of tumours of haematopoietic and lymphoid tissues. Lyon,
France: International Agency for Research on Cancer. Retrieved from
http://apps.who.int/bookorders/anglais/detartl.jsp?sesslan=l&codlan=l&codcol=70&cod
cch=3.
Jakobson, I.; Holmberg, B.; Ekner, A. (1986). Venous blood levels of inhaled trichloroethylene
in female rats and changes induced by interacting agents. Acta Pharmacol Toxicol, 59,
135-143. http://www.ncbi.nlm.nih.gov/pubmed/3776551
Jakobson, I.; Wahlberg, J. E.; Holmberg, B.; Johansson, G. (1982). Uptake via the blood and
elimination of 10 organic solvents following epicutaneous exposure of anesthetized
guinea pigs. Toxicol Appl Pharmacol, 63, 181-187. http://dx.doi.org/10.1016/0041-
008X(82)90039-4
Jakubowski, M. and Wieczorek, H. (1988). The effects of physical effort on pulmonary uptake of
selected organic compound vapours. Pol J Occup Med, 1, 62-71.
http://www.ncbi.nlm.nih.gov/pubmed/2980152
James, M.; Cornett, R.; Yan, Z.; Henderson, G.; Stacpoole, P. (1997). Glutathione-dependent
conversion to glyoxylate, a major pathway of dichloroacetate biotransformation in
hepatic cytosol from humans and rats, is reduced in dichloroacetate-treated rats. Drug
Metab Dispos, 25, 1223-1227. http://www.ncbi.nlm.nih.gov/pubmed/9351896
Jansson, C.; Johansson, A. L.; Bergdahl, I. A.; Dickman, P. W.; Plato, N.; Adami, J., . . .
Lagergren, J. (2005). Occupational exposures and risk of esophageal and gastric cardia
cancers among male Swedish construction workers. Cancer Causes Control, 16, 755-764.
http://dx.doi.org/10.1007/sl0552-005-1723-2
Jansson, C.; Plato, N.; Johansson, A. L.; Nyren, O.; Lagergren, J. (2006). Airborne occupational
exposures and risk of oesophageal and cardia adenocarcinoma. Occup Environ Med, 63,
107-112. http://dx.doi.org/10.1136/oem.2005.022467
Jasinska, M. (1965). [General anesthesia with trichloroethylene in children with congenital
developmental defects of the jaw and face], 18, 935-940.
http://www.ncbi.nlm.nih.gov/pubmed/5214455
Jaspers, R. M. A.; Muijser, H; Lammers, J. H. C. M.; Kulig, B. M. (1993). Mid-frequency
hearing loss and reduction of acoustic startle responding in rats following
56
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
trichloroethylene exposure. Neurotoxicol Teratol, 15, 407-412.
http://www.ncbi.nlm.nih.gov/pubmed/8302242
Jemal, A.; Siegel, R.; Ward, E.; Hao, Y.; Xu, J.; Murray, T.; Thun, M. J. (2008). Cancer
statistics, 2008. CA Cancer J Clin, 58, 71-96. http://dx.doi.org/10.3322/CA.2007.0010
Jensen, O. M.; Knudsen, J. B.; McLaughlin, J. K.; Sorensen, B. L. (1988). The Copenhagen case-
control study of renal pelvis and ureter cancer: role of smoking and occupational
exposures. Int J Cancer, 41, 557-561. http://www.ncbi.nlm.nih.gov/pubmed/3356489
Ji, B. T.; Blair, A.; Shu, X. O.; Chow, W. H.; Hauptmann, M.; Dosemeci, M., . . . Zheng, W.
(2008). Occupation and breast cancer risk among Shanghai women in a population-based
cohort study. Am J Ind Med, 51, 100-110. http://dx.doi.org/10.1002/ajim.20507
Ji, J. and Hemminki, K. (2005). Variation in the risk for liver and gallbladder cancers in
socioeconomic and occupational groups in Sweden with etiological implications. Int
Arch Occup Environ Health, 78, 641-649. http://dx.doi.org/10.1007/s00420-005-0015-l
Jia, H. L.; Ye, Q. H.; Qin, L. X.; Budhu, A.; Forgues, M.; Chen, Y., . . . Wang, X. W. (2007).
Gene expression profiling reveals potential biomarkers of human hepatocellular
carcinoma. Clin Cancer Res, 13, 1133-1139. http://dx.doi.org/10.1158/1078-0432.CCR-
06-1025
Jia, M.; Coats, B.; Chadha, M.; Frentzen, B.; Perez-Rodriguez, J.; Chadik, P., . . . Stacpoole, P.
(2006). Human kinetics of orally and intravenously administered low-dose 1,2-(13)C-
dichloroacetate. J Clin Pharmacol, 46, 1449-1459.
http://dx.doi.org/10.1177/0091270006292627
Jirtle, R. L.; Meyer, S. A.; Brockenbrough, J. S. (1991). Liver tumor promoter phenobarbital: A
biphasic modulator of hepatocyte proliferation. Prog Clin Biol Res, 369, 209-216.
Johnson, A. D.; VanDemark, W. L.; Gomes, W. R. (1967). The effect of antispermatogenic and
hormone treatments on testicular cholesterol. Fed Proc, 26, 645.
Johnson, P. D.; Dawson, B. V.; Goldberg, S. J. (1998). Cardiac teratogenicity of
trichloroethylene metabolites. J Am Coll Cardiol, 32, 540-545.
http://dx.doi.org/10.1016/S0735-1097(98)00232-0
Johnson, P. D.; Dawson, B. V.; Goldberg, S. J.; Mays, M. Z. (2004). Trichloroethylene: Johnson
et al.'s Response. Environ Health Perspect, 112, a608-a609.
http://dx.doi.org/10.1289/ehp. 112-a608
Johnson, P.; Dawson, B.; Goldberg, S. (1998). A review: trichloroethylene metabolites: potential
cardiac teratogens. Environ Health Perspect, 106 Suppl 4, 995-999.
http://www.ncbi.nlm.nih.gov/pubmed/9703484
Johnson, P.; Goldberg, S.; Mays, M.; Dawson, B. (2003). Threshold of trichloroethylene
contamination in maternal drinking waters affecting fetal heart development in the rat.
Environ Health Perspect, 111, 289-292. http://www.ncbi.nlm.nih.gov/pubmed/12611656
Johnsrud, E. K.; Koukouritaki, S. B.; Divakaran, K.; Brunengraber, L. L.; Hines, R. N.; p
McCarver, D. G. (2003). Human hepatic CYP2E1 expression during development. J
Pharmacol Exp Ther, 307, 402-407. http://dx.doi.org/10.1124/jpet.102.053124
Jones, T.; Qin, C.; Schaeffer, V.; Stevens, J. (1988). Immunohistochemical localization of
glutamine transaminase K, a rat kidney cysteine conjugate beta-lyase, and the
relationship to the segment specificity of cysteine conjugate nephrotoxicity. Mol
Pharmacol, 34, 621-627. http://www.ncbi.nlm.nih.gov/pubmed/3193955
57
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
Jorquera, R. and Tanguay, R. (2001). Fumarylacetoacetate, the metabolite accumulating in
hereditary tyrosinemia, activates the ERK pathway and induces mitotic abnormalities and
genomic instability. Hum Mol Genet, 10, 1741-1752.
http://www.ncbi.nlm.nih.gov/pubmed/11532983
Juntunen, J.; Antti-Poika, M.; Tola, S.; Partanen, T. (1982). Clinical prognosis of patients with
diagnosed chronic solvent intoxication. Acta Neurol Scand, 65, 488-503.
http://www.ncbi.nlm.nih.gov/pubmed/7113659
Juntunen, J.; Hupli, V.; Hernberg, S.; Luisto, M. (1980). Neurological picture of organic solvent
poisoning in industry: a retrospective clinical study of 37 patients. Int Arch Occup
Environ Health, 46, 219-231. http://www.ncbi.nlm.nih.gov/pubmed/7450887
Juran, B. D. and Lazaridis, K. N. (2006). Genomics and complex liver disease: Challenges and
opportunities. Hepatology, 44, 1380-1390. http://dx.doi.org/10.1002/hep.21453
Kadry, A.; Turkall, R.; Skowronski, G.; Abdel-Rahman, M. (1991). Soil adsorption alters
kinetics and bioavailability of trichloroethylene in orally exposed female rats. Toxicol
Lett, 58, 337-346. http://www.ncbi.nlm.nih.gov/pubmed/1957329
Kafer, E. (1986). Tests which distinguish induced crossing-over and aneuploidy from secondary
segregation in Aspergillus treated with chloral hydrate or gamma-rays. Mutat Res, 164,
145-166. http://dx.doi.org/10.1016/0165-l 161(86)90006-3
Kamijima, M.; Hisanaga, N.; Wang, H.; Nakajima, T. (2007). Occupational trichloroethylene
exposure as a cause of idiosyncratic generalized skin disorders and accompanying
hepatitis similar to drug hypersensitivities. Int Arch Occup Environ Health, 80, 357-370.
http://dx.doi.org/10.1007/s00420-006-0147-y
Kamijima, M.; Wang, H.; Huang, H.; Li, L.; Shibata, E.; Lin, B., . . . Nakajima, T. (2008).
Trichloroethylene causes generalized hypersensitivity skin disorders complicated by
hepatitis. J Occup Health, 50, 328-338. http://www.ncbi.nlm.nih.gov/pubmed/18540116
Kamijo, Y.; Hora, K.; Nakajima, T.; Kono, K.; Takahashi, K.; Ito, Y., . . . Aoyama, T. (2007).
Peroxisome proliferator-activated receptor alpha protects against glomerulonephritis
induced by long-term exposure to the plasticizer di-(2-ethylhexyl)phthalate. J Am Soc
Nephrol, 18, 176-188. http://dx.doi.org/10.1681/asn.2006060597
Kan, F.; Forkert, P.; Wade, M. (2007). Trichloroethylene exposure elicits damage in epididymal
epithelium and spermatozoa in mice. Histol Histopathol, 22, 977-988.
http://www.ncbi.nlm.nih.gov/pubmed/17523075
Kaneko, T.; Saegusa, M.; Tasaka, K.; Sato, A. (2000). Immunotoxicity of trichloroethylene: a
study with MRL-lpr/lpr mice. J Appl Toxicol, 20, 471-475.
http://www.ncbi.nlm.nih.gOv/pubmed/l 1180269
Kaneko, T.; Wang, P.-Y.; Sato, A. (1994). Enzymes induced by ethanol differently affect the
pharmacokinetics of trichloroethylene and 1,1,1-trichloroethane. Occup Environ Med, 51,
113-119. http://www.ncbi.nlm.nih.gov/pubmed/8111458
Kaneko, T.; Wang, P.-Y.; Sato, A. (1994). Partition coefficients of some acetate esters and
alcohols in water, blood, olive oil, and rat tissues. Occup Environ Med, 51, 68-72.
Kanisawa, M. and Suzuki, S. (1978). Induction of renal and hepatic tumors in mice by
ochratoxin A, a mycotoxin. Gann, 69, 599-600.
http://www.ncbi.nlm.nih.gov/pubmed/710809
58
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Kappas, A. (1989). On the mechanisms of induced aneuploidy in Aspergillus nidulans and
validation of tests for genomic mutations. Prog Clin Biol Res, 318, 377-384.
http://www.ncbi.nlm.nih.gov/pubmed/2696976
Karami, S.; Boffetta, P.; Brennan, P.; Stewart, P. A.; Zaridze, D.; Matveev, V., . . . Moore, L. E.
(2011). Renal cancer risk and occupational exposure to polycyclic aromatic hydrocarbons
and plastics. J Occup Environ Med, 53, 218-223.
http ://dx. doi. org/10.1097/JOM. ObO 13 e31820a40a3
Kargalioglu, Y.; McMillan, B. J.; Minear, R. A.; Plewa, M. J. (2002). Analysis of the
cytotoxicity and mutagenicity of drinking water disinfection by-products in Salmonella
typhimurium. Teratog Carcinog Mutagen, 22, 113-128.
http://dx.doi.org/10.1002/tcm.10010
Karpinets, T. V. and Foy, B. D. (2005). Tumorigenesis: the adaptation of mammalian cells to
sustained stress environment by epigenetic alterations and succeeding matched mutations.
Carcinogenesis, 26, 1323-1334. http://dx.doi.org/10.1093/carcin/bgi079
Karunanayake, C. P.; McDuffie, H. H.; Dosman, J. A.; Spinelli, J. J.; Pahwa, P. (2008).
Occupational exposures and non-Hodgkin's lymphoma: Canadian case-control study.
Environ Health, 7, 44. http://dx.doi.org/10.1186/1476-069X-7-44
Kato-Weinstein, J.; Lingohr, M. K.; Orner, G. A.; Thrall, B. D.; Bull, R. J. (1998). Effects of
dichloroacetate on glycogen metabolism inB6C3Fl mice. Toxicology, 130, 141-154.
http://dx.doi.org/10.1016/S0300-483X(98)00106-l
Kato-Weinstein, J.; Stauber, A. J.; Orner, G. A.; Thrall, B. D.; Bull, R. J. (2001). Differential
effects of dihalogenated and trihalogenated acetates in the liver of B6C3F1 mice. J Appl
Toxicol, 21, 81-89. http://dx.doi.org/10.1002/jat.717
Kauffmann, B.; White, K.; Sanders, V.; Douglas, K.; Sain, L.; Borzelleca, J.; Munson, A. (1982).
Humoral and cell-mediated immune status in mice exposed to chloral hydrate. Environ
Health Perspect, 44, 147-151. http://www.ncbi.nlm.nih.gov/pubmed/6979478
Kautiainen, A.; Vogel, J.; Turteltaub, K. (1997). Dose-dependent binding of trichloroethylene to
hepatic DNA and protein at low doses in mice. Chem Biol Interact, 106, 109-121.
http://www.ncbi.nlm.nih.gov/pubmed/9366897
Kawamoto, T.; Hobara, T.; Aoki, T.; Kobayashi, H.; Iwamoto, S.; Sakai, T., . . . Ogino, K.
(1988). Metabolism of chloral hydrate in the anoxic perfused liver. Toxicol Lett, 40, 225-
231. http://www.ncbi.nlm.nih.gov/pubmed/3354006
Kawamoto, T.; Hobara, T.; Nakamura, K.; Imamura, A.; Ogino, K.; Kobayashi, H., . . . Sakai, T.
(1988). Induction of cytochrome P-450, cytochrome b-5, NADPH-cytochrome c
reductase and change of cytochrome P-450 isozymes with long-term trichloroethylene
treatment. Toxicology, 53, 239-249. http://www.ncbi.nlm.nih.gov/pubmed/3145630
Keech, A.; Simes, J.; Barter, P.; Best, J.; Scott, R.; Taskinen, M. R.; Committee, F. M. (2006).
Correction to the FIELD study report. Lancet, 368, 1415.
http://dx.doi.org/10.1016/S0140-6736(06)69594-9
Keech, A.; Simes, R. J.; Barter, P.; Best, J.; Scott, R.; Taskinen, M. R., . . . investigators, F. s.
(2005). Effects of long-term fenofibrate therapy on cardiovascular events in 9795 people
with type 2 diabetes mellitus (the FIELD study): randomised controlled trial. Lancet, 366,
1849-1861. http://dx.doi.org/10.1016/S0140-6736(05)67667-2
Keil, D. E.; Peden-Adams, M. M.; Wallace, S.; Ruiz, P.; Gilkeson, G. S. (2009). Assessment of
trichloroethylene (TCE) exposure in murine strains genetically-prone and non-prone to
59
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
develop autoimmune disease. J Environ Sci Health A Tox Hazard Subst Environ Eng, 44,
443-453. http://dx.doi.org/10.1080/10934520902719738
Keller, D. A. and Hd'A, H. (1988). Mechanistic studies on chloral toxicity: Relationship to
trichloroethylene carcinogenesis. Toxicol Lett, 42, 183-191.
http://dx.doi.org/10.1016/0378-4274(88)90076-8
Kelsh, M. A.; Alexander, D. D.; Mink, P. J.; Mandel, J. H. (2010). Occupational
trichloroethylene exposure and kidney cancer: a meta-analysis. Epidemiology, 21, 95-
102. http://dx.doi.org/10.1097/EDE.0b013e3181c30e92
Kenck, C.; Wilhelm, M.; Bugert, P.; Staehler, G.; Kovacs, G. (1996). Mutation of the VHL gene
is associated exclusively with the development of non-papillary renal cell carcinomas. J
Pathol, 179, 157-161. http://dx.doi.org/10.1002/(SICI)1096-
9896(199606)179:2<l 57: :AID-PATH557>3.0.CO;2-S
Kernan, G. J.; Ji, B. T.; Dosemeci, M.; Silverman, D. T.; Balbus, J.; Zahm, S. H. (1999).
Occupational risk factors for pancreatic cancer: A case-control study based on death
certificates from 24 U.S. States. Am J Ind Med, 36, 260-270.
http://dx.doi.org/10.1002/(SICI)1097-0274(199908)36:2<260: :AID-AJIM5>3.0.CO;2-P
Keshava, N. and Caldwell, J. (2006). Key issues in the role of peroxisome proliferator-activated
receptor agonism and cell signaling in trichloroethylene toxicity. Environ Health
Perspect, 114, 1464-1470. http://www.ncbi.nlm.nih.gov/pubmed/16966106
Keshava, N.; Chiu, W. A.; Caldwell, J. C. (2007). PPARalpha and TCE: Keshava et al. respond.
Environ Health Perspect, 115, A15-A16. http://dx.doi.org/10.1289/ehp.115-al5
Kessler, W.; Numtip, W.; Grote, K.; Csanady, G. A.; Chahoud, I.; Filser, J. G. (2004). Blood
burden of di(2-ethylhexyl) phthalate and its primary metabolite mono(2-ethylhexyl)
phthalate in pregnant and nonpregnant rats and marmosets. Toxicol Appl Pharmacol, 195,
142-153. http://dx.doi.Org/10.1016/j.taap.2003.ll.014
Ketcha, M. M.; Stevens, D. K.; Warren, D. A.; Bishop, C. T.; Brashear, W. T. (1996).
Conversion of trichloroacetic acid to dichloroacetic acid in biological samples. J Anal
Toxicol, 20, 236-241. http://www.ncbi.nlm.nih.gov/pubmed/8835661
Keys, D. A.; Schultz, I. R.; Mahle, D. A.; Fisher, J. W. (2004). A quantitative description of
suicide inhibition of dichloroacetic acid in rats and mice. Toxicol Sci, 82, 381-393.
http://dx.doi.org/10.1093/toxsci/kfh276
Keys, D.; Bruckner, J.; Muralidhara, S.; Fisher, J. (2003). Tissue dosimetry expansion and cross-
validation of rat and mouse physiologically based pharmacokinetic models for
trichloroethylene. Toxicol Sci, 76, 35-50. http://dx.doi.org/10.1093/toxsci/kfg212
Kezic, S.; Monster, A.; Kriise, J.; Verberk, M. (2000). Skin absorption of some vaporous
solvents in volunteers. Int Arch Occup Environ Health, 73, 415-422.
http://www.ncbi.nlm.nih.gov/pubmed/11007346
Kezic, S.; Monster, A.; van de Gevel, I.; Kriise, J.; Opdam, J.; Verberk, M. (2001). Dermal
absorption of neat liquid solvents on brief exposures in volunteers. 62, 12-18.
http://www.ncbi.nlm.nih.gov/pubmed/11258863
Khan, M.; Kaphalia, B.; Prabhakar, B.; Kanz, M.; Ansari, G. (1995). Trichloroethene-induced
autoimmune response in female MRL +/+ mice. Toxicol Appl Pharmacol, 134, 155-160.
http://dx.doi.org/10.1006/taap.1995.1179
Khan, M.; Wu, X.; Ansari, G. (2001). Anti-malondialdehyde antibodies in MRL+/+ mice treated
with trichloroethene and dichloroacetyl chloride: possible role of lipid peroxidation in
60
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
autoimmunity. Toxicol Appl Pharmacol, 170, 88-92.
http://dx.doi.org/10.1006/taap.2000.9086
Khandwala, H. M.; McCutcheon, I. E.; Flyvbjerg, A.; Friend, K. E. (2000). The effects of
insulin-like growth factors on tumorigenesis and neoplastic growth. Endocr Rev, 21,215-
244. http://www.ncbi.nlm.nih.gov/pubmed/10857553
Kilburn, K. (2002). Do duration of exposure, proximity to electronic manufacturing plants, and
involvement in a lawsuit affect chlorinated solvent toxicity. Arch Environ Health, 57,
121-126. http://www.ncbi.nlm.nih.gov/pubmed/12194156
Kilburn, K. (2002). Is neurotoxicity associated with environmental trichloroethylene (TCE).
Arch Environ Health, 57, 113-120. http://www.ncbi.nlm.nih.gov/pubmed/12194155
Kilburn, K. and Thornton, J. (1996). Prediction equations for simple and visual two-choice
reactions times in environmental neurotoxicology. Arch Environ Health, 51, 439-444.
http://www.ncbi.nlm.nih.gov/pubmed/9012322
Kilburn, K. H. and Warshaw, R. H. (1992). Neurobehavioral effects of formaldehyde and
solvents on histology technicians: repeated testing across time. Environ Res, 58, 134-146.
http://dx.doi.org/10.1016/S0013-9351(05)80210-5
Kilburn, K. H. and Warshaw, R. H. (1992). Prevalence of symptoms of systemic lupus
erythematosus (SLE) and of fluorescent antinuclear antibodies associated with chronic
exposure to trichloroethylene and other chemicals in well water. Environ Res, 57, 1-9.
http://www.ncbi.nlm.nih.gov/pubmed/1740091
Kilburn, K. H. and Warshaw, R. H. (1993). Effects on neurobehavioral performance of chronic
exposure to chemically contaminated well water. Toxicol Ind Health, 9, 391-404.
http://www.ncbi.nlm.nih.gov/pubmed/8367882
Kilburn, K. H. and Warshaw, R. H. (1993). Neurobehavioral testing of subjects exposed
residentially to groundwater contaminated from an aluminum die-casting plant and local
referents. J Toxicol Environ Health, 39, 483-496.
http://www.ncbi.nlm.nih.gov/pubmed/8345533
Kilburn, K. H.; Warshaw, R.; Thornton, J. C. (1987). Formaldehyde impairs memory,
equilibrium, and dexterity in histology technicians: Effects which persist for days after
exposure. Arch Environ Occup Health, 42, 117-120.
http://www.ncbi.nlm.nih.gov/pubmed/3579365
Kim, C. F.; Jackson, E. L.; Woolfenden, A. E.; Lawrence, S.; Babar, I.; Vogel, S., . . . Jacks, T.
(2005). Identification of bronchioalveolar stem cells in normal lung and lung cancer.
Cell, 121, 823-835. http://dx.doi.Org/10.1016/j.cell.2005.03.032
Kim, D. and Ghanayem, B. (2006). Comparative metabolism and disposition of trichloroethylene
in Cyp2el-/-and wild-type mice. Drug Metab Dispos, 34, 2020-2027.
http://dx.doi.Org/10.l 124/dmd. 106.010538
Kim, H. J.; Bruckner, J. V.; Dallas, C. E.; Gallo, J. M. (1990). Effect of dosing vehicles on the
pharmacokinetics of orally administered carbon tetrachloride in rats. Toxicol Appl
Pharmacol, 102, 50-60.
Kim, H. J.; Odend'hal, S.; Bruckner, J. V. (1990). Effect of oral dosing vehicles on the acute
hepatotoxicity of carbon tetrachloride in rats. Toxicol Appl Pharmacol, 102, 34-49.
Kim, S.; Kim, D.; Pollack, G. M.; Collins, L. B.; Rusyn, I. (2009). Pharmacokinetic analysis of
trichloroethylene metabolism in male B6C3F1 mice: Formation and disposition of
trichloroacetic acid, dichloroacetic acid, S-(l,2-dichlorovinyl)glutathione and S-(l,2-
61
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
dichlorovinyl)-L-cysteine. Toxicol Appl Pharmacol, 238, 90-99.
http://dx.doi.Org/10.1016/j.taap.2009.04.019
Kim, S.; Kupke, K.; Ierardi-Curto, L.; Holme, E.; Greter, J.; Tanguay, R., . . . Levy, H. (2000).
Hepatocellular carcinoma despite long-term survival in chronic tyrosinaemia I. J Inherit
Metab Dis, 23, 791-804. http://www.ncbi.nlm.nih.gov/pubmed/11196105
Kimmerle, G. and Eben, A. (1973). Metabolism, excretion and toxicology of trichloroethylene
after inhalation: 1 experimental exposure on rats. Arch Toxicol, 30, 115-126.
http://www.ncbi.nlm.nih.gov/pubmed/4685263
Kimmerle, G. and Eben, A. (1973). Metabolism, excretion and toxicology of trichloroethylene
after inhalation: 2 experimental human exposure. Arch Toxicol, 30, 127-138.
http://www.ncbi.nlm.nih.gov/pubmed/4685264
King, G. S.; Smialek, J. E.; Troutman, W. G. (1985). Sudden death in adolescents resulting from
the inhalation of typewriter correction fluid. JAMA, 253, 1604-1606.
http://www.ncbi.nlm.nih.gov/pubmed/3974043
Kioski, C.; Heubach, S.; Petersen, N. (1990). The incidence of childhood leukemia and testicular
cancer in Pima County: 1970-1986.
Kioski, C.; Petersen, N.; Heubach, H.; Chapin, C. (1990). Cancer incidence and mortality in an
East Phoenix area overlaying groundwater contaminated with volatile organic
compounds.
Kiran, S.; Cocco, P.; Mannetje, A.; Satta, G.; D'Andrea, I.; Becker, N., . . . Boffetta, P. (2010).
Occupational exposure to ethylene oxide and risk of lymphoma. Epidemiology, 21, 905-
910. http://dx.doi.org/10.1097/EDE.0b013e3181f4cc0f
Kishi, R.; Harabuchi, I.; Ikeda, T.; Katakura, Y.; Miyake, H. (1993). Acute effects of
trichloroethylene on blood concentrations and performance decrements in rats and their
relevance to humans. Occup Environ Med, 50, 470-480.
http://www.ncbi.nlm.nih.gov/pubmed/8507600
Kitada, T.; Seki, S.; Kawakita, N.; Kuroki, T.; Monna, T. (1995). Telomere shortening in chronic
liver diseases. Biochem Biophys Res Commun, 211, 33-39.
http://dx.doi.org/10.1006/bbrc.1995.1774
Kjellstrand, P.; Bjerkemo, M.; Mortensen, I.; Mansson, L. (1981). Effects of long-term exposure
to trichloroethylene on the behavior of Mongolian gerbils (Meriones unguiculatus). J
Toxicol Environ Health, 8, 787-793.
Kjellstrand, P.; Holmquist, B.; Aim, P.; Kanje, M.; Romare, S.; Jonsson, I., . . . Bjerkemo, M.
(1983). Trichloroethylene: further studies of the effects on body and organ weights and
plasma butyrylcholinesterase activity in mice. Basic Clin Pharmacol Toxicol, 53, 375-
384. http://www.ncbi.nlm.nih.gov/pubmed/6659967
Kjellstrand, P.; Holmquist, B.; Mandahl, N.; Bjerkemo, M. (1983). Effects of continuous
trichloroethylene inhalation on different strains of mice. Basic Clin Pharmacol Toxicol,
53, 369-374. http://www.ncbi.nlm.nih.gov/pubmed/6659966
Kjellstrand, P.; Kanje, M.; Bjerkemo, M. (1987). Regeneration of the sciatic nerve in mice and
rats exposed to trichloroethylene. Toxicol Lett, 38, 187-191.
http://www.ncbi.nlm.nih.gov/pubmed/3629629
Kjellstrand, P.; Kanje, M.; Mansson, L.; Bjerkemo, M.; Mortensen, I.; Lanke, J.; Holmquist, B.
(1981). Trichloroethylene: effects on body and organ weights in mice, rats and gerbils.
Toxicology, 21, 105-115. http://www.ncbi.nlm.nih.gov/pubmed/7281199
62
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Kjellstrand, P.; Lanke, J.; Bjerkemo, M.; Zetterqvist, L.; Mansson, L. (1980). Irreversible effects
of trichloroethylene exposure on the central nervous system. Scand J Work Environ
Health, 6, 40-47. http://www.ncbi.nlm.nih.gov/pubmed/7384766
Klaunig, J. E.; Babich, M. A.; Baetcke, K. P.; Cook, J. C.; Corton, J. C.; David, R. M., . . .
Fenner-Crisp, P. A. (2003). PPARalpha agonist-induced rodent tumors: Modes of action
and human relevance. CritRev Toxicol, 33, 655-780.
http://dx.doi.org/10.1080/713608372
Kligerman, A. D.; Bryant, M. F.; Doerr, C. L.; Erexson, G. L.; Evansky, P. A.; Kwanyuen, P.;
McGee, J. K. (1994). Inhalation studies of the genotoxicity of trichloroethylene to
rodents. DNA Repair, 322, 87-96.
Kobayashi, M.; Ikeda, K.; Hosaka, T.; Sezaki, H.; Someya, T.; Akuta, N., . . . Kumada, H.
(2006). Dysplastic nodules frequently develop into hepatocellular carcinoma in patients
with chronic viral hepatitis and cirrhosis. Cancer, 106, 636-647.
http://dx.doi.org/10.1002/cncr.21607
Koch, R.; Schlegelmilch, R.; Wolf, H. U. (1988). Genetic effects of chlorinated ethylenes in the
yeast Saccharomyces cerevisiae. Mutat Res, 206, 209-216.
http://dx.doi.org/10.1016/0165-1218(88)90162-0
Kochen, W.; Kohlmiiller, D.; De Biasi, P.; Ramsay, R. (2003). The endogeneous formation of
highly chlorinated tetrahydro-beta-carbolines as a possible causative mechanism in
idiopathic Parkinson's disease. Adv Exp Med Biol, 527, 253-263.
http://www.ncbi.nlm.nih.gov/pubmed/15206739
Koizumi, A. (1989). Potential of physiologically based pharmacokinetics to amalgamate kinetic
data of trichloroethylene and tetrachloroethylene obtained in rats and man. Occup
Environ Med, 46, 239-249. http://www.ncbi.nlm.nih.gov/pubmed/2713280
Konietzko, H.; Elster, I.; Bencsath, A.; Drysch, K.; Weichardt, H. (1975). [EEG changes under
standardized trichloroethylene load (author's transl)]. Int Arch Occup Environ Health, 35,
257-264. http://www.ncbi.nlm.nih.gOv/pubmed/l 176181
Kopylev, L.; Chen, C.; White, P. (2007). Towards quantitative uncertainty assessment for cancer
risks: central estimates and probability distributions of risk in dose-response modeling.
Regul Toxicol Pharmacol, 49, 203-207.
Korrapati, M.; Lock, E.; Mehendale, H. (2005). Molecular mechanisms of enhanced renal cell
division in protection against S-l,2-dichlorovinyl-L-cysteine-induced acute renal failure
and death. Am J Physiol Renal Physiol, 289, F175-F185.
http://dx.doi.org/10.1152/ajprenal.00418.2004
Kozak, E. M. and Tate, S. S. (1982). Glutathione-degrading enzymes of microvillus membranes.
J Biol Chem, 257, 6322-6327. http://www.ncbi.nlm.nih.gov/pubmed/6122685
Krasowski, M. and Harrison, N. (2000). The actions of ether, alcohol and alkane general
anaesthetics on GABAA and glycine receptors and the effects of TM2 and TM3
mutations. Br J Pharmacol, 129, 731-743. http://dx.doi.org/10.1038/sj.bjp.0703087
Kraupp-Grasl, B.; Huber, W.; Putz, B.; Gerbracht, U.; Schulte-Hermann, R. (1990). Tumor
promotion by the peroxisome proliferator nafenopin involving a specific subtype of
altered foci in rat liver. Cancer Res, 50, 3701-3708.
http://www.ncbi.nlm.nih.gov/pubmed/1971196
Krause, R. J.; Lash, L. H.; Elfarra, A. A. (2003). Human kidney flavin-containing
monooxygenases and their potential roles in cysteine s-conjugate metabolism and
63
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
nephrotoxicity. J Pharmacol Exp Ther, 304, 185-191.
http://dx.doi.Org/10.l 124/j pet. 102.042911
Krejci, M.; Ridgewell, R.; Koechel, D. (1991). Acute effects of the D-isomer of S-(l,2-
dichlorovinyl)cysteine on renal function and ultrastructure in the pentobarbital-
anesthetized dog: site-specific toxicity involving the SI and S2 cells of the proximal
tubule. Toxicology, 69, 151-164. http://www.ncbi.nlm.nih.gov/pubmed/1949044
Krishnadasan, A.; Kennedy, N.; Zhao, Y.; Morgenstern, H.; Ritz, B. (2007). Nested case-control
study of occupational chemical exposures and prostate cancer in aerospace and radiation
workers. Am J Ind Med, 50, 383-390. http://dx.doi.org/10.1002/ajim.20458
Krishnadasan, A.; Kennedy, N.; Zhao, Y.; Morgenstern, H.; Ritz, B. (2008). Nested case-control
study of occupational physical activity and prostate cancer among workers using a job
exposure matrix. Cancer Causes Control, 19, 107-114. http://dx.doi.org/10.1007/sl0552-
007-9076-7
Kroneld, R. (1989). Volatile pollutants in the environment and human tissues. Bull Environ
Contam Toxicol, 42, 873-877. http://www.ncbi.nlm.nih.gov/pubmed/2743020
Kubo, Y.; Mitani, H.; Hino, O. (1994). Allelic loss at the predisposing gene locus in spontaneous
and chemically induced renal cell carcinomas in the Eker rat. Cancer Res, 54, 2633-2635.
http://www.ncbi.nlm.nih.gov/pubmed/8168090
Kudryavtsev, B. N.; Kudryavtseva, M. V.; Sakuta, G. A.; Stein, G. I. (1993). Human hepatocyte
polyploidization kinetics in the course of life cycle. Virchows Arch B Cell Pathol Incl
Mol Pathol, 64, 387-393. http://www.ncbi.nlm.nih.gov/pubmed/8148960
Kulig, B. M. (1987). The effects of chronic trichloroethylene exposure on neurobehavioral
functioning in the rat. Neurotoxicol Teratol, 9, 171-178.
http://www.ncbi.nlm.nih.gov/pubmed/3657753
Kuller, L. H.; Ockene, J. K.; Meilahn, E.; Wentworth, D. N.; Svendsen, K. H.; Neaton, J. D.
(1991). Cigarette smoking and mortality. MRFIT Research Group. Prev Med, 20, 638-
654. http://www.ncbi.nlm.nih.gov/pubmed/1758843
Kumar, M.; Kumar, R.; Hissar, S. S.; Saraswat, M. K.; Sharma, B. C.; Sakhuja, P.; Sarin, S. K.
(2007). Risk factors analysis for hepatocellular carcinoma in patients with and without
cirrhosis: a case-control study of 213 hepatocellular carcinoma patients from India. J
Gastroenterol Hepatol, 22, 1104-1111. http://dx.doi.0rg/lO.llll/j. 1440-
1746.2007.04908.x
Kumar, M.; Tewari, S.; Sharma, P.; Verma, V. K.; Chauhan, L. K.; Agarwal, S. K., . . . Goel, S.
K. (2009). Study of genetic polymorphism in solvent exposed population and its
correlation to in vitro effect of trichloroethylene on lymphocytes. J Environ Biol, 30, 685-
691. http://www.ncbi.nlm.nih.gov/pubmed/20136049
Kumar, P.; Prasad, A. K.; Dutta, K. K. (2000). Steroidogenic alterations in testes and sera of rats
exposed to trichloroethylene (TCE) by inhalation. Hum Exp Toxicol, 19, 117-121.
http://www.ncbi.nlm.nih.gov/pubmed/10773841
Kumar, P.; Prasad, A. K.; Saxena, D. K.; Manu, U.; Maji, B. K.; Dutta, K. K. (2000). Fertility
and general reproduction studies in trichloroethylene exposed rats. 43, 117-126.
Kumar, P.; Prasad, A.; Maji, B.; Mani, U.; Dutta, K. (2001). Hepatoxic alterations induced by
inhalation of trichl or ethylene (TCE) in rats. Biomed Environ Sci, 14, 325-332.
http://www.ncbi.nlm.nih.gov/pubmed/11862613
64
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
Kumar, P.; Prasad, A.; Mani, U.; Maji, B.; Dutta, K. (2001). Trichloroethylene induced testicular
toxicity in rats exposed by inhalation. Hum Exp Toxicol, 20, 585-589.
http://www.ncbi.nlm.nih.gov/pubmed/11926613
Kung, H.-C.; Hoyert, D. L.; Xu, J.; Murphy, S. L. (2008). Deaths: Final data for 2005, National
Vital Statistics Reports. Retrieved from
http://www.cdc.gov/nchs/data/nvsr/nvsr56/nvsr56_10.pdf
Kurasawa, K. (1988). Selective damage of pulmonary nonciliated bronchiolar epithelial (Clara)
cells by trichloroethylene in rats. Sangyo Eiseigaku Zasshi, 30, 121-133.
http://www.ncbi.nlm.nih.gov/pubmed/3418956
Kvam, B. M.; Romundstad, P. R.; Boffetta, P.; Andersen, A. (2005). Cancer in the Norwegian
printing industry. Scand J Work Environ Health, 31, 36-43.
http://www.ncbi.nlm.nih.gov/pubmed/15751617
Kylin, B.; Axell, K.; Samuel, H. E.; Lindborg, A. (1967). Effect of inhaled trichloroethylene on
the CNS: as measured by optokinetic nystagmus. Arch Environ Occup Health, 15, 48-52.
http://www.ncbi.nlm.nih.gov/pubmed/4951693
Kyrklund, T. and Haglid, K. (1991). Brain lipid composition in guinea pigs after intrauterine
exposure to perchloroethylene. Pharmacol Toxicol, 68, 146-148.
http://dx.doi.Org/10.llll/j.1600-0773.1991.tb02054.x
Kyrklund, T.; Kjellstrand, P.; Haglid, K. G. (1986). Fatty acid changes in rat brain ethanolamine
phosphoglycerides during and following chronic exposure to trichloroethylene. Toxicol
Appl Pharmacol, 85, 145-153. http://www.ncbi.nlm.nih.gov/pubmed/3764901
Kyrklund, T.; Kjellstrand, P.; Haglid, K. G. (1988). Effects of exposure to Freon 11, 1,1,1-
trichloroethane or perchloroethylene on the lipid and fatty-acid composition of rat
cerebral cortex. Scand J Work Environ Health, 14, 91-94.
La Vecchia, C.; Negri, E.; Franceschi, S.; D'Avanzo, B.; Boyle, P. (1994). A case-control study
of diabetes mellitus and cancer risk. Br J Cancer, 70, 950-953.
http://www.ncbi.nlm.nih.gov/pubmed/7947103
Lacey, J. V. J.; Garabrant, D. H.; Laing, T. J.; Gillespie, B. W.; Mayes, M. D.; Cooper, B. C.;
Schottenfeld, D. (1999). Petroleum distillate solvents as risk factors for undifferentiated
connective tissue disease (UCTD). Am J Epidemiol, 149, 761-770.
http://www.ncbi.nlm.nih.gov/pubmed/10206626
Lagakos, S. W.; Wessen, B. J.; Zelen, M. (1986). An analysis of contaminated well water and
health effects in Woburn, Massachusetts. J Am Stat Assoc, 81, 583-596.
Laham, S. (1970). Studies on placental transfer: Trichlorethylene. Ind Med Surg, 39, 46-49.
http://www.ncbi.nlm.nih.gov/pubmed/5263053
Lake, B. G.; Evans, J. G.; Cunninghame, M. E.; Price, R. J. (1993). Comparison of the hepatic
effects of nafenopin and WY-14,643 on peroxisome proliferation and cell replication in
the rat and Syrian hamster. Environ Health Perspect, 101, 241-247.
http://www.ncbi.nlm.nih.gov/pubmed/8013414
Lalwani, N. D.; Dethloff, L. A.; Haskins, J. R.; Robertson, D. G.; de la Iglesia, F. A. (1997).
Increased nuclear ploidy, not cell proliferation, is sustained in the peroxisome
proliferator-treated rat liver. Toxicol Pathol, 25, 165-176.
http://www.ncbi.nlm.nih.gov/pubmed/9125775
65
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Lamb, J. C. and Hentz, K. L. (2006). Toxicological review of male reproductive effects and
trichloroethylene exposure: assessing the relevance to human male reproductive health.
Reprod Toxicol, 22, 557-563. http://dx.doi.Org/10.1016/j.reprotox.2006.07.001
Land, P. C.; Owen, E. L.; Linde, H. W. (1981). Morphologic changes in mouse spermatozoa
after exposure to inhalational anesthetics during early spermatogenesis. Anesthesiology,
54, 53-56. http://www.ncbi.nlm.nih.gov/pubmed/6109470
Landrigan, P. J.; Stein, G. F.; Kominsky, J. R.; Ruhe, R. L.; Watanabe, A. S. (1987). Common-
source community and industrial exposure to trichloroethylene. Arch Environ Occup
Health, 42, 327-332. http://www.ncbi.nlm.nih.gov/pubmed/3439809
Lane, S. E.; Watts, R. A.; Bentham, G.; Innes, N. J.; Scott, D. G. I. (2003). Are environmental
factors important in primary systemic vasculitis? A case-control study. Arthritis Rheum,
48, 814-823. http://dx.doi.org/10.1002/art.10830
Langholz, B. and Goldstein, L. (1996). http://dx.doi.org/10.1214/ss/1032209663
Lapare, S.; Tardif, R.; Brodeur, J. (1995). Effect of various exposure scenarios on the biological
monitoring of organic solvents in alveolar air. II. 1,1,1-Trichloroethane and
trichloroethylene. Int Arch Occup Environ Health, 67, 375-394.
http://www.ncbi.nlm.nih.gov/pubmed/8567088
Lapis, K.; Zalatnai, A.; Timar, F.; Thorgeirsson, U. P. (1995). Quantitative evaluation of
lysozyme- and CD68-positive Kupffer cells in diethylnitrosamine-induced hepatocellular
carcinomas in monkeys. Carcinogenesis, 16, 3083-3085.
http://www.ncbi.nlm.nih.gov/pubmed/8603489
Larsen, G. L. and Stevens, J. L. (1986). Cysteine conjugate beta-lyase in the gastrointestinal
bacterium Eubacterium limosum. Mol Pharmacol, 29, 97-103.
http://www.ncbi.nlm.nih.gov/pubmed/3945231
Larson, J. and Bull, R. (1992). Species differences in the metabolism of trichloroethylene to the
carcinogenic metabolites trichloroacetate and dichloroacetate. Toxicol Appl Pharmacol,
115, 278-285. http://www.ncbi.nlm.nih.gov/pubmed/1641861
Larson, J. L. and Bull, R. J. (1989). Effect of ethanol on the metabolism of trichloroethylene. J
Toxicol Environ Health, 28, 395-406. http://dx.doi.org/10.1080/15287398909531359
Larson, J. L. and Bull, R. J. (1992). Metabolism and lipoperoxidative activity of trichloroacetate
and dichloroacetate in rats and mice. Toxicol Appl Pharmacol, 115, 268-277.
http://dx.doi.org/10.1016/0041-008X(92)90332-M
Lash, L. H.; Fisher, J. W.; Lipscomb, J. C.; Parker, J. C. (2000). Metabolism of
trichloroethylene. Environ Health Perspect, 108, 177-200.
Lash, L. H.; Jones, D. P.; Anders, M. W. (1988). Glutathione homeostasis and glutathione S-
conjugate toxicity in kidney. Rev Biochem Toxicol, 9, 29-67.
Lash, L. H.; Lipscomb, J. C.; Putt, D. A.; Parker, J. C. (1999). Glutathione conjugation of
trichloroethylene in human liver and kidney: Kinetics and individual variation. Drug
Metab Dispos, 27, 351-359. http://www.ncbi.nlm.nih.gov/pubmed/10064565
Lash, L. H.; Nelson, R. M.; Van Dyke, R. A.; Anders, M. W. (1990). Purification and
characterization of human kidney cytosolic cysteine conjugate beta-lyase activity. Drug
Metab Dispos, 18, 50-54. http://www.ncbi.nlm.nih.gov/pubmed/2139845
Lash, L. H.; Parker, J. C.; Scott, C. S. (2000). Modes of action of trichloroethylene for kidney
tumorigenesis. Environ Health Perspect, 108, 225-240.
http://www.ncbi.nlm.nih.gov/pubmed/10807554
66
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Lash, L. H.; Putt, D. A.; Huang, P.; Hueni, S. E.; Parker, J. C. (2007). Modulation of hepatic and
renal metabolism and toxicity of trichloroethylene and perchloroethylene by alterations in
status of cytochrome P450 and glutathione. Toxicology, 235, 11-26.
http://dx.doi.Org/10.1016/j.tox.2007.03.001
Lash, L. H.; Qian, W.; Putt, D. A.; Hueni, S. E.; Elfarra, A. A.; Krause, R. J.; Parker, J. C.
(2001). Renal and hepatic toxicity of trichloroethylene and its glutathione-derived
metabolites in rats and mice: sex-, species-, and tissue-dependent differences. J
Pharmacol Exp Ther, 297, 155-164. http://www.ncbi.nlm.nih.gov/pubmed/11259540
Lash, L. H.; Qian, W.; Putt, D. A.; Jacobs, K.; Elfarra, A. A.; Krause, R. J.; Parker, J. C. (1998).
Glutathione conjugation of trichloroethylene in rats and mice: Sex-, species-, and tissue-
dependent differences. DrugMetab Dispos, 26, 12-19.
http://www.ncbi.nlm.nih.gov/pubmed/9443846
Lash, L. H.; Visarius, T. M.; Sail, J. M.; Qian, W.; Tokarz, J. J. (1998). Cellular and subcellular
heterogeneity of glutathione metabolism and transport in rat kidney cells. Toxicology,
130, 1-15. http://www.ncbi.nlm.nih.gov/pubmed/9846992
Lash, L. H.; Xu, Y.; Elfarra, A. A.; Duescher, R. J.; Parker, J. C. (1995). Glutathione-dependent
metabolism of trichloroethylene in isolated liver and kidney cells of rats and its role in
mitochondrial and cellular toxicity. Drug Metab Dispos, 23, 846-853.
http://www.ncbi.nlm.nih.gov/pubmed/7493552
Lash, L.; Elfarra, A.; Anders, M. (1986). Renal cysteine conjugate beta-lyase. Bioactivation of
nephrotoxic cysteine S-conjugates in mitochondrial outer membrane. J Biol Chem, 261,
5930-5935. http://www.ncbi.nlm.nih.gov/pubmed/3700378
Lash, L.; Elfarra, A.; Rakiewicz-Nemeth, D.; Anders, M. (1990). Bioactivation mechanism of
cytotoxic homocysteine S-conjugates. Arch Biochem Biophys, 276, 322-330.
http://www.ncbi.nlm.nih.gov/pubmed/2154944
Lash, L.; Hueni, S.; Putt, D. (2001). Apoptosis, necrosis, and cell proliferation induced by S-
(l,2-dichlorovinyl)-L-cysteine in primary cultures of human proximal tubular cells.
Toxicol Appl Pharmacol, 177, 1-16. http://dx.doi.org/10.1006/taap.2001.9295
Lash, L.; Putt, D.; Brashear, W.; Abbas, R.; Parker, J.; Fisher, J. (1999). Identification of S-(l,2-
dichlorovinyl)glutathione in the blood of human volunteers exposed to trichloroethylene.
J Toxicol Environ Health A, 56, 1-21. http://www.ncbi.nlm.nih.gov/pubmed/9923751
Lash, L.; Putt, D.; Hueni, S.; Krause, R.; Elfarra, A. (2003). Roles of necrosis, Apoptosis, and
mitochondrial dysfunction in S-(l,2-dichlorovinyl)-L-cysteine sulfoxide-induced
cytotoxicity in primary cultures of human renal proximal tubular cells. J Pharmacol Exp
Ther, 305, 1163-1172. http://dx.doi.org/10.1124/jpet.102.046185
Lash, L.; Putt, D.; Parker, J. (2006). Metabolism and tissue distribution of orally administered
trichloroethylene in male and female rats: identification of glutathione- and cytochrome
P-450-derived metabolites in liver, kidney, blood, and urine. J Toxicol Environ Health A,
69, 1285-1309. http://dx.doi.org/10.1080/15287390500360133
Lash, L.; Sausen, P.; Duescher, R.; Cooley, A.; Elfarra, A. (1994). Roles of cysteine conjugate
beta-lyase and S-oxidase in nephrotoxicity: studies with S-(l,2-dichlorovinyl)-L-cysteine
and S-(l,2-dichlorovinyl)-L-cysteine sulfoxide. J Pharmacol Exp Ther, 269, 374-383.
http://www.ncbi.nlm.nih.gov/pubmed/8169843
Lash, T. L. and Green, L. C. (1993). Blink reflex measurement of effects of trichloroethylene
exposure on the trigeminal nerve [letter]. Muscle Nerve, 16, 217-220.
67
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Laslo-Baker, D.; Barrera, M.; Knittel-Keren, D.; Kozer, E.; Wolpin, J.; Khattak, S., . . . Koren,
G. (2004). Child neurodevelopmental outcome and maternal occupational exposure to
solvents. Arch Pediatr Adolesc Med, 158, 956-961.
http://dx.doi.org/10.1001/archpedi.158.10.956
Latendresse, J. R. and Pereira, M. A. (1997). Dissimilar characteristics of n-methyl-n-
nitrosourea-initiated foci and tumors promoted by dichloroacetic acid or trichloroacetic
acid in the liver of female B6C3F1 mice. Toxicol Pathol, 25, 433-440.
http://dx.doi.org/10.1177/019262339702500501
Lau, S. S.; Kleiner, H. E.; Monks, T. J. (1995). Metabolism as a determinant of species
susceptibility to 2,3,5-(triglutathion-S-yl)hydroquinone-mediated nephrotoxicity. The
role of N-acetylation and N-deacetylation. Drug Metab Dispos, 23, 1136-1142.
http://www.ncbi.nlm.nih.gov/pubmed/8654203
Laughter, A. R.; Dunn, C. S.; Swanson, C. L.; Howroyd, P.; Cattley, R. C.; Corton, J. C. (2004).
Role of the peroxisome proliferator-activated receptor alpha (PPARalpha) in responses to
trichloroethylene and metabolites, trichloroacetate and dichloroacetate in mouse liver.
Toxicology, 203, 83-98. http://dx.doi.Org/10.1016/j.tox.2004.06.014
Laureno, R. (1993). Trichloroethylene does not cause trigeminal neuropathy [letter]. Muscle
Nerve, 16, 217.
Lawrence, T.; Hageman, T.; Balkwill, F. (2007). Cancer. Sex, cytokines, and cancer. Science,
317, 51-52. http://dx.doi.org/10.1126/science.1146052
Le Mesurier, S. M.; Lykke, A. W. J.; Stewart, B. W. (1980). Reduced yield of pulmonary
surfactant: patterns of response following administration of chemicals to rats by
inhalation. Toxicol Lett, 5, 89-93. http://www.ncbi.nlm.nih.gov/pubmed/6892858
Leakey, J. E.; Seng, J. E.; Allaben, W. T. (2003). Body weight considerations in the B6C3F1
mouse and the use of dietary control to standardize background tumor incidence in
chronic bioassays. Toxicol Appl Pharmacol, 193, 237-265.
http://www.ncbi.nlm.nih.gov/pubmed/14644626
Leakey, J.; Seng, J.; Latendresse, J.; Hussain, N.; Allen, L.; Allaben, W. (2003). Dietary
controlled carcinogenicity study of chloral hydrate in male B6C3F1 mice. Toxicol Appl
Pharmacol, 193, 266-280. http://www.ncbi.nlm.nih.gov/pubmed/14644627
Leandri, M.; Schizzi, R.; Scielzo, C.; Favale, E. (1995). Electrophysiological evidence of
trigeminal root damage after trichloroethylene exposure. Muscle Nerve, 18, 467-468.
http://dx.doi.org/10.1002/mus.880180416
Leavitt, S. A.; DeAngelo, A. B.; George, M. H.; Ross, J. A. (1997). Assessment of the
mutagenicity of dichloroacetic acid in lacl transgenic B6C3F1 mouse liver.
Carcinogenesis, 18, 2101-2106. http://www.ncbi.nlm.nih.gov/pubmed/9395208
Ledda-Columbano, G. M.; Columbano, A.; Curto, M.; Ennas, M. G.; Coni, P.; Sarma, D. S.;
Pani, P. (1989). Further evidence that mitogen-induced cell proliferation does not support
the formation of enzyme-altered islands in rat liver by carcinogens. Carcinogenesis, 10,
847-850. http://www.ncbi.nlm.nih.gov/pubmed/2565171
Ledda-Columbano, G. M.; Coni, P.; Simbula, G.; Zedda, I.; Columbano, A. (1993).
Compensatory regeneration, mitogen-induced liver growth, and multistage chemical
carcinogenesis. Environ Health Perspect, 101 Suppl 5, 163-168.
http://www.ncbi.nlm.nih.gov/pubmed/8013404
68
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Ledda-Columbano, G. M.; Curto, M.; Piga, R.; Zedda, A. I.; Menegazzi, M.; Sartori, C., . . .
Columbano, A. (1998). In vivo hepatocyte proliferation is inducible through a TNF and
IL-6-independent pathway. Oncogene, 17, 1039-1044.
http://dx.doi.org/10.1038/sj.onc.1202018
Ledda-Columbano, G. M.; Perra, A.; Concas, D.; Cossu, C.; Molotzu, F.; Sartori, C., . . .
Columbano, A. (2003). Different effects of the liver mitogens triiodo-thyronine and
ciprofibrate on the development of rat hepatocellular carcinoma. Toxicol Pathol, 31, 113-
120. http://www.ncbi.nlm.nih.gov/pubmed/12597455
Lee, D.; Park, I.; Kim, J.; Lee, Y.; Kim, D.; Kang, S. (1998). Neurobehavioral changes in shoe
manufacturing workers. Neurotoxicol Teratol, 20, 259-263.
http://www.ncbi.nlm.nih.gov/pubmed/9638683
Lee, J. S.; Chu, I. S.; Heo, J.; Calvisi, D. F.; Sun, Z.; Roskams, T., . . . Thorgeirsson, S. S.
(2004). Classification and prediction of survival in hepatocellular carcinoma by gene
expression profiling. Hepatology, 40, 667-676. http://dx.doi.org/10.1002/hep.20375
Lee, J. S.; Chu, I. S.; Mikaelyan, A.; Calvisi, D. F.; Heo, J.; Reddy, J. K.; Thorgeirsson, S. S.
(2004). Application of comparative functional genomics to identify best-fit mouse
models to study human cancer. Nat Genet, 36, 1306-1311.
http://dx.doi.org/10.1038/ngl481
Lee, K.; Bruckner, J.; Muralidhara, S.; Gallo, J. (1996). Characterization of presystemic
elimination of trichloroethylene and its nonlinear kinetics in rats. Toxicol Appl
Pharmacol, 139, 262-271. http://dx.doi.org/10.1006/taap.1996.0165
Lee, K.; Muralidhara, S.; Schnellmann, R.; Bruckner, J. (2000). Contribution of direct solvent
injury to the dose-dependent kinetics of trichloroethylene: portal vein administration to
rats. Toxicol Appl Pharmacol, 164, 46-54. http://dx.doi.org/10.1006/taap.2000.8891
Lee, K.; Muralidhara, S.; White, C.; Bruckner, J. (2000). Mechanisms of the dose-dependent
kinetics of trichloroethylene: oral bolus dosing of rats. Toxicol Appl Pharmacol, 164, 55-
64. http://dx.doi.org/10.1006/taap.2000.8892
Lee, L. J.-H.; Chan, C.-C.; Chung, C.-W.; Ma, Y.-C.; Wang, G.-S.; Wang, J.-D. (2002). Health
risk assessment on residents exposed to chlorinated hydrocarbons contaminated in
groundwater of a hazardous waste site. J Toxicol Environ Health A, 65, 219-235.
http://dx.doi.org/10.1080/15287390252800828
Lee, L.; Chung, C.; Ma, Y.; Wang, G.; Chen, P.; Hwang, Y.; Wang, J. (2003). Increased
mortality odds ratio of male liver cancer in a community contaminated by chlorinated
hydrocarbons in groundwater. Occup Environ Med, 60, 364-369.
http://dx.doi.Org/10.l 136/oem. 60.5.364
Lehmann, I.; Rehwagen, M.; Diez, U.; Seiffart, A.; Rolle-Kampczyk, U.; Richter, M., . . .
Herbarth, O. (2001). Enhanced in vivo IgE production and T cell polarization toward the
type 2 phenotype in association with indoor exposure to VOC: Results of the LARS
study. Int JHyg Environ Health, 204, 211-221. http://dx.doi.org/10.1078/1438-4639-
00100
Lehmann, I.; Thoelke, A.; Rehwagen, M.; Rolle-Kampczyk, U.; Schlink, U.; Schulz, R., . . .
Herbarth, O. (2002). The influence of maternal exposure to volatile organic compounds
on the cytokine secretion profile of neonatal T cells. Environ Toxicol, 17, 203-210.
http ://dx. doi. org /10.1002/tox .10055
69
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Lehnart, S. E.; Mongillo, M.; Bellinger, A.; Lindegger, N.; Bi-Xing, C.; Hsueh, W., . . . Marks,
A. R. (2008). Leaky Ca2+ release channel/ryanodine receptor 2 causes
seizures and sudden cardiac death in mice. J Clin Invest, 118, 2230-2245.
Leigh, J. P. and Jiang, W. Y. (1993). Liver cirrhosis deaths within occupations and industries in
the California occupational mortality study. Addiction, 88, 767-779.
http://www.ncbi.nlm.nih.gov/pubmed/8329968
Lemke, L. E. and McQueen, C. A. (1995). Acetylation and its role in the mutagenicity of the
antihypertensive agent hydralazine. Drug Metab Dispos, 23, 559-565.
http://www.ncbi.nlm.nih.gov/pubmed/7587931
Leopardi, P.; Zijno, A.; Bassani, B.; Pacchierotti, F. (1993). In vivo studies on chemically
induced aneuploidy in mouse somatic and germinal cells. Mutat Res, 287, 119-130.
http://www.ncbi.nlm.nih.gov/pubmed/7683378
Leuschner, J. and Beuscher, N. (1998). Studies on the mutagenic and carcinogenic potential of
chloral hydrate. Arzneimittelforschung, 48, 961-968.
http://www.ncbi.nlm.nih.gov/pubmed/9825112
Leuschner, J. and Leuschner, F. (1991). Evaluation of the mutagenicity of chloral hydrate in
vitro and in vivo. Arzneimittelforschung, 41, 1101-1103.
http://www.ncbi.nlm.nih.gov/pubmed/1799393
Levy, D.; Lipton, R.; Holzman, P. (1981). Smooth pursuit eye movements: effects of alcohol and
chloral hydrate. J Psychiatr Res, 16, 1-11. http://www.ncbi.nlm.nih.gov/pubmed/7205696
Lewis, G. D.; Reynolds, R. C.; Johnson, A. R. (1984). Some effects of trichloroethylene on
mouse lungs and livers. Vascul Pharmacol, 15, 139-144.
http://www.ncbi.nlm.nih.gov/pubmed/6425 111
Lewis, R. J., Sr. (Ed.). (2001). Hawley's Condensed Chemical Dictionary (14 ed.). New York,
NY: John Wiley & Sons, Inc.
Lewitt, M.; Brismar, K.; Wang, J.; Wivall-Helleryd, I.; Sindelar, P.; Gonzalez, F., . . . Bobek, G.
(2001). Responses of insulin-like growth factor (IGF)-I and IGF-binding proteins to
nutritional status in peroxisome proliferator-activated receptor-alpha knockout mice.
Growth Horm IGF Res, 11, 303-313. http://dx.doi.org/10.1054/ghir.2001.0247
Li, H.; Dai, Y.; Huang, H.; Li, L.; Leng, S.; Cheng, J., . . . Zheng, Y. (2007). HLA-B*1301 as a
biomarker for genetic susceptibility to hypersensitivity dermatitis induced by
trichloroethylene among workers in China. Environ Health Perspect, 115, 1553-1556.
http://dx.doi.org/10.1289/ehp.10325
Li, T.; Schultz, I.; Keys, D. A.; Campbell, J. L.; Fisher, J. W. (2008). Quantitative evaluation of
dichloroacetic acid kinetics in human—a physiologically based pharmacokinetic modeling
investigation. Toxicology, 245, 35-48. http://dx.doi.Org/10.1016/j.tox.2007.12.010
Liang, J. C. and Brinkley, B. R. (1985). Chemical probes and possible targets for the induction of
aneuploidy. Basic Life Sci, 36, 491-505. http://www.ncbi.nlm.nih.gov/pubmed/4096703
Liang, J. C. and Pacchierotti, F. (1988). Cytogenetic investigation of chemically-induced
aneuploidy in mouse spermatocytes. Mutat Res-Fundam Mol Mech Mutagen, 201, 325-
335. http://dx.doi.org/10.1016/0027-5107(88)90021-8
Liangpunsakul, S.; Kolwankar, D.; Pinto, A.; Gorski, J.; Hall, S.; Chalasani, N. (2005). Activity
of CYP2E1 and CYP3A enzymes in adults with moderate alcohol consumption: a
comparison with nonalcoholics. Hepatology, 41, 1144-1150.
http://dx.doi.org/10.1002/hep.20673
70
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Libbrecht, L. (2006). Hepatic progenitor cells in human liver tumor development. World J
Gastroenterol, 12, 6261-6265. http://www.ncbi.nlm.nih.gov/pubmed/17072946
Lide, D. R. (Ed.). (1998). handbook of chemistry and physics (79th ed.). Boca Raton, FL: CRC
Press.
Lieber, C. (2004). The discovery of the microsomal ethanol oxidizing system and its physiologic
and pathologic role. Drug Metab Rev, 36, 511-529. http://dx.doi.org/10.1081/DMR-
200033441
Ligocki, M. P.; Leuenberger, C.; Pankow, J. F. (1985). Trace organic-compounds in rain: 2. Gas
scavenging of neutral organic-compounds. Atmos Environ, 19, 1609-1617.
Lilly, P. D.; Simmons, J. E.; Pegram, R. A. (1994). Dose-dependent vehicle differences in the
acute toxicity of bromodichloromethane. Fundam Appl Toxicol, 23, 132-140.
http://www.ncbi.nlm.nih.gov/pubmed/7958557
Lin, L. W.; Lin, X. Y.; He, Y. M.; Gao, S. D.; Lin, X. D. (2003). Biological characteristics of
HCC by ultrasound-guided aspiration biopsy and its clinical application. World J
Gastroenterol, 9, 941-945. http://www.ncbi.nlm.nih.gov/pubmed/12717834
Lin, Y. Z.; Brunt, E. M.; Bowling, W.; Hafenrichter, D. G.; Kennedy, S. C.; Flye, M. W.;
Ponder, K. P. (1995). Ras-transduced diethylnitrosamine-treated hepatocytes develop into
cancers of mixed phenotype in vivo. Cancer Res, 55, 5242-5250.
http://www.ncbi.nlm.nih.gov/pubmed/7585583
Lindbohm, M. L.; Sallmen, M.; Kyyronen, P.; Kauppinen, T.; Pukkala, E. (2009). Risk of liver
cancer and exposure to organic solvents and gasoline vapors among Finnish workers. Int
J Cancer, 124, 2954-2959. http://dx.doi.org/10.1002/ijc.24309
Lindbohm, M. L.; Taskinen, H.; Sallmen, M.; Hemminki, K. (1990). Spontaneous abortions
among women exposed to organic solvents. Am J Ind Med, 17, 449-463.
http://dx.doi.org/10.1002/ajim.4700170404
Lindstrom, K. (1973). Psychological performances of workers exposed to various solvents.
Scand J Work Environ Health, 10, 151-155.
Lindstrom, K. (1980). Changes in psychological performances of solvent-poisoned and solvent
exposed workers. Am J Ind Med, 1, 69-84.
http://www.ncbi.nlm.nih.gov/pubmed/7342756
Lindstrom, K.; Antti-Poika, M.; Tola, S.; Hyytiainen, A. (1982). Psychological prognosis of
diagnosed chronic organic solvent intoxication. Neurobehav Toxicol Teratol, 4, 581-588.
http://www.ncbi.nlm.nih.gov/pubmed/7177310
Lingohr, M.; Bull, R.; Kato-Weinstein, J.; Thrall, B. (2002). Dichloroacetate stimulates glycogen
accumulation in primary hepatocytes through an insulin-independent mechanism. Toxicol
Sci, 68, 508-515. http://www.ncbi.nlm.nih.gov/pubmed/12151648
Lingohr, M.; Thrall, B.; Bull, R. (2001). Effects of dichloroacetate (DCA) on serum insulin
levels and insulin-controlled signaling proteins in livers of male B6C3F1 mice. Toxicol
Sci, 59, 178-184. http://www.ncbi.nlm.nih.gov/pubmed/11134557
Lipscomb, J. C.; Confer, P. D.; Miller, M. R.; Stamm, S. C.; Snawder, J. E.; Bandiera, S. M.
(1998). Metabolism of trichloroethylene and chloral hydrate by the Japanese medada
(Oryzias latipes) in vitro. Environ Toxicol Chem, 17, 325-332.
Lipscomb, J. C.; Fisher, J. W.; Confer, P. D.; Byczkowski, J. Z. (1998). In vitro to in vivo
extrapolation for trichloroethylene metabolism in humans. Toxicol Appl Pharmacol, 152,
376-387. http://dx.doi.org/10.1006/taap.1998.8485
71
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Lipscomb, J. C.; Garrett, C. M.; Snawder, J. E. (1997). Cytochrome P450-dependent metabolism
of trichloroethylene: interindividual differences in humans. Toxicol Appl Pharmacol,
142, 311-318. http://dx.doi.org/10.1006/taap.1996.8040
Lipscomb, J. C.; Garrett, C. M.; Snawder, J. E. (1998). Use of kinetic and mechanistic data in
species extrapolation of bioactivation: cytochrome P-450 dependent trichloroethylene
metabolism at occupationally relevant concentrations. J Occup Health, 40, 110-117.
Lipscomb, J. C.; Mahle, D. A.; Brashear, W. T.; Barton, H. A. (1995). Dichloroacetic acid:
Metabolism in cytosol. Drug Metab Dispos, 23, 1202-1205.
http://www.ncbi.nlm.nih.gov/pubmed/8591719
Lipscomb, J. C.; Mahle, D. A.; Brashear, W. T.; Garrett, C. M. (1996). A species comparison of
chloral hydrate metabolism in blood and liver. Biochem Biophys Res Commun, 227,
340-350. http://dx.doi.org/10.1006/bbrc.1996.1511
Lipscomb, J. C.; Teuschler, L. K.; Swartout, J.; Popken, D.; Cox, T.; Kedderis, G. L. (2003). The
impact of cytochrome P450 2El-dependent metabolic variance on a risk-relevant
pharmacokinetic outcome in humans. Risk Anal, 23, 1221-1238.
http://www.ncbi.nlm.nih.gov/pubmed/14641897
Little, J. C.; Daisey, J. M.; Nazaroff, W. W. (1992). Transport of subsurface contaminants into
buildings: an exposure pathway for volatile organics. Environ Sci Technol, 26, 2058-
2066.
Liu, E.; Jelinek, J.; Pastore, Y. D.; Guan, Y.; Prchal, J. F.; Prchal, J. T. (2003). Discrimination of
polycythemias and thrombocytoses by novel, simple, accurate clonality assays and
comparison with PRV-1 expression and BFU-E response to erythropoietin. Blood, 101,
3294-3301. http://dx.doi.org/10.1182/blood-2002-07-2287
Liu, Y.; Bartlett, M. G.; White, C. A.; Muralidhara, S.; Bruckner, J. V. (2009). Presystemic
elimination of trichloroethylene in rats following environmentally relevant oral
exposures. Drug Metab Dispos, 37, 1994-1998.
http://dx.doi.Org/10.l 124/dmd. 109.028100
Liu, Y.-T.; Jin, C.; Chen, Z.; Cai, S.-X.; Yin, S.-N.; Li, G.-L., . . . Ikeda, M. (1988). Increased
subjective symptom prevalence among workers exposed to trichloroethylene at sub-OEL
levels. Tohoku J Exp Med, 155, 183-195. http://www.ncbi.nlm.nih.gov/pubmed/3212780
Lock, E.; Reed, C.; McMillan, J.; Oatis, J.; Schnellmann, R. (2007). Lack of formic acid
production in rat hepatocytes and human renal proximal tubule cells exposed to chloral
hydrate or trichloroacetic acid. Toxicology, 230, 234-243.
http://dx.doi.Org/10.1016/j.tox.2006.l 1.055
Lockey, J.; Kelly, C.; Cannon, G.; Colby, T.; Aldrich, V.; Livingston, G. (1987). Progressive
systemic sclerosis associated with exposure to trichloroethylene. J Occup Med, 29, 493-
496. http://www.ncbi.nlm.nih.gov/pubmed/3612322
Loeber, C. and Runyan, R. (1990). A comparison of fibronectin, laminin, and
galactosyltransferase adhesion mechanisms during embryonic cardiac mesenchymal cell
migration in vitro. Dev Biol, 140, 401-412.
http://www.ncbi.nlm.nih.gov/pubmed/2142656
Loeber, C.; Hendrix, M.; Diez De Pinos, S.; Goldberg, S. (1988). Trichloroethylene: a cardiac
teratogen in developing chick embryos. Pediatr Res, 24, 740-744.
http://www.ncbi.nlm.nih.gov/pubmed/3205631
72
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Lopreato, G.; Phelan, R.; Borghese, C.; Beckstead, M.; Mihic, S. (2003). Inhaled drugs of abuse
enhance serotonin-3 receptor function. Drug Alcohol Depend, 70, 11-15.
http://www.ncbi.nlm.nih.gov/pubmed/12681521
Lorente, C.; Cordier, S.; Bergeret, A.; De Walle, H.; Goujard, J.; Ayme, S., . . . Bianchi, F.
(2000). Maternal occupational risk factors for oral clefts. Occupational Exposure and
Congenital Malformation Working Group. Scand J Work Environ Health, 26, 137-145.
http://www.ncbi.nlm.nih.gov/pubmed/10817379
Lorimore, S.; Coates, P.; Wright, E. (2003). Radiation-induced genomic instability and bystander
effects: inter-related nontargeted effects of exposure to ionizing radiation. Oncogene, 22,
7058-7069. http://dx.doi.org/10.1038/sj.onc.1207044
Lowengart, R. A.; Peters, J. M.; Cicioni, C.; Buckley, J.; Bernstein, L.; Preston-Martin, S.;
Rappaport, E. (1987). Childhood leukemia and parents' occupational and home
exposures. J Natl Cancer Inst, 79, 39-46. http://www.ncbi.nlm.nih.gov/pubmed/3474448
Luebke, R. W.; Chen, D. H.; Dietert, R.; Yang, Y.; King, M.; Luster, M. I. (2006). The
comparative immunotoxicity of five selected compounds following developmental or
adult exposure. J Toxicol Environ Health B Crit Rev, 9, 1-26.
http://dx.doi.org/10.1080/15287390500194326
Luedde, T.; Beraza, N.; Kotsikoris, V.; van Loo, G.; Nenci, A.; De Vos, R., . . . Pasparakis, M.
(2007). Deletion of NEMO/IKKgamma in liver parenchymal cells causes steatohepatitis
and hepatocellular carcinoma. Cancer Cell, 11, 119-132.
http://dx.doi.Org/10.1016/j.ccr.2006.12.016
Lumpkin, M. H.; Bruckner, J. V.; Campbell, J. L.; Dallas, C. E.; White, C. A.; Fisher, J. W.
(2003). Plasma binding of trichloroacetic acid in mice, rats, and humans under cancer
bioassay and environmental exposure conditions. Drug Metab Dispos, 31, 1203-1207.
http://dx.doi.org/10.1124/dmd.31.10.1203
Lundberg, I.; Alfredsson, L.; Plato, N.; Sverdrup, B.; Klareskog, L.; Kleinau, S. (1994).
Occupation, occupational exposure to chemicals and rheumatological disease: A register
based cohort study. Scand J Rheumatol, 23, 305-310.
http://dx.doi.org/10.3109/03009749409099278
Lundgren, B. and DePierre, J. W. (1989). Proliferation of peroxisomes and induction of cytosolic
and microsomal epoxide hydrolases in different strains of mice and rats after dietary
treatment with clofibrate. Xenobiotica, 19, 867-881.
http://www.ncbi.nlm.nih.gov/pubmed/2815829
Lundgren, B.; Meijer, J.; DePierre, J. W. (1987). Induction of cytosolic and microsomal epoxide
hydrolases and proliferation of peroxisomes and mitochondria in mouse liver after dietary
exposure to p-chlorophenoxyacetic acid, 2,4-dichlorophenoxyacetic acid and 2,4,5-
trichlorophenoxyacetic acid. Biochem Pharmacol, 36, 815-821.
http://www.ncbi.nlm.nih.gov/pubmed/3032197
Lutz, W. K.; Gaylor, D. W.; Conolly, R. B.; Lutz, R. W. (2005). Nonlinearity and thresholds in
dose-response relationships for carcinogenicity due to sampling variation, logarithmic
dose scaling, or small differences in individual susceptibility. Toxicol Appl Pharmacol,
207, S565-S569. http://dx.doi.Org/10.1016/j.taap.2005.01.038
Lybarger, J. A.; Lichtveld, M. Y.; Amler, R. W. (1999). Biomedical testing of the kidney for
persons exposed to hazardous substances in the environment. Ren Fail, 21, 263-274.
73
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Lynch, A. and Parry, J. (1993). The cytochalasin-B micronucleus/kinetochore assay in vitro:
studies with 10 suspected aneugens. MutatRes, 287, 71-86.
http://www.ncbi.nlm.nih.gov/pubmed/7683386
Lynge, E.; Anttila, A.; Hemminki, K. (1997). Organic solvents and cancer. Cancer Causes
Control, 8, 406-419. http://dx.doi.Org/10.1023/A:1018461406120
Macchiarini, P. (2006). Primary tracheal tumours. Lancet Oncol, 7, 83-91.
http://dx.doi.org/10.1016/S1470-2045(05)70541-6
Macdonald, N.; Holden, P. R.; Roberts, R. A. (1999). Addition of peroxisome proliferator-
activated receptor alpha to guinea pig hepatocytes confers increased responsiveness to
peroxisome proliferators. Cancer Res, 59, 4776-4780.
http://www.ncbi.nlm.nih.gov/pubmed/10519382
Mackay, J.; Fox, V.; Griffiths, K.; Fox, D.; Howard, C.; Coutts, C., . . . Styles, J. (1995).
Trichloroacetic acid: investigation into the mechanism of chromosomal damage in the in
vitro human lymphocyte cytogenetic assay and the mouse bone marrow micronucleus
test. Carcinogenesis, 16, 1127-1133. http://www.ncbi.nlm.nih.gov/pubmed/7767975
Maeda, S.; Kamata, H.; Luo, J. L.; Leffert, H.; Karin, M. (2005). IKKbeta couples hepatocyte
death to cytokine-driven compensatory proliferation that promotes chemical
hepatocarcinogenesis. Cell, 121, 977-990. http://dx.doi.Org/10.1016/j.cell.2005.04.014
Magnusson, B. and Kligman, A. M. (1969). The identification of contact allergens by animal
assay. The guinea pig maximization test. J Invest Dermatol, 52, 268-276.
http://www.ncbi.nlm.nih.gov/pubmed/5774356
Mahle, D. A.; Gearhart, J. M.; Grigsby, C. C.; Mattie, D. R.; Barton, H. A.; Lipscomb, J. C.;
Cook, R. S. (2007). Age-dependent partition coefficients for a mixture of volatile organic
solvents in Sprague-Dawley rats and humans. J Toxicol Environ Health A, 70, 1745-
1751. http://dx.doi.org/10.1080/15287390701458991
Mahr, A. D.; Neogi, T.; Merkel, P. A. (2006). Epidemiology of Wegener's granulomatosis:
Lessons from descriptive studies and analyses of genetic and environmental risk
determinants. Clin Exp Rheumatol, 24, S82-S91.
http://www.ncbi.nlm.nih.gov/pubmed/16859601
Mailhes, J.; Aardema, M.; Marchetti, F. (1993). Investigation of aneuploidy induction in mouse
oocytes following exposure to vinblastine-sulfate, pyrimethamine, diethylstilbestrol
diphosphate, or chloral hydrate. Environ Mol Mutagen, 22, 107-114.
http://www.ncbi.nlm.nih.gov/pubmed/8359152
Maitre, A.; Hours, M.; Bonneterre, V.; Arnaud, J.; Arslan, M. T.; Carpentier, P., . . . de
Gaudemaris, R. (2004). Systemic sclerosis and occupational risk factors: role of solvents
and cleaning products. J Rheumatol, 31, 2395-2401.
http://www.ncbi.nlm.nih.gov/pubmed/15570640
Makwana, O.; King, N. M.; Ahles, L.; Selmin, O.; Granzier, H. L.; Runyan, R. B. (2010).
Exposure to low-dose trichloroethylene alters shear stress gene expression and function
in the developing chick heart. Cardiovasc Toxicol, 10, 100-107.
http://dx.doi.org/10.1007/sl2012-010-9066-y
Malarkey, D. E.; Johnson, K.; Ryan, L.; Boorman, G.; Maronpot, R. R. (2005). New insights into
functional aspects of liver morphology. Toxicol Pathol, 33, 27-34.
http://dx.doi.org/10.1080/01926230590881826
74
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Malherbe, P.; Alberati-Giani, D.; Kohler, C.; Cesura, A. M. (1995). Identification of a
mitochondrial form of kynurenine aminotransferase/glutamine transaminase K from rat
brain. FEBS Lett, 367, 141-144. http://www.ncbi.nlm.nih.gov/pubmed/7796908
Mallin, K. (1990). Investigation of a bladder cancer cluster in northwestern Illinois. Am J
Epidemiol, 132, S96-106. http://www.ncbi.nlm.nih.gov/pubmed/2356842
Mally, A.; Walker, C.; Everitt, J.; Dekant, W.; Vamvakas, S. (2006). Analysis of renal cell
transformation following exposure to trichloroethene in vivo and its metabolite S-
(dichlorovinyl)-L-cysteine in vitro. Toxicology, 224, 108-118.
http://dx.doi.Org/10.1016/j.tox.2006.04.036
Maloney, E. K. and Waxman, D. J. (1999). trans-Activation of PPARalpha and PPARgamma by
structurally diverse environmental chemicals. Toxicol Appl Pharmacol, 161, 209-218.
http://dx.doi.org/10.1006/taap.1999.8809
Maltoni, C.; Lefemine, G.; Cotti, G.; Perino, G. (1988). Long-term carcinogenicity bioassays on
trichloroethylene administered by inhalation to Sprague-Dawley rats and Swiss and
B6C3F1 mice. AnnN Y Acad Sci, 534, 316-342.
http://www.ncbi.nlm.nih.gov/pubmed/3389663
Maltoni, C.; Lefemine, G.; G, C. (1986). Experimental research on trichloroethylene
carcinogenesis. In M. M. C Maltoni (Ed.), Archives of Research on Industrial
Carcinogenesis (Vol. 5, pp. .). Princeton, NJ: Princeton Scientific Publishing.
Mandel, J. H.; Kelsh, M. A.; Mink, P. J.; Alexander, D. D.; Kalmes, R. M.; Weingart, M., . . .
Goodman, M. (2006). Occupational trichloroethylene exposure and non-Hodgkin's
lymphoma: a meta-analysis and review. Occup Environ Med, 63, 597-607.
http://dx.doi.Org/10.l 136/oem.2005.022418
Mandel, J. S. (2001). Renal cell cancer correlated with occupational exposure to
trichloroethylene. J Cancer Res Clin Oncol, 127, 265-268.
Mandel, J. S.; McLaughlin, J. K.; Schlehofer, B.; Mellemgaard, A.; Helmert, U.; Lindblad, P., . .
. Adami, H.-O. (1995). International renal-cell cancer study. IV. Occupation. Int J
Cancer, 61, 601-605. http://dx.doi.org/10.1002/ijc.2910610503
Manson, J.; Murphy, M.; Richdale, N.; Smith, M. (1984). Effects of oral exposure to
trichloroethylene on female reproductive function. Toxicology, 32, 229-242.
http://www.ncbi.nlm.nih.gov/pubmed/6474485
Marano, D.; Boice, J.; Fryzek, J.; Morrison, J.; Sadler, C.; McLaughlin, J. (2000). Exposure
assessment for a large epidemiological study of aircraft manufacturing workers. Appl
Occup Environ Hyg, 15, 644-656. http://dx.doi.org/10.1080/10473220050075653
Markwald, R. R.; Runyan, R. B.; Kitten, G. T.; Funderburg, F. M.; Bernanke, D. H.; Brayer, P.
R. (1984). Use of collagen gel cultures to study heart development: preteoglycan and
glycoprotein interactions during formation of endocardial cushion tissue. In R. L.
Trelstad (Ed.), The role of extracellular matrix in development (pp. 323-350): Alan R.
Liss.
Markwald, R.; Eisenberg, C.; Eisenberg, L.; Trusk, T.; Sugi, Y. (1996). Epithelial-mesenchymal
transformations in early avian heart development. Acta Anat, 156, 173-186.
http://www.ncbi.nlm.nih.gov/pubmed/9124035
Maronpot, R. R. (2007). Biological basis of differential susceptibility to hepatocarcinogenesis
among mouse strains. Paper presented at the 3rd Workshop of the Standing Committee
on Risk Analysis Issues and Reviews: Mouse liver tumors: Benefits and constraints on
75
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
use in human health risk assessment, qualititive and quantitative aspects, Washington,
DC.
Maronpot, R. R.; Flake, G.; Huff, J. (2004). Relevance of animal carcinogenesis findings to
human cancer predictions and prevention. Toxicol Pathol, 32, 40-48.
Maronpot, R. R.; Fox, T.; Malarkey, D. E.; Goldsworthy, T. L. (1995). Mutations in the ras
proto-oncogene: clues to etiology and molecular pathogenesis of mouse liver tumors.
Toxicology, 101, 125-156. http://dx.doi.org/10.1016/0300-483X(95)03112-S
Maronpot, R.; Anna, C.; Devereux, T.; Lucier, G.; Butterworth, B.; Anderson, M. (1995).
Considerations concerning the murine hepatocarcinogenicity of selected chlorinated
hydrocarbons. Prog Clin Biol Res, 391, 305-323.
http://www.ncbi.nlm.nih.gov/pubmed/8532724
Marrazzini, A.; Betti, C.; Bernacchi, F.; Barrai, I.; Barale, R. (1994). Micronucleus test and
metaphase analyses in mice exposed to known and suspected spindle poisons.
Mutagenesis, 9, 505-515. http://www.ncbi.nlm.nih.gov/pubmed/7854141
Marshall, E.; Gensburg, L.; Deres, D.; Geary, N.; Cayo, M. (1997). Maternal residential
exposure to hazardous wastes and risk of central nervous system and musculoskeletal
birth defects. Arch Environ Health, 52, 416-425.
http://www.ncbi.nlm.nih.gov/pubmed/9541362
Marsman, D. S.; Cattley, R. C.; Conway, J. G.; Popp, J. A. (1988). Relationship of hepatic
peroxisome proliferation and replicative DNA synthesis to the hepatocarcinogenicity of
the peroxisome proliferators di(2-ethylhexyl)phthalate and [4-chloro-6-(2,3-xylidino)-2-
pyrimidinylthio]acetic acid (Wy-14,643) in rats. Cancer Res, 48, 6739-6744.
http://www.ncbi.nlm.nih.gov/pubmed/3180084
Marsman, D. S.; Goldsworthy, T. L.; Popp, J. A. (1992). Contrasting hepatocytic peroxisome
proliferation, lipofuscin accumulation and cell turnover for the hepatocarcinogens Wy-
14,643 and clofibric acid. Carcinogenesis, 13, 1011-1017.
http://dx.doi.Org/10.1093/carcin/13.6.1011
Martinez-Larranaga, M. R.; Anadon, A.; Diaz, M. J.; Fernandez, R.; Sevil, B.; Fernandez-Cruz,
M. L., . . . Anton, R. (1996). Induction of cytochrome P4501A1 and P4504A1 activities
and peroxisomal proliferation by fumonisin Bl. Toxicol Appl Pharmacol, 141, 185-194.
http://dx.doi.org/10.1006/taap.1996.0275
Mather, G.; Exon, J.; Koller, L. (1990). Subchronic 90 day toxicity of dichloroacetic and
trichloroacetic acid in rats. Toxicology, 64, 71-80.
http://www.ncbi.nlm.nih.gov/pubmed/2219134
Mattioli, S.; Truffelli, D.; Baldasseroni, A.; Risi, A.; Marchesini, B.; Giacomini, C., . . . Buiatti,
E. (2002). Occupational risk factors for renal cell cancer: a case—control study in
northern Italy. J Occup Environ Med, 44, 1028-1036.
http://www.ncbi.nlm.nih.gov/pubmed/12448354
Mazzullo, M.; Bartoli, S.; Bonora, B.; Colacci, A.; Lattanzi, G.; Niero, A., . . . Grilli, S. (1992).
In vivo and in vitro interaction of trichloroethylene with macromolecules from various
organs of rat and mouse. Res Commun Mol Pathol Pharmacol, 76, 192-208.
http://www.ncbi.nlm.nih.gov/pubmed/1376481
McCarthy, T. B. and Jones, R. D. (1983). Industrial gassing poisonings due to trichl or ethylene,
perchlorethylene, and 1,1,1-trichloroethane, 1961-80. Occup Environ Med, 40, 450-455.
http://www.ncbi.nlm.nih.gov/pubmed/6626474
76
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
McCarver, D. G. and Hines, R. N. (2002). The ontogeny of human drug-metabolizing enzymes:
Phase II conjugation enzymes and regulatory mechanisms. J Pharmacol Exp Ther, 300,
361-366. http://dx.doi.Org/10.1124/jpet.300.2.361
McCarver, D. G.; Byun, R.; Hines, R. N.; Hichme, M.; Wegenek, W. (1998). A genetic
polymorphism in the regulatory sequences of human CYP2E1: Association with
increased chlorzoxazone hydroxylation in the presence of obesity and ethanol intake.
Toxicol Appl Pharmacol, 152, 276-281. http://dx.doi.org/10.1006/taap.1998.8532
McConnell, G.; Ferguson, D. M.; Pearson, C. R. (1975). Chlorinated hydrocarbons and the
environment. Endeavour, 34, 13-18. http://www.ncbi.nlm.nih.gov/pubmed/54249
McCredie, M. and Stewart, J. H. (1993). Risk factors for kidney cancer in New South Wales. IV.
Occupation. Br JInd Med, 50, 349-354. http://www.ncbi.nlm.nih.gov/pubmed/8494775
McCunney, R. J. (1988). Diverse manifestations of trichloroethylene. Br J Ind Med, 45, 122-126.
http://dx.doi.Org/10.1136/oem.45.2.122
McDaniel, T.; Martin, P.; Ross, N.; Brown, S.; Lesage, S.; Pauli, B. (2004). Effects of
chlorinated solvents on four species of North American amphibians. Arch Environ
Contam Toxicol, 47, 101-109. http://www.ncbi.nlm.nih.gov/pubmed/15346783
McEntee, G. P.; Batts, K. A.; Katzmann, J. A.; Ilstrup, D. M.; Nagorney, D. M. (1992).
Relationship of nuclear DNA content to clinical and pathologic findings in patients with
primary hepatic malignancy. Surgery, 111, 376-379.
http://www.ncbi.nlm.nih.gov/pubmed/1313603
McGregor, D.; Reynolds, D.; Zeiger, E. (1989). Conditions affecting the mutagenicity of
trichloroethylene in Salmonella. Environ Mol Mutagen, 13, 197-202.
http://www.ncbi.nlm.nih.gov/pubmed/2651116
Mcguire, V.; Nelson, L. M.; Koepsell, T. D.; Checkoway, H.; Longstreth, W. T.; Jr. (1998).
Assessment of occupational exposures in community-based case-control studies. Annu
Rev Public Health, 19, 35-53.
McKinney, P. A.; Alexander, F. E.; Cartwright, R. A.; Parker, L. (1991). Parental occupations of
children with leukaemia in west Cumbria, north Humberside, and Gateshead. Br Med J,
302, 681-687. http://dx.doi.org/10.1136/bmj.302.6778.681
McLaughlin, J. and Blot, W. (1997). A critical review of epidemiology studies of
trichloroethylene and perchloroethylene and risk of renal-cell cancer. Int Arch Occup
Environ Health, 70, 222-231. http://www.ncbi.nlm.nih.gov/pubmed/9342621
McLaughlin, J. K.; Hrubec, Z.; Blot, W. J.; Fraumeni, J. F. (1995). Smoking and cancer
mortality among U.S. veterans: a 26-year follow-up. Int J Cancer, 60, 190-193.
http://www.ncbi.nlm.nih.gov/pubmed/7829214
McLaughlin, J. K.; Lipworth, L.; Tarone, R. E. (2006). Epidemiologic aspects of renal cell
carcinoma. Semin Oncol, 33, 527-533.
http://dx.doi.Org/10.1053/j.seminoncol.2006.06.010
McLean, A. J.; Cogger, V. C.; Chong, G. C.; Warren, A.; Markus, A. M.; Dahlstrom, J. E.; Le
Couteur, D. G. (2003). Age-related pseudocapillarization of the human liver. J Pathol,
200, 112-117. http://dx.doi.org/10.1002/path.1328
McMichael, A. J.; Spirtas, R.; Gamble, J. F.; Tousey, P. M. (1976). Mortality among rubber
workers: relationship to specific jobs. J Occup Environ Med, 18, 178-185.
McNutt, L. A. (2003). Estimating the relative risk in cohort studies and clinical trials of common
outcomes. Am J Epidemiol, 157, 940-943. http://dx.doi.org/10.1093/aje/kwg074
77
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
MDPH. (Massachusetts Department of Public Health), (1997). Woburn childhood leukemia
follow-up study: Information booklet. Boston, MA: Massachusetts Department of Public
Health, Bureau of Environmental Health Assessment.
MDPH. (Massachusetts Department of Public Health). (1997). Woburn childhood leukemia
follow-up study: Volume I: Analyses. Boston, MA: Massachusetts Department of Public
Health, Bureau of Environmental Health Assessment. Retrieved from
http://www.mass.gov/Eeohhs2/docs/dph/environmental/investigations/woburn_cancer_le
ukemia_follow_up_study_1997.pdf.
MDPH. (Massachusetts Department of Public Health). (1997). Woburn childhood leukemia
follow-up study: Volume II: Appendix. Boston, MA: Massachusetts Department of
Public Health, Bureau of Environmental Health Assessment.
Mehendale, H. (2000). PPAR-alpha: a key to the mechanism of hepatoprotection by clofibrate.
Toxicol Sci, 57, 187-190. http://www.ncbi.nlm.nih.gov/pubmed/11006348
Melchiorri, C.; Bolondi, L.; Chieco, P.; Pagnoni, M.; Gramantieri, L.; Barbara, L. (1994).
Diagnostic and prognostic value of DNA ploidy and cell nuclearity in ultrasound-guided
liver biopsies. Cancer, 74, 1713-1719. http://www.ncbi.nlm.nih.gov/pubmed/8082072
Melchiorri, C.; Chieco, P.; Zedda, A. I.; Coni, P.; Ledda-Columbano, G. M.; Columbano, A.
(1993). Ploidy and nuclearity of rat hepatocytes after compensatory regeneration or
mitogen-induced liver growth. Carcinogenesis, 14, 1825-1830.
Mellemgaard, A.; Engholm, G.; McLaughlin, J. K.; Olsen, J. H. (1994). Occupational risk factors
for renal-cell carcinoma in Denmark. Scand J Work Environ Health, 20, 160-165.
http://www.ncbi.nlm.nih.gov/pubmed/7973487
Melnick, R. L.; Huff, J.; Barrett, J. C.; Maronpot, R. R.; Lucier, G.; Portier, C. J. (1993). Cell
proliferation and chemical carcinogenesis: symposium overview. Environ Health
Perspect, 101 Suppl 5, 3-7. http://www.ncbi.nlm.nih.gov/pubmed/7912189
Melnick, R. L.; Kohn, M. C.; Portier, C. J. (1996). Implications for risk assessment of suggested
nongenotoxic mechanisms of chemical carcinogenesis. Environ Health Perspect, 104
Suppl 1, 123-134. http://www.ncbi.nlm.nih.gov/pubmed/8722116
Melnick, R. L.; Thayer, K. A.; Bucker, J. R. (2007). Conflicting views on chemical
carcinogenesis arising from the design and evaluation of rodent carcinogenicity studies.
Environ Health Perspect, 116, 130-135. http://dx.doi.org/10.1289/ehp.9989
Melnick, R.; Jameson, C.; Goehl, T.; Maronpot, R.; Collins, B.; Greenwell, A., . . . Agarwal, D.
(1987). Application of microencapsulation for toxicology studies. II. Toxicity of
microencapsulated trichloroethylene in Fischer 344 rats. Fundam Appl Toxicol, 8, 432-
442. http://www.ncbi.nlm.nih.gOv/pubmed/3111923
Menegazzi, M.; Carcereri-De Prati, A.; Suzuki, H.; Shinozuka, H.; Pibiri, M.; Piga, R., . . .
Ledda-Columbano, G. M. (1997). Liver cell proliferation induced by nafenopin and
cyproterone acetate is not associated with increases in activation of transcription factors
NF-kappaB and AP-1 or with expression of tumor necrosis factor alpha. Hepatology, 25,
585-592. http://dx.doi.org/10.1002/hep.510250316
Mensing, T.; Welge, P.; Voss, B.; Fels, L.; Fricke, H.; Briining, T.; Wilhelm, M. (2002). Renal
toxicity after chronic inhalation exposure of rats to trichloroethylene. Toxicol Lett, 128,
243-247. http://www.ncbi.nlm.nih.gOv/pubmed/l 1869834
Merdink, J. L.; Bull, R. J.; Schultz, I. R. (2000). Trapping and identification of the
dichloroacetate radical from the reductive dehalogenation of trichloroacetate by mouse
78
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
and rat liver microsomes. Free Radic Biol Med, 29, 125-130.
http://dx.doi.org/10.1016/S0891-5849(00)00330-0
Merdink, J. L.; Robison, L. M.; Stevens, D. K.; Hu, M.; Parker, J. C.; Bull, R. J. (2008). Kinetics
of chloral hydrate and its metabolites in male human volunteers. Toxicology, 245, ISO-
MO. http://dx.doi.Org/10.1016/j.tox.2007.12.018
Merdink, J.; Gonzalez-Leon, A.; Bull, R.; Schultz, I. (1998). The extent of dichloroacetate
formation from trichloroethylene, chloral hydrate, trichloroacetate, and trichloroethanol
inB6C3Fl mice. Toxicol Sci, 45, 33-41. http://dx.doi.org/10.1006/toxs.1998.2500
Merdink, J.; Stenner, R.; Stevens, D.; Parker, J.; Bull, R. (1999). Effect of enterohepatic
circulation on the pharmacokinetics of chloral hydrate and its metabolites in F344 rats. J
Toxicol Environ Health A, 57, 357-368. http://www.ncbi.nlm.nih.gov/pubmed/10405189
Mergler, D.; Huel, G.; Bowler, R.; Frenette, B.; Cone, J. (1991). Visual dysfunction among
former microelectronics assembly workers. Arch Environ Health, 46, 326-334.
http://www.ncbi.nlm.nih.gov/pubmed/1772256
Merrick, B. A.; Robinson, M.; Condie, L. W. (1989). Differing hepatotoxicity and lethality after
subacute trichloroethylene exposure in aqueous or corn oil gavage vehicles in B6C3F1
mice. J Appl Toxicol, 9, 15-21. http://www.ncbi.nlm.nih.gov/pubmed/2926094
Messing, K.; Punnett, L.; Bond, M.; Alexanderson, K.; Pyle, J.; Zahm, S., . . . de Grosbois, S.
(2003). Be the fairest of them all: challenges and recommendations for the treatment of
gender in occupational health research. Am J Ind Med, 43, 618-629.
http://dx.doi.org/10.1002/ajim.10225
Mester, B.; Meters, A.; Deeg, E.; Eisner, G.; Becker, N.; Seidler, A. (2006). Occupation and
malignant lymphoma: A population based case control study in Germany. Occup Environ
Med, 63, 17-26. http://dx.doi.org/10.1136/oem.2005.020453
Meyer, K.; Lee, J.-S.; Dyck, P. A.; Cao, W.-Q.; Rao, M. S.; Thorgeirsson, S. S.; Reddy, J. K.
(2003). Molecular profiling of hepatocellular carcinomas developing spontaneously in
acyl-CoA oxidase deficient mice: Comparison with liver tumors induced in wild-type
mice by a peroxisome proliferator and a genotoxic carcinogen. Carcinogenesis, 24, 975-
984. http://dx.doi.org/10.1093/carcin/bgg040
Mhiri, C.; Choyakh, F.; Ben, H. M.; Feki, I.; Ben, M. M.; Zouari, N. (2004). Trigeminal
somatosensory evoked potentials in trichloroethylene-exposed workers. 9, 102-107.
Michalik, L.; Desvergne, B.; Dreyer, C.; Gavillet, M.; Laurini, R.; Wahli, W. (2002). PPAR
expression and function during vertebrate development. Int J Dev Biol, 46, 105-114.
http://www.ncbi.nlm.nih.gov/pubmed/11902671
Michel, C.; Desdouets, C.; Slaoui, M.; Isaacs, K. R.; Roberts, R. A.; Boitier, E. (2007).
Diethylnitrosamine initiation does not alter clofibric acid-induced hepatocarcinogenesis
in the rat. Toxicol Sci, 99, 58-69. http://dx.doi.org/10.1093/toxsci/kfml68
Mikol, Y. B.; Hoover, K. L.; Creasia, D.; Poirier, L. A. (1983). Hepatocarcinogenesis in rats fed
methyl-deficient, amino acid-defined diets. Carcinogenesis, 4, 1619-1629.
http://www.ncbi.nlm.nih.gov/pubmed/6317218
Miligi, L.; Costantini, A. S.; Benvenuti, A.; Kriebel, D.; Bolejack, V.; Tumino, R., . . . Vineis, P.
(2006). Occupational exposure to solvents and the risk of lymphomas. Epidemiology, 17,
552-561. http://dx.doi.org/10.1097/01.ede.0000231279.30988.4d
Miller, B. M. and Adler, I.-D. (1992). Aneuploidy induction in mouse spermatocytes.
Mutagenesis, 7, 69-76.
79
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Miller, R. E. and Guengerich, F. P. (1982). Oxidation of trichloroethylene by liver microsomal
cytochrome P-450: evidence for chlorine migration in a transition state not involving
trichloroethylene oxide. Biochemistry, 21, 1090-1097.
http://dx.doi.org/10.1021/bi00534a041
Miller, R. E. and Guengerich, F. P. (1983). Metabolism of trichloroethylene in isolated
hepatocytes, microsomes, and reconstituted enzyme systems containing cytochrome P-
450. Cancer Res, 43, 1145-1152. http://www.ncbi.nlm.nih.gov/pubmed/6825087
Miller, R. T.; Shah, R. S.; Cattley, R. C.; Popp, J. A. (1996). The peroxisome proliferations WY-
14,643 and methylclofenapate induce hepatocyte ploidy alterations and ploidy-specific
DNA synthesis in F344 rats. Toxicol Appl Pharmacol, 138, 317-323.
http://www.ncbi.nlm.nih.gov/pubmed/8658533
Mills, N. C. and Means, A. R. (1972). Sorbitol dehydrogenase of rat testis: changes of activity
during development, after hypophysectomy and following gonadotrophic hormone
administration. Endocrinology, 91, 147-156.
http://www.ncbi.nlm.nih.gov/pubmed/4337211
Minamishima, Y. A.; Nakayama, K.; Nakayama, K.-I. (2002). Recovery of liver mass without
proliferation of hepatocytes after partial hepatectomy in Skp2-deficient mice. Cancer Res,
62, 995-999. http://www.ncbi.nlm.nih.gov/pubmed/11861371
Mirer, F. E. (2010). New evidence on the health hazards and control of metalworking fluids
since completion of the OSHA advisory committee report. Am J Ind Med, 53, 792-801.
http://dx.doi.org/10.1002/ajim.20853
Mirsalis, J. C.; Tyson, C. K.; Steinmetz, K. L.; Loh, E. K.; Hamilton, C. M.; Bakke, J. P.;
Spalding, J. W. (1989). Measurement of unscheduled DNA synthesis and S-phase
synthesis in rodent hepatocytes following in vivo treatment: Testing of 24 compounds.
Environ Mol Mutagen, 14, 155-164. http://dx.doi.org/10.1002/em.2850140305
Mishima, N.; Hoffman, S.; Hill, E. G.; Krug, E. L. (2006). Chick embryos exposed to
trichloroethylene in an ex ovo culture model show selective defects in early endocardial
cushion tissue formation. Birth Defects Res A Clin Mol Teratol, 76, 517-527.
http://dx.doi.org/10.1002/bdra.20283
Mitchell, A. B. S. and Parsons-Smith, B. G. (1969). Trichloroethylene neuropathy. Br Med J, 1,
422-423.
Miura, Y.; Anzai, K.; Takahashi, S.; TBTIFL, O. (1997). A novel lipophilic spin probe for the
measurement of radiation damage in mouse brain using in vivo electron spin resonance
(ESR). FEBS Lett, 419, 99-102. http://www.ncbi.nlm.nih.gov/pubmed/9426228
Mjaatvedt, C.; Krug, E.; Markwald, R. (1991). An antiserum (ESI) against a particulate form of
extracellular matrix blocks the transition of cardiac endothelium into mesenchyme in
culture. DevBiol, 145, 219-230. http://www.ncbi.nlm.nih.gov/pubmed/2040370
Mjaatvedt, C.; Lepera, R.; Markwald, R. (1987). Myocardial specificity for initiating endothelial-
mesenchymal cell transition in embryonic chick heart correlates with a particulate
distribution of fibronectin. Dev Biol, 119, 59-67.
http://www.ncbi.nlm.nih.gov/pubmed/3539667
Moghaddam, A. P.; Abbas, R.; Fisher, J. W.; Lipscomb, J. C. (1997). The role of mouse
intestinal microflora in the metabolism of trichloroethylene, an in vivo study. Hum Exp
Toxicol, 16, 629-635. http://dx.doi.org/10.1177/096032719701601101
80
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Moghaddam, A. P.; Abbas, R.; Fisher, J. W.; Stavrou, S.; Lipscomb, J. C. (1996). Formation of
dichloroacetic acid by rat and mouse gut microflora, an in vitro study. Biochem Biophys
Res Commun, 228, 639-645. http://dx.doi.org/10.1006/bbrc.1996.1709
Moinzadeh, P.; Breuhahn, K.; Stiitzer, H.; Schirmacher, P. (2005). Chromosome alterations in
human hepatocellular carcinomas correlate with aetiology and histological grade—results
of an explorative CGH meta-analysis. Br J Cancer, 92, 935-941.
http://dx.doi.org/10.1038/sj.bjc.6602448
Monster, A. C. (1979). Difference in uptake, elimination, and metabolism in exposure to
trichloroethylene, 1,1,1-trichloroethane, and tetrachloroethylene. Int Arch Occup Environ
Health, 42, 311-317. http://www.ncbi.nlm.nih.gov/pubmed/422272
Monster, A. C. and Houtkooper, J. M. (1979). Estimation of individual uptake of
trichloroethylene, 1,1,1-trichloroethane and tetrachloroethylene from biological
parameters. Int Arch Occup Environ Health, 42, 319-323.
http://www.ncbi.nlm.nih.gov/pubmed/422273
Monster, A. C.; Boersma, G.; Duba, W. C. (1976). Pharmacokinetics of trichloroethylene in
volunteers, influence of workload and exposure concentration. Int Arch Occup Environ
Health, 38, 87-102. http://www.ncbi.nlm.nih.gov/pubmed/1002309
Monster, A. C.; Boersma, G.; Duba, W. C. (1979). Kinetics of trichloroethylene in repeated
exposure of volunteers. Int Arch Occup Environ Health, 42, 283-292.
http://www.ncbi.nlm.nih.gov/pubmed/422269
Moore, L. E.; Boffetta, P.; Karami, S.; Brennan, P.; Stewart, P. S.; Hung, R., . . . Rothman, N.
(2010). Occupational trichloroethylene exposure and renal carcinoma risk: Evidence of
genetic susceptibility by reductive metabolism gene variants. Cancer Res, 70, 6527-6536.
http://dx.doi.org/10.1158/0008-5472.CAN-09-4167
Moore, L. E.; Wilson, R. T.; Campleman, S. L. (2005). Lifestyle factors, exposures, genetic
susceptibility, and renal cell cancer risk: A review. Cancer Invest, 23, 240-255.
http://dx.doi.org/10.1081/cnv-200055962
Moore, M. M. and Harrington-Brock, K. (2000). Mutagenicity of trichloroethylene and its
metabolites: Implications for the risk assessment of trichloroethylene. Environ Health
Perspect, 108, 215-223. http://www.ncbi.nlm.nih.gov/pubmed/10807553
Morata, T. C.; Fiorini, A. C.; Fischer, F. M.; Colacioppo, S.; Wallingford, K. M.; Krieg, E. F., . .
. Cesar, C. L. (1997). Toluene-induced hearing loss among rotogravure printing workers.
Scand J Work Environ Health, 23, 289-298.
http://www.ncbi.nlm.nih.gov/pubmed/9322820
Morata, T. C.; Johnson, A. C.; Nylen, P.; Svensson, E. B.; Cheng, J.; Krieg, E. F., . . . Franks, J.
(2002). Audiometric findings in workers exposed to low levels of styrene and noise. J
Occup Environ Med, 44, 806-814.
http://dx.doi.org/10.1097/01.jom.0000026645.83602.60
Morgan, J. and Cassady, R. (2002). Community cancer assessment in response to long-time
exposure to perchlorate and trichloroethylene in drinking water. J Occup Environ Med,
44, 616-621. http://www.ncbi.nlm.nih.gov/pubmed/12134524
Morgan, R. W.; Kelsh, M. A.; Zhao, K.; et al. (2000). Erratum. Epidemiology, 11, 360.
Morgan, R. W.; Kelsh, M. A.; Zhao, K.; Heringer, S. (1998). Mortality of aerospace workers
exposed to Trichloroethylene. Epidemiology, 9, 424-431.
http://www.ncbi.nlm.nih.gov/pubmed/9647907
81
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Morgan, R. W.; Kelsh, M. A.; Zhao, K.; Heringer, S. (2000). Mortality of aerospace workers
exposed to trichloroethylene. Epidemiology, 11, 360.
http://www.ncbi.nlm.nih.gov/pubmed/9647907
Morgenstern, H. (1998). Ecologic studies. In S. Greenland (Ed.), Modern epidemiology (pp. 459-
480): Lippincott -Raven.
Morgenstern, H.; Froines, J.; Ritz, B.; Young, B. (1997). Epidemiologic study to determine
possible adverse effects to Rocketdyne/Atomics International workers from exposure to
ionizing radiation. Berkeley, CA: Public Health Institute. Retrieved from
http://www.mapcruzin.com/rocketdyne-cancer/radiation.pdf.
Morgenstern, H.; Froines, J.; Ritz, B.; Young, B. (1999). Epidemiologic study to determine
possible adverse effects to Rocketdyne/Atomics International workers from exposure to
selected chemicals. Berkeley, CA: Public Health Institute. Retrieved from
http://www.ssflpanel.org/files/UCLA_chem.pdf.
Morimura, K.; Cheung, C.; Ward, J. M.; Reddy, J. K.; Gonzalez, F. J. (2006). Differential
susceptibility of mice humanized for peroxisome proliferator-activated receptor alpha to
Wy-14,643-induced liver turnorigenesis. Carcinogenesis, 27, 1074-1080.
http://dx.doi.org/10.1093/carcin/bgi329
Mori oka, I.; Miyai, N.; Yamamoto, H.; Miyashita, K. (2000). Evaluation of combined effect of
organic solvents and noise by the upper limit of hearing. Ind Health, 38, 252-257.
http://www.ncbi.nlm.nih.gov/pubmed/10812851
Moriya, M.; Ohta, T.; Watanabe, K.; Miyazawa, T.; Kato, K.; Shirasu, Y. (1983). Further
mutagenicity studies on pesticides in bacterial reversion assay systems. Mutat Res, 116,
185-216. http://www.ncbi.nlm.nih.gov/pubmed/6339892
Morrow, L.; Ryan, C.; Goldstein, G.; Hodgson, M. (1989). A distinct pattern of personality
disturbance following exposure to mixtures of organic solvents. J Occup Med, 31, 743-
746. http://www.ncbi.nlm.nih.gov/pubmed/2795250
Morrow, L.; Steinhauer, S.; Hodgson, M. (1992). Delay in P300 latency in patients with organic
solvent exposure. Arch Neurol, 49, 315-320.
http://www.ncbi.nlm.nih.gov/pubmed/1536637
Mortelmans, K.; Haworth, S.; Lawlor, T.; Speck, W.; Tainer, B.; Zeiger, E. (1986). Salmonella
mutagenicity tests: II. Results from the testing of 270 chemicals. Environ Mutagen, 8, 1-
119. http://dx.doi.org/10.1002/em.2860080702
Morton, L. D.; Youssef, A. F.; Lloyd, E.; Kiorpes, A. L.; Goldsworthy, T. L.; Fort, F. L. (2002).
Evaluation of carcinogenic responses in the Eker rat following short-term exposure to
selected nephrotoxins and carcinogens. Toxicol Pathol, 30, 559-564.
http://www.ncbi.nlm.nih.gov/pubmed/12371664
Morton, L. M.; Hartge, P.; Holford, T. R.; Holly, E. A.; Chiu, B. C.; Vineis, P., . . . Zheng, T.
(2005). Cigarette smoking and risk of non-Hodgkin lymphoma: A pooled analysis from
the International Lymphoma Epidemiology Consortium (interlymph). Cancer Epidemiol
Biomarkers Prev, 14, 925-933. http://dx.doi.org/10.1158/1055-9965.EPI-04-0693
Morton, L. M.; Holford, T. R.; Leaderer, B.; Zhang, Y.; Zahm, S. H.; Boyle, P., . . . Zheng, T.
(2003). Alcohol use and risk of non-Hodgkin's lymphoma among Connecticut women
(United States). Cancer Causes Control, 14, 687-694.
http://www.ncbi.nlm.nih.gov/pubmed/14575367
82
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Morton, L. M.; Turner, J. J.; Cerhan, J. R.; Linet, M. S.; Treseler, P. A.; Clarke, C. A., . . .
Weisenburger, D. D. (2007). Proposed classification of lymphoid neoplasms for
epidemiologic research from the Pathology Working Group of the International
Lymphoma Epidemiology Consortium (InterLymph). Blood, 110, 695-708.
http://dx.doi.org/10.1182/blood-2006-ll-051672
Moser, V. C.; Cheek, B. M.; MacPhail, R. C. (1995). A multidisciplinary approach to
toxicological screening III Neurobehavioral toxicity. J Toxicol Environ Health, 45, 173-
210. http://www.ncbi.nlm.nih.gov/pubmed/7783252
Moser, V. C.; MacPhail, R. C.; Gennings, C. (2003). Neurobehavi oral evaluations of mixtures of
trichloroethylene, heptachlor, and di(2-ethylhexyl)phthlate in a full-factorial design.
Toxicology, 188, 125-137. http://www.ncbi.nlm.nih.gov/pubmed/12767685
Moser, V. C.; Phillips, P. M.; McDaniel, K. L.; MacPhail, R. C. (1999). Behavioral evaluation of
the neurotoxicity produced by dichloroacetic acid in rats. Neurotoxicol Teratol, 21,719-
731. http://www.ncbi.nlm.nih.gov/pubmed/10560779
Moslen, M. T.; Reynolds, E. S.; Boor, P. J.; Bailey, K.; Szabo, S. (1977). Trichloroethylene-
induced deactivation of cytochrome P-450 and loss of liver glutathione in vivo. Res
Commun Mol Pathol Pharmacol, 16, 109-120.
http://www.ncbi.nlm.nih.gov/pubmed/841173
Motohashi, Y. and Miyazaki, Y. (1990). Time-dependent effects of trichloroethylene on motor
activity in rats. Chronobiol Int, 7, 193-201.
http://www.ncbi.nlm.nih.gov/pubmed/2268881
Mueller, W. F.; Coulston, F.; Korte, F. (1982). Comparative metabolism of 14C-
trichloroethylene in chimpanzees, baboons, and rhesus monkeys. Chemosphere, 11, 215-
218.
Muijser, H.; Lammers, J.; Kullig, B. (2000). Effects of exposure to trichloroethylene and noise
on hearing in rats. Noise Health, 2, 57-66.
http://www.ncbi.nlm.nih.gov/pubmed/12689479
Muller, G.; Spassovski, M.; Henschler, D. (1972). Trichloroethylene exposure and
trichloroethylene metabolites in urine and blood. Arch Toxicol, 29, 335-340.
Muller, G.; Spassovski, M.; Henschler, D. (1974). Metabolism of trichloroethylene in man II
Pharmacokinetics of metabolites. Arch Toxicol, 32, 283-295.
http://www.ncbi.nlm.nih.gov/pubmed/4480028
Muller, G.; Spassowski, M.; Henschler, D. (1975). Metabolism of trichloroethylene in man III
Interaction of trichloroethylene and ethanol. Arch Toxicol, 33, 173-189.
http://www.ncbi.nlm.nih.gov/pubmed/1173750
Murphy, J. P. (1990). Exposure to wells G and H in Woburn, Massachusetts. Boston, MA:
Massachusetts Department of Public Health. Retrieved from
http://www.mass.gov/Eeohhs2/docs/dph/environmental/investigations/woburn_exposure_
wells.pdf.
Nagaya, T.; Ishikawa, N.; Hata, H. (1989). Sister-chromatid exchanges in lymphocytes of
workers exposed to trichloroethylene. Mutat Res, 222, 279-282.
Nagaya, T.; Ishikawa, N.; Hata, H. (1989). Urinary total protein and "beta"-2-microglobulin in
workers exposed to trichloroethylene. Environ Res, 50, 86-92.
http://www.ncbi.nlm.nih.gov/pubmed/2676510
83
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Nagaya, T.; Ishikawa, N.; Hata, H. (1990). No change in serum dopamine-"beta"-hydroxylase
activity in workers exposed to trichloroethylene. Toxicol Lett, 54, 221-227.
http://www.ncbi.nlm.nih.gov/pubmed/2260120
Nagaya, T.; Ishikawa, N.; Hata, H.; Otobe, T. (1993). Subclinical and reversible hepatic effects
of occupational exposure to trichloroethylene. Int Arch Occup Environ Health, 64, 561 -
563. http://dx.doi.org/10.1007/BF00517701
Nakai, J. S.; Stathopulos, P. B.; Campbell, G. L.; Chu, I.; Li-Muller, A.; Aucoin, R. (1999).
Penetration of chloroform, trichloroethylene, and tetrachloroethylene through human
skin. J Toxicol Environ Health A, 58, 157-170.
http://dx.doi.org/10.1080/009841099157368
Nakajima, T. and Aoyama, T. (2000). Polymorphism of drug-metabolizing enzymes in relation
to individual susceptibility to industrial chemicals. Ind Health, 38, 143-152.
http://www.ncbi.nlm.nih.gov/pubmed/10812837
Nakajima, T.; Kamijo, Y.; Usuda, N.; Liang, Y.; Fukushima, Y.; Kametani, K., . . . Aoyama, T.
(2000). Sex-dependent regulation of hepatic peroxisome proliferation in mice by
trichloroethylene via peroxisome proliferator-activated receptor alpha (PPARalpha).
Carcinogenesis, 21, 677-682. http://www.ncbi.nlm.nih.gov/pubmed/10753203
Nakajima, T.; Koyama, Y.; Sato, A. (1982). Dietary modification of metabolism and toxicity of
chemical substances - with special reference to carbohydrate. Biochem Pharmacol, 31,
1005-1011. http://www.ncbi.nlm.nih.gov/pubmed/7082357
Nakajima, T.; Moriguchi, M.; Mitsumoto, Y.; Sekoguchi, S.; Nishikawa, T.; Takashima, H., . . .
Kagawa, K. (2004). Centrosome aberration accompanied with p53 mutation can induce
genetic instability in hepatocellular carcinoma. Mod Pathol, 17, 722-727.
http://dx.doi.org/10.1038/modpathol.3800115
Nakajima, T.; Okino, T.; Okuyama, S.; Kaneko, T.; Yonekura, I.; Sato, A. (1988). Ethanol-
induced enhancement of trichloroethylene metabolism and hepatotoxicity: difference
from the effect of phenobarbital. Toxicol Appl Pharmacol, 94, 227-237.
http://www.ncbi.nlm.nih.gov/pubmed/3388420
Nakajima, T.; Wang, R. S.; Katakura, Y.; Kishi, R.; Elovaara, E.; Park, S. S., . . . Vainio, H.
(1992). Sex-, age- and pregnancy-induced changes in the metabolism of toluene and
trichloroethylene in rat liver in relation to the regulation of cytochrome P450IIE1 and
P450IIC11 content. J Pharmacol Exp Ther, 261, 869-874.
http://www.ncbi.nlm.nih.gov/pubmed/1602390
Nakajima, T.; Wang, R.-S.; Elovaara, E.; Park, S. S.; Gelboin, H. V.; Vainio, H. (1992). A
comparative study on the contribution of cytochrome P450 isozymes to metabolism of
benzene, toluene and trichloroethylene in rat liver. Biochem Pharmacol, 43, 251-257.
http://www.ncbi.nlm.nih.gov/pubmed/1739413
Nakajima, T.; Wang, R.-S.; Elovaara, E.; Park, S. S.; Gelboin, H. V.; Vainio, H. (1993).
Cytochrome P450-related differences between rats and mice in the metabolism of
benzene, toluene and trichloroethylene in liver microsomes. Biochem Pharmacol, 45,
1079-1085. http://dx.doi.org/10.1016/0006-2952(93)90252-R
Nakajima, T.; Wang, R.-S.; Murayama, N.; Sato, A. (1990). Three forms of trichloroethylene-
metabolizing enzymes in rat liver induced by ethanol, phenobarbital, and 3-
methylcholanthrene. Toxicol Appl Pharmacol, 102, 546-552.
http://www.ncbi.nlm.nih.gov/pubmed/2315921
84
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Nakajima, T.; Yamanoshita, O.; Kamijima, M.; Kishi, R.; Ichihara, G. (2003). Generalized skin
reactions in relation to trichloroethylene exposure: a review from the viewpoint of drug-
metabolizing enzymes. J Occup Health, 45, 8-14.
http://www.ncbi.nlm.nih.gov/pubmed/14605423
Nakajima, Y.; Mironov, V.; Yamagishi, T.; Nakamura, H.; Markwald, R. (1997). Expression of
smooth muscle alpha-actin in mesenchymal cells during formation of avian endocardial
cushion tissue: a role for transforming growth factor beta3. Dev Dyn, 209, 296-309.
http://dx.doi.org/10.1002/(SICI)1097-0177(199707)209:3<296:: AID-
AJA5>3.0.CO;2-D
Nakanishi, S.; Shiohara, E.; Tsukada, M.; Yamazaki, Y.; Okumura, K. (1978). Acetaldehyde
level in the blood and liver aldehyde dehydrogenase activities in trichloroethylene-treated
rats. Arch Toxicol, 41, 207-214. http://www.ncbi.nlm.nih.gov/pubmed/736791
Narotsky, M. G. and Kavlock, R. J. (1995). A multidisciplinary approach to toxicological
screening: II. Developmental toxicity. J Toxicol Environ Health, 45, 145-171.
http://dx.doi.org/10.1080/15287399509531987
Narotsky, M. G.; Weller, E. A.; Chinchilli, V. M.; Kavlock, R. J. (1995). Nonadditive
developmental toxicity in mixtures of trichloroethylene, Di(2-ethylhexyl) phthalate, and
heptachlor in a 5 x 5 x 5 design. Fundam Appl Toxicol, 27, 203-216.
http://dx.doi.Org/10.1093/toxsci/27.2.203
Natarajan, A.; Duivenvoorden, W.; Meijers, M.; Zwanenburg, T. (1993). Induction of mitotic
aneuploidy using Chinese hamster primary embryonic cells. Test results of 10 chemicals.
MutatRes, 287, 47-56. http://www.ncbi.nlm.nih.gov/pubmed/7683384
Naugler, W. E.; Sakurai, T.; Kim, S.; Maeda, S.; Kim, K.; Elsharkawy, A. M.; Karin, M. (2007).
Gender disparity in liver cancer due to sex differences in MyD88-dependent IL-6
production. Science, 317, 121-124. http://dx.doi.org/10.1126/science.1140485
NCI. (National Institutes of Health, National Cancer Institute). (1976). Carcinogenesis bioassay
of trichloroethylene. (NCI-CG-TR-2). Bethesda, MD: National Cancer Institute.
Retrieved from http://ntp.niehs.nih.gov/ntp/htdocs/LT_rpts/tr002.pdf.
NCI. (National Institutes of Health, National Cancer Institute). (2008). SEER cancer statistics
review, 1975-2005. Bethesda, MD. Retrieved from http://seer.cancer.gov/csr/1975_2005.
Neghab, M. and Stacey, N. (1997). Toluene-induced elevation of serum bile acids: relationship
to bile acid transport. J Toxicol Environ Health, 52, 249-268.
http://www.ncbi.nlm.nih.gov/pubmed/9316646
Neghab, M.; Qu, S.; Bai, C. L.; Caples, J.; Stacey, N. H. (1997). Raised concentration of serum
bile acids following occupational exposure to halogenated solvents, 1,1,2-trichloro-1,2,2-
trifluoroethane and trichloroethylene. Int Arch Occup Environ Health, 70, 187-194.
http://dx.doi.org/10.1007/s004200050205
Nelson, G. M.; Swank, A. E.; Brooks, L. R.; Bailey, K. C.; George, S. E. (2001). Metabolism,
microflora effects, and genotoxicity in haloacetic acid-treated cultures of rat cecal
microbiota. Toxicol Sci, 60, 232-241. http://www.ncbi.nlm.nih.gov/pubmed/11248134
Nelson, H. H.; Wiencke, J. K.; Christiani, D. C.; Cheng, T. J.; Zuo, Z.-F.; Schwartz, B. S., . . .
Kelsey, K. T. (1995). Ethnic differences in the prevalence of the homozygous deleted
genotype of glutathione S-transferase theta. Carcinogenesis, 16, 1243-1245.
http://www.ncbi.nlm.nih.gov/pubmed/7767992
85
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Nelson, J. and Zenick, H. (1986). The effect of trichloroethylene on male sexual behavior:
possible opioid role. Neurobehav Toxicol Teratol, 8, 441-445.
http://www.ncbi.nlm.nih.gov/pubmed/3785506
Nelson, L.; Longstreth, W.; Koepsell, T.; Checkoway, H.; van Belle, G. (1994). Completeness
and accuracy of interview data from proxy respondents: demographic, medical, and life-
style factors. Epidemiology, 5, 204-217. http://www.ncbi.nlm.nih.gov/pubmed/8172996
Nelson, M. A. and Bull, R. J. (1988). Induction of strand breaks in DNA by trichloroethylene
and metabolites in rat and mouse liver in vivo. Toxicol Appl Pharmacol, 94, 45-54.
http://dx.doi.org/10.1016/0041-008X(88)90335-3
Nelson, M. A.; Lansing, A. J.; Sanchez, I. M.; Bull, R. J.; Springer, D. L. (1989). Dichloroacetic
acid and trichloroacetic acid-induced DNA strand breaks are independent of peroxisome
proliferation. Toxicology, 58, 239-248. http://dx.doi.org/10.1016/0300-483X(89)90139-
X
Nelson, M. A.; Sanchez, I. M.; Bull, R. J.; Sylvester, S. R. (1990). Increased expression of c-myc
and c-H-ras in dichloroacetate and trichloroacetate-induced liver tumors in B6C3F1 mice.
Toxicology, 64, 47-57.
Nesslany, F. and Marzin, D. (1999). A micromethod for the in vitro micronucleus assay.
Mutagenesis, 14, 403-410. http://www.ncbi.nlm.nih.gov/pubmed/10390508
Nestmann, E. R.; Chu, I.; Kowbel, D. J.; Matula, T. I. (1980). Short-lived mutagen in Salmonella
produced by reaction of trichloroacetic acid and dimethyl sulphoxide. Can J Genet Cytol,
22, 35-40. http://dx.doi.org/10.1139/g80-006
Ni, Y. C.; Kadlubar, F. F.; Fu, P. P. (1995). Formation of malondialdehyde-modified 2'-
deoxyguanosinyl adduct from metabolism of chloral hydrate by mouse liver microsomes.
Biochem Biophys Res Commun, 216, 1110-1117.
http://dx.doi.org/10.1006/bbrc.1995.2735
Ni, Y. C.; Wong, T. Y.; Lloyd, R. V.; Heinze, T. M.; Shelton, S.; Casciano, D., . . . Fu, P. P.
(1996). Mouse liver microsomal metabolism of chloral hydrate, trichloroacetic acid, and
trichloroethanol leading to induction of lipid peroxidation via a free radical mechanism.
Drug Metab Dispos, 24, 81-90. http://www.ncbi.nlm.nih.gov/pubmed/8825194
Ni, Y.; Wong, T.; Kadlubar, F.; Fu, P. (1994). Hepatic metabolism of chloral hydrate to free
radical(s) and induction of lipid peroxidation. Biochem Biophys Res Commun, 204, 937-
943. http://dx.doi.org/10.1006/bbrc.1994.2550
Nickerson, M. L.; Jaeger, E.; Shi, Y.; Durocher, J. A.; Mahurkar, S.; Zaridze, D., . . . Moore, L.
E. (2008). Improved identification of von Hippel-Lindau gene alterations in clear cell
renal tumors. Clin Cancer Res, 14, 4726-4734. http://dx.doi.org/10.1158/1078-
0432.CCR-07-4921
Nickoloff, B. J.; Ben-Neriah, Y.; Pikarsky, E. (2005). Inflammation and cancer: is the link as
simple as we think. J Invest Dermatol, 124, x-xiv. http://dx.doi.0rg/lO.llll/j.0022-
202X.2005.23724.x
Niederlehner, B.; Cairns, J.; Smith, E. (1998). Modeling acute and chronic toxicity of nonpolar
narcotic chemicals and mixtures to Ceriodaphnia dubia. Ecotoxicol Environ Saf, 39, 136-
146. http://dx.doi.org/10.1006/eesa.1997.1621
NIEHS. (National Institutes of Health, National Institute for Environmental Health Sciences).
(1998). Tumor incidence in control animals by route and vehicle of administration,
B6C3F1 mice. Research Triangle Park, NC.
86
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Nietert, P. J.; Sutherland, S. E.; Silver, R. M.; Pandey, J. P.; Knapp, R. G.; Hoel, D. G.;
Dosemeci, M. (1998). Is occupational organic solvent exposure a risk factor for
scleroderma? Arthritis Rheum, 41, 1111-1118. http://dx.doi.org/10.1002/1529-
0131(199806)41:6<1111 ::AID-ART19>3.0.CO;2-J
Niklasson, M.; Tham, R.; Larsby, B.; Eriksson, B. (1993). Effects of toluene, styrene,
trichloroethylene, and trichloroethane on the vestibulo- and opto-oculo motor system in
rats. Neurotoxicol Teratol, 15, 327-334. http://www.ncbi.nlm.nih.gov/pubmed/8277926
NIOSH. (National Institute for Occupational Safety and Health). (1992). Health Hazard
Evaluation Report HETA 86-469-2189, James River Corporation, Newnan, Georgia.
(HETA 86-469-2189). Springfield, VA.
NIOSH. (National Institute for Occupational Safety and Health). (1998). Criteria for a
recommended standard: Occupational exposure to metalworking fluids. (98-102).
Atlanta, GA: Author. Retrieved from http://www.cdc.gov/niosh/98-102.html.
Nissen, S. E.; Nicholls, S. J.; Wolski, K.; Howey, D. C.; McErlean, E.; Wang, M. D., . . . Russo,
J. M. (2007). Effects of a potent and selective PPAR-alpha agonist in patients with
atherogenic dyslipidemia or hypercholesterolemia: two randomized controlled trials.
JAMA, 297, 1362-1373. http://dx.doi.org/10.1001/jama.297.12.1362
Nitcheva, D. K.; Piegorsch, W. W.; West, R. W. (2007). On use of the multistage dose-response
model for assessing laboratory animal carcinogenicity. Regul Toxicol Pharmacol, 48,
135-147. http://dx.doi.Org/10.1016/j.yrtph.2007.03.002
Nohammer, C.; Brunner, F.; Wolkart, G.; Staber, P.; Steyrer, E.; Gonzalez, F., . . . Hoefler, G.
(2003). Myocardial dysfunction and male mortality in peroxisome proliferator-activated
receptor alpha knockout mice overexpressing lipoprotein lipase in muscle. Lab Invest, 83,
259-269. http://www.ncbi.nlm.nih.gov/pubmed/12594240
Noland-Gerbec, E. A.; Pfohl, R. J.; Taylor, D. H.; Bull, R. J. (1986). 2-Deoxyglucose uptake in
the developing rat brain upon pre- and postnatal exposure to trichloroethylene.
Neurotoxicology, 7, 157-164. http://www.ncbi.nlm.nih.gov/pubmed/3822256
Nomiyama, H. and Nomiyama, K. (1979). Pathway and rate of metabolism of trichloroethylene
in rats and rabbits. Ind Health, 17, 29-37.
Nomiyama, K. and Nomiyama, H. (1971). Metabolism of trichloroethylene in human: sex
difference in urinary excretion of trichloroacetic acid and trichloroethanol. Int Arch
Occup Environ Health, 28, 37-48. http://www.ncbi.nlm.nih.gov/pubmed/5088555
Nordlund, L. A.; Carstensen, J. M.; Pershagen, G. (1997). Cancer incidence in female smokers: a
26-year follow-up. Int J Cancer, 73, 625-628.
http://www.ncbi.nlm.nih.gov/pubmed/9398036
Nordlund, L. A.; Carstensen, J. M.; Pershagen, G. (1999). Are male and female smokers at equal
risk of smoking-related cancer: evidence from a Swedish prospective study. Scand J
Public Health, 27, 56-62. http://www.ncbi.nlm.nih.gov/pubmed/10847673
Nordstrom, M.; Hardell, L.; Magnuson, A.; Hagberg, H.; Rask-Andersen, A. (1998).
Occupational exposures, animal exposure and smoking as risk factors for hairy cell
leukaemia evaluated in a case-control study. Br J Cancer, 77, 2048-2052.
http://www.ncbi.nlm.nih.gov/pubmed/9667691
NRC. (National Research Council). (1983). Risk assessment in the federal government:
Managing the process. Washington, DC: National Academy Press.
87
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
NRC. (National Research Council). (1988). Health risks of radon and other internally deposited
alpha-emitters: BEIRIV. (PB88-225594). Washington, DC: National Research Council;
National Academy Press. Retrieved from
http://www.nap.edu/openbook.php?isbn=0309037972.
NRC. (National Research Council). (1993). Pesticides in the diets of infants and children.
Washington, DC: National Academy Press.
NRC. (National Research Council). (1994). Science and judgment in risk assessment.
Washington, DC: National Academy Press.
NRC. (National Research Council). (1997). Environmental epidemiology, volume 2: Use of the
gray literature and other data in environmental epidemiology (Vol. 2). Washington, DC:
National Academies Press.
NRC. (National Research Council). (2006). Assessing the human health risks of
trichloroethylene: Key scientific issues. Washington, DC: The National Academies
Press. Retrieved from http://nae.edu/nae/naepcms.nsf/weblinks/MKEZ-
6SSHPD?OpenDocument.
NRC. (National Research Council). (2009). Contaminated water supplies at Camp Lejeune:
Assessing potential health effects. Washington, DC: National Academies Press.
NTP. (National Toxicology Program). (1988). Toxicology and carcinogenesis studies of
trichloroethylene (CAS No. 79-01-6) in four strains of rats (ACI, August, Marshall,
Osborne-Mendel)(gavage studies). Research Triangle Park, NC: Public Health Service,
U.S. Department of Health and Human Services. Retrieved from
http://ntp.niehs.nih.gov/ntp/htdocs/LT_rpts/tr273.pdf.
NTP. (National Toxicology Program). (1990). Carcinogenesis studies of trichloroethylene
(without epichlorohydrin) (CAS no 79-01-6) in F344/N rats and B6C3F mice (gavage
studies). Research Triangle Park, NC: Author.
Nunes, J.; Ehrich, M.; Robertson, J. (2001). Toxicosis associated with dual oral exposure of rats
to lead and trichloroethylene. Toxicol Pathol, 29, 451-457.
http://www.ncbi.nlm.nih.gov/pubmed/11560250
Nutley, E.; Tcheong, A.; Allen, J.; Collins, B.; Ma, M.; Lowe, X., . . . Wyrobek, A. (1996).
Micronuclei induced in round spermatids of mice after stem-cell treatment with chloral
hydrate: evaluations with centromeric DNA probes and kinetochore antibodies. Environ
Mol Mutagen, 28, 80-89. http://dx.doi.org/10.1002/(SICI)1098-
2280(1996)28:2<80::AID-EM3>3.0.CO;2-I
NYSDEC. (New York State Department of Environmental Conservation). (2007). Project
information summary: Solvents, contamination and cleanup at Endicott. Albany, NY.
Retrieved from http://www.dec.ny.gov/chemical/24887.html.
Nysdoh. (2005). Trichloroethene air criteria document. Draft report. Retrieved from
file://L:\Lab\NCEA\TCE Documents\TCE PDFs4\ NYSDH 2005 Draft Trichloroethene
Air Criteria.pdf.
NYSDOH. (New York State Department of Health). (2006). Trichloroethene air criteria
document: Final report. Troy, NY.
O'Donnell, L.; McLachlan, R. I.; Wreford, N. G.; Robertson, D. M. (1994). Testosterone
promotes the conversion of round spermatids between stages VII and VIII of the rat
spermatogenic cycle. Endocrinology, 135, 2608-2614.
http://www.ncbi.nlm.nih.gov/pubmed/7988449
88
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Odum, J.; Foster, J. R.; Green, T. (1992). A mechanism for the development of Clara cell lesions
in the mouse lung after exposure to trichloroethylene. Chem Biol Interact, 83, 135-153.
http://www.ncbi.nlm.nih.gov/pubmed/1505057
Ogata, M.; Takatsuka, Y.; Tomokuni, K. (1971). Excretion of organic chlorine compounds in the
urine of persons exposed to vapours of trichloroethylene and tetrachloroethylene. Occup
Environ Med, 28, 386-391. http://dx.doi.Org/10.1136/oem.28.4.386
Ohta, M.; Saito, T.; Saito, K.; Kurasaki, M.; Hosokawa, T. (2001). Effect of trichloroethylene on
spatiotemporal pattern of LTP in mouse hippocampal slices. Int J Neurosci, 111, 257-
271. http://www.ncbi.nlm.nih.gov/pubmed/11912680
Okino, T.; Nakajima, T.; Nakano, M. (1991). Morphological and biochemical analyses of
trichloroethylene hepatotoxicity: differences in ethanol- and phenobarbital-pretreated
rats. Toxicol Appl Pharmacol, 108, 379-389.
http://www.ncbi.nlm.nih.gov/pubmed/2020965
Olivier, P.; Plancke, M. O.; Marzin, D.; Clavey, V.; Sauzieres, J.; Fruchart, J. C. (1988). Effects
of fenofibrate, gemfibrozil and nicotinic acid on plasma lipoprotein levels in normal and
hyperlipidemic mice. A proposed model for drug screening. Atherosclerosis, 70, 107-
114. http://www.ncbi.nlm.nih.gov/pubmed/3162679
Olshan, A. F.; De Roos, A. J.; Teschke, K.; Neglia, J. P.; Stram, D. O.; Pollock, B. H.;
Castleberry, R. P. (1999). Neuroblastoma and parental occupation. Cancer Causes
Control, 10, 539-549. http://www.ncbi.nlm.nih.gov/pubmed/10616823
Olsson, A.; Skogh, T.; Axelson, O.; Wingren, G. (2004). Occupations and exposures in the work
environment as determinants for rheumatoid arthritis. Occup Environ Med, 61, 233-238.
http://www.ncbi.nlm.nih.gov/pubmed/14985518
Olsson, A.; Skogh, T.; Wingren, G. (2000). Occupational determinants for rheumatoid arthritis.
Scand J Work Environ Health, 26, 243-249.
http://www.ncbi.nlm.nih.gov/pubmed/10901117
Oneta, C.; Lieber, C.; Li, J.; Ruttimann, S.; Schmid, B.; Lattmann, J., . . . Seitz, H. (2002).
Dynamics of cytochrome P4502E1 activity in man: induction by ethanol and
disappearance during withdrawal phase. J Hepatol, 36, 47-52.
http://www.ncbi.nlm.nih.gov/pubmed/11804663
Ono, Y.; Somiya, I.; Kawamura, M. (1991). The evaluation of genotoxicity using DNA repairing
test for chemicals produced in chlorination and ozonation processes. Water Sci Technol,
23, 329-338.
Opdam, J. J. (1989). Intra and interindividual variability in the kinetics of a poorly and highly
metabolising solvent. Br J Ind Med, 46, 831-845.
http://www.ncbi.nlm.nih.gov/pubmed/2611156
Opdam, J. J. (1989). Respiratory input in inhalation experiments. Occup Environ Med, 46, 145-
156. http://dx.doi.Org/10.l 136/oem.46.3.145
Orsi, L.; Monnereau, A.; Dananche, B.; Berthou, C.; Fenaux, P.; Marit, G., . . . Clavel, J. (2010).
Occupational exposure to organic solvents and lymphoid neoplasms in men: Results of a
French case-control study. Occup Environ Med, 67, 664-672.
http://dx.doi.Org/10.l 136/oem.2009.049460
Oshiro, W.; Krantz, Q.; Bushnell, P. (2004). A search for residual behavioral effects of
trichloroethylene (TCE) in rats exposed as young adults. Neurotoxicol Teratol, 26, 239-
251. http://dx.doi.Org/10.1016/j.ntt.2003.12.005
89
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
Ou, J.; Ou, Z.; McCarver, D.; Hines, R.; Oldham, K.; Ackerman, A.; Pritchard, K. (2003).
Trichloroethylene decreases heat shock protein 90 interactions with endothelial nitric
oxide synthase: implications for endothelial cell proliferation. Toxicol Sci, 73, 90-97.
http://dx.doi. org/10.1093/toxsci/kfg062
Owens, A. H. and Marshall, E. K. (1955). Further studies on the metabolic rate of chloral hydrate
and trichloroethanol. Johns Hopkins Med J, 97, 320-326.
Palmer, C. N.; Hsu, M. H.; Griffin, K. J.; Raucy, J. L.; Johnson, E. F. (1998). Peroxisome
proliferator activated receptor-alpha expression in human liver. Mol Pharmacol, 53, 14-
22. http://www.ncbi.nlm.nih.gov/pubmed/9443928
Pannett, B.; Coggon, D.; Acheson, E. (1985). A job-exposure matrix for use in population based
studies in England and Wales. Br J Ind Med, 42, 777-783.
http://www.ncbi.nlm.nih.gov/pubmed/4063222
Parent, M. E.; Hua, Y.; Siemiatycki, J. (2000). Occupational risk factors for renal cell carcinoma
in Montreal. Am J Ind Med, 38, 609-618.
http://www.ncbi.nlm.nih.gov/pubmed/11071683
Parent, M. E.; Siemiatycki, J.; Fritschi, L. (2000). Workplace exposures and oesophageal cancer.
Occup Environ Med, 57, 325-334. http://www.ncbi.nlm.nih.gov/pubmed/10769298
Paris, France: Commission of the European Communities.
Park, P. A.; Gilmour, M. I.; Selgrade, M. K. (1993). Pulmonary defenses to streptococcal
infection following acute exposure to trichloroethylene (TCE) or chloroform (CHC13).
Toxicologist, 13, 107.
Park, S.; Osterloh, J.; Vamvakas, S.; Hashmi, M.; Anders, M.; Cashman, J. (1992). Flavin-
containing monooxygenase-dependent stereoselective S-oxygenation and cytotoxicity of
cysteine S-conjugates and mercapturates. Chem Res Toxicol, 5, 193-201.
http://www.ncbi.nlm.nih.gov/pubmed/1643249
Parker, M. J.; Tarlow, M. J.; Milne Anderson, J. (1984). Glue sniffing and cerebral infarction.
Arch Dis Child, 59, 675-677. http://www.ncbi.nlm.nih.gov/pubmed/6465942
Parkin, D. M.; Whelan, S. L.; Ferlay, J.; Teppo, L.; Thomas, D. B. (Eds.). (2002). Cancer
incidence in five continents: Volume VIII. Lyon, France: International Agency for
Research on Cancer.
Parkinson, A.; Mudra, D. R.; Johnson, C.; Dwyer, A.; Carroll, K. M. (2004). The effects of
gender, age, ethnicity, and liver cirrhosis on cytochrome P450 enzyme activity in human
liver microsomes and inducibility in cultured human hepatocytes. Toxicol Appl
Pharmacol, 199, 193-209. http://dx.doi.Org/10.1016/j.taap.2004.01.010
Parnell, M.; Koller, L.; Exon, J.; Arnzen, J. (1986). Trichloroacetic acid effects on rat liver
peroxisomes and enzyme-altered foci. Environ Health Perspect, 69, 73-79.
http://www.ncbi.nlm.nih.gov/pubmed/2880713
Parrish, J. M.; Austin, E. W.; Stevens, D. K.; Kinder, D. H.; Bull, R. J. (1996). Haloacetate-
induced oxidative damage to DNA in the liver of male B6C3F1 mice. Toxicology, 110,
103-111. http://dx.doi.org/10.1016/0300-483X(96)03342-2
Parry, E. M.; Hague, A.; Parry, J. M. (1990). A study of mitotic division fidelity and numerical
chromosome changes in ageing Syrian hamster dermal cells. Mutat Res, 237, 83-93.
http://www.ncbi.nlm.nih.gov/pubmed/2366802
90
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Parry, J. M.; Parry, E. M.; Bourner, R.; Doherty, A.; Ellard, S.; O'Donovan, J., . . . Hess, P.
(1996). The detection and evaluation of aneugenic chemicals. Mutat Res, 353, 11-46.
http://dx.doi. org/10.1016/0027-5107(95)00242-1
Partanen, T.; Heikkila, P.; Hernberg, S.; Kauppinen, T.; Moneta, G.; Ojajarvi, A. (1991). Renal
cell cancer and occupational exposure to chemical agents. Scand J Work Environ Health,
17, 231-239. http://www.ncbi.nlm.nih.gov/pubmed/1925434
Pastino, G. M.; Yap, W. Y.; Carroquino, M. (2000). Human variability and susceptibility to
trichloroethylene. Environ Health Perspect, 108 Suppl 2, 201-214.
http://www.ncbi.nlm.nih.gov/pubmed/10807552
Paykoc, Z. V. and Powell, J. F. (1945). The excretion of sodium trichloroacetate. J Pharmacol
Exp Ther, 85, 289-293.
Peden-Adams, M. M.; Eudaly, J. G.; Lee, A. M.; Miller, J.; Keil, D. E.; Gilkeson, G. S. (2008).
Lifetime exposure to trichloroethylene (TCE) does not accelerate autoimmune disease in
MRL +/+ mice. J Environ Sci Health A Tox Hazard Subst Environ Eng, 43, 1402-1409.
http://www.ncbi.nlm.nih.gov/pubmed/18780217
Peden-Adams, M.; Eudaly, J.; Heesemann, L.; Smythe, J.; Miller, J.; Gilkeson, G.; Keil, D.
(2006). Developmental immunotoxicity of trichloroethylene (TCE): studies in B6C3F1
mice. J Environ Sci Health A Tox Hazard Subst Environ Eng, 41, 249-271.
http://dx.doi.org/10.1080/10934520500455289
Pelekis, M.; Gephart, L. A.; Lerman, S. E. (2001). Physiological-model-based derivation of the
adult and child pharmacokinetic intraspecies uncertainty factors for volatile organic
compounds. Regul Toxicol Pharmacol, 33, 12-20.
http://dx.doi.org/10.1006/rtph.2000.1436
Pellizzari, E. D.; Hartwell, T. D.; 3rd, H. B.; Waddell, R. D.; Whitaker, D. A.; Erickson, M. D.
(1982). Purgeable organic compounds in mother's milk. Bull Environ Contam Toxicol,
28, 322-328. http://www.ncbi.nlm.nih.gov/pubmed/7082873
Perbellini, L.; Olivato, D.; Zedde, A.; Miglioranzi, R. (1991). Acute trichloroethylene poisoning
by ingestion: clinical and pharmacokinetic aspects. Intensive Care Med, 17, 234-235.
http://www.ncbi.nlm.nih.gov/pubmed/1744309
Percy, C.; 3rd, S. E.; Gloeckler, L. (1981). Accuracy of cancer death certificates and its effect on
cancer mortality statistics. Am J Public Health, 71, 242-250.
http://www.ncbi.nlm.nih.gov/pubmed/7468855
Percy, C.; Ries, L. G.; Van Holten, V. D. (1990). The accuracy of liver cancer as the underlying
cause of death on death certificates. Public Health Rep, 105, 361-367.
http://www.ncbi.nlm.nih.gov/pubmed/2116637
Pereira, M. A. (1996). Carcinogenic activity of dichloroacetic acid and trichloroacetic acid in the
liver of female B6C3F1 mice. Fundam Appl Toxicol, 31, 192-199.
http://dx.doi.org/10.1006/faat.1996.0091
Pereira, M. A. and Phelps, J. B. (1996). Promotion by dichloroacetic acid and trichloroacetic acid
of N-methyl-N-nitrosourea-initiated cancer in the liver of female B6C3F1 mice. Cancer
Lett, 102, 133-141. http://dx.doi.org/10.1016/0304-3835(96)04156-0
Pereira, M. A.; Kramer, P. M.; Conran, P. B.; Tao, L. (2001). Effect of chloroform on
dichloroacetic acid and trichloroacetic acid-induced hypomethylation and expression of
the c-myc gene and on their promotion of liver and kidney tumors in mice.
Carcinogenesis, 22, 1511-1519. http://www.ncbi.nlm.nih.gov/pubmed/11532874
91
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Pereira, M. A.; Li, K.; Kramer, P. M. (1997). Promotion by mixtures of dichloroacetic acid and
trichloroacetic acid of N-methyl-N-nitrosourea-initiated cancer in the liver of female
B6C3F1 mice. Cancer Lett, 115, 15-23. http://dx.doi.org/10.1016/S0304-3835(97)04699-
5
Pereira, M. A.; Wang, W.; Kramer, P. M.; Tao, L. (2004). Prevention by methionine of
dichloroacetic acid-induced liver cancer and DNA hypomethylation in mice. Toxicol Sci,
77, 243-248. http://dx.doi.org/10.1093/toxsci/kfh031
Pereira, M.; Wang, W.; Kramer, P.; Tao, L. (2004). DNA hypomethylation induced by non-
genotoxic carcinogens in mouse and rat colon. Cancer Lett, 212, 145-151.
http://dx.doi.Org/10.1016/j.canlet.2004.03.024
Perkins, N. D. (2004). NF-kappaB: tumor promoter or suppressor. Trends Cell Biol, 14, 64-69.
http://dx.doi.Org/10.1016/j.tcb.2003.12.004
Perkins, N. D. and Gilmore, T. D. (2006). Good cop, bad cop: the different faces of NF-kappaB.
Cell Death Differ, 13, 759-772. http://dx.doi.org/10.1038/sj.cdd.4401838
Perocco, P. and Prodi, G. (1981). DNA damage by haloalkanes in human lymphocytes cultured
in vitro. Cancer Lett, 13, 213-218. http://dx.doi.org/10.1016/0304-3835(81)90020-3
Perrot, N.; Nalpas, B.; Yang, C. S.; Beaune, P. H. (1989). Modulation of cytochrome P450
isozymes in human liver, by ethanol and drug intake. Eur J Clin Invest, 19, 549-555.
http://www.ncbi.nlm.nih.gov/pubmed/2515975
Perry, S.; Schofield, M.; MacFarlane, M.; Lock, E.; King, L.; Gibson, G.; Goldfarb, P. (1993).
Isolation and expression of a cDNA coding for rat kidney cytosolic cysteine conjugate
beta-lyase. Mol Pharmacol, 43, 660-665. http://www.ncbi.nlm.nih.gov/pubmed/8502223
Persson, B. and Fredrikson, M. (1999). Some risk factors for non-Hodgkin's lymphoma. Int J
Occup Med Environ Health, 12, 135-142.
http://www.ncbi.nlm.nih.gov/pubmed/10465904
Persson, B.; Dahlander, A. M.; Fredriksson, M.; Brage, H. N.; Ohlson, C. G.; Axelson, O.
(1989). Malignant lymphomas and occupational exposures. Br J Ind Med, 46, 516-520.
http://www.ncbi.nlm.nih.gov/pubmed/2775671
Persson, B.; Fredriksson, M.; Olsen, K.; Boeryd, B.; Axelson, O. (1993). Some occupational
exposures as risk factors for malignant lymphomas. Cancer, 72, 1773-1778.
http://www.ncbi.nlm.nih.gov/pubmed/8348507
Pesch, B.; Haerting, J.; Ranft, U.; Klimpel, A.; Oelschlagel, B.; Schill, W. (2000). Occupational
risk factors for urothelial carcinoma: agent-specific results from a case-control study in
Germany. MURC Study Group. Multicenter Urothelial and Renal Cancer. Int J
Epidemiol, 29, 238-247.
Pesch, B.; Haerting, J.; Ranft, U.; Klimpel, A.; Oelschlagel, B.; Schill, W. (2000). Occupational
risk factors for renal cell carcinoma: Agent-specific results from a case-control study in
Germany. Int J Epidemiol, 29, 1014-1024. http://dx.doi.Org/10.1093/ije/29.6.1014
Peter, R.; Bocker, R.; Beaune, P.; Iwasaki, M.; Guengerich, F.; Yang, C. (1990). Hydroxylation
of chlorzoxazone as a specific probe for human liver cytochrome P-450IIE1. Chem Res
Toxicol, 3, 566-573. http://www.ncbi.nlm.nih.gov/pubmed/2103328
Peters, J. M.; Cattley, R. C.; Gonzalez, F. J. (1997). Role of PPAR alpha in the mechanism of
action of the nongenotoxic carcinogen and peroxisome proliferator Wy-14,643.
Carcinogenesis, 18, 2029-2033.
92
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
Peters, J. M.; Preston-Martin, S.; Yu, M. C. (1981). Brain tumors of children and occupational
exposures of parents. Science, 213, 235-237.
Peters, J.; Garabrant, D.; Wright, W.; Bernstein, L.; Mack, T. (1985). Uses of a cancer registry in
the assessment of occupational cancer risks. 69, 157-161.
http://www.ncbi.nlm.nih.gov/pubmed/3834325
Pfaffenberger, C. D.; Peoples, A. J.; Enos, H. F. (1980). Distribution of volatile halogenated
organic compounds between rat blood serum and adipose tissue. Int J Environ Anal
Chem, 8, 55-65. http://www.ncbi.nlm.nih.gov/pubmed/7399799
Phillips, T. and Macdonald, R. (1971). Comparative effect of pethidine, trichloroethylene, and
Entonox on fetal and neonatal acid-base and P02. Br Med J, 3, 558-560.
http://www.ncbi.nlm.nih.gov/pubmed/5571451
Piegorsch, W. W. and Bailer, A. J. (1997). Statistics for environmental biology and toxicology.
London, England: Chapman & Hall.
Pikarsky, E.; Porat, R. M.; Stein, I.; Abramovitch, R.; Amit, S.; Kasem, S., . . . Ben-Neriah, Y.
(2004). NF-kappaB functions as a tumour promoter in inflammation-associated cancer.
Nature, 431, 461-466. http://dx.doi.org/10.1038/nature02924
Plewa, M. J.; Kargalioglu, Y.; Vankerk, D.; Minear, R. A.; Wagner, E. D. (2002). Mammalian
cell cytotoxicity and genotoxicity analysis of drinking water disinfection by-products.
Environ Mol Mutagen, 40, 134-142. http://dx.doi.org/10.1002/em.10092
Poet, T.; Corley, R.; Thrall, K.; Edwards, J.; Tanojo, H.; Weitz, K., . . . Wester, R. (2000).
Assessment of the percutaneous absorption of trichloroethylene in rats and humans using
MS/MS real-time breath analysis and physiologically based pharmacokinetic modeling.
Toxicol Sci, 56, 61-72. http://www.ncbi.nlm.nih.gov/pubmed/10869454
Pogribny, I. P.; Tryndyak, V. P.; Woods, C. G.; Witt, S. E.; Rusyn, I. (2007). Epigenetic effects
of the continuous exposure to peroxisome proliferator WY-14,643 in mouse liver are
dependent upon peroxisome proliferator activated receptor alpha. Mutat Res-Fundam
Mol Mech Mutagen, 625, 62-71. http://dx.doi.Org/10.1016/j.mrfmmm.2007.05.004
Pohl, H. R.; Roney, N.; Wilbur, S.; Hansen, H.; De Rosa, C. T. (2003). Six interaction profiles
for simple mixtures. Chemosphere, 53, 183-197.
Pollack, G. M.; Li, R. C.; Ermer, J. C.; Shen, D. D. (1985). Effects of route of administration and
repetitive dosing on the disposition kinetics of di(2-ethylhexyl) phthalate and its mono-
de-esterified metabolite in rats. Toxicol Appl Pharmacol, 79, 246-256.
http://dx.doi. org/10.1016/0041-008X(85)90346-1
Popper, H. (1986). Aging and the liver. In H. Popper & F. Schaffner (Eds.), Progress in liver
diseases: Volume VIII (pp. 659-683). Orlando, FL: Grune and Stratton.
Porru, S.; Placidi, D.; Carta, A.; Gelatti, U.; Ribero, M. L.; Tagger, A., . . . Donato, F. (2001).
Primary liver cancer and occupation in men: a case-control study in a high-incidence area
in Northern Italy. Int J Cancer, 94, 878-883.
http://www.ncbi.nlm.nih.gov/pubmed/11745492
Porter, R. (1993). Follow-up of childhood leukemia incidence rate in Maricopa County
standardized incidence ratio of area F to area R for 1987.
Portier, C. J. and Bailer, A. J. (1989). Testing for increased carcinogenicity using a survival-
adjusted quantal response test. Toxicol Sci, 12, 731-737.
93
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Portier, C. J.; Hedges, J. C.; Hoel, D. G. (1986). Age-specific models of mortality and tumor
onset for historical control animals in the National Toxicology Program's carcinogenicity
experiments. Cancer Res, 46, 4372-4378.
Povey, A.; Guppy, M.; Wood, M.; Knight, C.; Black, C.; Silman, A. (2001). Cytochrome P2
polymorphisms and susceptibility to scleroderma following exposure to organic solvents.
Arthritis Rheum, 44, 662-665. http://dx.doi.org/10.1002/1529-
0131(200103)44:3<662::AID-ANR115>3.0.CO;2-P
Powell, J. F. (1945). Trichlorethylene: absorption, elimination and metabolism. Occup Environ
Med, 2, 142-145.
Prasanna, H. R.; Hart, R. W.; Magee, P. N. (1989). Effect of food restriction on the metabolism
of dimethylnitrosamine (NDMA) in rats. Anticancer Res, 9, 93-96.
http://www.ncbi.nlm.nih.gov/pubmed/2705761
Pravecek, T. L.; Channel, S. R.; Schmidt, W. J.; Kidney, J. K. (1996). Cytotoxicity and
metabolism of dichloroacetic acid and trichloroacetic acid in B6C3F1 mouse liver tissue.
9, 261-269.
Prendergast, J. A.; Jones, R. A.; Jenkins, L. J. J.; Siegel, J. (1967). Effects on experimental
animals of long-term inhalation of trichloroethylene, carbon tetrachloride, 1,1,1-
trichloroethane, dichlorodifluoromethane, and 1,1-dichloroethylene. Toxicol Appl
Pharmacol, 10, 270-289. http://www.ncbi.nlm.nih.gov/pubmed/4962252
Price, P. J.; Hassett, C. M.; Mansfield, J. I. (1978). Transforming activities of trichloroethylene
and proposed industrial alternatives. In Vitro Cell Dev Biol Anim, 14, 290-293.
Price, R. G.; Taylor, S. A.; Chivers, I.; Arce-Tomas, M.; Crutcher, E.; Franchini, I., . . .
Eisenberger, U. (1996). Development and validation of new screening tests for
nephrotoxic effects. Hum Exp Toxicol, 1, S10-S19.
Price, R. G.; Wedeen, R.; Lichtveld, M. Y.; Lybarger, J. A.; Porter, G. A.; Fels, L., . . . Finn, W.
F. (1999). Urinary biomarkers: Roles in risk assessment to environmental and
occupational nephrotoxins: Monitoring of effects and evaluation of mechanisms of
toxicity. Ren Fail, 21, xiii-xviii.
Prout, M. S.; Provan, W. M.; Green, T. (1985). Species differences in response to
trichloroethylene: I pharmacokinetics in rats and mice. Toxicol Appl Pharmacol, 79, 389-
400. http://www.ncbi.nlm.nih.gov/pubmed/4035686
Pryor, G. T.; Dickinson, J.; Howd, R. A.; Rebert, C. S. (1983). Transient cognitive deficits and
high-frequency hearing loss in weanling rats exposed to toluene. Neurotoxicol Teratol, 5,
53-57.
Pryor, G. T.; Rebert, C. S.; Howd, R. A. (1987). Hearing loss in rats caused by inhalation of
mixed xylenes and styrene. J Appl Toxicol, 7, 55-61.
Purdue, M. P.; Severson, R. K.; Colt, J. S.; Stewart, P.; De Roos, A. J.; Cerhan, J. R., . . . Blair,
A. (2009). Degreasing and risk of non-Hodgkin lymphoma. Occup Environ Med, 66,
557-560. http://dx.doi.org/10.1136/oem.2008.040386
Purdue, M.; Bakke, B.; Stewart, P.; De Roos, A.; Schenk, M.; Lynch, C., . . . Colt, J. (2011). A
case-control study of occupational exposure to trichloroethylene and non-Hodgkin
lymphoma. Environ Health Perspect, 119, 232-238.
http://dx.doi.org/10.1289/ehp.1002106
Raaschou-Nielsen, O.; Hansen, J.; Christensen, J.; Blot, W.; McLaughlin, J.; Olsen, J. (2001).
Urinary concentrations of trichloroacetic acid in Danish workers exposed to
94
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
trichloroethylene, 1947-1985. Am J Ind Med, 39, 320-327.
http://www.ncbi.nlm.nih.gov/pubmed/11241565
Raaschou-Nielsen, O.; Hansen, J.; McLaughlin, J.; Kolstad, H.; Christensen, J.; Tarone, R.;
Olsen, J. (2003). Cancer risk among workers at Danish companies using
trichloroethylene: A cohort study. Am J Epidemiol, 158, 1182-1192.
http://www.ncbi.nlm.nih.gov/pubmed/14652303
Raaschou-Nielsen, O.; Hansen, J.; Thomsen, B.; Johansen, I.; Lipworth, L.; McLaughlin, J.;
Olsen, J. (2002). Exposure of Danish workers to trichloroethylene, 1947-1989. Appl
Occup Environ Hyg, 17, 693-703. http://dx.doi.org/10.1080/10473220290096186
Radican, L.; Blair, A.; Stewart, P.; Wartenberg, D. (2008). Mortality of aircraft maintenance
workers exposed to trichloroethylene and other hydrocarbons and chemicals: extended
follow-up. J Occup Environ Med, 50, 1306-1319.
http://dx.doi.org/10.1097/JOM.0b013e3181845f7f
Radican, L.; Wartenberg, D.; Rhoads, G.; Schneider, D.; Wedeen, R.; Stewart, P.; Blair, A.
(2006). A retrospective occupational cohort study of end-stage renal disease in aircraft
workers exposed to trichloroethylene and other hydrocarbons. J Occup Environ Med, 48,
1-12. http://www.ncbi.nlm.nih.gov/pubmed/16404204
Raijmakers, M. T. M.; Steegers, E. A. P.; Peters, W. H. M. (2001). Glutathione S-transferases
and thiol concentrations in embryonic and early fetal tissues. Hum Reprod, 16, 2445-
2450. http://www.ncbi.nlm.nih.gOv/pubmed/l 1679536
Rake, J. P.; Visser, G.; Labrune, P.; Leonard, J. V.; Ullrich, K.; Smit, G. P. (2002). Glycogen
storage disease type I: diagnosis, management, clinical course and outcome. Results of
the European Study on Glycogen Storage Disease Type I (ESGSD I). Eur J Pediatr, 161
Suppl 1, S20-S34. http://dx.doi.org/10.1007/s00431-002-0999-4
Rakoff-Nahoum, S. and Medzhitov, R. (2007). Regulation of spontaneous intestinal
tumorigenesis through the adaptor protein MyD88. Science, 317, 124-127.
http://dx.doi.Org/10.l 126/science.l 140488
Rakyan, V. K.; Blewitt, M. E.; Druker, R.; Preis, J. I.; Whitelaw, E. (2002). Metastable epialleles
in mammals. Trends in Genetics, 18, 348-351.
http://www.ncbi.nlm.nih.gov/pubmed/12127774
Rakyan, V. K.; Chong, S.; Champ, M. E.; Cuthbert, P. C.; Morgan, H. D.; Luu, K. V.; Whitelaw,
E. (2003). Transgenerational inheritance of epigenetic states at the murine Axin(Fu)
allele occurs after maternal and paternal transmission. PNAS, 100, 2538-2543.
http://dx.doi.org/10.1073/pnas.0436776100
Ramanakumar, A. V.; Nadon, L.; Siemiatycki, J. (2008). Exposures in painting related
occupations and risk of selected cancers: results from a case-control study in Montreal.
Am J Ind Med, 51, 419-427. http://dx.doi.org/10.1002/ajim.20564
Ramdhan, D. H.; Kamijima, M.; Wang, D.; Ito, Y.; Naito, H.; Yanagiba, Y., . . . Nakajima, T.
(2010). Differential response to trichloroethylene-induced hepatosteatosis in wild-type
and PPARalpha-humanized mice. Environ Health Perspect, 118, 1557-1563.
http://dx.doi.org/10.1289/ehp.1001928
Ramdhan, D. H.; Kamijima, M.; Yamada, N.; Ito, Y.; Yanagiba, Y.; Nakamura, D., . . .
Nakajima, T. (2008). Molecular mechanism of trichloroethylene-induced hepatotoxicity
mediated by CYP2E1. Toxicol Appl Pharmacol, 231, 300-307.
http://dx.doi.Org/10.1016/j.taap.2008.04.020
95
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
Ramsdell, A. and Markwald, R. (1997). Induction of endocardial cushion tissue in the avian
heart is regulated, in part, by TGFbeta-3-mediated autocrine signaling. Dev Biol, 188, 64-
74. http://dx.doi.org/10.1006/dbio.1997.8637
Ramsey, J. C. and Andersen, M. E. (1984). A physiologically based description of the inhalation
pharmacokinetics of styrene in rats and humans. Toxicol Appl Pharmacol, 73, 159-175.
http://dx.doi. org/10.1016/0041 -008X(84)90064-4
Rao, M. S.; Tatematsu, M.; Subbarao, V.; Ito, N.; Reddy, J. K. (1986). Analysis of peroxisome
proliferator-induced preneoplastic and neoplastic lesions of rat liver for placental form of
glutathione S-transferase and gamma-glutamyltranspeptidase. Cancer Res, 46, 5287-
5290. http://www.ncbi.nlm.nih.gov/pubmed/2875789
Rapson, W. H.; Nazar, M. A.; Butsky, V. V. (1980). Mutagenicity produced by aqueous
chlorination of organic compounds. Bull Environ Contam Toxicol, 24, 590-596.
http://dx.doi.org/10.1007/BF01608160
Rasmussen, K. and Sabroe, S. (1986). Neuropsychological symptoms among metal workers
exposed to halogenated hydrocarbons. Scand J Public Health, 14, 161-168.
http://dx.doi.org/10.1177/140349488601400309
Rasmussen, K.; Arlien-S0borg, P.; Sabroe, S. (1993). Clinical neurological findings among metal
degreasers exposed to chlorinated solvents. Acta Neurol Scand, 87, 200-204.
http://www.ncbi.nlm.nih.gov/pubmed/8475690
Rasmussen, K.; Brogren, C.-H.; Sabroe, S. (1993). Subclinical affection of liver and kidney
function and solvent exposure. Int Arch Occup Environ Health, 64, 445-448.
http://dx.doi.org/10.1007/BF00517951
Rasmussen, K.; Jeppesen, H. J.; Sabroe, S. (1993). Solvent-induced chronic toxic
encephalopathy. Am J Ind Med, 23, 779-792. http://dx.doi.org/10.1002/ajim.4700230511
Rasmussen, K.; Jeppesen, H.; Sabroe, S. (1993). Psychometric tests for assessment of brain
function after solvent exposure. Am J Ind Med, 24, 553-565.
http://www.ncbi.nlm.nih.gov/pubmed/8266931
Rasmussen, K.; Sabroe, S.; Wohlert, M.; Ingerslev, H. J.; Kappel, B.; Nielsen, J. (1988). A
genotoxic study of metal workers exposed to trichloroethylene Sperm parameters and
chromosome aberrations in lymphocytes. Int Arch Occup Environ Health, 60, 419-423.
http://www.ncbi.nlm.nih.gov/pubmed/3410552
Raunio, H.; Hakkola, J.; Hukkanen, J.; Pelkonen, O.; Edwards, R.; Boobis, A.; Anttila, S. (1998).
Expression of xenobiotic metabolising cytochrome P450's in human pulmonary tissue.
Arch Toxicol Suppl, 20, 465-469.
Ray, W. A.; Griffin, M. R.; Schaffner, W. (1987). Psychotropic drug use and the risk of hip
fracture. 316/7 (363-369).
Reamon-Buettner, S. M. and Borlak, J. (2007). A new paradigm in toxicology and teratology:
altering gene activity in the absence of DNA sequence variation. Reprod Toxicol, 24, 20-
30. http://dx.doi.Org/10.1016/j.reprotox.2007.05.002
Rebert, C. S.; Boyes, W. K.; Pryor, G. T.; Svensgaard, D. J.; Kassay, K. M.; Gordon, G. R.;
Shinsky, N. (1993). Combined effects of solvents on the rat's auditory system: styrene
and trichloroethylene. Int J Psychophysiol, 14, 49-59.
http://www.ncbi.nlm.nih.gov/pubmed/8432679
96
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
Rebert, C. S.; Day, V. L.; Matteucci, M. J.; Pryor, G. T. (1991). Sensory-evoked potentials in
rats chronically exposed to trichloroethylene: predominant auditory dysfunction.
Neurotoxicol Teratol, 13, 83-90. http://www.ncbi.nlm.nih.gov/pubmed/2046630
Rebert, C. S.; Schwartz, R. W.; Svendsgaard, D. J.; Pryor, G. T.; Boyes, W. K. (1995).
Combined effects of paired solvents on the rat's auditory system. Toxicology, 105, 345-
354. http://www.ncbi.nlm.nih.gov/pubmed/8571371
Reif, J.; Burch, J.; Nuckols, J.; Metzger, L.; Ellington, D.; Anger, W. (2003). Neurobehavioral
effects of exposure to trichloroethylene through a municipal water supply. Environ Res,
93, 248-258. http://www.ncbi.nlm.nih.gov/pubmed/14615234
Rennix, C. P.; Quinn, M. M.; Amoroso, P. J.; Eisen, E. A.; Wegman, D. H. (2005). Risk of breast
cancer among enlisted Army women occupationally exposed to volatile organic
compounds. Am J Ind Med, 48, 157-167.
Reznik, G.; Ward, J. M.; Hardisty, J. F.; Russfield, A. (1979). Renal carcinogenic and
nephrotoxic effects of the flame retardant tris(2,3-dibromopropyl) phosphate in F344 rats
and (C57BL/6N X C3H/HeN)Fl mice. J Natl Cancer Inst, 63, 205-212.
http://www.ncbi.nlm.nih.gov/pubmed/286830
Rhee, I.; Bachman, K.; Park, B.; Jair, K.; Yen, R.; Schuebel, K., . . . Vogelstein, B. (2002).
DNMT1 and DNMT3b cooperate to silence genes in human cancer cells. Nature, 416,
552-556. http://dx.doi.org/10.1038/416552a
Rhomberg, L. R. (2000). Dose-response analyses of the carcinogenic effects of trichloroethylene
in experimental animals. Environ Health Perspect, 108, 343-358.
http://www.ncbi.nlm.nih.gov/pubmed/10807564
Rice, D. C. and Barone, S., Jr. (2000). Critical periods of vulnerability for the developing
nervous system: Evidence from humans and animal models. Environ Health Perspect,
108, 511-533. http://www.ncbi.nlm.nih.gov/pubmed/10852851
Richard, A. and Hunter, E. (1996). Quantitative structure-activity relationships for the
developmental toxicity of haloacetic acids in mammalian whole embryo culture.
Teratology, 53, 352-360. http://dx.doi.org/10.1002/(SICI)1096-
9926(199606)53:6<3 52: :AID-TERA6>3.0.CO;2-l
Richardson, D. B.; Terschiiren, C.; Hoffmann, W. (2008). Occupational risk factors for non-
Hodgkin's lymphoma: a population-based case-control study in Northern Germany. Am J
Ind Med, 51, 258-268. http://dx.doi.org/10.1002/ajim.20552
Richardson, F. C.; Boucheron, J. A.; Dyroff, M. C.; Popp, J. A.; Swenberg, J. A. (1986).
Biochemical and morphologic studies of heterogeneous lobe responses in
hepatocarcinogenesis. Carcinogenesis, 7, 247-251.
http://www.ncbi.nlm.nih.gov/pubmed/2868806
Richmond, R. E.; Carter, J. H.; Carter, H. W.; Daniel, F. B.; DeAngelo, A. B. (1995).
Immunohistochemical analysis of dichloroacetic acid (DCA)-induced
hepatocarcinogenesis in male Fischer (F344) rats. Cancer Lett, 92, 67-76.
http://dx.doi.org/10.1016/0304-3835(94)03756-9
Riederer, F.; Luborzewski, A.; God, R.; Bringmann, G.; Scholz, J.; Feineis, D.; Moser, A.
(2002). Modification of tyrosine hydroxylase activity by chloral derived beta-carbolines
in vitro. JNeurochem, 81, 814-819. http://www.ncbi.nlm.nih.gov/pubmed/12065640
97
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Ripp, S. L.; Itagaki, K.; Philpot, R. M.; Elfarra, A. A. (1999). Species and sex differences in
expression of flavin-containing monooxygenase form 3 in liver and kidney microsomes.
Drug Metab Dispos, 27, 46-52. http://www.ncbi.nlm.nih.gov/pubmed/9884308
Ripp, S. L.; Overby, L. H.; Philpot, R. M.; Elfarra, A. A. (1997). Oxidation of cysteine S-
conjugates by rabbit liver microsomes and cDNA-expressed flavin-containing mono-
oxygenases: studies with S-(l,2-dichlorovinyl)-L-cysteine, S-(l,2,2-trichlorovinyl)-L-
cysteine, S-allyl-L-cysteine, and S-benzyl-L-cysteine. Mol Pharmacol, 51, 507-515.
http://www.ncbi.nlm.nih.gov/pubmed/9058607
Ristoff, E. and Larsson, A. (2007). Inborn errors in the metabolism of glutathione. Orphanet J
RareDis, 2, 16. http://dx.doi.org/10.1186/1750-1172-2-16
Ritz, B. (1999). Cancer mortality among workers exposed to chemicals during uranium
processing. J Occup Environ Med, 41, 556-566.
http://www.ncbi.nlm.nih.gov/pubmed/10412097
Ritz, B. (1999). Radiation exposure and cancer mortality in uranium processing workers.
Epidemiology, 10, 531-538. http://www.ncbi.nlm.nih.gov/pubmed/10468427
Ritz, B.; Morgenstern, H.; Crawford-Brown, D.; Young, B. (2000). The effects of internal
radiation exposure on cancer mortality in nuclear workers at Rocketdyne/Atomics
International. Environ Health Perspect, 108, 743-751.
http://www.ncbi.nlm.nih.gov/pubmed/10964795
Ritz, B.; Morgenstern, H.; Froines, J.; Young, B. B. (1999). Effects of exposure to external
ionizing radiation on cancer mortality in nuclear workers monitored for radiation at
Rocketdyne/Atomics International. Am J Ind Med, 35, 21-31.
http://www.ncbi.nlm.nih.gov/pubmed/9884742
Robbiano, L.; Baroni, D.; Carrozzino, R.; Mereto, E.; Brambilla, G. (2004). DNA damage and
micronuclei induced in rat and human kidney cells by six chemicals carcinogenic to the
rat kidney. Toxicology, 204, 187-195. http://dx.doi.Org/10.1016/j.tox.2004.06.057
Robbiano, L.; Mereto, E.; Migliazzi Morando, A.; Pastore, P.; Brambilla, G. (1998). Increased
frequency of micronucleated kidney cells in rats exposed to halogenated anaesthetics.
MutatRes, 413, 1-6. http://www.ncbi.nlm.nih.gov/pubmed/9602852
Robins, J. (1987). A graphical approach to the identification and estimation of causal parameters
in mortality studies with sustained exposure periods. J Clin Epidemiol, 2, 139S-161S.
http://www.ncbi.nlm.nih.gov/pubmed/3667861
Rodenbeck, S.; Sanderson, L.; Rene, A. (2000). Maternal exposure to trichloroethylene in
drinking water and birth-weight outcomes. Arch Environ Health, 55, 188-194.
http://www.ncbi.nlm.nih.gov/pubmed/10908102
Rodriguez, C. E.; Mahle, D. A.; Gearhart, J. M.; Mattie, D. R.; Lipscomb, J. C.; Cook, R. S.;
Barton, H. A. (2007). Predicting age-appropriate pharmacokinetics of six volatile organic
compounds in the rat utilizing physiologically based pharmacokinetic modeling. Toxicol
Sci, 98, 43-56. http://dx.doi.org/10.1093/toxsci/kfm082
Rosenbaum, A. (2005). The HAPEM5 user's guide: Hazardous air pollutant exposure model,
Version 5. San Francisco, CA: ICF Consulting.
Roskams, T. A.; Theise, N. D.; Balabaud, C.; Bhagat, G.; Bhathal, P. S.; Bioulac-Sage, P., . . .
West, A. B. (2004). Nomenclature of the finer branches of the biliary tree: canals,
ductules, and ductular reactions in human livers. Hepatology, 39, 1739-1745.
http ://dx. doi. org /10.1002/hep .20130
98
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
Roskams, T.; Yang, S. Q.; Koteish, A.; Durnez, A.; DeVos, R.; Huang, X., . . . Diehl, A. M.
(2003). Oxidative stress and oval cell accumulation in mice and humans with alcoholic
and nonalcoholic fatty liver disease. Am J Pathol, 163, 1301-1311.
http://dx.doi.Org/l 0.1016/S0002-9440(10)63489-X
Rossi, A. M.; Migliore, L.; Barale, R.; Loprieno, N. (1983). In vivo and in vitro mutagenicity
studies of a possible carcinogen, trichloroethylene, and its two stabilizers,
epichlorohydrin and 1,2-epoxybutane. Birth Defects Res B Dev Reprod Toxicol, 3, 75-
87.
Rothman, K. J. (1986). Modern epidemiology. New York, NY: Little Brown & Co.
Rothman, K. J. and Greenland, S. (1998). Modern epidemiology (2nd ed.). Philadelphia, PA:
Lippincott-Raven Publishers.
Rothman, K. J.; Greenland, S.; Lash, T. L. (2008). Case-control studies, chapter 8 Modern
epidemiology (3rd ed., pp. 111-127). Philadelphia, PA: Lippincott Williams & Wilkins.
Ruckenstein, M. J. (2004). Autoimmune inner ear disease. Curr Opin Otolaryngol Head Neck
Surg, 12, 426-430.
Rufer, E. S.; Hacker, T. A.; Flentke, G. R.; Drake, V. J.; Brody, M. J.; Lough, J.; Smith, S. M.
(2010). Altered cardiac function and ventricular septal defect in avian embryos exposed
to low-dose trichloroethylene. Toxicol Sci, 113, 444-452.
http://dx.doi.org/10.1093/toxsci/kfp269
Rufer, E. S.; Hacker, T.; Lough, J.; Smith, S. M. (2008). Low-dose trichloroethylene exposure
during valvuloseptal morphogenesis causes ventricular septal defects in hatched chicks.
Toxicologist, 102, 314.
Ruijten, M. W. M. M.; Verberk, M. M.; Salle, H. J. A. (1991). Nerve function in workers with
long term exposure to trichloroethene. Occup Environ Med, 48, 87-92.
http://www.ncbi.nlm.nih.gov/pubmed/1998613
Russo, A. and Levis, A. (1992). Detection of aneuploidy in male germ cells of mice by means of
a meiotic micronucleus assay. MutatRes, 281, 187-191.
http://www.ncbi.nlm.nih.gov/pubmed/1371841
Russo, A. and Levis, A. G. (1992). Further evidence for the aneuploidogenic properties of
chelating agents: induction of micronuclei in mouse male germ cells by EDTA. Environ
Mol Mutagen, 19, 125-131. http://www.ncbi.nlm.nih.gov/pubmed/1541253
Russo, A.; Pacchierotti, F.; Metalli, P. (1984). Nondisjunction induced in mouse spermatogenesis
by chloral hydrate, a metabolite of trichloroethylene. Environ Mutagen, 6, 695-703.
http://www.ncbi.nlm.nih.gov/pubmed/6479114
Russo, A.; Stocco, A.; Majone, F. (1992). Identification of kinetochore-containing (CREST+)
micronuclei in mouse bone marrow erythrocytes. Mutagenesis, 7, 195-197.
Rusyn, I.; Bradham, C. A.; Cohn, L.; Schoonhoven, R.; Swenberg, J. A.; Brenner, D. A.;
Thurman, R. G. (1999). Corn oil rapidly activates nuclear factor-kappaB in hepatic
Kupffer cells by oxidant-dependent mechanisms. Carcinogenesis, 20, 2095-2100.
http://www.ncbi.nlm.nih.gov/pubmed/10545411
Rusyn, I.; Peters, J. M.; Cunningham, M. L. (2006). Modes of action and species-specific effects
of di-(2-ethylhexyl)phthalate in the liver. Crit Rev Toxicol, 36, 459-479.
http://dx.doi.org/10.1080/10408440600779065
99
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Sabel, G. V. and Clark, T. P. (1984). Volatile organic compounds as indicators of municipal
solid waste leachate contamination. 2, 119-130. (as cited in U.S. EPA 2001a in NRC
2006).
Sagawa, K.; Nishitani, H.; Kawai, H.; Kuge, Y.; Ikeda, M. (1973). Transverse lesion of spinal
cord after accidental exposure to trichloroethylene. Int Arch Arbeitsmed, 31, 257-264.
http://www.ncbi.nlm.nih.gov/pubmed/4767703
Saghir, S. and Schultz, I. (2002). Low-dose pharmacokinetics and oral bioavailability of
dichloroacetate in naive and GST-zeta-depleted rats. Environ Health Perspect, 110, 757-
763. http://www.ncbi.nlm.nih.gov/pubmed/12153755
Saihan, E.; Burton, J.; Heaton, K. (1978). A new syndrome with pigmentation, scleroderma,
gynaecomastia, Raynaud's phenomenon and peripheral neuropathy. Br J Dermatol, 99,
437-440. http://www.ncbi.nlm.nih.gov/pubmed/213095
Saillenfait, A. M.; Langonne, I.; Sabate, J. P. (1995). Developmental toxicity of
trichloroethylene, tetrachloroethylene and four of their metabolites in rat whole embryo
culture. Arch Toxicol, 70, 71-82. http://dx.doi.org/10.1007/BF02733666
Sakurai, T.; Maeda, S.; Chang, L.; Karin, M. (2006). Loss of hepatic NF-kappa B activity
enhances chemical hepatocarcinogenesis through sustained c-Jun N-terminal kinase 1
activation. PNAS, 103, 10544-10551. http://dx.doi.org/10.1073/pnas.0603499103
Sallmen, M.; Lindbohm, M. L.; Anttila, A.; Kyyronen, P.; Taskinen, H.; Nykyri, E.; Hemminki,
K. (1998). Time to pregnancy among the wives of men exposed to organic solvents.
Occup Environ Med, 55, 24-30. http://www.ncbi.nlm.nih.gov/pubmed/9536159
Sallmen, M.; Lindbohm, M. L.; Kyyronen, P.; Nykyri, E.; Anttila, A.; Taskinen, H.; Hemminki,
K. (1995). Reduced fertility among women exposed to organic solvents. Am J Ind Med,
27, 699-713. http://dx.doi.org/10.1002/ajim.4700270506
Salvan, A. (1990). Occupational exposure and cancer mortality at an electrical manufacturing
plant: A case-control study. University of California, Los Angeles, CA.
Salvan, A.; Stayner, L.; Steenland, K.; Smith, R. (1995). Selecting an exposure lag period.
Epidemiology, 6, 387-390. http://www.ncbi.nlm.nih.gov/pubmed/7548346
Salvini, M.; Binaschi, S.; Riva, M. (1971). Evaluation of the psychophysiological functions in
humans exposed to trichloroethylene. Occup Environ Med, 28, 293-295.
http://www.ncbi.nlm.nih.gov/pubmed/5557850
Sanchez, I. M. and Bull, R. J. (1990). Early induction of reparative hyperplasia in the liver of
B6C3F1 mice treated with dichloroacetate and trichloroacetate. Toxicology, 64, 33-46.
http://www.ncbi.nlm.nih.gov/pubmed/2219131
Sanders, V. M.; Tucker, A. N; White, K. L. J.; Kauffmann, B. M.; Hallett, P.; Carchman, R. A., .
. . Munson, A. E. (1982). Humoral and cell-mediated immune status in mice exposed to
trichloroethylene in the drinking water. Toxicol Appl Pharmacol, 62, 358-368.
http://www.ncbi.nlm.nih.gov/pubmed/7071855
Sanders, V.; Kauffmann, B.; White, K.; Douglas, K.; Barnes, D.; Sain, L., . . . Munson, A.
(1982). Toxicology of chloral hydrate in the mouse. Environ Health Perspect, 44, 137-
146. http://www.ncbi.nlm.nih.gov/pubmed/7084146
Santibanez, M.; Vioque, J.; Alguacil, J.; Barber, X.; de la Hera, G.; Kauppinen, T. (2008).
Occupational exposures and risk of oesophageal cancer by histological type: a case-
control study in eastern Spain. Occup Environ Med, 65, 774-781.
http://dx.doi.org/10.! 136/oem.2007.037929
100
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Sanz, N.; Diez-Fernandez, C.; Cascales, M. (1996). Variations of hepatic antioxidant systems
and DNA ploidy in rats aged 2 to 8 months. Biochim Biophys Acta, 1315, 123-130.
http://www.ncbi.nlm.nih.gov/pubmed/8608169
Sapkota, A.; Williams, D.; Buckley, T. J. (2005). Tollbooth workers and mobile source-related
hazardous air pollutants: how protective is the indoor environment? Environ Sci Technol,
39, 2936-2943. http://www.ncbi.nlm.nih.gov/pubmed/15926536
Sarangapani, R.; Gentry, P. R.; Covington, T. R.; Teeguarden, J. G.; 3rd, C. H. (2003).
Evaluation of the potential impact of age- and gender-specific lung morphology and
ventilation rate on the dosimetry of vapors. Inhal Toxicol, 15, 987-1016.
http://dx.doi.org/10.1080/713857276
Sargent, L. M.; Zhou, X.; Keck, C. L.; Sanderson, N. D.; Zimonjic, D. B.; Popescu, N. C.;
Thorgeirsson, S. S. (1999). Nonrandom cytogenetic alterations in hepatocellular
carcinoma from transgenic mice overexpressing c-Myc and transforming growth factor-
alpha in the liver. Am J Pathol, 154, 1047-1055.
http://www.ncbi.nlm.nih.gov/pubmed/10233843
Sato, A. (1993). Confounding factors in biological monitoring of exposure to organic solvents.
Int Arch Occup Environ Health, 65, S61-S67.
http://www.ncbi.nlm.nih.gov/pubmed/8406940
Sato, A. and Nakajima, T. (1978). Differences following skin or inhalation exposure in the
absorption and excretion kinetics of trichloroethylene and toluene. Occup Environ Med,
35, 43-49. http://www.ncbi.nlm.nih.gov/pubmed/629888
Sato, A. and Nakajima, T. (1979). A vial-equilibration method to evaluate the drug-metabolizing
enzyme activity for volatile hydrocarbons. Toxicol Appl Pharmacol, 47, 41-46.
http://www.ncbi.nlm.nih.gov/pubmed/34248
Sato, A. and Nakajima, T. (1985). Enhanced metabolism of volatile hydrocarbons in rat liver
following food deprivation, restricted carbohydrate intake and administration of ethanol,
phenobarbital, polychlorinated biphenyl and 3-methylcholanthrene: A comparative study.
Xenobiotica, 15, 67. http://dx.doi.org/10.3109/00498258509045336
Sato, A.; Endoh, K.; Johanson, G. (1991). The Use of Models to Investigate the Toxicokinetic
Behavior of Organic Solvents. Lewis Publishers, Chelsea, Michigan, 131-152.
Sato, A.; Endoh, K.; Kaneko, T.; Johanson, G. (1991). A simulation study of physiological
factors affecting pharmacokinetic behaviour of organic solvent vapours. Occup Environ
Med, 48, 342-347. http://www.ncbi.nlm.nih.gov/pubmed/2039747
Sato, A.; Nakajima, T.; Fujiwara, Y.; Murayama, N. (1977). A pharmacokinetic model to study
the excretion of trichloroethylene and its metabolites after an inhalation exposure. Occup
Environ Med, 34, 56-63. http://www.ncbi.nlm.nih.gov/pubmed/843464
Sato, A.; Nakajima, T.; Koyama, Y. (1980). Effects of chronic ethanol consumption on hepatic
metabolism of aromatic and chlorinated hydrocarbons in rats. Br J Ind Med, 37, 382-386.
http://dx.doi.Org/10.l 136/oem.37.4.382
Sato, A.; Nakajima, T.; Koyama, Y. (1981). Dose-related effects of a single dose of ethanol on
the metabolism in rat liver of some aromatic and chlorinated hydrocarbons. Toxicol Appl
Pharmacol, 60, 8-15. http://dx.doi.org/10.1016/0041-008X(81)90129-0
Sato, A.; Nakajima, T.; Koyama, Y. (1983). Interaction between ethanol and carbohydrate on the
metabolism in rat liver of aromatic and chlorinated hydrocarbons. Toxicol Appl
Pharmacol, 68, 242-249. http://dx.doi.org/10.1016/0041-008X(83)90008-X
101
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
Saunders, R. A. (1967). A new hazard in closed environmental atmospheres. Arch Environ
Health, 14, 380-384. http://www.ncbi.nlm.nih.gov/pubmed/4951670
Sausen, P. and Elfarra, A. (1991). Reactivity of cysteine S-conjugate sulfoxides: formation of S-
[l-chloro-2-(S-glutathionyl)vinyl]-L-cysteine sulfoxide by the reaction of S-(l,2-
dichlorovinyl)-L-cysteine sulfoxide with glutathione. Chem Res Toxicol, 4, 655-660.
http://www.ncbi.nlm.nih.gov/pubmed/1807449
Sausen, P. J. and Elfarra, A. A. (1990). Cysteine conjugate S-oxidase. Characterization of a
novel enzymatic activity in rat hepatic and renal microsomes. J Biol Chem, 265, 6139-
6145. http://www.ncbi.nlm.nih.gov/pubmed/2318851
Savolainen, H.; Pfaffli, P.; Tengen, M.; Vainio, H. (1977). Trichloroethylene and 1,1,1-
trichloroethane: effects on brain and liver after five days intermittent inhalation. Arch
Toxicol, 38, 229-237. http://www.ncbi.nlm.nih.gov/pubmed/578726
Saxena, D. K.; Srivastava, R. S.; Lai, B.; Chandra, S. V. (1987). The effect of lead exposure on
the testis of growing rats. Exp Toxicol Pathol, 31, 249-252.
Saygun, M.; Cakmak, A.; Ekici, A.; Pinar, T.; Bulcun, E.; Ulu, N.; Ekici, M. (2007). Five annual
observations of respiratory findings in gun factory workers exposed to solvents. J Occup
Environ Med, 49, 909-912. http://dx.doi.org/10.1097/JOM.0b013e318124a8f3
Sbrana, I.; Di Sibio, A.; Lomi, A.; Scarcelli, V. (1993). C-mitosis and numerical chromosome
aberration analyses in human lymphocytes: 10 known or suspected spindle poisons.
MutatRes, 287, 57-70. http://www.ncbi.nlm.nih.gov/pubmed/7683385
Scharf, J. G.; Dombrowski, F.; Ramadori, G. (2001). The IGF axis and hepatocarcinogenesis. J
Clin Pathol, 54, 138-144. http://www.ncbi.nlm.nih.gov/pubmed/11376124
Schatten, H. and Chakrabarti, A. (1998). Centrosome structure and function is altered by chloral
hydrate and diazepam during the first reproductive cell cycles in sea urchin eggs. Eur J
Cell Biol, 75, 9-20.
Scheff, P. A. and Wadden, R. A. (1993). Receptor modeling of volatile organic compounds 1
Emission inventory and validation. Environ Sci Technol, 27, 617-625.
http://dx.doi.org/10.1021/es00041a005
Schenk, M.; Purdue, M.; Colt, J.; Hartge, P.; Blair, A.; Stewart, P., . . . Severson, R. (2009).
Occupation/industry and risk of non-Hodgkin's lymphoma in the United States. Occup
Environ Med, 66, 23-31. http://dx.doi.org/10.1136/oem.2007.036723
Schlehofer, B.; Heuer, C.; Blettner, M.; Niehoff, D.; Wahrendorf, J. (1995). Occupation,
smoking and demographic factors, and renal cell carcinoma in Germany. Int J Epidemiol,
24, 51-57. http://www.ncbi.nlm.nih.gov/pubmed/7797356
Schlesselman, J. J. and Stolley, P. D. (1982). Case control studies: design, conduct, analysis.
New York: Oxford University Press.
Schlipkoter, H. W.; Fodor, G. G.; Winneke, G. (1974). Zum Einfluss neurotoxischer
Industriegifte auf unfallgenerierendes Fehlverhalten bei Kontroll- und Steuertatigkeiten.
Wilhelmshaven, Dortmund: Wirtschaftsverlag Nordwest.
Schmucker, D. L. (2001). Liver function and phase I drug metabolism in the elderly: a paradox.
Drugs Aging, 18, 837-851. http://www.ncbi.nlm.nih.gov/pubmed/11772124
Schmucker, D. L. (2005). Age-related changes in liver structure and function: Implications for
disease. Exp Gerontol, 40, 650-659. http://dx.doi.Org/10.1016/j.exger.2005.06.009
102
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
Schmucker, D. L.; Mooney, J. S.; Jones, A. L. (1978). Stereological analysis of hepatic fine
structure in the Fischer 344 rat. Influence of sublobular location and animal age. J Cell
Biol, 78, 319-337. http://www.ncbi.nlm.nih.gov/pubmed/690169
Schraml, P.; Zhaou, M.; Richter, J.; Briining, T.; Pommer, M.; Sauter, G., . . . Moch, H. (1999).
[Analysis of kidney tumors in trichloroethylene exposed workers by comparative
genomic hybridization and DNA sequence analysis], Verh Dtsch Ges Pathol, 83, 218-
224. http://www.ncbi.nlm.nih.gov/pubmed/10714214
Schulte-Hermann, R.; Timmermann-Trosiener, I.; Schuppler, J. (1986). Facilitated expression of
adaptive responses to phenobarbital in putative pre-stages of liver cancer. Carcinogenesis,
7, 1651-1655. http://www.ncbi.nlm.nih.gov/pubmed/2875809
Schultz, I. R.; Merdink, J. L.; Gonzalez-Leon, A.; Bull, R. J. (1999). Comparative toxicokinetics
of chlorinated and brominated haloacetates in F344 rats. Toxicol Appl Pharmacol, 158,
103-114. http://dx.doi.org/10.1006/taap.1999.869
Schultz, I. R.; Merdink, J. L.; Gonzalez-Leon, A.; Bull, R. J. (2002). Dichloroacetate
toxicokinetics and disruption of tyrosine catabolism in B6C3F1 mice: Dose-response
relationships and age as a modifying factor. Toxicology, 173, 229-247.
http://dx.doi.org/10.1016/S0300-483X(02)00034-3
Schumucker, D. L.; Mooney, J. S.; Jones, A. L. (1977). Age-related changes in the hepatic
endoplasmic reticulum: a quantitative analysis. Science, 197, 1005-1008.
http://www.ncbi.nlm.nih.gov/pubmed/887935
Schwartz-Arad, D.; Zajicek, G.; Bartfeld, E. (1989). The streaming liver IV: DNA content of the
hepatocyte increases with its age. 9, 93-99.
http://www.ncbi.nlm.nih.gov/pubmed/2709954
Schwetz, B. A.; Leong, B. K. J.; Gehring, P. J. (1975). The effect of maternally inhaled
trichloroethylene, perchloroethylene, methyl chloroform, and methylene chloride on
embryonal and fetal development in mice and rats. Toxicol Appl Pharmacol, 32, 84-96.
http://dx.doi. org/10.1016/0041 -008X(75)90197-0
Scott, C. and Chiu, W. (2006). Trichloroethylene cancer epidemiology: a consideration of select
issues. Environ Health Perspect, 114, 1471-1478.
http://www.ncbi.nlm.nih.gov/pubmed/16966107
Scott, J. E.; Forkert, P. G.; Oulton, M.; Rasmusson, M.-G.; Temple, S.; Fraser, M. O.;
Whitefield, S. (1988). Pulmonary toxicity of trichloroethylene: induction of changes in
surfactant phospholipids and phospholipase A2 activity in the mouse lung. Exp Mol
Pathol, 49, 141-150. http://www.ncbi.nlm.nih.gov/pubmed/3396665
Seelbach, A.; Fissler, B.; Madle, S. (1993). Further evaluation of a modified micronucleus assay
with V79 cells for detection of aneugenic effects. Mutat Res, 303, 163-169.
http://www.ncbi.nlm.nih.gov/pubmed/7694134
Seidler, A.; Mohner, M.; Berger, J.; Mester, B.; Deeg, E.; Eisner, G., . . . Becker, N. (2007).
Solvent exposure and malignant lymphoma: A population-based case-control study in
Germany. J Occup Med Toxicol, 2, 2. http://dx.doi.org/10.1186/1745-6673-2-2
Seitz, C. S.; Lin, Q.; Deng, H.; Khavari, P. A. (1998). Alterations in NF-kappaB function in
transgenic epithelial tissue demonstrate a growth inhibitory role for NF-kappaB. PNAS,
95, 2307-2312. http://www.ncbi.nlm.nih.gov/pubmed/9482881
103
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
Seitz, H. K. and Stickel, F. (2006). Risk factors and mechanisms of hepatocarcinogenesis with
special emphasis on alcohol and oxidative stress. Biol Chem, 387, 349-360.
http://dx.doi.org/10.1515/bc.2006.047
Selden, A. and Ahlborg, G. (1991). Mortality and cancer morbidity after exposure to military
aircraft fuel. Aviat Space Environ Med, 62, 789-794.
http://www.ncbi.nlm.nih.gov/pubmed/1930063
Selden, A.; Hultberg, B.; Ulander, A.; Ahlborg, G. J. (1993). Trichloroethylene exposure in
vapour degreasing and the urinary excretion of N-acetyl-"beta"-D-glucosaminidase. Arch
Toxicol, 67, 224-226. http://www.ncbi.nlm.nih.gov/pubmed/8494503
Selgrade, M. K. and Gilmour, M. I. (2010). Suppression of pulmonary host defenses and
enhanced susceptibility to respiratory bacterial infection in mice following inhalation
exposure to trichloroethylene and chloroform. J Immunotoxicol, 7, 350-356.
http://dx.doi.org/10.3109/1547691X.2010.520139
Sellers, E. M.; Lang, M.; Koch-Weser, J.; LeBlanc, E.; Kalant, H. (1972). Interaction of chloral
hydrate and ethanol in man I Metabolism. Clin Pharmacol Ther, 13, 37-49.
http://www.ncbi.nlm.nih.gov/pubmed/4550322
Sellers, E.; Lang-Sellers, M.; Koch-Weser, J. (1978). Comparative metabolism of chloral hydrate
and triclofos. J Clin Pharmacol, 18, 457-461.
http://www.ncbi.nlm.nih.gov/pubmed/711926
Selmin, O. I.; Thorne, P. A.; Caldwell, P. T.; Taylor, M. R. (2008). Trichloroethylene and
trichloroacetic acid regulate calcium signaling pathways in murine embryonal carcinoma
cells pl9. Cardiovasc Toxicol, 8, 47-56. http://dx.doi.org/10.1007/sl2012-008-9014-2
Seng, J.; Agrawal, N.; Horsley, E.; Leakey, T.; Scherer, E.; Xia, S., . . . Leakey, J. (2003).
Toxicokinetics of chloral hydrate in ad libitum-fed, dietary-controlled, and calorically
restricted male B6C3F1 mice following short-term exposure. Toxicol Appl Pharmacol,
193, 281-292. http://www.ncbi.nlm.nih.gov/pubmed/14644628
Seo, M.; Ikeda, K.; Okamura, T.; Kida, K.; Satoh, M.; Inagaki, N., . . . Nagase, H. (2008).
Enhancing effect of chlorinated organic solvents on histamine release and inflammatory
mediator production. Toxicology, 243, 75-83. http://dx.doi.Org/10.1016/j.tox.2007.09.024
Seppalainen, A. M. and Antti-Poika, M. (1983). Time course of electrophysiological findings for
patients with solvent poisoning. A descriptive study. Scand J Work Environ Health, 9,
15-24. http://www.ncbi.nlm.nih.gov/pubmed/6857184
Sexton, K.; Adgate, J. L.; Church, T. R.; Ashley, D. L.; Needham, L. L.; Ramachandran, G., . . .
Ryan, A. D. (2005). Children's exposure to volatile organic compounds as determined by
longitudinal measurements in blood. Environ Health Perspect, 113, 342-349.
http://dx.doi.org/10.1289/ehp.7412
Sexton, K.; Mongin, S. J.; Adgate, J. L.; Pratt, G. C.; Ramachandran, G.; Stock, T. H.; Morandi,
M. T. (2007). Estimating volatile organic compound concentrations in selected
microenvironments using time-activity and personal exposure data. J Toxicol Environ
Health A, 70, 465-476. http://dx.doi.org/10.1080/15287390600870858
Shafer, T. J.; Bushnell, P. J.; Benignus, V. A.; Woodward, J. J. (2005). Perturbation of voltage-
sensitive Ca2+ channel function by volatile organic solvents. J Pharmacol Exp Ther, 315,
1109-1118. http://dx.doi.org/10.! 124/jpet. 105.090027
104
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Shah, J. J. and Singh, H. B. (1988). Distribution of volatile organic chemicals in outdoor and
indoor air: a national VOCs database. Environ Sci Technol, 22, 1381-1388.
http://dx.doi.org/10.1021/es00177a001
Shah, Y. M.; Morimura, K.; Yang, Q.; Tanabe, T.; Takagi, M.; Gonzalez, F. J. (2007).
Peroxisome proliferator-activated receptor alpha regulates a microRNA-mediated
signaling cascade responsible for hepatocellular proliferation. Mol Cell Biol, 27, 4238-
4247. http://dx.doi.org/10.1128/MCB.00317-07
Shankar, K.; Vaidya, V.; Corton, J.; Bucci, T.; Liu, J.; Waalkes, M.; Mehendale, H. (2003).
Activation of PPAR-alpha in streptozotocin-induced diabetes is essential for resistance
against acetaminophen toxicity. FASEB J, 17, 1748-1750.
http://dx.doi.org/10.1096/fj.02-l 186fje
Shannon, H.; Haines, T.; Bernholz, C.; Julian, J.; Verma, D.; Jamieson, E.; Walsh, C. (1988).
Cancer morbidity in lamp manufacturing workers. Am J Ind Med, 14, 281-290.
http://www.ncbi.nlm.nih.gov/pubmed/3189346
Shao, J.; Stapleton, P. L.; Lin, Y. S.; Gallagher, E. P. (2007). Cytochrome p450 and glutathione
s-transferase mRNA expression in human fetal liver hematopoietic stem cells. Drug
Metab Dispos, 35, 168-175. http://dx.doi.org/10.1124/dmd.106.012757
Sharpe, C. R.; Rochon, J. E.; Adam, J. M.; Suissa, S. (1989). Case-control study of hydrocarbon
exposures in patients with renal cell carcinoma. CMAJ, 140, 1309-1318.
http://www.ncbi.nlm.nih.gov/pubmed/2720514
Shearer, B. G. and Hoekstra, W. J. (2003). Recent advances in peroxisome proliferator-activated
receptor science. Curr Med Chem, 10, 267-280.
http://www.ncbi.nlm.nih.gov/pubmed/12570700
Sheffels, P.; Schroeder, J.; Altuntas, T.; Liggitt, H.; Kharasch, E. (2004). Role of cytochrome
P4503 A in cysteine S-conjugates sulfoxidation and the nephrotoxicity of the sevoflurane
degradation product fluoromethyl-2,2-difluoro-l-(trifluoromethyl)vinyl ether (compound
A) in rats. Chem Res Toxicol, 17, 1177-1189. http://dx.doi.org/10.1021/tx049899e
Sher, T.; Yi, H. F.; McBride, O. W.; Gonzalez, F. J. (1993). cDNA cloning, chromosomal
mapping, and functional characterization of the human peroxisome proliferator activated
receptor. Biochemistry, 32, 5598-5604.
Sherins, R. J. and Hodgen, G. D. (1976). Testicular gamma glutamyl-transpeptidase: an index of
Sertoli cell function in man. Journal of Reproduction and Fertility, 48, 191-193.
http://www.ncbi.nlm.nih.gov/pubmed/9507
Shiao, Y.; Kamata, S.; Li, L.; Hooth, M.; DeAngelo, A.; Anderson, L.; Wolf, D. (2002).
Mutations in the VHL gene from potassium bromate-induced rat clear cell renal tumors.
Cancer Lett, 187, 207-214. http://www.ncbi.nlm.nih.gov/pubmed/12359370
Shiao, Y.; Rice, J.; Anderson, L.; Diwan, B.; Hard, G. (1998). von Hippel-Lindau gene
mutations in N-nitrosodimethylamine-induced rat renal epithelial tumors. J Natl Cancer
Inst, 90, 1720-1723. http://www.ncbi.nlm.nih.gov/pubmed/9827526
Shih, C.; Chen, H.; Chiu, T. (2001). Acute exposure to trichloroethylene differentially alters the
susceptibility to chemoconvulsants in mice. Toxicology, 162, 35-42.
http://www.ncbi.nlm.nih.gOv/pubmed/l 1311456
Shimada, T.; Swanson, A. F.; Leber, P.; Williams, G. M. (1985). Activities of chlorinated ethane
and ethylene compounds in the Salmonella/rat microsome mutagenesis and rat
105
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
hepatocyte/DNA repair assays under vapor phase exposure conditions. Cell Biol Toxicol,
1, 159-179. http://dx.doi.org/10.1007/BF00120162
Shimada, T.; Yamazaki, H.; Mimura, M.; Inui, Y.; Guengerich, F. P. (1994). Interindividual
variations in human liver cytochrome P-450 enzymes involved in the oxidation of drugs,
carcinogens and toxic chemicals: studies with liver microsomes of 30 Japanese and 30
Caucasians. J Pharmacol Exp Ther, 270, 414-423.
http://www.ncbi.nlm.nih.gov/pubmed/8035341
Shindell, S. and Ulrich, S. (1985). A cohort study of employees of a manufacturing plant using
trichlorethylene. J Occup Environ Med, 27, 577-579.
http://www.ncbi.nlm.nih.gov/pubmed/4032094
Shirasu, Y.; Moriya, M.; Kato, K.; Furuhashi, A.; Kada, T. (1976). Mutagenicity screening of
pesticides in the microbial system. Mutat Res, 40, 19-30. http://dx.doi.org/10.1016/0165-
1218(76)90018-5
Short, B. G. (1993). Cell proliferation and renal carcinogenesis. Environ Health Perspect, 101
Suppl 5, 115-120. http://www.ncbi.nlm.nih.gov/pubmed/7516872
Shroads, A. L.; Guo, X.; Dixit, V.; Liu, H. P.; James, M. O.; Stacpoole, P. W. (2008). Age-
dependent kinetics and metabolism of dichloroacetate: possible relevance to toxicity. J
Pharmacol Exp Ther, 324, 1163-1171. http://dx.doi.org/10.1124/jpet.107.134593
Shu, X. O.; Perentesis, J. P.; Wen, W.; Buckley, J. D.; Boyle, E.; Ross, J. A., . . . Group, C. s. O.
(2004). Parental exposure to medications and hydrocarbons and ras mutations in children
with acute lymphoblastic leukemia: a report from the Children's Oncology Group. Cancer
Epidemiol Biomarkers Prev, 13, 1230-1235.
http://www.ncbi.nlm.nih.gov/pubmed/15247135
Shu, X. O.; Stewart, P.; Wen, W. Q.; Han, D.; Potter, J. D.; Buckley, J. D., . . . Robison, L. L.
(1999). Parental occupational exposure to hydrocarbons and risk of acute lymphocytic
leukemia in offspring. Cancer Epidemiol Biomarkers Prev, 8, 783-791.
http://www.ncbi.nlm.nih.gov/pubmed/10498397
Shultz, J. and Weiner, H. (1979). Alteration of the enzymology of chloral hydrate reduction in
the presence of ethanol. Biochem Pharmacol, 28, 3379-3384.
http://www.ncbi.nlm.nih.gov/pubmed/43731
Siegel, J. H. and Kasmin, F. E. (1997). Biliary tract diseases in the elderly: management and
outcomes. Gut, 41, 433-435. http://www.ncbi.nlm.nih.gov/pubmed/9391238
Siemiatycki, J. (1991). Risk factors for cancer in the workplace. Boca Raton, FL: CRC Press.
Siemiatycki, J.; Dewar, R.; Nadon, L.; Gerin, M. (1994). Occupational risk factors for bladder
cancer: results from a case-control study in Montreal, Quebec, Canada. Am J Epidemiol,
140, 1061-1080. http://www.ncbi.nlm.nih.gov/pubmed/7998589
Siemiatycki, J.; Fritschi, L.; Nadon, L.; Gerin, M. (1997). Reliability of an expert rating
procedure for retrospective assessment of occupational exposures in community-based
case-control studies. Am J Ind Med, 31, 280-286.
http://www.ncbi.nlm.nih.gov/pubmed/9055950
Siemiatycki, J.; Wacholder, S.; Richardson, L.; Dewar, R.; Gerin, M. (1987). Discovering
carcinogens in the occupational environment. Methods of data collection and analysis of
a large case-referent monitoring system. Scand J Work Environ Health, 13, 486-492.
http://www.ncbi.nlm.nih.gov/pubmed/3433050
106
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Sigal, S. H.; Rajvanshi, P.; Gorla, G. R.; Sokhi, R. P.; Saxena, R.; Gebhard, D. R., . . . Gupta, S.
(1999). Partial hepatectomy-induced polyploidy attenuates hepatocyte replication and
activates cell aging events. Am J Physiol, 276, G1260-G1272.
http://www.ncbi.nlm.nih.gov/pubmed/10330018
Simmon, V. F.; Kauhanen, K.; Tardiff, R. G. (1977). Mutagenic activity of chemicals identified
in drinking water. In B. B. F. S. D Scott (Ed.), Progress in Genetic Toxicology (Vol. 2,
pp. 249-268). New York; Amsterdam: Elsevier/North Holland Press.
Simmons, J. E.; Boyes, W. K.; Bushnell, P. J.; Raymer, J. H.; Limsakun, T.; McDonald, A., . . .
Evans, M. V. (2002). A physiologically based pharmacokinetic model for
trichloroethylene in the male long-evans rat. Toxicol Sci, 69, 3-15.
http://www.ncbi.nlm.nih.gov/pubmed/12215655
Singh, M. and Sinha, U. (1976). Chloral hydrate induced haploidization in Aspergillus nidulans.
Experientia, 32, 1144-1145.
Singh, M. and Sinha, U. (1979). Mitotic haploidization and growth of Aspergillus nidulans on
media containing chloral hydrate. 14, 1-4.
Sinks, T.; Lushniak, B.; Haussler, B.; Sniezek, J.; Deng, J.; Roper, P., . . . Coates, R. (1992).
Renal cell cancer among paperboard printing workers. Epidemiology, 3, 483-489.
http://www.ncbi.nlm.nih.gov/pubmed/1420513
Skender, L. J.; Karacic, V.; Prpic-Majic, D. (1991). A comparative study of human levels of
trichloroethylene and tetrachloroethylene after occupational exposure. Arch Environ
Occup Health, 46, 174-178. http://www.ncbi.nlm.nih.gov/pubmed/2039273
Slikker, J. W.; Andersen, M. E.; Bogdanffy, M. S.; Bus, J. S.; Cohen, S. D.; Conolly, R. B., . . .
Wallace, K. (2004). Dose-dependent transitions in mechanisms of toxicity: Case studies.
Toxicol Appl Pharmacol, 201, 226-294. http://dx.doi.Org/10.1016/j.taap.2004.06.027
Slikker, J. W.; Andersen, M. E.; Bogdanffy, M. S.; Bus, J. S.; Cohen, S. D.; Conolly, R. B., . . .
Wallace, K. (2004). Dose-dependent transitions in mechanisms of toxicity. Toxicol Appl
Pharmacol, 201, 203-225. http://dx.doi.Org/10.1016/j.taap.2004.06.019
Sliwinska-Kowalska, M.; Jedlinska, U.; Rzadzinska, A.; Rajkowska, E. (1999). [Hair cell
regeneration in a chicken's inner ear after damage due to exposure to industrial noise],
Otolaryngol Pol, 53, 469-477. http://www.ncbi.nlm.nih.gov/pubmed/10581958
Smith, G. (1970). The investigation of the mental effects of trichl or ethylene. Ergonomics, 13,
580-586. http://www.ncbi.nlm.nih.gov/pubmed/5494866
Smith, M. K.; Randall, J. L.; Read, E. J.; Stober, J. A. (1989). Teratogenic activity of
trichloroacetic acid in the rat. Teratology, 40, 445-451.
http://dx.doi.org/10.1002/tera.1420400506
Smith, M.; Randall, J.; Read, E.; Stober, J. (1992). Developmental toxicity of dichloroacetate in
the rat. Teratology, 46, 217-223. http://dx.doi.org/10.1002/tera.1420460305
Snedcor, G. W. and Cochran, W. G. (1980). Statistical methods (7th ed.). Ames, IA: Iowa State
University Press.
Snyder, R.; Pullman, J.; Carter, J.; Carter, H.; DeAngelo, A. (1995). In vivo administration of
dichloroacetic acid suppresses spontaneous apoptosis in murine hepatocytes. Cancer Res,
55, 3702-3705. http://www.ncbi.nlm.nih.gov/pubmed/7641179
Song, B. J.; Veech, R. L.; Saenger, P. (1990). Cytochrome P450IIE1 is elevated in lymphocytes
from poorly controlled insulin-dependent diabetics. J Clin Endocrinol Metab, 71, 1036-
1040. http://www.ncbi.nlm.nih.gov/pubmed/2205622
107
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Soni, M.; Mangipudy, R.; Mumtaz, M.; Mehendale, H. (1998). Tissue repair response as a
function of dose during trichloroethylene hepatotoxicity. Toxicol Sci, 42, 158-165.
http://dx.doi.org/10.1006/toxs.1998.2427
Soni, M.; Ramaiah, S.; Mumtaz, M.; Clewell, H.; Mehendale, H. (1999). Toxicant-inflicted
injury and stimulated tissue repair are opposing toxicodynamic forces in predictive
toxicology. Regul Toxicol Pharmacol, 29, 165-174.
http://dx.doi.org/10.1006/rtph.1998.1280
Sonnenfeld, N.; Hertz-Picciotto, I.; Kaye, W. E. (2001). Tetrachloroethylene in drinking water
and birth outcomes at the US Marine Corps Base at Camp Lejeune, North Carolina. Am J
Epidemiol, 154, 902-908. http://www.ncbi.nlm.nih.gov/pubmed/11700244
Sora, S. and Agostini Carbone, M. L. (1987). Chloral hydrate, methylmercury hydroxide and
ethidium bromide affect chromosomal segregation during meiosis of Saccharomyces
cerevisiae. Mutat Res Lett, 190, 13-17. http://dx.doi.org/10.1016/0165-7992(87)90075-3
Soresi, M.; Giannitrapani, L.; D'Antona, F.; Florena, A. M.; La Spada, E.; Terranova, A., . . .
Montalto, G. (2006). Interleukin-6 and its soluble receptor in patients with liver cirrhosis
and hepatocellular carcinoma. World J Gastroenterol, 12, 2563-2568.
http://www.ncbi.nlm.nih.gov/pubmed/16688802
Soucek, B. and Vlachova, D. (1960). Excretion of trichloroethylene metabolites in human urine.
Occup Environ Med, 17, 60-64. http://www.ncbi.nlm.nih.gov/pubmed/13832993
Spiegelhalter, D.; Thomas, A.; Best, N. (2003). WinBugs version 1.4 user manual. Cambridge,
UK: MRC Biostatistics Unit.
Spielberg, S. P. (1996). N-acetyltransferases: pharmacogenetics and clinical consequences of
polymorphic drug metabolism. J Pharmacokinet Biopharm, 24, 509-519.
http://www.ncbi.nlm.nih.gov/pubmed/9131488
Spirtas, R.; Stewart, P. A.; Lee, J. S.; Marano, D. E.; Forbes, C. D.; Grauman, D. J., . . . Cohen, J.
L. (1991). Retrospective cohort mortality study of workers at an aircraft maintenance
facility. I. Epidemiological results. Br J Ind Med, 48, 515-530.
http://dx.doi.Org/10.l 136/oem. 48.8.515
Squillace, P. J.; Moran, M. J.; Price, C. V. (2004). VOCs in shallow groundwater in new
residential/commercial areas of the United States. Environ Sci Technol, 38, 5327-5338.
SRI International. (1992). 1992 directory of chemical producers United States of America.
Menlo Park, CA: Author.
Srivastava, S. and Srivastava, S. (1991). Effect of di(2-ethylhexyl)phthalate on 17P-
hydroxysteroid dehydrogenase activity in testis of rat. Toxicol Lett, 57, 235-239.
http://dx.doi.org/10.1016/0378-4274(91)90150-5
Staats, D. A.; Fisher, J. W.; Connolly, R. B. (1991). Gastrointestinal absorption of xenobiotics in
physiologically based pharmacokinetic models: A two-compartment description. Drug
Metab Dispos, 19, 144-148. http://www.ncbi.nlm.nih.gov/pubmed/1673388
Stadler, J. C. and Kennedy, G. L. J. (1996). Evaluation of the sensory irritation potential of
volatile organic chemicals from carpets—alone and in combination. Food Chem Toxicol,
34, 1125-1130. http://www.ncbi.nlm.nih.gov/pubmed/9119325
Stanley, L. A.; Blackburn, D. R.; Devereaux, S.; Foley, J.; Lord, P. G.; Maronpot, R. R., . . .
Anderson, M. W. (1994). Ras mutations in methylclofenapate-induced B6C3F1 and
C57BL/10J mouse liver tumours. Carcinogenesis, 15, 1125-1131.
http://www.ncbi.nlm.nih.gov/pubmed/8020144
108
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Staples, R. E. (1974). Detection of visceral alterations in mammalian fetuses. Teratology, 9,
A3 7-A3 8.
Stauber, A. J. and Bull, R. J. (1997). Differences in phenotype and cell replicative behavior of
hepatic tumors induced by dichloroacetate (DCA) and trichloroacetate (TCA). Toxicol
Appl Pharmacol, 144, 235-246. http://dx.doi.org/10.1006/taap.1997.8159
Stauber, A. J.; Bull, R. J.; Thrall, B. D. (1998). Dichloroacetate and trichloroacetate promote
clonal expansion of anchorage-independent hepatocytes in vivo and in vitro. Toxicol
Appl Pharmacol, 150, 287-294. http://dx.doi.org/10.1006/taap.1998.8417
Steinberg, A. (1993). Should chloral hydrate be banned. Pediatrics, 92, 442-446.
http://www.ncbi.nlm.nih.gov/pubmed/8361800
Stemhagen, A.; Slade, J.; Altman, R.; Bill, J. (1983). Occupational risk factors and liver cancer:
A retrospective case-control study of primary liver cancer in New Jersey. Am J
Epidemiol, 117, 443-454. http://www.ncbi.nlm.nih.gov/pubmed/6837558
Stemmer, K.; Ellinger-Ziegelbauer, H.; Ahr, H. J.; Dietrich, D. R. (2007). Carcinogen-specific
gene expression profiles in short-term treated Eker and wild-type rats indicative of
pathways involved in renal tumorigenesis. Cancer Res, 67, 4052-4068.
http://dx.doi.org/10.1158/0008-5472.CAN-06-3587
Stenner, R.; Merdink, J.; Stevens, D.; Springer, D.; Bull, R. (1997). Enterohepatic recirculation
of trichloroethanol glucuronide as a significant source of trichloroacetic acid. Metabolites
of trichloroethylene. Drug Metab Dispos, 25, 529-535.
http://www.ncbi.nlm.nih.gov/pubmed/9152590
Stephens, E. A.; Taylor, J. A.; Kaplan, N.; Yang, C. H.; Hsieh, L. L.; Lucier, G. W.; Bell, D. A.
(1994). Ethnic variation in the CYP2E1 gene: polymorphism analysis of 695 African-
Americans, European-Americans and Taiwanese. Pharmacogenetics, 4, 185-192.
Stephenson, R. P. (1956). A modification of receptor theory. Br J Pharmacol, 11, 379-393.
Steup, D. R.; Hall, P.; McMillan, D. A.; Sipes, I. G. (1993). Time course of hepatic injury and
recovery following coadministration of carbon tetrachloride and trichloroethylene in
Fischer-344 rats. Toxicol Pathol, 21, 327-334.
http://www.ncbi.nlm.nih.gov/pubmed/8248722
Steup, D. R.; Wiersma, D.; McMillan, D. A.; Sipes, I. G. (1991). Pretreatment with drinking
water solutions containing trichloroethylene or chloroform enhances the hepatotoxicity of
carbon tetrachloride in Fischer 344 rats. Toxicol Sci, 16, 798-809.
http://www.ncbi.nlm.nih.gov/pubmed/1884917
Stevens, J. (1985). Cysteine conjugate beta-lyase activities in rat kidney cortex: subcellular
localization and relationship to the hepatic enzyme. Biochem Biophys Res Commun, 129,
499-504. http://www.ncbi.nlm.nih.gov/pubmed/4015643
Stevens, J. and Jakoby, W. B. (1983). Cysteine conjugate beta-lyase. Mol Pharmacol, 23, 761-
765.
Stevens, J.; Ayoubi, N.; Robbins, J. (1988). The role of mitochondrial matrix enzymes in the
metabolism and toxicity of cysteine conjugates. J Biol Chem, 263, 3395-3401.
http://www.ncbi.nlm.nih.gov/pubmed/3343250
Stevens, J.; Hatzinger, P.; Hayden, P. (1989). Quantitation of multiple pathways for the
metabolism of nephrotoxic cysteine conjugates using selective inhibitors of L-alpha-
hydroxy acid oxidase (L-amino acid oxidase) and cysteine conjugate beta-lyase. Drug
Metab Dispos, 17, 297-303. http://www.ncbi.nlm.nih.gov/pubmed/2568912
109
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
Stevens, J.; Robbins, J.; Byrd, R. (1986). A purified cysteine conjugate beta-lyase from rat
kidney cytosol. Requirement for an alpha-keto acid or an amino acid oxidase for activity
and identity with soluble glutamine transaminase K. J Biol Chem, 261, 15529-15537.
http://www.ncbi.nlm.nih.gov/pubmed/3782077
Stewart, B. W.; Le Mesurier, S. M.; Lykke, A. W. J. (1979). Correlation of biochemical and
morphological changes induced by chemical injury to the lung. Chem Biol Interact, 26,
321-338. http://www.ncbi.nlm.nih.gov/pubmed/509695
Stewart, P. A.; Lee, J. S.; Marano, D. E.; Spirtas, R.; Forbes, C. D.; Blair, A. (1991).
Retrospective cohort mortality study of workers at an aircraft maintenance facility II
Exposures and their assessment. Occup Environ Med, 48, 531-537.
http://www.ncbi.nlm.nih.gov/pubmed/1878309
Stewart, P. A.; Lees, P. S.; Francis, M. (1996). Quantification of historical exposures in
occupational cohort studies. Scand J Work Environ Health, 22, 405-414.
http://www.ncbi.nlm.nih.gov/pubmed/9000307
Stewart, P. and Dosemeci, M. (2005). Exposure estimation in occupational epidemiology studies.
Third Meeting on Assessing Human Health Risks of Trichloroethylene. Irvine, C A.
Stewart, R. D.; Dodd, H. C.; Gay, H. H.; Erley, D. S. (1970). Experimental human exposure to
trichloroethylene. Arch Environ Health, 20, 64-71.
http://www.ncbi.nlm.nih.gov/pubmed/5409662
Stewart, R. D.; Hake, C. L.; Peterson, J. E. (1974). "Degreaser's flush". Arch Environ Health, 29,
1-5.
Stewart, W. F. and Correa-Villaseor, A. (1991). False positive exposure errors and low exposure
prevalence in community-based case-control studies. Appl Occup Environ Hyg, 6, 534-
540.
Stopps, G. J. and McLaughlin, M. (1967). Psychophysiological testing of human subjects
exposed to solvent vapors. Am Ind Hyg Assoc J, 28, 43-50.
http://www.ncbi.nlm.nih.gov/pubmed/6032934
Stott, W. T.; Quast, J. F.; Watanabe, P. G. (1982). The pharmacokinetics and macromolecular
interactions of trichloroethylene in mice and rats. Toxicol Appl Pharmacol, 62, 137-151.
http://www.ncbi.nlm.nih.gov/pubmed/7064149
Stuckhardt, J. L. and Poppe, S. M. (1984). Fresh visceral examination of rat and rabbit fetuses
used in teratogenicity testing. Teratog Carcinog Mutagen, 4, 181-188.
http://www.ncbi.nlm.nih.gov/pubmed/6145223
Styles, J. A.; Kelly, M.; Pritchard, N. R.; Ekcombe, C. R. (1988). A species comparison of acute
hyperplasia induced by the peroxisome proliferator methylclofenapate: Involvement of
the binucleated hepatocyte. Carcinogenesis, 9, 1647-1655.
http://dx.doi.Org/10.1093/carcin/9.9.1647
Styles, J. A.; Wyatt, I.; Coutts, C. (1991). Trichloroacetic acid: studies on uptake and effects on
hepatic DNA and liver growth in mouse. Carcinogenesis, 12, 1715-1719.
http://www.ncbi.nlm.nih.gov/pubmed/1893532
Su, Q. and Bannasch, P. (2003). Relevance of hepatic preneoplasia for human
hepatocarcinogenesis. Toxicol Pathol, 31, 126-133.
http://www.ncbi.nlm.nih.gov/pubmed/12597457
110
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Subramoniam, A.; Goel, S. K.; Pandya, K. P.; Seth, P. K. (1989). Influence of trichloroethylene
treatment on phosphoinositides in rat brain. Toxicol Lett, 49, 55-60.
http://www.ncbi.nlm.nih.gov/pubmed/2554538
Sung, T.; Chen, P.; Jyuhn-Hsiarn Lee, L.; Lin, Y.; Hsieh, G.; Wang, J. (2007). Increased
standardized incidence ratio of breast cancer in female electronics workers. BMC Public
Health, 7, 102. http://dx.doi.org/10.1186/1471-2458-7-102
Sung, T.; Wang, J.; Chen, P. (2008). Increased risk of cancer in the offspring of female
electronics workers. Reprod Toxicol, 25, 115-119.
http ://dx. doi. org /10.1016/j. reprotox.2007.08.004
Supplementary data for TCE assessment: Cancer rodents input data. (2011).
Supplementary data for TCE assessment: Cancer rodents model selections. (2011).
Supplementary data for TCE assessment: Cancer rodents plots. (2011).
Supplementary data for TCE assessment: Cancer rodents results. (2011).
Supplementary data for TCE assessment: Cancer rodents uncertainty analysis. (2011).
Supplementary data for TCE assessment: Cancer rodents uncertainty CSF-inhalation histograms,
inhalation bioassays. (2011).
Supplementary data for TCE assessment: Cancer rodents uncertainty CSF-inhalation historams,
oral bioassays. (2011).
Supplementary data for TCE assessment: Cancer rodents uncertainty CSF-oral histograms,
inhalation bioassay. (2011).
Supplementary data for TCE assessment: Cancer rodents uncertainty CSF-oral histograms, oral
bioassays. (2011).
Supplementary data for TCE assessment: Hack mouse population calibration evaluation. (2011).
Supplementary data for TCE assessment: Hack mouse subject calibration. (2011).
Supplementary data for TCE assessment: Human posteriors by subject. (2011).
Supplementary data for TCE assessment: Mouse posteriors by subject. (2011).
Supplementary data for TCE assessment: Non-cancer HECs plots from human inhalation studies.
(2011).
Supplementary data for TCE assessment: Non-cancer HECs plots from rodent inhalation studies.
(2011).
Supplementary data for TCE assessment: Non-cancer HECs plots from rodent oral studies.
(2011).
Supplementary data for TCE assessment: Non-cancer HEDs altPOD plots from rodent inhalation
studies. (2011).
Supplementary data for TCE assessment: Non-cancer HEDs plots from human inhalation studies.
(2011).
Supplementary data for TCE assessment: Non-cancer HEDs plots from rodent inhalation studies.
(2011).
Supplementary data for TCE assessment: Non-cancer HEDs plots from rodent oral studies.
(2011).
Supplementary data for TCE assessment: Non-cancer input data contin. (2011).
Supplementary data for TCE assessment: Non-cancer input data dichot. (2011).
Supplementary data for TCE assessment: Non-cancer plots contin. (2011).
Supplementary data for TCE assessment: Non-cancer plots dichot. (2011).
Supplementary data for TCE assessment: Non-cancer results contin. (2011).
Ill
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Supplementary data for TCE assessment: Non-cancer results dichot. (2011).
Supplementary data for TCE assessment: Rat population example. (2011).
Supplementary data for TCE assessment: Rat posterior by subject. (2011).
Supplementary data for TCE assessment: Rodents time to tumor results. (2011).
Sverdrup, B.; Kallberg, H.; Bengtsson, C.; Lundberg, I.; Padyukov, L.; Alfredsson, L., . . .
Group, E. I. o. R. A. S. (2005). Association between occupational exposure to mineral oil
and rheumatoid arthritis: results from the Swedish EIRA case-control study. Arthritis Res
Ther, 7, R1296-R1303. http://dx.doi.org/10.1186/arl824
Swaen, G. M. (1995). Increased incidence of renal cell tumours in a cohort of cardboard workers
exposed to trichloroethylene [letter to the editor] [see Henschler, 1995, for a reply]. Arch
Toxicol, 70, 127-128.
Swan, S.; Shaw, G.; Harris, J.; Neutra, R. (1989). Congenital cardiac anomalies in relation to
water contamination, Santa Clara County, California, 1981-1983. Am J Epidemiol, 129,
885-893. http://www.ncbi.nlm.nih.gov/pubmed/2784935
Sweeney, C.; Farrow, D.; Schwartz, S.; Eaton, D.; Checkoway, H.; Vaughan, T. (2000).
Glutathione S-transferase Ml, Tl, and PI polymorphisms as risk factors for renal cell
carcinoma: A case-control study. Cancer Epidemiol Biomarkers Prev, 9, 449-454.
http://www.ncbi.nlm.nih.gov/pubmed/10794492
Sweeney, L. M.; Kirman, C. R.; Gargas, M. L.; Dugard, P. H. (2009). Contribution of
trichloroacetic acid to liver tumors observed in perchloroethylene (perc)-exposed mice.
Toxicology, 260, 77-83. http://dx.doi.Org/10.1016/j.tox.2009.03.008
Sweet, C. W. and Vermette, S. J. (1992). Toxic volatile organic compounds in urban air in
Illinois. Environ Sci Technol, 26, 165-173. http://dx.doi.org/10.1021/es00025a020
Szyf, M. (2007). The dynamic epigenome and its implications in toxicology. Toxicol Sci, 100, 7-
23. http://dx.doi.org/10.1093/toxsci/kfml77
Szymanska, K.; Moore, L. E.; Rothman, N.; Chow, W. H.; Waldman, F.; Jaeger, E., . . . Brennan,
P. (2010). TP53, EGFR, and KRAS mutations in relation to VHL inactivation and
lifestyle risk factors in renal-cell carcinoma from central and eastern Europe. 293, 92-98.
http://dx.doi.Org/10.1016/j.canlet.2009.l 1.024
'tMannetje, A.; Dryson, E.; Walls, C.; McLean, D.; McKenzie, F.; Maule, M., . . . Pearce, N.
(2008). High risk occupations for non-Hodgkin's lymphoma in New Zealand: case-
control study. Occup Environ Med, 65, 354-363.
http://dx.doi.Org/10.l 136/oem. 2007.035014
'tMannetje, A.; Steenland, K.; Attfield, M.; Boffetta, P.; Checkoway, H.; DeKlerk, N.; Koskela,
R. (2002). Exposure-response analysis and risk assessment for silica and silicosis
mortality in a pooled analysis of six cohorts. Occup Environ Med, 59, 723-728.
http://www.ncbi.nlm.nih.gov/pubmed/12409529
Takashima, K.; Ito, Y.; Gonzalez, F. J.; Nakajima, T. (2008). Different mechanisms of DEHP-
induced hepatocellular adenoma tumorigenesis in wild-type and Ppar alpha-null mice. J
Occup Health, 50, 169-180. http://www.ncbi.nlm.nih.gov/pubmed/18403868
Takeda, S.; Kondo, M.; Kumada, T.; Koshikawa, T.; Ueda, R.; Nishio, M., . . . Nakao, A. (1996).
Allelic-expression imbalance of the insulin-like growth factor 2 gene in hepatocellular
carcinoma and underlying disease. Oncogene, 12, 1589-1592.
http://www.ncbi.nlm.nih.gov/pubmed/8622877
112
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Takubo, K. and Kaminishi, M. (2001). [Diseases of the digestive tract and telomere lengths:
significance and problems of telomere measurement], Nippon Shokakibyo Gakkai
Zasshi, 98, 144-150. http://www.ncbi.nlm.nih.gov/pubmed/11235188
Tanaka, K.; Smith, P. F.; Stromberg, P. C.; Eydelloth, R. S.; Herold, E. G.; Grossman, S. J., . . .
Keenan, K. P. (1992). Studies of early hepatocellular proliferation and peroxisomal
proliferation in Sprague-Dawley rats treated with tumorigenic doses of clofibrate.
Toxicol Appl Pharmacol, 116, 71-77. http://www.ncbi.nlm.nih.gov/pubmed/1529455
Tang, X.; Li, L.; Huang, J.; Deng, Y. (2002). Guinea pig maximization test for trichloroethylene
and its metabolites. Biomed Environ Sci, 15, 113-118.
http://www.ncbi.nlm.nih.gov/pubmed/12244752
Tang, X.; Que, B.; Song, X.; Li, S.; Yang, X.; Wang, H., . . . Li, L. (2008). Characterization of
liver injury associated with hypersensitive skin reactions induced by trichloroethylene in
the guinea pig maximization test. J Occup Health, 50, 114-121.
http://www.ncbi.nlm.nih.gov/pubmed/18403861
Tanguay, R.; Jorquera, R.; Poudrier, J.; St-Louis, M. (1996). Tyrosine and its catabolites: from
disease to cancer. Acta Biochim Pol, 43, 209-216.
http://www.ncbi.nlm.nih.gov/pubmed/8790725
Tao, L.; Kramer, P. M.; Ge, R.; Pereira, M. A. (1998). Effect of dichloroacetic acid and
trichloroacetic acid on DNA methylation in liver and tumors of female B6C3F1 mice.
Toxicol Sci, 43, 139-144. http://dx.doi.org/10.1006/toxs.1998.2449
Tao, L.; Li, Y.; Kramer, P.; Wang, W.; Pereira, M. (2004). Hypomethylation of DNA and the
insulin-like growth factor-II gene in dichloroacetic and trichloroacetic acid-promoted
mouse liver tumors. Toxicology, 196, 127-136.
http://dx.doi.Org/10.1016/j.tox.2003.l 1.011
Tao, L.; Wang, W.; Li, L.; Kramer, P.; Pereira, M. (2004). Effect of dibromoacetic acid on DNA
methylation, glycogen accumulation, and peroxisome proliferation in mouse and rat liver.
Toxicol Sci, 82, 62-69. http://dx.doi.org/10.1093/toxsci/kfh266
Tao, L.; Yang, S.; Xie, M.; Kramer, P. M.; Pereira, M. A. (2000). Effect of trichloroethylene and
its metabolites, dichloroacetic acid and trichloroacetic acid, on the methylation and
expression of c-jun and c-myc protooncogenes in mouse liver: Prevention by methionine.
Toxicol Sci, 54, 399-407. http://www.ncbi.nlm.nih.gov/pubmed/10774822
Tao, L.; Yang, S.; Xie, M.; Kramer, P. M.; Pereira, M. A. (2000). Hypomethylation and
overexpression of c-jun and c-myc protooncogenes and increased DNA methyltransferase
activity in dichloroacetic and trichloroacetic acid-promoted mouse liver tumors. Cancer
Lett, 158, 185-193. http://dx.doi.org/10.1016/S0304-3835(00)00518-8
Taskinen, H.; Anttila, A.; Lindbohm, M.-L.; Sallmen, M.; Hemminki, K. (1989). Spontaneous
abortions and congenital malformations among the wives of men occupationally exposed
to organic solvents. Scand J Work Environ Health, 15, 345-352.
http://www.ncbi.nlm.nih.gov/pubmed/2799322
Taskinen, H.; Kyyronen, P.; Hemminki, K. (1994). Laboratory work and pregnancy outcome. J
Occup Med, 36, 311-319. http://www.ncbi.nlm.nih.gov/pubmed/8195901
Tateishi, M.; Suzuki, S.; Shimizu, H. (1978). Cysteine conjugate beta-lyase in rat liver. A novel
enzyme catalyzing formation of thiol-containing metabolites of drugs. J Biol Chem, 253,
8854-8859. http://www.ncbi.nlm.nih.gov/pubmed/721818
113
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Tateishi, T.; Nakura, H.; Asoh, M.; Watanabe, M.; Tanaka, M.; Kumai, T., . . . Kobayashi, S.
(1997). A comparison of hepatic cytochrome P450 protein expression between infancy
and postinfancy. Life Sci, 61, 2567-2574. http://dx.doi.org/10.1016/S0024-
3205(97)01011-4
Tatham, L.; Tolbert, P.; Kjeldsberg, C. (1997). Occupational risk factors for subgroups of non-
Hodgkin's lymphoma. Epidemiology, 8, 551-558.
http://www.ncbi.nlm.nih.gov/pubmed/9270958
Taylor, D. H.; Lagory, K. E.; Zaccaro, D. J.; Pfohl, R. J.; Laurie, R. D. (1985). Effect of
trichloroethylene on the exploratory and locomotor activity of rats exposed during
development. Sci Total Environ, 47, 415-420.
http://www.ncbi.nlm.nih.gov/pubmed/4089608
Templin, M. V.; Parker, J. C.; Bull, R. J. (1993). Relative formation of dichloroacetate and
trichloroacetate from trichloroethylene in male B6C3F1 mice. Toxicol Appl Pharmacol,
123, 1-8. http://dx.doi.org/10.1006/taap.1993.1214
Templin, M. V.; Parker, J. C.; Bull, R. J. (1995). Erratum for: relative formation of
dichloroacetate and trichloroacetate from trichloroethylene in male B6C3F1 mice.
Toxicol Appl Pharmacol, 133, 177.
Templin, M. V.; Stevens, D. K.; Stenner, R. D.; Bonate, P. L.; Tuman, D.; Bull, R. J. (1995).
Factors affecting species differences in the kinetics of metabolites of trichloroethylene. J
Toxicol Environ Health, 44, 435-447. http://www.ncbi.nlm.nih.gov/pubmed/7723076
Tennant, R. W.; Elwell, M. R.; Spalding, J. W.; Griesemer, R. A. (1991). Evidence that toxic
injury is not always associated with induction of chemical carcinogenesis. Mol Carcinog,
4, 420-440. http://dx.doi.org/10.1002/mc.2940040604
TERRACINI, B. and PARKER, V. (1965). A PATHOLOGICAL STUDY ON THE TOXICITY
OF S-DICHLOROVINYL-L-CYSTEINE. Food Cosmet Toxicol, 3, 67-74.
http://www.ncbi.nlm.nih.gov/pubmed/14342797
Teschke, K.; Olshan, A. F.; Daniels, J. L.; De Roos, A. J.; Parks, C. G.; Schulz, M.; Vaughan, T.
L. (2002). Occupational exposure assessment in case-control studies: opportunities for
improvement. Occup Environ Med, 59, 575-593; discussion 594.
http://www.ncbi.nlm.nih.gov/pubmed/12205230
Thai, S.; Allen, J.; DeAngelo, A.; George, M.; Fuscoe, J. (2001). Detection of early gene
expression changes by differential display in the livers of mice exposed to dichloroacetic
acid. Carcinogenesis, 22, 1317-1322. http://www.ncbi.nlm.nih.gov/pubmed/11470764
Thai, S.-F.; Allen, J. W.; DeAngelo, A. B.; George, M. H.; Fuscoe, J. C. (2003). Altered gene
expression in mouse livers after dichloroacetic acid exposure. Mutat Res Rev Mutat Res,
543, 167-180. http://dx.doi.org/10.1016/S1383-5742(03)00014-0
Tham, R.; Bunnfors, I.; Eriksson, B.; Larsby, B.; Lindgren, S.; Odkvist, L. M. (1984). Vestibulo-
ocular disturbances in rats exposed to organic solvents. Acta Pharmacol Toxicol, 54, 58-
63. http://dx.doi.Org/10.llll/j.1600-0773.1984.tb01895.x
Tham, R.; Larsby, B.; Odkvist, L. M.; Norlander, B.; Hyden, D.; Aschan, G.; Bertler, A. (1979).
The influence of trichloroethylene and related drugs on the vestibular system. Basic Clin
Pharmacol Toxicol, 44, 336-342. http://www.ncbi.nlm.nih.gov/pubmed/314225
Tharappel, J. B.; Cunningham, M. L.; Spear, B. T.; Glauert, H. P. (2001). Differential activation
of hepatic NF-kB in rats and hamsters by the peroxisome proliferators Wy-14,643,
gemfibrozil, and dibutyl phthalate. Toxicol Sci, 62, 20-27.
114
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Thiele, D. L.; Eigenbrodt, E. H.; Ware, A. J. (1982). Cirrhosis after repeated trichloroethylene
and 1,1,1-trichloroethane exposure. Gastroenterology, 83, 926-929.
http://www.ncbi.nlm.nih.gov/pubmed/7106523
Thorgeirsson, S. S. (2006). Genomic decoding of hepatocellular carcinoma. Gastroenterology,
131, 1344-1346. http://dx.doi.Org/10.1053/j.gastro.2006.08.060
Thrall, K. and Poet, T. (2000). Determination of biokinetic interactions in chemical mixtures
using real-time breath analysis and physiologically based pharmacokinetic modeling. J
Toxicol Environ Health A, 59, 653-670. http://www.ncbi.nlm.nih.gov/pubmed/10839498
Till, C.; Koren, G.; Rovet, J. F. (2001). Prenatal exposure to organic solvents and child
neurobehavioral performance. Neurotoxicol Teratol, 23, 235-245.
http://dx.doi.org/10.1016/S0892-0362(01)00141-6
Till, C.; Westall, C. A.; Koren, G.; Nulman, I.; Rovet, J. F. (2005). Vision abnormalities in
young children exposed prenatally to organic solvents. Neurotoxicology, 26, 599-613.
http://dx.doi.Org/10.1016/j.neuro.2005.05.011
Till, C.; Westall, C. A.; Rovet, J. F.; Koren, G. (2001). Effects of maternal occupational exposure
to organic solvents on offspring visual functioning: A prospective controlled study.
Teratology, 64, 134-141. http://dx.doi.org/10.1002/tera.1056
Tola, S.; Vilhunen, R.; Jarvinen, E.; Korkala, M.-L. (1980). A cohort study on workers exposed
to trichloroethylene. J Occup Environ Med, 22, 737-740.
Toma, M.; Grosser, M.; Herr, A.; Aust, D.; Meye, A.; Hoefling, C., . . . Baretton, G. (2008). Loss
of heterozygosity and copy number abnormality in clear cell renal cell carcinoma
discovered by high-density affymetrix 10K single nucleotide polymorphism mapping
array. Neoplasia, 10, 634-642. http://www.ncbi.nlm.nih.gov/pubmed/18592004
Tomisawa, H.; Ichihara, S.; Fukazawa, H.; Ichimoto, N.; Tateishi, M.; Yamamoto, I. (1986).
Purification and characterization of human hepatic cysteine-conjugate beta-lyase.
Biochem J, 235, 569-575. http://www.ncbi.nlm.nih.gov/pubmed/3741406
Tomisawa, H.; Suzuki, S.; Ichihara, S.; Fukazawa, H.; Tateishi, M. (1984). Purification and
characterization of C-S lyase from Fusobacterium varium . A C-S cleavage enzyme of
cysteine conjugates and some S-containing amino acids. J Biol Chem, 259, 2588-2593.
http://www.ncbi.nlm.nih.gov/pubmed/6698982
Tong, Z.; Board, P. G.; Anders, M. W. (1998). Glutathione transferase zeta catalyses the
oxygenation of the carcinogen dichloroacetic acid to glyoxylic acid. Biochem J, 331 ( Pt
2), 371-374. http://www.ncbi.nlm.nih.gov/pubmed/9531472
Toraason, M.; Clark, J.; Dankovic, D.; Mathias, P.; Skaggs, S.; Walker, C.; Werren, D. (1999).
Oxidative stress and DNA damage in Fischer rats following acute exposure to
trichloroethylene or perchloroethylene. Toxicology, 138, 43-53.
http://dx.doi.org/10.1016/S0300-483X(99)00083-9
Toth, B.; Hornung, D.; Scholz, C.; Djalali, S.; Friese, K.; Jeschke, U. (2007). Peroxisome
proliferator-activated receptors: new players in the field of reproduction. Am J Reprod
Immunol, 58, 289-310. http://dx.doi.Org/10.llll/j.1600-0897.2007.00514.x
Triebig, G.; Bestler, W.; Baumeister, P.; Valentin, H. (1983). Investigations on neurotoxicity of
chemical substances at the workplace: 4. Determination of the motor and sensory nerve
conduction velocity in persons occupationally exposed to a mixture of organic solvents.
Int Arch Occup Environ Health, 52, 139-150.
115
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Triebig, G.; Essing, H.; Schaller, K.; Valentin, H. (1976). [Biochemical and psychological
examinations of trichloroenthylene exposed volunteers (author's transl)]. Zentralbl
Bakteriol, Parasitenkd, Infektionskrankh Hyg, Abt 1: Orig, Reihe B, 163, 383-416.
http://www.ncbi.nlm.nih.gov/pubmed/1020533
Triebig, G.; Lehrl, S.; Kinzel, W.; Erzigkeit, H.; Galster, J.; Schaller, K. (1977).
[Psychopathometrical results of follow-up studies of trichloroethylene-exposed persons
(author's transl)]. Zentralbl Bakteriol, Parasitenkd, Infektionskrankh Hyg, Abt 1: Orig,
Reihe B, 164, 314-377. http://www.ncbi.nlm.nih.gov/pubmed/329621
Triebig, G.; Schaller, K.; Erzigkeit, H.; Valentin, H. (1977). [Biochemical investigations and
psychological studies of persons chronically exposed to trichloroethylene with regard to
non-exposure intervals (author's transl)]. Int Arch Occup Environ Health, 38, 149-162.
http://www.ncbi.nlm.nih.gov/pubmed/832911
Triebig, G.; Trautner, P.; Weltle, D.; Saure, E.; Valentin, H. (1982). [Investigations on
neurotoxicity of chemical substances at the workplace. III. Determination of the motor
and sensory nerve conduction velocity in persons occupationally exposed to
trichloroethylene], Int Arch Occup Environ Health, 51, 25-34.
http://www.ncbi.nlm.nih.gov/pubmed/7152699
Troster, A. I. and Ruff, R. M. (1990). Neuropsychological sequelae of exposure to the
chlorinated hydrocarbon solvents trichloroethylene and trichloroethane. Arch Clin
Neuropsychol, 5, 31-47. http://www.ncbi.nlm.nih.gov/pubmed/14589542
Truhaut, R. (1951). [Contribution to the study of toxicology of trichloroethylene; establishment
of a method of determination applicable to atmosphere and biological media], Ann Pharm
Fr, 9, 175-187. http://www.ncbi.nlm.nih.gov/pubmed/14838437
Tryndyak, V. P.; Muskhelishvili, L.; Kovalchuk, O.; Rodriguez-Juarez, R.; Montgomery, B.;
Churchwell, M. I., . . . Pogribny, I. P. (2006). Effect of long-term tamoxifen exposure on
genotoxic and epigenetic changes in rat liver: implications for tamoxifen-induced
hepatocarcinogenesis. Carcinogenesis, 27, 1713-1720.
http://dx.doi.org/10.1093/carcin/bgl050
Tu, A. S.; Murray, T. A.; Hatch, K. M.; Sivak, A.; Milman, H. A. (1985). In vitro transformation
of BALB/c-3T3 cells by chlorinated ethanes and ethylenes. Cancer Lett, 28, 85-92.
http://dx.doi.org/10.1016/0304-3835(85)90096-5
Tucker, A. N.; Sanders, V. M.; Barnes, D. W.; Bradshaw, T. J.; White, K. L. J.; Sain, L. E., . . .
Munson, A. E. (1982). Toxicology of trichloroethylene in the mouse. Toxicol Appl
Pharmacol, 62, 351-357. http://www.ncbi.nlm.nih.gov/pubmed/7071854
Tugwood, J. D.; Aldridge, T. C.; Lambe, K. G.; Macdonald, N.; Woodyatt, N. J. (1996).
Peroxisome proliferator-activated receptors: structures and function. Ann N Y Acad Sci,
804, 252-265.
Tzeng, H. F.; Blackburn, A. C.; Board, P. G.; Anders, M. W. (2000). Polymorphism- and
species-dependent inactivation of glutathione transferase zeta by dichloroacetate. Chem
Res Toxicol, 13, 231-236. http://dx.doi.org/10.1021/tx990175q
U.S. EPA. (U.S. Environmental Protection Agency). (1980). Ambient water quality criteria for
trichloroethylene. (EPA/440/5-80/077). Washington, DC: U.S. Environmental Protection
Agency, Office of Water Regulations and Standards. Retrieved from
http://nepis.epa.gov/Exe/ZyPURL.cgi?Dockey=2000M4KP.txt.
116
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
U.S. EPA. (U.S. Environmental Protection Agency). (1984). Health assessment document for
1,1,1-trichloroethane (methyl chloroform) [final report],
U.S. EPA. (U.S. Environmental Protection Agency). (1985). Health assessment document for
trichloroethylene [final report], (EPA/600/8-82/006F). Research Triangle Park, NC: U.S.
Environmental Protection Agency, Office of Research and Development. Retrieved from
http://nepis.epa.gov/Exe/ZyPURL.cgi?Dockey=2000BFCV.txt.
U.S. EPA. (U.S. Environmental Protection Agency). (1987). Addendum to the health assessment
document for trichloroethylene: updated carcinogenicity assessment for trichloroethylene
[review draft], (EPA/600/8-82/006FA). Washington, DC: U.S. Environmental Protection
Agency, Office of Research and Development. Retrieved from
http://nepis. epa.gov/Exe/ZyPURL. cgi?Dockey=3 00016WQ.txt.
U.S. EPA. (U.S. Environmental Protection Agency). (1992). Dermal exposure assessment:
Principles and applications. (EPA/600/8-91/01 IB). Washington, DC: U.S. Environmental
Protection Agency, Office of Health and Environmental Assessment. Retrieved from
http: //cfpub. epa. gov/ncea/cfm / recordi spl ay. cfm? dei d= 1218 8.
U.S. EPA. (U.S. Environmental Protection Agency). (1994). Interim policy for particle size and
limit concentration issues in inhalation toxicity studies. Washington, DC: U.S.
Environmental Protection Agency, Office of Pesticide Products, Health Effects Division.
Retrieved from http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=l86068.
U.S. EPA. (U.S. Environmental Protection Agency). (1994). Methods for derivation of inhalation
reference concentrations and application of inhalation dosimetry. (EPA/600/8-90/066F).
Washington, DC: U.S. Environmental Protection Agency, Office of Research and
Development. Retrieved from
http://cfpub. epa.gov/ncea/cfm/recordisplay. cfm?deid=71993.
U.S. EPA. (U.S. Environmental Protection Agency). (1995). National public water system
supervision program: FY 1994 compliance report. (EPA 812-R-95-001). Washington,
DC: Author. Retrieved from
http://www.ntis.gov/search/product.aspx?ABBR=PB95270534.
U.S. EPA. (U.S. Environmental Protection Agency). (1997). Exposure factors handbook.
(EPA/600/P-95/002FA-C). Washington, DC: U.S. Environmental Protection Agency,
Office of Research and Development. Retrieved from
http://cfpub. epa.gov/ncea/cfm/recordisplay. cfm?deid=12464.
U.S. EPA. (U.S. Environmental Protection Agency). (1997). Exposure factors handbook: Vol 1 -
General factors. (EPA/600/P-95/002FA). Washington, DC: U.S. Environmental
Protection Agency, Office of Research and Development. Retrieved from http://epa-
prgs.ornl.gov/radionuclides/EFH_Final_l 997_EPA600P95002Fa.pdf.
U.S. EPA. (U.S. Environmental Protection Agency). (1998). EPA data file: Office of
groundwater and drinking water. Washington, DC: Author.
U.S. EPA. (U.S. Environmental Protection Agency). (2000). Benchmark dose technical guidance
document [external review draft], (EPA/630/R-00/001). Washington, DC: U.S.
Environmental Protection Agency, Risk Assessment Forum. Retrieved from
http://www.epa.gov/raf/publications/benchmark-dose-doc-draft.htm.
U.S. EPA. (U.S. Environmental Protection Agency). (2000). Science policy council handbook:
Risk characterization. (EPA 100-B-00-002). Washington, D.C.: U.S. Environmental
117
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Protection Agency, Office of Research and Development, Office of Science Policy.
Retrieved from http://www.epa.gov/osa/spc/pdfs/rchandbk.pdf.
U.S. EPA. (U.S. Environmental Protection Agency). (2001). Sources, emission and exposure for
trichloroethylene (TCE) and related chemicals. (EPA/600/R-00/009). Washington, DC:
U.S. Environmental Protection Agency, Office of Research and Development.
U.S. EPA. (U.S. Environmental Protection Agency). (2002). A review of the reference dose and
reference concentration processes. (EPA/630/P-02/0002F). Washington, DC: U.S.
Environmental Protection Agency, Risk Assessment Forum. Retrieved from
http://cfpub. epa.gov/ncea/cfm/recordisplay. cfm?deid=51717.
U.S. EPA. (U.S. Environmental Protection Agency). (2002). Health assessment of 1,3-butadiene.
(EPA/600/P-98/001F). Washington, DC: U.S. Environmental Protection Agency, Office
of Research and Development.
U.S. EPA. (U.S. Environmental Protection Agency). (2002). OSWER draft guidance for
evaluating the vapor intrusion to indoor air pathway from groundwater and soils
(subsurface vapor intrusion guidance). (EPA 530-D-02-004). Washington, DC: Author.
Retrieved from http://epa.gov/osw/hazard/correctiveaction/eis/vapor/guidance.pdf.
U.S. EPA. (U.S. Environmental Protection Agency). (2003). Framework for cumulative risk
assessment. (EPA/630/P-02/00IF). Washington, DC: U.S. Environmental Protection
Agency, Risk Assessment Forum. Retrieved from
http://nepis. epa.gov/Exe/ZyPURL. cgi?Dockey=3 0004TJH.txt.
U.S. EPA. (U.S. Environmental Protection Agency). (2003). Occurrence estimation methodology
and occurrence findings report of the six-year review of existing National Primary
Drinking Water Regulations. (EPA-815/R-03-006). Washington, D.C.: Author.
U.S. EPA. (U.S. Environmental Protection Agency). (2003). Toxicological review of
dichloroacetic acid (CAS No. 79-43-6). (EPA 635/R-03/007). Washington, DC: Author.
Retrieved from http://www.epa.gov/iris/subst/0654.htm.
U.S. EPA. (U.S. Environmental Protection Agency). (2004). Estimated per capita water ingestion
and body weight in the United States: An update. (EPA-822/R-00-001). Washington,
D.C.: Author. Retrieved from
http://www.epa.gov/waterscience/criteria/drinking/percapita/2004.pdf.
U.S. EPA. (U.S. Environmental Protection Agency). (2005). Documentation of environmental
indicator determination: RCRA corrective action. IBM-Endicott. New York, NY:
Author. Retrieved from http://www.epa.gov/Region2/waste/ibm_e725.pdf.
U.S. EPA. (U.S. Environmental Protection Agency). (2005). Guidance on selecting age groups
for monitoring and assessing childhood exposures to environmental contaminants (Final).
(EPA/630/P-03/003F). Washington, DC: U.S. Environmental Protection Agency, Risk
Assessment Forum. Retrieved from
http://cfpub. epa.gov/ncea/cfm/recordisplay. cfm?deid=146583.
U.S. EPA. (U.S. Environmental Protection Agency). (2005). Guidelines for carcinogen risk
assessment. (EPA/630/P-03/001F). Washington, DC: U.S. Environmental Protection
Agency, Risk Assessment Forum. Retrieved from http://www.epa.gov/cancerguidelines/.
U.S. EPA. (U.S. Environmental Protection Agency). (2005). Supplemental guidance for
assessing susceptibility from early-life exposure to carcinogens. (EPA/630/R-03/003F).
Washington, DC: U.S. Environmental Protection Agency, Risk Assessment Forum.
118
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
Retrieved from http://www.epa.gov/cancerguidelines/guidelines-carcinogen-
supplement.htm.
U.S. EPA. (U.S. Environmental Protection Agency). (2006). Peer review handbook (3rd edition).
(EPA/100/B-06/002). Washington, DC: U.S. Environmental Protection Agency, Science
Policy Council. Retrieved from http://www.epa.gov/peerreview/.
U.S. EPA. (U.S. Environmental Protection Agency). (2006). Technology transfer network 1999
national-scale air toxics assessment, from
http://www.epa.gov/ttn/atw/natal999/index.html
U.S. EPA. (U.S. Environmental Protection Agency). (2007). AirData: Access to Air Pollution
Data (Version 2007). Research Triangle Park, NC: Office of Air Quality Planning and
Standards. Retrieved from http://www.epa.gov/air/data/index.html
U.S. EPA. (U.S. Environmental Protection Agency). (2008). l,l,2-Trichloro-l,2,2-
trifluoroethane. Washington, D.C.: Author. Retrieved from
http://www.epa.gov/IRIS/subst/0123.htm.
U.S. EPA. (U.S. Environmental Protection Agency). (2008). Child-specific exposure factors
handbook. (EPA/600/R-06/096F). Washington, DC: U.S. Environmental Protection
Agency, National Center for Environmental Assessment. Retrieved from
http://cfpub. epa.gov/ncea/cfm/recordisplay. cfm?deid= 199243.
U.S. EPA. (U.S. Environmental Protection Agency). (2008). Factoids: Drinking water and
ground water statistics for 2008. (EPA-816/K-08-004). Washington, D.C.: Author.
Retrieved from http://www.epa.gov/ogwdw000/databases/pdfs/data_factoids_2008.pdf.
U.S. GAO. (U.S. Government Accountability Office). (2007). Testimony before the
subcommittee on oversight and investigations, committee on energy and commerce,
house of representatives: Defense health care: Issues related to past drinking water
contamination at marine corps base Camp Lejeune. Washington, DC: Author.
U.S. GAO. (U.S. Government Accountability Office). Report to congressional committees:
Defense health care: Activities related to past drinking water contamination at marine
corps base Camp Lejeune. Washington, DC: Author.
Ulrich, R. G. (2003). The toxicogenomics of nuclear receptor agonists. Curr Opin Chem Biol, 7,
505-510. http://www.ncbi.nlm.nih.gov/pubmed/12941426
Ulrich, R. G.; Rockett, J. C.; Gibson, G. G.; Pettit, S. D. (2004). Overview of an interlaboratory
collaboration on evaluating the effects of model hepatotoxicants on hepatic gene
expression. Environ Health Perspect, 112, 423-427.
http://www.ncbi.nlm.nih.gov/pubmed/15033591
Umezu, T.; Yonemoto, J.; Soma, Y.; Miura, T. (1997). Behavioral effects of trichloroethylene
and tetrachloroethylene in mice. Pharmacol Biochem Behav, 58, 665-671.
http://dx.doi.Org/l 0.1016/S0091 -3057(97)00046-4
Urabe, Y.; Nouso, K.; Higashi, T.; Nakatsukasa, H.; Hino, N.; Ashida, K., . . . Tsuji, T. (1996).
Telomere length in human liver diseases. 16, 293-297.
http://www.ncbi.nlm.nih.gov/pubmed/8938628
USAF. (U.S. Air Force). (2004). Determination of partition coefficients for a mixture of volatile
organic compounds in rats and humans at different life stages. (AFRL-HE-WP-TR-2005-
0012). Wright-Patterson Air Force Base, OH: U.S. Air Force Research Laboratory.
119
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
USGS. (U.S. Geological Survey). (2006). Volatile organic compounds in the nation's ground
water and drinking water supply wells. Reston, VA: Author. Retrieved from
http://pubs.usgs.gov/circ/circl292/pdf/circularl292.pdf.
Vagnarelli, P.; De Sario, A.; De Carli, L. (1990). Aneuploidy induced by chloral hydrate
detected in human lymphocytes with the Y97 probe. Mutagenesis, 5, 591-592.
http://www.ncbi.nlm.nih.gov/pubmed/2263216
Vaidya, V.; Shankar, K.; Lock, E.; Bucci, T.; Mehendale, H. (2003). Renal injury and repair
following S-l, 2 dichlorovinyl-L-cysteine administration to mice. Toxicol Appl
Pharmacol, 188, 110-121. http://www.ncbi.nlm.nih.gov/pubmed/12691729
Vaidya, V.; Shankar, K.; Lock, E.; Bucci, T.; Mehendale, H. (2003). Role of tissue repair in
survival from s-(l,2-dichlorovinyl)-L-cysteine-induced acute renal tubular necrosis in the
mouse. Toxicol Sci, 74, 215-227. http://dx.doi.org/10.1093/toxsci/kfglll
Vaidya, V.; Shankar, K.; Lock, E.; Dixon, D.; Mehendale, H. (2003). Molecular mechanisms of
renal tissue repair in survival from acute renal tubule necrosis: role of ERK1/2 pathway.
Toxicol Pathol, 31, 604-618. http://www.ncbi.nlm.nih.gov/pubmed/14585728
Vainio, H.; Linnainmaa, K.; Kahonen, M.; Nickels, J.; Hietanen, E.; Marniemi, J.; Peltonen, P.
(1983). Hypolipidemia and peroxisome proliferation induced by phenoxyacetic acid
herbicides in rats. Biochem Pharmacol, 32, 2775-2779.
http://www.ncbi.nlm.nih.gov/pubmed/6626247
Vainio, H.; Nickels, J.; Linnainmaa, K. (1982). Phenoxy acid herbicides cause peroxisome
proliferation in Chinese hamsters. Scand J Work Environ Health, 8, 70-73.
http://www.ncbi.nlm.nih.gov/pubmed/7134925
Valic, E.; Waldhor, T.; Konnaris, C.; Michitsch, A.; Wolf, C. (1997). Acquired dyschromatopsia
in combined exposure to solvents and alcohol. Int Arch Occup Environ Health, 70, 403-
406. http://www.ncbi.nlm.nih.gov/pubmed/9439987
Vamvakas, S.; Bittner, D.; Koster, U. (1993). Enhanced expression of the protooncogenes c-myc
and c-fos in normal and malignant renal growth. Toxicol Lett, 67, 161-172.
http://www.ncbi.nlm.nih.gov/pubmed/8451757
Vamvakas, S.; Bruning, T.; Thomasson, B.; Lammert, M.; Baumuller, A.; Bolt, H. M., . . . Ulm,
K. (1998). Renal cell cancer correlated with occupational exposure to trichloroethene. J
Cancer Res Clin Oncol, 124, 374-382.
Vamvakas, S.; Dekant, W.; Berthold, K.; Schmidt, S.; Wild, D.; Henschler, D. (1987).
Enzymatic transformation of mercapturic acids derived from halogenated alkenes to
reactive and mutagenic intermediates. Biochem Pharmacol, 36, 2741-2748.
http://dx.doi.org/10.1016/0006-2952(87)90258-9
Vamvakas, S.; Dekant, W.; Henschler, D. (1989). Assessment of unscheduled DNA synthesis in
a cultured line of renal epithelial cells exposed to cysteine S-conjugates of haloalkenes
and haloalkanes. MutatRes, 222, 329-335. http://dx.doi.org/10.1016/0165-
1218(89)90108-0
Vamvakas, S.; Dekant, W.; Schiffmann, D.; Henschler, D. (1988). Induction of unscheduled
DNA synthesis and micronucleus formation in Syrian hamster embryo fibroblasts treated
with cysteine S-conjugates of chlorinated hydrocarbons. Cell Biol Toxicol, 4, 393-403.
http://dx.doi.org/10.1007/BF00117768
120
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Vamvakas, S.; Elfarra, A.; Dekant, W.; Henschler, D.; Anders, M. (1988). Mutagenicity of
amino acid and glutathione S-conjugates in the Ames test. Mutat Res, 206, 83-90.
http://www.ncbi.nlm.nih.gov/pubmed/2901035
Vamvakas, S.; Richter, H.; Bittner, D. (1996). Induction of dedifferentiated clones of LLC-PK1
cells upon long-term exposure to dichlorovinylcysteine. Toxicology, 106, 65-74.
http://www.ncbi.nlm.nih.gov/pubmed/8571403
van Bree, L.; Commandeur, J.; Lamberts, B.; Cornelissen, M.; van Roon, M.; Laterveer, H.; de
Vries, J. (1990). Induction of drug metabolism enzymes by dihalogenated biphenyls. J
Biochem Toxicol, 5, 57-63. http://dx.doi.org/10.1002/jbt.2570050109
Van Hummelen, P. and Kirsch-Volders, M. (1992). Analysis of eight known or suspected
aneugens by the in vitro human lymphocyte micronucleus test. Mutagenesis, 7, 447-455.
http://dx.doi.Org/10.1093/mutage/7.6.447
van Lieshout, E. M. and Peters, W. H. (1998). Age and gender dependent levels of glutathione
and glutathione S-transferases in human lymphocytes. Carcinogenesis, 19, 1873-1875.
http://www.ncbi.nlm.nih.gov/pubmed/9806172
van Lieshout, E.; Knapen, M.; Lange, W.; Steegers, E.; Peters, W. (1998). Localization of
glutathione S-transferases alpha and pi in human embryonic tissues at 8 weeks gestational
age. Hum Reprod, 13, 1380-1386. http://www.ncbi.nlm.nih.gov/pubmed/9647576
Van Duuren, B. L.; Goldschmidt, B. M.; Loewengart, G.; Smith, A. C.; Melchionne, S.;
Seidman, I.; Roth, D. (1979). Carcinogenicity of halogenated olefinic and aliphatic
hydrocarbons in mice. J Natl Cancer Inst, 63, 1433-1439.
http://www.ncbi.nlm.nih.gov/pubmed/292813
Vartiainen, T.; Pukkala, E.; Rienoja, T.; Strandman, T.; Kaksonen, K. (1993). Population
exposure to tri- and tetrachloroethene and cancer risk: Two cases of drinking water
pollution. Chemosphere, 27, 1171-1181. http://dx.doi.org/10.1016/0045-6535(93)90165-
2
Veeramachaneni, D.; Palmer, J.; Amann, R. (2001). Long-term effects on male reproduction of
early exposure to common chemical contaminants in drinking water. Hum Reprod, 16,
979-987. http://www.ncbi.nlm.nih.gOv/pubmed/l 1331649
Verma, Y. and Rana, S. (2003). Gender differences in the metabolism of benzene, toluene and
trichloroethylene in rat with special reference to certain biochemical parameters. J
Environ Biol, 24, 135-140. http://www.ncbi.nlm.nih.gov/pubmed/12974453
Vernon, R. J. and Ferguson, R. K. (1969). Effects of trichloroethylene on visual-motor
performance. Arch Environ Occup Health, 18, 894-900.
http://www.ncbi.nlm.nih.gov/pubmed/5770690
Vesterberg, O. and Astrand, I. (1976). Exposure to trichloroethylene monitored by analysis of
metabolites in blood and urine. J Occup Med, 18, 224-226.
http://www.ncbi.nlm.nih.gov/pubmed/1271124
Vesterberg, O.; Gorczak, J.; Krasts, M. (1976). Exposure to trichloroethylene II Metabolites in
blood and urine. Scand J Work Environ Health, 4, 212-219.
Vickers, A. E. and Lucier, G. W. (1996). Estrogen receptor levels and occupancy in hepatic
sinusoidal endothelial and Kupffer cells are enhanced by initiation with
diethylnitrosamine and promotion with 17alpha-ethinylestradiol in rats. Carcinogenesis,
17, 1235-1242. http://www.ncbi.nlm.nih.gov/pubmed/8681437
121
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Vieira, I.; Sonnier, M.; Cresteil, T. (1996). Developmental expression of CYP2E1 in the human
liver: Hypermethylation control of gene expression during the neonatal period. Eur J
Biochem, 238, 476-483. http://dx.doi.Org/10.llll/j.1432-1033.1996.0476z.x
Vielhauer, V.; Sarafoff, M.; Gais, P.; Rabes, H. M. (2001). Cell type-specific induction of 06-
alkylguanine-DNA alkyltransferase mRNA expression in rat liver during regeneration,
inflammation and preneoplasia. J Cancer Res Clin Oncol, 127, 591-602.
http://www.ncbi.nlm.nih.gov/pubmed/11599795
Vig, P.; Russo, F. P.; Edwards, R. J.; Tadrous, P. J.; Wright, N. A.; Thomas, H. C., . . . Forbes, S.
J. (2006). The sources of parenchymal regeneration after chronic hepatocellular liver
injury in mice. Hepatology, 43, 316-324. http://dx.doi.org/10.1002/hep.21018
Villaschi, S.; Giovanetti, A.; Lombardi, C. C.; Nicolai, G.; Garbati, M.; Andreozzi, U. (1991).
Damage and repair of mouse bronchial epithelium following acute inhalation of
trichloroethylene. Exp Lung Res, 17, 601-614.
http://www.ncbi.nlm.nih.gov/pubmed/1860457
Vineis, P.; Miligi, L.; Costantini, A. (2007). Exposure to solvents and risk of non-Hodgkin
lymphoma: clues on putative mechanisms. Cancer Epidemiol Biomarkers Prev, 16, 381-
384. http://dx.doi.org/10.1158/1055-9965.EPI-07-0124
Vogelstein, B. and Kinzler, K. W. (2004). Cancer genes and the pathways they control. Nat Med,
10, 789-799. http://dx.doi.org/10.1038/nml087
Vollmar, B.; Pradarutti, S.; Richter, S.; Menger, M. D. (2002). In vivo quantification of ageing
changes in the rat liver from early juvenile to senescent life. 22, 330-341.
http://www.ncbi.nlm.nih.gov/pubmed/12296967
Von Tungeln, L. S.; Yi, P.; Bucci, T. J.; Samokyszyn, V. M.; Chou, M. W.; Kadlubar, F. F.; Fu,
P. P. (2002). Tumorigenicity of chloral hydrate, trichloroacetic acid, trichloroethanol,
malondialdehyde, 4-hydroxy-2-nonenal, crotonaldehyde, and acrolein in the B6C3F1
neonatal mouse. Cancer Lett, 185, 13-19. http://dx.doi.org/10.1016/S0304-
3835(02)00231-8
Voss, C.; Zerban, H.; Bannasch, P.; Berger, M. R. (2005). Lifelong exposure to di-(2-
ethylhexyl)-phthalate induces tumors in liver and testes of Sprague-Dawley rats.
Toxicology, 206, 359-371. http://dx.doi.Org/10.1016/j.tox.2004.07.016
Voss, K. A.; Riley, R.; Dunn, C.; Christopher Corton, J. (2006). The role of tumor necrosis factor
alpha and the peroxisome proliferator-activated receptor alpha in modulating the effects
of fumonisin in mouse liver. Toxicology, 222, 165-174.
http://dx.doi.Org/10.1016/j.tox.2006.02.012
Wada, N.; Marsman, D. S.; Popp, J. A. (1992). Dose-related effects of the hepatocarcinogen,
Wy-14,643, on peroxisomes and cell replication. Fundam Appl Toxicol, 18, 149-154.
http://www.ncbi.nlm.nih.gov/pubmed/1601204
Wainfan, E. and Poirier, L. A. (1992). Methyl groups in carcinogenesis: effects on DNA
methylation and gene expression. Cancer Res, 52, 2071s-2077s.
http://www.ncbi.nlm.nih.gov/pubmed/1544143
Walgren, J. E.; Kurtz, D. T.; McMillan, J. M. (2000). Expression of PPAR(alpha) in human
hepatocytes and activation by trichloroacetate and dichloroacetate. Res Commun Mol
Pathol Pharmacol, 108, 116-132. http://www.ncbi.nlm.nih.gov/pubmed/11758968
Walgren, J. E.; Kurtz, D. T.; McMillan, J. M. (2000). The effect of the trichloroethylene
metabolites trichloroacetate and dichloroacetate on peroxisome proliferation and DNA
122
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
synthesis in cultured human hepatocytes. Cell Biol Toxicol, 16, 257-273.
http://dx.doi.Org/10.1023/A:1007638227821
Wallace, L. A.; Pellizzari, E.; Hartwell, T.; Rosenzweig, M.; Erickson, M.; Sparacino, C.; Zelon,
H. (1984). Personal exposure to volatile organic compounds: 1. Direct measurements in
breathing-zone air, drinking water, food and exhaled breath. Environ Res, 35, 293-319.
Walles, S. A. S. (1986). Induction of single-strand breaks in dna of mice by trichloroethylene and
tetrachloroethylene. Toxicol Lett, 31, 31-35. http://dx.doi.org/10.1016/0378-
4274(86)90191-8
Wang, G. and Stacey, N. H. (1990). Elevation of individual serum bile acids on exposure to
trichloroethylene or "alpha"-naphthylisothiocyanate. Toxicol Appl Pharmacol, 105, 209-
215. http://www.ncbi.nlm.nih.gov/pubmed/2219116
Wang, G.; Ansari, G. A.; Khan, M. F. (2007). Involvement of lipid peroxidation-derived
aldehyde-protein adducts in autoimmunity mediated by trichloroethene. J Toxicol
Environ Health A, 70, 1977-1985. http://dx.doi.org/10.1080/15287390701550888
Wang, G.; Cai, P.; Ansari, G. A. S.; Khan, M. F. (2007). Oxidative and nitrosative stress in
trichloroethene-mediated autoimmune response. Toxicology, 229, 186-193.
http://dx.doi.Org/10.1016/j.tox.2006.10.014
Wang, G.; Konig, R.; Ansari, G. A.; Khan, M. F. (2008). Lipid peroxidation-derived aldehyde-
protein adducts contribute to trichloroethene-mediated autoimmunity via activation of
CD4+ T cells. Free Radic Biol Med, 44, 1475-1482.
http://dx.doi.Org/10.1016/j.freeradbiomed.2008.01.012
Wang, G.; Takano, T.; Tomita, K.; Nakata, K.; Nakamura, K. (1993). [Interaction of
trichloroethylene and isopropyl alcohol in the perfused rat liver], Nippon Eiseigaku
Zasshi, 48, 1000-1005. http://www.ncbi.nlm.nih.gov/pubmed/8107291
Wang, J.-D. (2004). Increased mortality odds ratio of male liver cancer in a community
contaminated by chlorinated hydrocarbons in groundwater. Symposium on New Scientific
Research Related to the Health Effects of Trichloroethylene. Washington, DC.
Wang, J.-L.; Chen, W.-L.; Tsai, S.-Y.; Sung, P.-Y.; Huang, R.-N. (2001). An in vitro model for
evaluation of vaporous toxicity of trichloroethylene and tetrachloroethylene to CHO-K1
cells. Chem Biol Interact, 137, 139-154. http://dx.doi.org/10.1016/S0009-
2797(01)00226-5
Wang, R.; Nakajima, T.; Honma, T. (1999). Trichloroethylene inhibits aldehyde dehydrogenase
only for aliphatic aldehydes of short chains in rats. Toxicology, 132, 9-18.
http://www.ncbi.nlm.nih.gov/pubmed/10199577
Wang, R.; Zhang, Y.; Lan, Q.; Holford, T. R.; Leaderer, B.; Zahm, S. H., . . . Zheng, T. (2009).
Occupational exposure to solvents and risk of non-Hodgkin lymphoma in Connecticut
women. Am J Epidemiol, 169, 176-185. http://dx.doi.org/10.1093/aje/kwn300
Wang, Z.; Hall, S. D.; Maya, J. F.; Li, L.; Asghar, A.; Gorski, J. C. (2003). Diabetes mellitus
increases the in vivo activity of cytochrome P450 2E1 in humans. Br J Clin Pharmacol,
55, 77-85. http://www.ncbi.nlm.nih.gov/pubmed/12534643
Wanson, J.-C.; Bernaert, D.; Penasse, W.; Mosselmans, R.; Bannasch, P. (1980). Separation in
distinct subpopulations by elutriation of liver cells following exposure of rats to N-
nitrosomorpholine. Cancer Res, 40, 459-471.
Ward, E. M.; Thun, M. J.; Hannan, L. M.; Jemal, A. (2006). Interpreting cancer trends. Ann N Y
Acad Sci, 1076, 29-53. http://dx.doi.org/10.1196/annals.1371.048
123
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Ward, J. M.; Peters, J. M.; Perella, C. M.; Gonzalez, F. J. (1998). Receptor and nonreceptor-
mediated organ-specific toxicity of di(2-ethylhexyl)phthalate (DEHP) in peroxisome
proliferator-activated receptor alpha-null mice. Toxicol Pathol, 26, 240-246.
Warr, T.; Parry, E.; Parry, J. (1993). A comparison of two in vitro mammalian cell cytogenetic
assays for the detection of mitotic aneuploidy using 10 known or suspected aneugens.
MutatRes, 287, 29-46. http://www.ncbi.nlm.nih.gov/pubmed/7683382
Warren, D. A.; Graeter, L. J.; Channel, S. R.; Eggers, J. S.; Goodyear, C. D.; Macmahon, K. L., .
. . Baker, W. H. (2006). Trichloroethylene, trichloroacetic acid, and dichloroacetic acid:
do they affect eye development in the Sprague-Dawley rat. Int J Toxicol, 25, 279-284.
http://dx.doi.org/10.1080/10915810600745975
Wartenberg, D.; Reyner, D.; Scott, C. S. (2000). Trichloroethylene and cancer: Epidemiologic
evidence. Environ Health Perspect, 108, 161-176.
http://www.ncbi.nlm.nih.gov/pubmed/10807550
Waseem, M.; Ali, M.; Dogra, S.; Dutta, K.; Kaw, J. (2001). Toxicity of trichloroethylene
following inhalation and drinking contaminated water. J Appl Toxicol, 21, 441-444.
http://www.ncbi.nlm.nih.gov/pubmed/11746189
Waskell, L. (1978). A study of the mutagenicity of anesthetics and their metabolites. DNA
Repair, 57, 141-153. http://www.ncbi.nlm.nih.gov/pubmed/351387
Watanabe, S. and Fukui, T. (2000). Suppressive effect of curcumin on trichloroethylene-induced
oxidative stress. JNutr Sci Vitaminol, 46, 230-234.
http://www.ncbi.nlm.nih.gov/pubmed/11234915
Watanabe, T.; Shimada, Ft.; Tanaka, Y. (1978). Human hepatocytes and aging: a
cytophotometrical analysis in 35 sudden-death cases. Virchows Arch B Cell Pathol Incl
Mol Pathol, 27, 307-316. http://www.ncbi.nlm.nih.gov/pubmed/98902
Watson, R. E. (2002). Effects of Phenobarbital on DNA Methylation in GC-Rich Regions of
Hepatic DNA from Mice That Exhibit Different Levels of Susceptibility to Liver
Tumorigenesis. Toxicol Sci, 68, 51-58. http://dx.doi.Org/10.1093/toxsci/68.l.51
Watson, R. E.; Jacobson, C. F.; Williams, A. L.; Howard, W. B.; DeSesso, J. M. (2006).
Trichloroethylene-contaminated drinking water and congenital heart defects: a critical
analysis of the literature. Reprod Toxicol, 21, 117-147.
http://dx.doi.Org/10.1016/j.reprotox.2005.07.013
Waxman, S. and Wurmbach, E. (2007). De-regulation of common housekeeping genes in
hepatocellular carcinoma. BMC Genomics, 8, 243. http://dx.doi.org/10.1186/1471-2164-
8-243
Weiderpass, E.; Pukkala, E.; Kauppinen, T.; Mutanen, P.; Paakkulainen, H.; Vasama-Neuvonen,
K., . . . Partanen, T. (1999). Breast cancer and occupational exposures in women in
Finland. Am J Ind Med, 36, 48-53. http://www.ncbi.nlm.nih.gov/pubmed/10361586
Weiderpass, E.; Vainio, H.; Kauppinen, T.; Vasama-Neuvonen, K.; Partanen, T.; Pukkala, E.
(2003). Occupational exposures and gastrointestinal cancers among Finnish women. J
Occup Environ Med, 45, 305-315. http://www.ncbi.nlm.nih.gov/pubmed/12661188
Weidman, J. R.; Dolinoy, D. C.; Murphy, S. K.; Jirtle, R. L. (2007). Cancer susceptibility:
epigenetic manifestation of environmental exposures. Cancer J, 13, 9-16.
http://dx.doi.org/10.1097/PPO.0b013e31803c71f2
Weinberg, H. S.; Krasner, S. W.; Richardson, S. D.; Thruston, A. D. J. (2002). The occurrence of
disinfection by-products (DBPs) of health concern in drinking water: results of a
124
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
nationwide DBP occurrence study. (EPA/600/R-02/068). Athens, GA: U.S.
Environmental Protection Agency, National Exposure Research Laboratory.
Weisel, C. P. and Jo, W.-K. (1996). Ingestion, inhalation, and dermal exposures to chloroform
and trichloroethene from tap water. Environ Health Perspect, 104, 48-51.
Weisel, C. P.; Kim, H.; Haltmeier, P.; Klotz, J. B. (1999). Exposure estimates to disinfection by-
products of chlorinated drinking water. Environ Health Perspect, 107, 103-110.
http://www.ncbi.nlm.nih.gov/pubmed/9924004
Weisenburger, D. D. (1992). Pathological classification of non-Hodgkin's lymphoma for
epidemiological studies. Cancer Res, 52, 5456s-5462s.
Weiss, J. R.; Moysich, K. B.; Swede, H. (2005). Epidemiology of male breast cancer. Cancer
Epidemiol Biomarkers Prev, 14, 20-26. http://www.ncbi.nlm.nih.gov/pubmed/15668471
Weiss, P. A. (1971). A cell is not an island entire of itself. Perspect Biol Med, 14, 182-205.
http://www.ncbi.nlm.nih.gov/pubmed/4925931
Weistenhofer, W.; Blaszkewicz, M.; Bolt, H. M.; Golka, K. (2008). N-acetyltransferase-2 and
medical history in bladder cancer cases with a suspected occupational disease (BK 1301)
in Germany. J Toxicol Environ Health A, 71, 906-910.
http://dx.doi.org/10.1080/15287390801988681
Wells, G. M.; Schroth, W.; Brauch, H.; Ross, E. A. (2009). Bilateral renal-cell carcinoma
associated with an acquired VHL mutation and long-term trichloroethylene exposure.
Clin Nephrol, 71, 708-713. http://www.ncbi.nlm.nih.gov/pubmed/19473641
Wendling, O.; Chambon, P.; Mark, M. (1999). Retinoid X receptors are essential for early mouse
development and placentogenesis. PNAS, 96, 547-551.
http://www.ncbi.nlm.nih.gov/pubmed/9892670
Werner, H. and Le Roith, D. (2000). New concepts in regulation and function of the insulin-like
growth factors: implications for understanding normal growth and neoplasia. Cell Mol
Life Sci, 57, 932-942. http://www.ncbi.nlm.nih.gov/pubmed/10950308
Werner, M.; Birner, G.; Dekant, W. (1995). The role of cytochrome P4503A1/2 in the sex-
specific sulfoxidation of the hexachlorobutadiene metabolite, N-acetyl-S-
(pentachlorobutadienyl)-L-cysteine in rats. Drug Metab Dispos, 23, 861-868.
http://www.ncbi.nlm.nih.gov/pubmed/7493554
Werner, M.; Birner, G.; Dekant, W. (1996). Sulfoxidation of mercapturic acids derived from tri-
and tetrachloroethene by cytochromes P450 3 A: A bioactivation reaction in addition to
deacetylation and cysteine conjugate beta-lyase mediated cleavage. Chem Res Toxicol, 9,
41-49. http://dx.doi.org/10.1021/tx950075u
Werner, M.; Guo, Z.; Birner, G.; Dekant, W.; Guengerich, F. (1995). The sulfoxidation of the
hexachlorobutadiene metabolite N-acetyl-S-(l,2,3,4,4-pentachlorobutadienyl)-L-cysteine
is catalyzed by human cytochrome P450 3A enzymes. Chem Res Toxicol, 8, 917-923.
http://www.ncbi.nlm.nih.gov/pubmed/8555406
West, G. B.; Woodruff, W. H.; Brown, J. H. (2002). Allometric scaling of metabolic rate from
molecules and mitochondria to cells and mammals. PNAS, 99, 2473-2478.
Westergren, I.; Kjellstrand, P.; Linder, L.; Johansson, B. (1984). Reduction of brain specific
gravity in mice prenatally exposed to trichloroethylene. Toxicol Lett, 23, 223-226.
http://www.ncbi.nlm.nih.gov/pubmed/6506097
125
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
Wheeler, C. W. and Guenthner, T. M. (1990). Spectroscopic quantitation of cytochrome P-450 in
human lung microsomes. J Biochem Toxicol, 5, 269-272.
http://www.ncbi.nlm.nih.gov/pubmed/2096222
Wheeler, C. W.; Wrighton, S. A.; Guenthner, T. M. (1992). Detection of human lung
cytochromes P-450 that are immunochemically related to cytochrome P450IIE1 and
cytochrome P-450TTTA Biochem Pharmacol, 44, 183-186.
http://www.ncbi.nlm.nih.gov/pubmed/1632834
White, A. E.; Takehisa, S.; Eger EI, 1.1.; Wolff, S.; Stevens, W. C. (1979). Sister chromatid
exchanges induced by inhaled anesthetics. 50, 426-430.
White, J. F. and Carlson, G. P. (1981). Epinephrine-induced cardiac arrhythmias in rabbits
exposed to trichloroethylene: potentiation by ethanol. Toxicol Appl Pharmacol, 60, 466-
471.
White, J. F. and Carlson, G. P. (1981). Epinephrine-induced cardiac arrhythmias in rabbits
exposed to trichloroethylene: role of trichloroethylene metabolites. Toxicol Appl
Pharmacol, 60, 458-465.
White, K.; David, D.; Butterworth, L.; Klykken, P. (2000). Assessment of autoimmunity-
inducing potential using the brown Norway rat challenge model. Toxicol Lett, 112-113,
443-451. http://www.ncbi.nlm.nih.gov/pubmed/10720764
White, R.; Feldman, R.; Eviator, I.; Jabre, J.; Niles, C. (1997). Hazardous waste and
neurobehavioral effects: a developmental perspective. Environ Res, 73, 113-124.
http://dx.doi.org/10.1006/enrs.1997.3699
WHO. (World Health Organization). (1977). Manual of the international statistical classification
of diseases, injuries, and causes of death. Geneva, Switzerland: Author.
WHO. (World Health Organization). (1985). Environmental health criteria: Trichloroethylene.
Geneva, Switzerland: Author.
Wideroff, L.; Gridley, G.; Mellemkjaer, L.; Chow, W. H.; Linet, M.; Keehn, S., . . . Olsen, J. H.
(1997). Cancer incidence in a population-based cohort of patients hospitalized with
diabetes mellitus in Denmark. J Natl Cancer Inst, 89, 1360-1365.
http://www.ncbi.nlm.nih.gov/pubmed/9308706
Wiesenhiitter, B.; Selinski, S.; Golka, K.; Briining, T.; Bolt, H. M. (2007). Re-assessment of the
influence of polymorphisms of phase-II metabolic enzymes on renal cell cancer risk of
trichloroethylene-exposed workers. Int Arch Occup Environ Health, 81, 247-251.
http://dx.doi.org/10.1007/s00420-007-0200-5
Wigle, D. T.; Turner, M. C.; Gomes, J.; Parent, M. E. (2008). Role of hormonal and other factors
in human prostate cancer. J Toxicol Environ Health B Crit Rev, 11, 242-259.
http://dx.doi.org/10.1080/10937400701873548
Wilcosky, T. C.; Checkoway, H.; Marshall, E. G.; Tyroler, H. A. (1984). Cancer mortality and
solvent exposures in the rubber industry. AIHA J, 45, 809-811.
http://www.ncbi.nlm.nih.gov/pubmed/6549104
Wilkens, L.; Flemming, P.; Gebel, M.; Bleck, J.; Terkamp, C.; Wingen, L., . . . Schlegelberger,
B. (2004). Induction of aneuploidy by increasing chromosomal instability during
dedifferentiation of hepatocellular carcinoma. PNAS, 101, 1309-1314.
http://dx.doi.org/10.1073/pnas.0305817101
126
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
DRAFT- DO NOT QUOTE OR CITE
Williams, A. L. and DeSesso, J. M. (2008). Trichloroethylene and ocular malformations:
analysis of extant literature. Int J Toxicol, 27, 81-95.
http://dx.doi.org/10.1080/10915810701876760
Williams, F.; Meenagh, A.; Darke, C.; Acosta, A.; Daar, A. S.; Gorodezky, C., . . . Middleton, D.
(2001). Analysis of the distribution of HLA-B alleles in populations from five continents.
Hum Immunol, 62, 645-650. http://www.ncbi.nlm.nih.gov/pubmed/11390040
Williams, P.; Benton, L.; Warmerdam, J.; Sheehan, P. (2002). Comparative risk analysis of six
volatile organic compounds in California drinking water. Environ Sci Technol, 36, 4721-
4728. http://www.ncbi.nlm.nih.gov/pubmed/12487291
Wilson, J. G. (1965). Methods for administering agents and detecting malformations in
experimental animals. In J. Warkany (Ed.), Teratology: Principles and techniques (8 ed.,
pp. 262-267). Chicago, IL: University of Chicago Press.
Wilson, R.; Donahue, M.; Gridley, G.; Adami, J.; El Ghormli, L.; Dosemeci, M. (2008). Shared
occupational risks for transitional cell cancer of the bladder and renal pelvis among men
and women in Sweden. Am J Ind Med, 51, 83-99. http://dx.doi.org/10.1002/ajim.20522
Windemuller, F. J. B. and Ettema, J. H. (1978). Effects of combined exposure to
trichloroethylene and alcohol on mental capacity. Int Arch Occup Environ Health, 41, 77-
85. http://www.ncbi.nlm.nih.gov/pubmed/631909
Windham, G. C.; Shusterman, D.; Swan, S. H.; Fenster, L.; Eskenazi, B. (1991). Exposure to
organic solvents and adverse pregnancy outcome. Am J Ind Med, 20, 241-259.
http://dx.doi.org/10.1002/ajim.4700200210
Windham, G. C.; Zhang, L.; Gunier, R.; Croen, L. A.; Grether, J. K. (2006). Autism spectrum
disorders in relation to distribution of hazardous air pollutants in the San Francisco Bay
area. Environ Health Perspect, 114, 1438-1444.
http://www.ncbi.nlm.nih.gov/pubmed/16966102
Winneke, G. (1982). Acute behavioral effects of exposure to some organic solvents -
psychophysiological aspects. Acta Neurol Scand, 92, 117-129.
http://www.ncbi.nlm.nih.gov/pubmed/6962647
Winneke, G.; Fodor, G.; Schlipkoter, H. (1978). Carbon monoxide, trichloroethylene, and
alcohol: reliability and validity of neurobehavioral effects.
Winneke, G.; Kramer, U.; Kastka, J. (1976). Zur Beeinflussung psychomotor Leistungen Durch
Alkohol und durch verschiedene Losungsmitteldampfe. In M. Horvath (Ed.), Adverse
effects of environmental chemicals and psychotropic drugs (Vol. 2, pp. 99-110).
Amsterdam, The Netherlands: Elsevier.
Withey, J. R. and Karpinski, K. (1985). The fetal distribution of some aliphatic chlorinated
hydrocarbons in the rat after vapor phase exposure. Biol Res Pregnancy Perinatol, 6, 79-
88. http://www.ncbi.nlm.nih.gov/pubmed/4016168
Withey, J. R.; Collins, B. T.; Collins, P. G. (1983). Effect of vehicle on the pharmacokinetics and
uptake of four halogenated hydrocarbons from the gastrointestinal tract of the rat. J Appl
Toxicol, 3, 249-253. http://dx.doi.org/10.1002/jat.2550030506
Wolf, D. C.; Goldsworthy, T. L.; Janszen, D. B.; Harden, R.; Donner, E. M.; David, C. S.;
Everitt, J. I. (2000). Promotion by sodium barbital induces early development but does
not increase the multiplicity of hereditary renal tumors in Eker rats. Carcinogenesis, 21,
1553-1558. http://www.ncbi.nlm.nih.gov/pubmed/10910958
127
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
Wolff, D. L. and Siegmund, R. (1978). The circadian dependent effect of trichloroethylene on
spontaneous locomotor activity and of tetrachloroethane on mortality in mice. 97, 345-
352.
Wolfgang, G.; Gandolfi, A.; Stevens, J.; Brendel, K. (1989). In vitro and in vivo nephrotoxicity
of the L and D isomers of S-(l,2-dichlorovinyl)-cysteine. Toxicology, 58, 33-42.
http://www.ncbi.nlm.nih.gov/pubmed/2815092
Wolfgang, G.; Gandolfi, A.; Stevens, J.; Brendel, K. (1989). N-acetyl S-(l,2-dichlorovinyl)-L-
cysteine produces a similar toxicity to S-(l,2-dichlorovinyl)-L-cysteine in rabbit renal
slices: differential transport and metabolism. Toxicol Appl Pharmacol, 101, 205-219.
http://www.ncbi.nlm.nih.gov/pubmed/2815079
Wood, L. D.; Parsons, D. W.; Jones, S.; Lin, J.; Sjoblom, T.; Leary, R. J., . . . Vogelstein, B.
(2007). The genomic landscapes of human breast and colorectal cancers. Science, 318,
1108-1113. http://dx.doi.Org/10.l 126/science.l 145720
Woodruff, T. J.; Axelrad, D. A.; Caldwell, J.; Morello-Frosch, R.; Rosenbaum, A. (1998). Public
health implications of 1990 air toxics concentrations across the United States. Environ
Health Perspect, 106, 245-251. http://www.ncbi.nlm.nih.gov/pubmed/9518474
Woods, C. G.; Burns, A. M.; Bradford, B. U.; Ross, P. K.; Kosyk, O.; Swenberg, J. A., . . .
Rusyn, I. (2007). WY-14,643-induced cell proliferation and oxidative stress in mouse
liver are independent of NADPH oxidase. Toxicol Sci, 98, 366-374.
http://www.ncbi.nlm.nih.gov/pubmed/17483499
Woods, C. G.; Heuvel, J. P.; Rusyn, I. (2007). Genomic profiling in nuclear receptor-mediated
toxicity. Toxicol Pathol, 35, 474-494. http://dx.doi.org/10.1080/01926230701311351
WOOLF, B. (1955). On estimating the relation between blood group and disease. Ann Hum
Genet, 19, 251-253. http://www.ncbi.nlm.nih.gov/pubmed/14388528
Woolhiser, M. R.; Krieger, S. M.; Thomas, J.; Hotchkiss, J. A. (2006). Trichloroethylene (TCE):
Immunotoxicity potential in CD rats following a 4-week vapor inhalation exposure.
Midland, MI.
Wright, P. F. A.; Thomas, W. D.; Stacey, N. H. (1991). Effects of trichloroethylene on hepatic
and splenic lymphocytotoxic activities in rodents. Toxicology, 70, 231-242.
http://www.ncbi.nlm.nih.gov/pubmed/1763417
Wu, C. and Schaum, J. (2000). Exposure assessment of trichloroethylene. Environ Health
Perspect, 108(Suppl 2), 359-363.
Wu, C. and Schaum, J. (2001). Sources, emissions and exposures for trichloroethylene (TCE)
and related chemicals. (EPA/600/R-00/099). Washington, DC: U.S. Environmental
Protection Agency.
Wu, K. L. and Berger, T. (2007). Trichloroethylene metabolism in the rat ovary reduces oocyte
fertilizability. Chem Biol Interact, 170, 20-30.
http://dx.doi.Org/10.1016/j.cbi.2007.06.038
Wu, K. L. and Berger, T. (2008). Ovarian gene expression is stable after exposure to
trichloroethylene. Toxicol Lett, 177, 59-65. http://dx.doi.Org/10.1016/j.toxlet.2007.12.008
Wu, X. Z. and Chen, D. (2006). Origin of hepatocellular carcinoma: role of stem cells. J
Gastroenterol Hepatol, 21, 1093-1098. http://dx.doi.0rg/lO.llll/j.1440-
1746.2006.04485.x
Xu, H.; Tanphaichitr, N.; Forkert, P.; Anupriwan, A.; Weerachatyanukul, W.; Vincent, R., . . .
Wade, M. (2004). Exposure to trichloroethylene and its metabolites causes impairment of
128
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
DRAFT- DO NOT QUOTE OR CITE
sperm fertilizing ability in mice. Toxicol Sci, 82, 590-597.
http://dx.doi.org/10.1093/toxsci/kfh277
Xu, W. and Adler, I. (1990). Clastogenic effects of known and suspect spindle poisons studied
by chromosome analysis in mouse bone marrow cells. Mutagenesis, 5, 371-374.
http://www.ncbi.nlm.nih.gov/pubmed/2398818
Xu, X.; Yang, R.; Wu, N.; Zhong, P.; Ke, Y.; Zhou, L., . . . Wu, B. (2009). Severe
hypersensitivity dermatitis and liver dysfunction induced by occupational exposure to
trichloroethylene. Ind Health, 47, 107-112.
http://www.ncbi.nlm.nih.gov/pubmed/19367038
Yamamura, K.; Ikeda, T.; Sadmoto, T.; Maehara, N.; Harabuchi, I.; Takashima, H.; Kiyosawa,
H. (1983). Effects of trichloroethylene exposure on hearing: an investigation of cochlear
microphonics and action potential of the guinea pig. Eur J Appl Physiol, 52, 47-50.
http://www.ncbi.nlm.nih.gov/pubmed/6686127
Yang, Q.; Ito, S.; Gonzalez, F. J. (2007). Hepatocyte-restricted constitutive activation of PPAR
alpha induces hepatoproliferation but not hepatocarcinogenesis. Carcinogenesis, 28,
1171-1177. http://dx.doi.org/10.1093/carcin/bgm046
Yano, M.; Yamamoto, T.; Kobayashi, S.; Ikeda, Y.; Matsuzaki, M. (2008). Defective Ca2+
cycling as a key pathogenic mechanism of heart failure. Circ J, 72 Suppl A, A22-A30.
http://www.ncbi.nlm.nih.gov/pubmed/18772523
Yauck, J.; Malloy, M.; Blair, K.; Simpson, P.; McCarver, D. (2004). Proximity of residence to
trichloroethylene-emitting sites and increased risk of offspring congenital heart defects
among older women. Birth Defects Res A Clin Mol Teratol, 70, 808-814.
http://dx.doi.org/10.1002/bdra.20060
Yeh, M. M.; Larson, A. M.; Campbell, J. S.; Fausto, N.; Rulyak, S. J.; Swanson, P. E. (2007).
The expression of transforming growth factor-alpha in cirrhosis, dysplastic nodules, and
hepatocellular carcinoma: an immunohistochemical study of 70 cases. Am J Surg Pathol,
31, 681-689. http://dx.doi.org/10.1097/PAS.0b013e31802ff7aa
Yeldandi, A. V.; Milano, M.; Subbarao, V.; Reddy, J. K.; Rao, M. S. (1989). Evaluation of liver
cell proliferation during ciprofibrate-induced hepatocarcinogenesis. Cancer Lett, 47, 21-
27. http://www.ncbi.nlm.nih.gov/pubmed/2636030
Yeung, R. S.; Xiao, G. H.; Everitt, J. I.; Jin, F.; Walker, C. L. (1995). Allelic loss at the tuberous
sclerosis 2 locus in spontaneous tumors in the Eker rat. Mol Carcinog, 14, 28-36.
http://www.ncbi.nlm.nih.gov/pubmed/7546222
Yoon, H. S.; Monks, T. J.; Walker, C. L.; Lau, S. S. (2001). Transformation of kidney epithelial
cells by a quinol thioether via inactivation of the tuberous sclerosis-2 tumor suppressor
gene. Mol Carcinog, 31, 37-45. http://www.ncbi.nlm.nih.gov/pubmed/11398196
Yoon, J. S.; Mason, J. M.; Valencia, R.; Woodruff, R. C.; Zimmering, S. (1985). Chemical
mutagenesis testing in Drosophila. IV. Results of 45 coded compounds tested for the
National Toxicology Program. Environ Mutagen, 7, 349-367.
http://dx.doi.org/10.1002/em.2860070310
Yoon, M.; Madden, M. C.; Barton, H. A. (2007). Extrahepatic metabolism by CYP2E1 in PBPK
modeling of lipophilic volatile organic chemicals: Impacts on metabolic parameter
estimation and prediction of dose metrics. J Toxicol Environ Health A, 70, 1527-1541.
Yoshida, M.; Fukabori, S.; Hara, K.; Yuasa, H.; Nakaaki, K.; Yamamura, Y.; Yoshida, K.
(1996). Concentrations of trichloroethylene and its metabolites in blood and urine after
129
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
acute poisoning by ingestion. Hum Exp Toxicol, 15, 254-258.
http://www.ncbi.nlm.nih.gov/pubmed/8839215
Yoshioka, T.; Krauser, J. A.; Guengerich, F. P. (2002). Microsomal oxidation of
tribromoethylene and reactions of tribromoethylene oxide. Chem Res Toxicol, 15, 1414-
1420. http://www.ncbi.nlm.nih.gov/pubmed/12437332
Yu, K. O.; Barton, H. A.; Mahle, D. A.; Frazier, J. M. (2000). In vivo kinetics of trichloroacetate
in male Fischer 344 rats. Toxicol Sci, 54, 302-311.
http://www.ncbi.nlm.nih.gov/pubmed/10774812
Yu, M. C.; Garabrant, D. H.; Peters, J. M.; Mack, T. M. (1988). Tobacco, alcohol, diet,
occupation, and carcinoma of the esophagus. Cancer Res, 48, 3843-3848.
http://www.ncbi.nlm.nih.gov/pubmed/3378219
Yu, S.; Rao, S.; Reddy, J. (2003). Peroxisome proliferator-activated receptors, fatty acid
oxidation, steatohepatitis and hepatocarcinogenesis. Curr Mol Med, 3, 561-572.
http://www.ncbi.nlm.nih.gov/pubmed/14527087
Yuan, J. M.; Castelao, J. E.; Gago-Dominguez, M.; Yu, M. C.; Ross, R. K. (1998). Tobacco use
in relation to renal cell carcinoma. Cancer Epidemiol Biomarkers Prev, 7, 429-433.
http://www.ncbi.nlm.nih.gov/pubmed/9610793
Yue, F.; Huang, H. L.; Huang, J. X.; Liang, L. Y.; Huang, Z. L.; Wei, Q. Y.; Chen, X. M. (2007).
HLA-DM polymorphism and risk of trichloroethylene induced medicamentosa-like
dermatitis. Biomed Environ Sci, 20, 506-511.
http://www.ncbi.nlm.nih.gov/pubmed/18348411
Zack, M.; Cannon, S.; Loyd, D.; Heath, C. W.; Falletta, J. M.; Jones, B., . . . Crowley, S. (1980).
Cancer in children of parents exposed to hydrocarbon-related industries and occupations.
Am J Epidemiol, 111, 329-336. http://www.ncbi.nlm.nih.gov/pubmed/7361757
Zahm, S. H. (1992). Computerized occupational referent population system (CORPS): Study
documentation. Rockville, MD: National Cancer Institute & National Institute for
Occupational Safety and Health.
Zajicek, G. and Schwartz-Arad, D. (1990). Streaming liver. VII: DNA turnover in acinus zone-3.
10, 137-140. http://www.ncbi.nlm.nih.gov/pubmed/2385154
Zajicek, G.; Arber, N.; Schwartz-Arad, D. (1991). Streaming liver. VIII: Cell production rates
following partial hepatectomy. 11, 347-351.
http://www.ncbi.nlm.nih.gov/pubmed/1779714
Zajicek, G.; Schwartz-Arad, D.; Bartfeld, E. (1989). The streaming liver. V: Time and age-
dependent changes of hepatocyte DNA content, following partial hepatectomy. 9, 164-
171. http://www.ncbi.nlm.nih.gov/pubmed/2747441
Zandi-Nejad, K.; Eddy, A.; Glassock, R.; Brenner, B. (2004). Why is proteinuria an ominous
biomarker of progressive kidney disease. Kidney Int SupplS76-S89.
http://dx.doi.Org/10.l 111/j. 1523-1755.2004.09220.X
Zanelli, U.; Puccini, P.; Acerbi, D.; Ventura, P.; Gervasi, P. G. (1996). Induction of peroxisomal
beta-oxidation and P-450 4A-dependent activities by pivalic and trichloroacetic acid in
rat liver and kidney. Arch Toxicol, 70, 145-149.
Zeise, L.; Wilson, R.; Crouch, E. A. C. (1987). Dose-response relationships for carcinogens: a
review. Environ Health Perspect, 73, 259-308.
130
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
DRAFT- DO NOT QUOTE OR CITE
Zenick, H.; Blackburn, K.; Hope, E.; Richdale, N.; Smith, M. K. (1984). Effects of
trichloroethylene exposure on male reproductive function in rats. Toxicology, 31, 237-
250. http://www.ncbi.nlm.nih.gov/pubmed/6740699
Zeppa, P.; Benincasa, G.; Troncone, G.; Lucariello, A.; Zabatta, A.; Cochand-Priollet, B., . . .
Palombini, L. (1998). Retrospective evaluation of DNA ploidy of hepatocarcinoma on
cytologic samples. Diagn Cytopathol, 19, 323-329.
http://www.ncbi.nlm.nih.gov/pubmed/9812223
Zhang, B.; Pan, X.; Cobb, G. P.; Anderson, T. A. (2007). microRNAs as oncogenes and tumor
suppressors. Dev Biol, 302, 1-12. http://dx.doi.Org/10.1016/j.ydbio.2006.08.028
Zhang, G. and Stevens, J. (1989). Transport and activation of S-(l,2-dichlorovinyl)-L-cysteine
and N-acetyl-S-(l,2-dichlorovinyl)-L-cysteine in rat kidney proximal tubules. Toxicol
Appl Pharmacol, 100, 51-61. http://www.ncbi.nlm.nih.gov/pubmed/2763302
Zhang, S. H.; Miao, D. Y.; Liu, A. L.; Zhang, L.; Wei, W.; Xie, H.; Lu, W. Q. (2010).
Assessment of the cytotoxicity and genotoxicity of haloacetic acids using microplate-
based cytotoxicity test and CHO/HGPRT gene mutation assay. Mutat Res, 703, 174-179.
http://dx.doi.Org/10.1016/j.mrgentox.2010.08.014
Zhang, Y.; Cantor, K. P.; Lynch, C. F.; Zheng, T. (2004). A population-based case-control study
of occupation and renal cell carcinoma risk in Iowa. J Occup Environ Med, 46, 235-240.
http://www.ncbi.nlm.nih.gov/pubmed/15091286
Zhao, Y.; Krishnadasan, A.; Kennedy, N.; Morgenstern, H.; Ritz, B. (2005). Estimated effects of
solvents and mineral oils on cancer incidence and mortality in a cohort of aerospace
workers. Am J Ind Med, 48, 249-258. http://dx.doi.org/10.1002/ajim.20216
Zhou, H.; Randers-Pehrson, G.; Geard, C. R.; Brenner, D. J.; Hall, E. J.; Hei, T. K. (2003).
Interaction between radiation-induced adaptive response and bystander mutagenesis in
mammalian cells. RadiatRes, 160, 512-516.
http://www.ncbi.nlm.nih.gov/pubmed/14565832
Zhou, Y. C. and Waxman, D. J. (1998). Activation of peroxisome proliferator-activated receptors
by chlorinated hydrocarbons and endogenous steroids. Environ Health Perspect, 106
Suppl 4, 983-988. http://www.ncbi.nlm.nih.gov/pubmed/9703482
Zhu, J.; Newhook, R.; Marro, L.; Chan, C. (2005). Selected volatile organic compounds in
residential air in the city of Ottawa, Canada. Environ Sci Technol, 39, 3964-3971.
http://www.ncbi.nlm.nih.gov/pubmed/15984771
Zielinska-Psuja, B.; Orlowski, J.; Kowalowka, J.; Zieba-Proc, B. (1998). Metabolic implications
of benzene and trichloroethylene. Part I. The effect trichloroethylene on phenol
elimination in rats. Bromatol Chem Toksykol, 31, 13-20.
Zielinski, A. (1973). General health state of women professionally exposed to trichloroethylene
vapours. Med Pr, 24, 263-271.
Zordan, M.; Osti, M.; Pesce, M.; Costa, R. (1994). Chloral hydrate is recombinogenic in the
wing spot test in Drosophila melanogaster. Mutat Res, 322, 111-116.
http://www.ncbi.nlm.nih.gov/pubmed/7519317
131
------- |