DRAFT - DO NOT CITE OR QUOTE EPA/635/R-09/001D
www.epa.gov/iris
TOXICOLOGICAL REVIEW
OF
1,1,2,2-TETRACHLOROETHANE
(CAS No. 79-34-5)
In Support of Summary Information on the
Integrated Risk Information System (IRIS)
July 2010
NOTICE
This document is an Interagency Science Discussion/Final Agency 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 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 review 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.
DRAFT - DO NOT CITE OR QUOTE
-------
CONTENTS - TOXICOLOGICAL REVIEW OF
1,1,2,2-TETRACHLOROETHANE (CAS No. 79-34-5)
LIST OF TABLES vi
LIST OF FIGURES ix
LIST OF ABBREVIATIONS AND ACRONYMS x
FOREWORD xn
AUTHORS, CONTRIBUTORS, AND REVIEWERS xm
1. INTRODUCTION 1
2. CHEMICAL AND PHYSICAL INFORMATION 3
3. TOXICOKINETICS 5
3.1. ABSORPTION 5
3.1.1. Oral Exposure 5
3.1.2. Inhalation Exposure 6
3.2. DISTRIBUTION 6
3.3. METABOLISM 7
3.4. ELIMINATION 10
3.5. PHYSIOLOGICALLY BASED TOXICOKINETIC MODELS 11
4. HAZARD IDENTIFICATION 13
4.1. STUDIES IN HUMANS—EPIDEMIOLOGY, CASE REPORTS, CLINICAL
CONTROLS 13
4.1.1. Oral Exposure 13
4.1.2. Inhalation Exposure 13
4.2. SUBCHRONIC AND CHRONIC STUDIES AND CANCER BIOASSAYS IN
ANIMALS—ORAL AND INHALATION 16
4.2.1. Oral Exposure 16
4.2.1.1. Subchronic Studies 16
4.2.1.2. Chronic Studies 27
4.2.2. Inhalation Exposure 32
4.2.2.1. Subchronic Studies 32
4.2.2.2. Chronic Studies 34
4.3. REPRODUCTIVE/DEVELOPMENTAL STUDIES—ORAL AND
INHALATION 35
4.3.1. Oral Exposure 35
4.3.2. Inhalation Exposure 38
4.4. OTHER DURATION- OR ENDPOINT-SPECIFIC STUDIES 38
4.4.1. Acute Studies (Oral and Inhalation) 39
4.4.1.1. Oral Studies 39
4.4.1.2. Inhalation Studies 40
4.4.2. Short-term Studies (Oral and Inhalation) 42
4.4.2.1. Oral Studies 42
4.4.2.2. Short-term Inhalation Studies 48
4.4.3. Acute Injection Studies 49
111 DRAFT - DO NOT CITE OR QUOTE
-------
4.4.4. Immunotoxicological Studies 50
4.5. MECHANISTIC DATA AND OTHER STUDIES IN SUPPORT OF THE MODE
OF ACTION 50
4.5.1. Genotoxicity 51
4.5.2. Short-Term Tests of Carcinogenicity 54
4.6. SYNTHESIS OF MAJOR NONCANCER EFFECTS 56
4.6.1. Oral 56
4.6.1.1. Human Data 56
4.6.1.2. Animal Data 56
4.6.2. Inhalation 62
4.6.2.1. Human Data 62
4.6.2.2. Animal Data 64
4.6.3. Mode of action Information 69
4.7. EVALUATION OF CARCINOGENICITY 70
4.7.1. Summary of Overall Weight of Evidence 70
4.7.2. Synthesis of Human, Animal, and Other Supporting Evidence 71
4.7.3. Mode of action Information 73
4.8. SUSCEPTIBLE POPULATIONS AND LIFE STAGES 74
4.8.1. Possible Childhood Susceptibility 74
4.8.2. Possible Gender Differences 74
4.8.3. Other Susceptible Populations 75
5. DOSE-RESPONSE ASSESSMENTS 76
5.1. ORAL REFERENCE DOSE (RfD) 76
5.1.1. Subchronic Oral RfD 76
5.1.1.1. Choice of Principal Study and Critical Effect—with Rationale and
Justification 76
5.1.1.2. Methods of Analysis—Including Models (PBPK, BMD, etc.) 79
5.1.1.3. RfD Derivation—Including Application of Uncertainty Factors (UFs) 81
5.1.2. Chronic Oral RfD 82
5.1.2.1. Choice of Principal Study and Critical Effect - with Rationale and
Justification 82
5.1.2.2. Methods of Analysis—Including Models (PBPK, BMD, etc.) 82
5.1.2.3. RfD Derivation—Including Application of UFs 83
5.1.3. RfD Comparison Information 84
5.1.4. Previous RfD Assessment 90
5.2. INHALATION REFERENCE CONCENTRATION (RfC) 90
5.2.1. Choice of Principal Study and Critical Effect—with Rationale and
Justification 90
5.2.2. Methods of Analysis—Including Models (PBPK, BMD, etc.) 91
5.2.3. Previous RfC Assessment 92
5.3. UNCERTAINTIES IN THE INHALATION REFERENCE CONCENTRATION
(RfC) AND ORAL REFERENCE DOSE (RfD) 92
5.4.1. Choice of Study/Data—with Rationale and Justification 95
5.4.2. Dose-response Data 95
5.4.3. Dose Adjustments and Extrapolation Method(s) 96
5.4.4. Oral Slope Factor and Inhalation Unit Risk 98
5.4.5. Uncertainties in Cancer Risk Values 98
5.4.6. Previous Cancer Assessment 101
iv DRAFT - DO NOT CITE OR QUOTE
-------
6. MAJOR CONCLUSIONS IN THE CHARACTERIZATION OF HAZARD AND
DOSE RESPONSE 102
6.1. HUMAN HAZARD POTENTIAL 102
6.2. DOSE RESPONSE 103
6.2.1. Noncancer/Oral 103
6.2.2. Noncancer/Inhalation 109
6.2.3. Cancer/Oral and Inhalation 110
7. REFERENCES 113
APPENDIX A. SUMMARY OF EXTERNAL PEER REVIEW AND PUBLIC
COMMENTS AND DISPOSITION A-l
APPENDIX B. BENCHMARK DOSE MODELING RESULTS FOR THE DERIVATION
OFTHERFD B-A
APPENDIX C. BENCHMARK DOSE MODELING RESULTS FOR THE DERIVATION
OF THE ORAL SLOPE FACTOR C-l
v DRAFT - DO NOT CITE OR QUOTE
-------
LIST OF TABLES
2-1. Chemical and physical properties of 1,1,2,2-tetrachloroethane .
4-1. Final body weights (g) and percent change compared to controls in F344/N rats
exposed to 1,1,2,2-tetrachloroethane in feed for 14 weeks 17
4-2a. Absolute liver weights (g) and percent change compared to controls in F344/N
rats exposed to 1,1,2,2-tetrachloroethane in feed for 14 weeks 17
4-2b. Relative liver weight (mg organ weight/g body weight) and percent change compared
to controls in F344/N rats exposed to 1,1,2,2-tetrachloroethane in feed for 14 weeks 18
4-3. Serum chemistry and hematology changes in rats exposed to dietary 1,1,2,2-tetra-
chloroethane for 14 weeks 19
4-4. Incidences of selected histopathological lesions in rats exposed to dietary 1,1,2,2-tetra-
chlorethane for 14 weeks 21
4-5. Final body weights (g) and percent change compared to controls in B6C3Fi mice exposed
to 1,1,2,2-tetrachloroethane in feed for 14 weeks 23
4-6a. Absolute liver weights (g) and percent change compared to controls in B6C3Fi
mice exposed to 1,1,2,2-tetrachloroethane in feed for 14 weeks 24
4-6b. Relative liver weights (mg organ weight/g body weight) and percent change compared
to controls in B6C3Fi mice exposed to 1,1,2,2-tetrachloroethane in feed for 14 weeks ... 24
4-7. Selected clinical chemistry changes in male mice exposed to dietary 1,1,2,2-tetra-
chloroethane for 14 weeks 25
4-8. Selected clinical chemistry changes in female mice exposed to dietary 1,1,2,2-tetra-
chloroethane for 14 weeks 26
4-9. Incidences of selected histopathological lesions in mice exposed to dietary 1,1,2,2-tetra-
chloroethane for 14 weeks 27
4-10. Incidence of neoplasms in male Osborne-Mendel rats exposed to 1,1,2,2-tetrachloro-
ethane in feed for 78 weeks 29
4-11. Incidence of neoplasms in female Osborne-Mendel rats exposed to 1,1,2,2-tetra-
chloroethane in feed for 78 weeks 30
4-12. Incidence of hepatocelluar carcinomas in male and female B6C3Fi mice exposed to
1,1,2,2-tetrachloroethane in feed for 78 weeks 32
4-13. Incidence of additional neoplasms in male and female B6C3Fi mice exposed to
1,1,2,2-tetrachloroethane in feed for 78 weeks 32
vi DRAFT - DO NOT CITE OR QUOTE
-------
4-14. Effects of acute (60 minutes) 1,1,2,2-tetrachloroethane treatment on rat liver 39
4-15. Results of in vitro and in vivo genotoxicity studies of 1,1,2,2-tetrachloroethane 51
4-16. Pulmonary adenomas from 1,1,2,2-tetrachloroethane exposure in mice 55
4-17. Pulmonary adenomas from 1,1,2,2-tetrachloroethane exposure in A/St mice 55
4-18. Summary of noncancer results of major studies for oral exposure of animals to
1,1,2,2-tetrachloroethane 57
4-19. Summary of noncancer results of major human studies of inhalation exposure to
1,1,2,2-tetrachloroethane 63
4-20. Summary of noncancer results of major studies for inhalation exposure of animals to
1,1,2,2-tetrachloroethane 65
5-1. Summary of BMD model results for rats exposed to 1,1,2,2-tetrachloroethane in the
diet for 14 weeks 80
5-2. Best-fitting BMD model predictions for relative liver weight in rats exposed to
1,1,2,2-tetrachloroethane in the diet for 14 weeks Error! Bookmark not defined.
5-3. Potential PODs with applied UFs and resulting subchronic RfDs 87
5-4. Incidences of hepatocellular carcinomas in B6C3Fi mice used for dose-response
assessment of 1,1,2,2-tetrachloroethane 96
5-5. HEDs corresponding to duration-adjusted TWA doses in mice 97
5-6. Summary of human equivalent BMDs and BMDLs based on hepatocellular
carcinoma incidence data in female B6C3Fi mice 98
5-7. Summary of uncertainty in the 1,1,2,2-tetrachloroethane cancer risk assessment 99
B-l. BMD modeling results based on incidence of hepatocytocellular vacuolization in
male rats exposed to 1,1,2,2-tetrachloroethane in the diet for 14 weeks B-A
B-2. BMD modeling results based on incidence of hepatocytocelluar vacuolization in
female rats exposed to 1,1,2,2-tetrachloroethane in the diet for 14 weeks B-B
B-3. Summary of BMD modeling results based on mean absolute liver weights in male
rats administered 1,1,2,2-tetrachloroethane in the diet for 14 weeks ....B-Error! Bookmark
not defined.
B-4. Summary of BMD modeling results based on mean absolute liver weights in female
rats administered 1,1,2,2-tetrachloroethane in the diet for 14 weeks ....B-Error! Bookmark
not defined.
vn DRAFT - DO NOT CITE OR QUOTE
-------
B-5. Summary of BMD modeling results based on mean relative liver weights in male
rats administered 1,1,2,2-tetrachloroethane in the diet for 14 weeks B-21
B-6. Summary of BMD modeling results based on mean relative liver weights in female
rats administered 1,1,2,2-tetrachloroethane in the diet for 14 weeks ....B-Error! Bookmark
not defined.
B-7. Summary of BMD modeling results based on mean serum ALT levels in male rats
administered 1,1,2,2-tetrachloroethane in the diet for 14 weeks B-27
B-8. Summary of BMD modeling results based on mean serum ALT levels in female rats
administered 1,1,2,2-tetrachloroethane in the diet for 14 weeks.... B-Error! Bookmark not
defined.
B-9. Summary of BMD modeling results on mean serum SDH levels in male rats
administered 1,1,2,2-tetrachloroethane in the diet for 14 weeks.... B-Error! Bookmark not
defined.
B-10. Summary of BMD modeling results on mean serum SDH levels in female rats
administered 1,1,2,2-tetrachloroethane in the diet for 14 weeks.. B-Error! Bookmark not
defined.
B-l 1. Summary of BMD modeling results based on mean serum bile acids in male rats
administered 1,1,2,2-tetrachloroethane in the diet for 14 weeks.. B-Error! Bookmark not
defined.
B-12. Summary of BMD modeling results based on mean serum bile acids in female rats
administered 1,1,2,2-tetrachloroethane in the diet for 14 weeks.. B-Error! Bookmark not
defined.
B-l 3. BMD modeling results for decreases in mean weights of fetuses from rat dams
exposed to 1,1,2,2-tetrachloroethane in the diet on GDs 4-20 B-Error! Bookmark not
defined.
C-l. Data used for dose-response assessment of hepatocellular carcinomas in B6C3Fi mice
administered 1,1,2,2-tetrachloroethane via gavage for 78 weeks C-l
C-2. Summary of human equivalent BMDs and BMDLs based on hepatocellular
carcinoma incidence in B6C3Fi mice administered 1,1,2,2-tetrachloroethane via
gavage for 78 weeks C-Error! Bookmarknot defined.
C-3. BMD modeling results based on incidence of hepatocellular carcinomas in male
B6C3Fi mice administered 1,1,2,2-tetrachloroethane via gavage for 78 weeks C-2
C-4. BMD modeling results based on incidence of hepatocellular carcinomas in female
B6C3Fi mice administered 1,1,2,2-tetrachloroethane via gavage for 78 weeks C-Error!
Bookmark not defined.
vin DRAFT - DO NOT CITE OR QUOTE
-------
LIST OF FIGURES
2-1. Structure of 1,1,2,2-tetrachloroethane 3
3-1. Suggested metabolic pathways of 1,1,2,2-tetrachloroethane 7
5-1. Exposure response array for subchronic and chronic oral exposure to
1,1,2,2-tetrachloroethane 86
5-2. PODs for selected endpoints (with critical effect circled) from Table 5-3 with
corresponding applied UFs and derived sample subchronic inhalation reference values
(RfVs) 88
5-3. PODs for selected endpoints (with critical effect circled) from Table 5-3 with
corresponding applied UFs and derived sample chronic inhalation RfVs 89
6-1. PODs for selected endpoints (with critical effect circled) with corresponding applied
UFs and derived sample subchronic inhalation RfVs 105
6-2. PODs for selected endpoints (with critical effect circled) from Table 5-3 with
corresponding applied UFs and derived sample subchronic inhalation RfVs 108
ix DRAFT - DO NOT CITE OR QUOTE
-------
LIST OF ABBREVIATIONS AND ACRONYMS
ACTH adrenocorticotropic hormone
AIC Akaike's Information Criterion
ALP alkaline phosphatase
ALT alanine aminotransferase
AST aspartate aminotransferase
ATP adenosine triphosphate
AUC area under the curve
BMD benchmark dose
BMDL 95% confidence limit (lower bound) on the benchmark dose
BMDS benchmark dose software
BMR benchmark response
CASRN Chemical Abstracts Service Registry Number
CHO Chinese hamster ovary
CNS central nervous system
DEN diethylnitrosamine
FEL frank effect level
FOB functional observational battery
G6Pase glucose-6-phosphatase
GD gestation day
GST glutathione S-transferase
Hb hemoglobin
HED human equivalent dose
i.p. intraperitoneal
IU International units
LCso median lethal concentration
LDso median lethal dose
LOAEL lowest-observed-adverse-effect level
mA milliampere
NCI National Cancer Institute
NOAEL no-observed-adverse-effect level
NTP National Toxicology Program
PBPK physiologically based pharmacokinetic
PBTK physiologically based toxicokinetic
PCNA proliferating cell nuclear antigen
POD point of departure
RBC red blood cell
RfC reference concentration
RfD reference dose
RfV reference value
SCE sister chromatid exchange
SD standard deviation
SDH sorbitol dehydrogenase
TWA time-weighted average
UDS unscheduled DNA synthesis
UF uncertainty factor
x
DRAFT - DO NOT CITE OR QUOTE
-------
U.S. EPA U.S. Environmental Protection Agency
WBC white blood cell
xi DRAFT - DO NOT CITE OR QUOTE
-------
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 subchronic and chronic
exposure to 1,1,2,2-tetrachloroethane. It is not intended to be a comprehensive treatise on the
chemical or toxicological nature of 1,1,2,2-tetrachloroethane.
The intent of Section 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).
xn DRAFT - DO NOT CITE OR QUOTE
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
CHEMICAL MANAGER/AUTHOR
Martin W. Gehlhaus, M.H.S.
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Washington, DC
AUTHORS
Ambuja Bale, Ph.D.
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Washington, DC
Geoffrey W. Patton, Ph.D.
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Washington, DC
Susan Rieth, M.S.
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Washington, DC
TedBerner, M.S.
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Washington, DC
Karen Hogan, M.S.
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Washington, DC
CONTRACTING SUPPORT
Mark Osier, Ph.D.
Environmental Science Center
Syracuse Research Corporation
Syracuse, NY
Stephen Bosch
Environmental Science Center
Syracuse Research Corporation
Syracuse, NY
xiii DRAFT - DO NOT CITE OR QUOTE
-------
Marc Odin, M.S.
Environmental Science Center
Syracuse Research Corporation
Syracuse, NY
REVIEWERS
This document has been provided for review to EPA scientists, interagency reviewers
from other federal agencies and White House offices, and the public, and peer reviewed by
independent scientists external to EPA. A summary and EPA's disposition of the comments
received from the independent external peer reviewers and from the public is included in
Appendix A.
INTERNAL EPA REVIEWERS
Joyce M. Donohue, Ph.D.
Office of Water
Office of Science and Technology (OST)
Health and Ecological Criteria Division (HECD)
Lynn Flowers, Ph.D., DABT
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC
Chris Cubbison
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH
xiv DRAFT - DO NOT CITE OR QUOTE
-------
1 1. INTRODUCTION
2
3
4 This document presents background information and justification for the Integrated Risk
5 Information System (IRIS) Summary of the hazard and dose-response assessment of
6 1,1,2,2-tetrachloroethane. IRIS Summaries may include oral reference dose (RfD) and
7 inhalation reference concentration (RfC) values for chronic and other exposure durations, and a
8 carcinogenicity assessment.
9 The RfD and RfC, if derived, provide quantitative information for use in risk assessments
10 for health effects known or assumed to be produced through a nonlinear (presumed threshold)
11 mode of action. The RfD (expressed in units of mg/kg-day) is defined as an estimate (with
12 uncertainty spanning perhaps an order of magnitude) of a daily exposure to the human
13 population (including sensitive subgroups) that is likely to be without an appreciable risk of
14 deleterious effects during a lifetime. The inhalation RfC (expressed in units of mg/m3) is
15 analogous to the oral RfD, but provides a continuous inhalation exposure estimate. The
16 inhalation RfC considers toxic effects for both the respiratory system (portal-of-entry) and for
17 effects peripheral to the respiratory system (extrarespiratory or systemic effects). Reference
18 values are generally derived for chronic exposures (up to a lifetime), but may also be derived for
19 acute (<24 hours), short-term (>24 hours up to 30 days), and subchronic (>30 days up to 10% of
20 lifetime) exposure durations, all of which are derived based on an assumption of continuous
21 exposure throughout the duration specified. Unless specified otherwise, the RfD and RfC are
22 derived for chronic exposure duration.
23 The carcinogenicity assessment provides information on the carcinogenic hazard
24 potential of the substance in question and quantitative estimates of risk from oral and inhalation
25 exposure may be derived. The information includes a weight of evidence judgment of the
26 likelihood that the agent is a human carcinogen and the conditions under which the carcinogenic
27 effects may be expressed. Quantitative risk estimates may be derived from the application of a
28 low-dose extrapolation procedure. If derived, the oral slope factor is a plausible upper bound on
29 the estimate of risk per mg/kg-day of oral exposure. Similarly, an inhalation unit risk is a
30 plausible upper bound on the estimate of risk per ug/m3 air breathed.
31 Development of these hazard identification and dose-response assessments for
32 1,1,2,2-tetrachloroethane has followed the general guidelines for risk assessment as set forth by
33 the National Research Council (NRC, 1983). The U.S. Environmental Protection Agency (U.S.
34 EPA) guidelines and Risk Assessment Forum Technical Panel Reports that may have been used
35 in the development of this assessment include the following: Guidelines for Mutagenicity Risk
36 Assessment (U.S. EPA, 1986), Recommendations for and Documentation of Biological Values
37 for Use in Risk Assessment (U.S. EPA, 1988), Guidelines for Developmental Toxicity Risk
3 8 Assessment (U. S. EPA, 1991 a), Interim Policy for Particle Size and Limit Concentration Issues
DRAFT - DO NOT CITE OR QUOTE
-------
1 in Inhalation Toxicity (U.S. EPA, 1994a), Methods for Derivation of Inhalation Reference
2 Concentrations and Application of Inhalation Dosimetry (U.S. EPA, 1994b), Use of the
3 Benchmark Dose Approach in Health Risk Assessment (U.S. EPA, 1995), Guidelines for
4 Reproductive Toxicity Risk Assessment (U.S. EPA, 1996), Guidelines for Neurotoxicity Risk
5 Assessment (U.S. EPA, 1998a), Science Policy Council Handbook. Risk Characterization (U.S.
6 EPA, 2000a), Benchmark Dose Technical Guidance Document (U. S. EPA, 2000b),
7 Supplementary Guidance for Conducting Health Risk Assessment of Chemical Mixtures (U.S.
8 EPA, 2000c), A Review of the Reference Dose and Reference Concentration Processes (U.S.
9 EPA, 2002), Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005a), Supplemental
10 Guidance for Assessing Susceptibility from Early-Life Exposure to Carcinogens (U.S. EPA,
11 2005b), Science Policy Council Handbook: Peer Review (U. S. EPA, 2006a), and A Framework
12 for Assessing Health Risks of Environmental Exposures to Children (U.S. EPA, 2006b).
13 The literature search strategy employed for this compound was based on the Chemical
14 Abstracts Service Registry Number (CASRN) and at least one common name. Any pertinent
15 scientific information submitted by the public to the IRIS Submission Desk was also considered
16 in the development of this document. The relevant literature was reviewed through May, 2009.
17 Portions of this document were developed under a Memorandum of Understanding,
18 signed November 4, 2004, with the Agency for Toxic Substances and Disease Registry
19 (ATSDR).
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
10
12
14
16
17
18
2. CHEMICAL AND PHYSICAL INFORMATION
1,1,2,2-Tetrachloroethane (1,1,2,2TCE; CASRN 79-34-5) is a synthetic halogenated
hydrocarbon that is a colorless, nonflammable liquid at room temperature. It is highly volatile,
somewhat soluble in water, and miscible with many organic solvents. The structure of
1,1,2,2-tetrachloroethane is shown below (Figure 2-1), and the chemical and physical properties
are presented in Table 2-1.
Cl Cl
H-
Cl
-C H
Cl
Figure 2-1. Structure of l,l?2,2-tetrachloroethane.
Table 2-1. Chemical and physical properties of l,l?2,2-tetrachloroethane
Characteristic
Chemical name
Synonym(s)
Chemical formula
CASRN
Molecular weight
Color
Freezing point
Boiling point
Density at 20 °C
Odor threshold:
Water
Air
Solubility:
Water
Organic solvents
Information
1 , 1 ,2,2-Tetrachloroethane
Acetylene tetrachloride; sym-tetrachloroethane; s-tetrachloro-
ethane; tetrachlorethane; l,l-dichloro-2,2-dichloroethane
C2H2C14
79-34-5
167.85
Colorless
-43.8°C
-36°C
145. 1°C
146.2°C
146. 5°C
1.594
1.595
0.50 ppm
1.5 ppm
3-5 ppm
2.87 g/L (20°C)
2.85 g/L (25°C)
Miscible with ethanol, methanol, ether, acetone, benzene,
petroleum, carbon tetrachloride, carbon disulfide, dimethyl
formamide, oils
Reference
HSDB, 2009; CAS, 1994
CAS, 1994
CAS, 1994
HSDB, 2009; CAS, 1994;
Lide, 1993;Riddicketal,
1986
Hawley, 1981
Riddicketal, 1986
Lide, 1993
Riddicketal., 1986
Lide, 1993
Merck Index, 1989
Riddicketal., 1986
Lide, 1993
HSDB, 2009; Amoore and
Hautala, 1983
Amoore and Hautala, 1983
HSDB, 2009
Riddicketal., 1986
Merck Index, 1989
HSDB, 2009; Merck Index,
1989; Hawley, 1981
DRAFT - DO NOT CITE OR QUOTE
-------
Table 2-1. Chemical and physical properties of l,l?2,2-tetrachloroethane
Characteristic
Vapor pressure
Partition
coefficients:
log Kow
logKoc
Henry's law constant
Flash point
Conversions:
ppm to mg/m3
mg/m3 to ppm
Information
5.95mmHg(25°C)
9mmHg(30°C)
2.39
1.66
2.78
4.7 x 10~4 atm-m3/mol
4.55 x lO'4 atm-m3/mol
1.80 x 10"3 atm-m3/mol
None - nonflammable
1 ppm = 6.87 mg/m3
1 mg/m3 = 0. 146 ppm
Reference
Riddicketal., 1986
HSDB, 2009; Flick, 1985
Hansch and Leo, 1 985
Chiouetal., 1979
ASTER, 1995
Mackay and Shiu, 1981
HSDB, 2009
ASTER, 1995
HSDB, 2009; Hawley, 1981
Calculated
Calculated
1
2
3
4
5
9
10
11
12
13
In the past, the major use for 1,1,2,2-tetrachloroethane was in the production of
trichloroethylene, tetrachloroethylene, and 1,2-dichloroethylene (Archer, 1979). With the
development of new processes for manufacturing chlorinated ethylenes and the availability of
less toxic solvents, the production of 1,1,2,2-tetrachloroethane as a commercial end-product in
the United States and Canada has steadily declined since the late 1960s, and production ceased
by the early-1990s (HSDB, 2009; Environment Canada and Health Canada, 1993).
1,1,2,2-Tetrachloroethane may still appear as a chemical intermediate in the production of a
variety of other common chemicals. It was also used as a solvent, in cleaning and degreasing
metals, in paint removers, varnishes, and lacquers, in photographic films, and as an extractant for
oils and fats (Hawley, 1981). Although at one time it was used as an insecticide, fumigant, and
weed killer (Hawley, 1981), it presently is not registered for any of these purposes. It was once
used as an ingredient in an insect repellent, but registration was canceled in the late 1970s.
DRAFT - DO NOT CITE OR QUOTE
-------
1 3. TOXICOKINETICS
2
3
4 1,1,2,2-Tetrachloroethane is well absorbed from the respiratory and gastrointestinal tracts
5 in both humans and laboratory animals and is extensively metabolized and excreted, chiefly as
6 metabolites, in the urine and breath. The metabolism of 1,1,2,2-tetrachloroethane in rats and
7 mice results in the production of trichloroethanol, trichloroacetic acid, and dichloroacetic acid.
8 The dichloroacetic acid is then broken down to glyoxalic acid, oxalic acid, and carbon dioxide.
9 When 1,1,2,2-tetrachloroethane undergoes reductive or oxidative metabolism, reactive radical
10 and acid chloride intermediates, respectively, are produced.
11
12 3.1. ABSORPTION
13 3.1.1. Oral Exposure
14 There are no known studies that quantify absorption following oral exposure in humans.
15 However, the health effects resulting from ingestion of large amounts of 1,1,2,2-tetrachloro-
16 ethane in humans (Section 4.1.1) indicate that 1,1,2,2-tetrachloroethane is absorbed following
17 oral exposure.
18 Observations in animals indicate that the oral absorption of 1,1,2,2-tetrachloroethane is
19 rapid and extensive. Cottalasso et al. (1998) reported hepatic effects, including increases in
20 serum aspartate aminotransferase (AST) and alanine aminotransferase (ALT), a decrease in
21 microsomal glucose-6-phosphatase (G6Pase) activity, and an increase in triglyceride levels, only
22 15-30 minutes following a single oral exposure in rats. Following a single oral exposure of male
23 Osborne-Mendel rats and B6C3Fi mice to 150 mg/kg of radiolabeled 1,1,2,2-tetrachloroethane in
24 corn oil, only 4-6% of the activity was recovered in the feces 72 hours postexposure while >90%
25 of the administered activity was found in both species as metabolites, indicating that the
26 compound was nearly completely absorbed in both rats and mice within 72 hours (Dow
27 Chemical Company, 1988). Mitoma et al. (1985) exposed groups of male Osborne-Mendel rats
28 to 25 or 100 mg/kg and B6C3Fi mice to 50 or 200 mg/kg of 1,1,2,2-tetrachloroethane in corn oil
29 gavage 5 days/week for 4 weeks, followed by a single radiolabeled dose of the compound, and
30 evaluated the disposition of the radiolabeled 1,1,2,2-tetrachloroethane over the next 48 hours.
31 While absorption was not quantified, 79% of the dose was metabolized in rats and 68% was
32 metabolized in mice, suggesting that at least those levels of compound had been absorbed within
33 48 hours.
34
DRAFT - DO NOT CITE OR QUOTE
-------
1 3.1.2. Inhalation Exposure
2 While studies of the systemic toxicity of 1,1,2,2-tetrachloroethane following inhalation in
3 humans are indicative of some level of systemic absorption, comparatively few studies have
4 quantitatively addressed this issue. A study in volunteers was carried out in which a bulb
00
5 containing [ Cl]-labeled 1,1,2,2-tetrachloroethane was inserted into their mouths; they
6 immediately inhaled deeply, held their breaths for 20 seconds, and then exhaled through a trap
7 containing granulated charcoal. The study showed that approximately 96% of a single breath of
8 1,1,2,2-tetrachloroethane was absorbed systemically (Morgan etal., 1970). Two subjects were
9 reported to retain approximately 40-60% of inspired 1,1,2,2-tetrachloroethane after a 30-minute
10 exposure of up to 2,300 mg/m3 (Lehmann et al., 1936), but additional details were not provided.
11 The total body burden of 1,1,2,2-tetrachloroethane in male Osborne-Mendel rats and
12 B6C3Fi mice exposed to a vapor concentration of 10 ppm (68.7 mg/m3) for 6 hours (Dow
13 Chemical Company, 1988) was 38.7 (imol equivalents/kg in rats (9.50 umol equivalents and
14 using a body weight of 245 g from the study) and 127 umol equivalents/kg in mice (3.059 umol
15 equivalents and using a body weight of 24.1 g from the study), indicating that while absorption
16 occurred in both species, mice absorbed proportionally more 1,1,2,2-tetrachloroethane on a per-
17 body-weight basis. Ikeda and Ohtsuji (1972) detected metabolites, measured as total
18 trichlorocompounds, trichloroacetic acid, and trichloroethanol, in the urine of rats exposed to 200
19 ppm (1,370 mg/m3) 1,1,2,2-tetrachloroethane, indicating that absorption had occurred; however,
20 they did not provide a quantitative estimate of absorption rate or fraction. Similarly, Gargas and
21 Anderson (1989) followed the elimination of 1,1,2,2-tetrachloroethane as exhaled breath from
22 the blood after a 6-hour exposure to 350 ppm (2,400 mg/m3), but did not provide quantitative
23 estimates of absorption.
24
25 3.2. DISTRIBUTION
26 No studies measuring the distribution of 1,1,2,2-tetrachloroethane in humans following
27 inhalation or oral exposure were located. Following absorption in animals, 1,1,2,2-tetrachloro-
28 ethane appears to be distributed throughout the body, but may selectively accumulate to a degree
29 in certain cells and tissues. The human blood-air partition coefficient for 1,1,2,2-tetrachloro-
30 ethane has been reported to be in the range of 72.6-116 (Meulenberg and Vijverberg, 2000;
31 Gargas et al., 1989; Morgan et al., 1970). The tissue:air partition coefficients for 1,1,2,2-tetra-
32 chloroethane in rats have been reported to be 142 (blood), 3,767 (fat), 196 (liver), and
33 101 (muscle) (Meulenberg and Vijverberg, 2000; Gargas et al., 1989), indicating that
34 1,1,2,2-tetrachloroethane may partition into fatty tissues, consistent with its low water solubility.
35 Following a single intravenous injection of radiolabeled 1,1,2,2-tetrachloroethane,
36 Eriksson and Brittebo (1991) reported a selective uptake of nonvolatile radioactivity in the
37 mucosal tissues of olfactory and tracheobronchial regions of the respiratory tract and in the
38 mucosae of the oral cavity, tongue, nasopharynx, esophagus, and cardiac region of the
6 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
forestomach. High levels of radioactivity were also found in the liver, bile, inner zone of the
adrenal cortices, and interstitium of the testes, although the levels were not quantified.
3.3. METABOLISM
No studies were located that investigated the metabolism of 1,1,2,2-tetrachloroethane in
humans. Information regarding 1,1,2,2-tetrachloroethane metabolism in animals is summarized
below, and a suggested metabolic scheme based on in vivo and in vitro data is presented in
Figure 3-1.
ci
10
11
12
13
14
15
16
17
18
19
20
21
22
free radical
CI
CI
CI
CI CI
\ /
reductive /\
dechlorination cf CI
1,1,2,2-tetrachloroethane
hydrolytic
avage
CI
>
/
cr
Trichloroethylene
y
CI
CI
trichloroacetaldehyde fa
/
CI OH
1 1
CI— 1
CI
_ Non-enzymatic UAI
dehydrochlorination 1
CI *
\
o cr
JdUUII
CI
/
CI
\ y \
CI O
// o
CI /' v\
Cl \ \s
nH
CI OH
o
J/
oxalic acid
trichloroethanol
Trichloroacetic acid
XI
. \
CI O
\ _
-------
1 Yllner, 1971). Dichloroacetic acid was identified as the major urinary metabolite in mice treated
2 with 1,1,2,2-tetrachloroethane by intraperitoneal (i.p.) injection (Yllner et al., 1971) and in in
3 vitro systems with rat liver microsomal and nuclear cytochrome P450 (Casciola and Ivanetich,
4 1984; Halpert, 1982; Halpert and Neal, 1981). Dichloroacetic acid can be further metabolized to
5 glyoxylic acid, formic acid, and carbon dioxide (Yllner, 1971), with carbon dioxide a potential
6 major component of the end products (Yllner, 1971). Other pathways involve the formation of
7 trichloroethylene via dehydrochlorination or tetrachloroethylene via oxidation as initial
8 metabolites. Trichloroethylene and tetrachloroethylene are further metabolized to trichloro-
9 ethanol and trichloroacetic acid, and oxalic acid and trichloroacetic acid, respectively (Mitoma et
10 al., 1985; Ikeda and Ohtsuji, 1972; Yllner et al., 1971). 1,1,2,2-Tetrachloroethane may also form
11 free radicals by undergoing reductive dechlorination (ATSDR, 1996). The formation of free
12 radical intermediates during 1,1,2,2-tetrachloroethane metabolism has been demonstrated in
13 spin-trapping experiments (Paolini et al., 1992; Tomasi et al., 1984).
14 Metabolism of 1,1,2,2-tetrachloroethane is generally extensive, with 68-95% of a total
15 administered dose found as metabolites (Dow Chemical Company, 1988; Mitoma et al., 1985;
16 Yllner, 1971). Mice given a single 0.21-0.32 g/kg i.p. dose of [14C]-labeled 1,1,2,2-tetrachloro-
17 ethane eliminated 45-61% of the administered radioactivity as carbon dioxide in expired air and
18 23-34% of the radioactivity in urine in the following 3 days (Yllner et al., 1971). Dichloroacetic
19 acid, trichloroacetic acid, trichloroethanol, oxalic acid, glyoxylic acid, and urea accounted for 27,
20 4, 10, 7, 0.9, and 2% of the mean urinary radioactivity excreted by the mice in 24 hours,
21 respectively (Yllner et al., 1971). Yllner et al. (1971) also demonstrated that 20-23% of the
22 [14C]-tetrachloroethane was converted to glycine following the simultaneous i.p. injection of
23 [14C]-tetrachloroethane and sodium benzoate and the estimation of [14C]-hippuric acid in the
24 urine. In rats, trichloroethanol appeared to be present as a urinary metabolite at approximately
25 fourfold greater levels than trichloroacetic acid following a single 8-hour inhalation exposure
26 (Ikeda and Ohtsuji, 1972). Several studies have reported that metabolism of 1,1,2,2-tetrachloro-
27 ethane is greater in mice than in rats, with magnitudes of the reported difference generally in the
28 range of a 1.1-3.5-fold greater metabolic activity, on a per-kg basis, in mice (Dow Chemical
29 Company, 1988; Mitoma et al., 1985; Milman et al., 1984).
30 As indicated above, cytochrome P450-based metabolism of 1,1,2,2-tetrachloroethane to
31 dichloroacetic acid has been demonstrated in vitro. Multiple P450 isozymes are likely to be
32 involved, as demonstrated by studies reporting increased metabolism and covalent binding of
33 metabolites following pretreatment with phenobarbital (Casciola and Ivanetich, 1984; Halpert,
34 1982), xylene (Halpert, 1982), or ethanol (Sato et al., 1980). The isozymes induced by
35 phenobarbital, xylene, and ethanol include members of the CYP2A, CYP2B, CYP2E, and
36 CYP3A subfamilies (Omiecinski et al., 1999; Nebert et al., 1987).
37 1,1,2,2-Tetrachloroethane has also been reported to produce inactivation of cytochrome
38 P450. 1,1,2,2-Tetrachloroethane effectively inactivated the major phenobarbital-inducible P450
8 DRAFT - DO NOT CITE OR QUOTE
-------
1 isozyme, but not the major P450 isozyme induced by p-naphthoflavone, in rat liver in vitro
2 (Halpert et al, 1986). Rat liver nuclear cytochrome P450 levels were reduced following in vitro
3 incubation with 1,1,2,2-tetrachloroethane and a NADPH-generating system (Casciola and
4 Ivanetich, 1984). In an in vivo study, cytochrome P450 activity was evaluated in male and
5 female Swiss albino mice 24 hours after a single 0, 300, or 600 mg/kg i.p. dose of 1,1,2,2-tetra-
6 chloroethane (Paolini et al., 1992). 1,1,2,2-Tetrachloroethane treatment statistically significantly
7 (p < 0.01) reduced total cytochrome P450 activity 44 and 37% in males and females, respectively,
8 at 300 mg/kg and 85 and 74% in males and females, respectively, at 600 mg/kg. Treatment with
9 600 mg/kg statistically significantly reduced the microsomal activity of P450 isozymes 3 A, 2E1,
10 1A2, 2B1, and 1A1 in both genders, and 300 mg/kg reduced the activity of P4503A in both sexes
11 and P4502B1 in males. Heme content was reduced 13 and 33% at 300 and 600 mg/kg,
12 respectively, and may have contributed to the decrease in CYP450 levels. The 600 mg/kg dose
13 also reduced the activity of glutathione S-transferase (GST) toward l-chloro-2,4-dinitrobenzene,
14 a general GST substrate, in both genders.
15 Due to the extensive metabolism of 1,1,2,2 tetrachloroethane to products such as
16 trichloroethylene and dichloroacetic acid, the relevance of 1,1,2,2-tetrachloroethane interactions
17 with GST is important. Studies of human GST-zeta polymorphic variants show different
18 enzymatic activities toward and inhibition by dichloroacetic acid that could reasonably affect the
19 metabolism of 1,1,2,2-tetrachloroethane (Lantum et al., 2002; Blackburn et al., 2001, 2000;
20 Tzeng et al., 2000). Dichloroacetic acid may covalently bind to GST-zeta (Anderson et al.,
21 1999) and inhibit its own metabolism, leading to an increase in the amount of unmetabolized
22 dichloroacetic acid as the dose and/or duration increases (U.S. EPA, 2003).
23 Data indicate that 1,1,2,2-tetrachlorethane can be metabolized to dichloroacetic acid
24 (ATSDR, 1996; Yllner, 1971), suggesting a potential role for this metabolite in some of the
25 cancer and noncancer effects observed following exposure to 1,1,2,2 tetrachloroethane.
26 Following an intravenous injection of radiolabeled 1,1,2,2-tetrachloroethane, radioactivity could
27 not be extracted from epithelium of the respiratory and upper alimentary tracts, or from the liver,
28 adrenal cortex, or testes (Eriksson and Brittebo, 1991). The presence of tissue-bound metabolites
29 in the epithelial linings in the upper respiratory tract may demonstrate a first-pass effect by the
30 respiratory tract (Eriksson and Brittebo, 1991). In addition, the presence of irreversible tissue-
31 bound metabolites demonstrates the metabolism of 1,1,2,2-tetrachloroethane to reactive
32 metabolites (Eriksson and Brittebo, 1991). However, the identities of the bound metabolites and
33 modified proteins or phospholipids were not identified. The presence of radiolabel in the
34 proteins may have been radiolabeled incorporated glycine.
35 Dow Chemical Company (1988) observed radiolabel in hepatic DNA, although the
36 presence of the radiolabel in the hepatic DNA likely represented the incorporation of single
37 [14C]-atoms via normal biosynethetic pathways. Mice were found to have approximately a
38 1.9-fold greater extent of [14C] activity irreversibly associated with hepatic macromolecules than
9 DRAFT - DO NOT CITE OR QUOTE
-------
1 rats, which the study authors noted was consistent with the greater metabolism, on a per-kg basis,
2 in mice compared to rats. After a 4-week oral exposure to unlabeled 1,1,2,2-tetrachloroethane
3 followed by a single oral dose of labeled 1,1,2,2-tetrachloroethane, Mitoma et al. (1985) also
4 reported greater levels of hepatic protein-binding in the tissue of mice compared to rats, and the
5 differences were on the order of twofold greater binding in mice, which would be consistent both
6 with the Dow Chemical Company (1988) studies and with the observed differences in
7 metabolism of the two species discussed above. This may also be related to the 3.2-3.5-fold
8 greater absorption, on a per-kg basis, of mice compared to rats following inhalation exposure
9 (Dow Chemical Company, 1988).
10 The kinetic constants of 1,1,2,2-tetrachloroethane metabolism in rats exposed to 350 ppm
11 of the chemical for 6 hours were determined in gas uptake studies performed by Gargas and
12 Anderson (1989). The rate of exhalation of 1,1,2,2-tetrachloroethane was measured and,
13 combined with previously published values for partition coefficients for blood/air, liver/blood,
14 muscle/blood, and fat/blood, allowed the estimation of the disposition of the chemical in rat
15 (Gargas et al., 1989). A Km of 4.77 uM and a Vmax of 12 mg/hour (scaled to a 1-kg rat) were
16 measured.
17
18 3.4. ELIMINATION
19 Morgan et al. (1970) reported that the urinary excretion rate of 1,1,2,2-tetrachloroethane
20 in humans was 0.015% of the absorbed dose/minute. No other studies measuring the elimination
21 of 1,1,2,2-tetrachloroethane in humans have been reported.
22 Available animal data indicate that following absorption into the body, 1,1,2,2-tetra-
23 chloroethane is eliminated mainly as metabolites in urine, as carbon dioxide, or as unchanged
24 compound in expired air (Gargas and Anderson, 1989; Dow Chemical Company, 1988; Mitoma
25 et al., 1985; Ikeda and Ohtsuji, 1972; Yllner et al., 1971). The patterns of elimination in rats and
26 mice are qualitatively similar (Dow Chemical Company, 1988; Mitoma et al., 1985), although
27 covalent binding is somewhat greater in mice than rats. Elimination is fairly rapid, with
28 significant amounts present in the urine and expired air at 48-72 hours postexposure (Dow
29 Chemical Company, 1988; Mitoma et al., 1985; Ikeda and Ohtsuji, 1972; Yllner et al., 1971).
30 Only one study quantitatively evaluated the elimination of 1,1,2,2-tetrachloroethane
31 following inhalation exposure. Dow Chemical Company (1988) followed the excretion of
32 1,1,2,2-tetrachloroethane for 72 hours following exposure of rats and mice to vapor
33 concentrations of 10 ppm (68.7 mg/m3) [14C]-l,l,2,2-tetrachloroethane for 6 hours. More than
34 90% of the absorbed dose was metabolized in both species. The percentage of recovered
35 radioactivity reported in rats was 33% in breath (25% as CC>2 and 8% as unchanged compound),
36 19% in urine, and 5% in feces. In mice, the percentage of recovered radioactivity was 34% in
37 breath (32% as CC>2 and 2% as unchanged compound), 26% in urine, and 6% in feces.
10 DRAFT - DO NOT CITE OR QUOTE
-------
1 Radioactivity in urine and feces was nonvolatile (inferred by the researchers to be product(s) of
2 metabolism), but was not otherwise characterized.
3 With regard to oral exposure, the excretion of 1,1,2,2-tetrachloroethane was followed for
4 72 hours following oral administration of 150 mg/kg doses to rats and mice (Dow Chemical
5 Company, 1988). Greater than 90% of the absorbed dose was detected as metabolites in both
6 species. In rats, 41% was excreted in breath (32% as CC>2 and 9% as unchanged compound),
7 23% in urine, and 4% in feces. In mice, 51% was excreted in breath (50% as CC>2 and 1% as
8 unchanged compound), 22% in urine, and 6% in feces. Radioactivity in urine and feces was
9 nonvolatile (inferred by the researchers to be product(s) of metabolism), but was not otherwise
10 characterized. Mitoma et al. (1985) found that mice given an oral dose of 1,1,2,2-tetrachloro-
11 ethane excreted about 10% of the dose unchanged in the breath, and the rest was metabolized
12 and excreted in the breath as carbon dioxide (10%) or in the urine and feces (30%, measured
13 together), or retained in the carcass (27%) after 48 hours. Rats showed similar patterns of
14 excretion (Mitoma etal., 1985). The most comprehensive study of the metabolism and excretion
15 of 1,1,2,2-tetrachloroethane was an i.p. study in mice using [14C]-labeled 1,1,2,2-tetrachloro-
16 ethane. Yllner (1971) showed that after 72 hours, about 4% of the radioactivity was expired
17 unchanged in the breath, 50% was expired as carbon dioxide, 28% was excreted in the urine, 1%
18 was excreted in the feces, and 16% remained in the carcass.
19 Delays in elimination may be the result of covalent binding of 1,1,2,2-tetrachloroethane
20 metabolites, as reflected in high levels of compound detected in the carcasses of animals.
21 Mitoma et al. (1985) reported a 30.75% retention in the carcass of rats and a 27.44% retention in
22 the carcass of mice 48 hours after exposure to a single labeled dose of 25 m/kg in rats and 50
23 mg/kg in mice 1,1,2,2-tetrachloroethane. Dow Chemical Company (1988) reported 30%
24 retention in the carcass in rats exposed to 10 ppm by inhalation, 25% in mice exposed to 10 ppm
25 by inhalation, 23% in rats exposed to 150 mg/kg by gavage, and 17.3% in mice exposed to
26 150 mg/kg by gavage. Colacci et al. (1987) reported covalent binding of radiolabeled
27 1,1,2,2-tetrachloroethane to DNA, RNA, and protein in the liver, kidneys, lung, and stomach of
28 rats and mice exposed to a single intravenous dose and analyzed 22 hours postexposure. In vitro
29 binding to calf thymus DNA was found to be greatest when the microsomal fraction was present,
30 and was inhibited by SKF-525A, indicating that metabolic activation was likely required for
31 DNA binding (Colacci et al., 1987). However, Collaci et al. (1987) did not distinguish between
32 covalent binding and whether the presence of radiolabel in the DNA, RNA, and protein was the
33 result of incorporated radiolabeled carbon into the biomolecules through normal biochemical
34 processes.
35
36 3.5. PHYSIOLOGICALLY BASED TOXICOKINETIC MODELS
37 No physiologically based toxicokinetic (PBTK) models for 1,1,2,2-tetrachloroethane
38 were located for humans. Muelenberg et al. (2003) used saline:air, rat brain:air, and olive oil:air
11 DRAFT - DO NOT CITE OR QUOTE
-------
1 partition coefficients to model 28 chemicals from three distinct chemical classes, including
2 alkylbenzenes, chlorinated hydrocarbons, and ketones. The saline:air, rat brain:air, and olive
3 oil:air partition coefficients derived for 1,1,2,2-tetrachloroethane were 35.6 ± 6.05, 344 ± 21.0,
4 and 10,125 ± 547, respectively. The brain partition coefficients for the 28 chemicals were
5 predicted with accuracy within a factor of 2.5 for 95% of the chemicals. While the study
6 demonstrates the ability to predict rat brain partition coefficients using a bilinear equation, the
7 utility of the information for this assessment is limited. Similarly, several physiologically based
8 pharmacokinetic (PBPK) investigations of 1,1,2,2-tetrachloroethane exposure in fish (McKim et
9 al., 1999; Nichols et al, 1993) provide little utility for this assessment. In sum, adequate
10 information for PBTK modeling of 1,1,2,2-tetrachloroethane remains a research need.
11 Chiu and White (2006) presented an analysis of steady-state solutions to a PBPK model
12 for a generic volatile organic chemical (VOC) metabolized in the liver. The only parameters
13 used to determine the system state for a given oral dose rate or inhalation exposure concentration
14 were the blood-air partition coefficient, metabolic constants, and the rates of blood flow to the
15 liver and of alveolar ventilation. At exposures where metabolism is close to linear (i.e.,
16 unsaturated), it was demonstrated that only the effective first order metabolic rate constant was
17 needed. Additionally, it was found that the relationship between cumulative exposure and
18 average internal dose (e.g., areas under the curve [AUCs]) remains the same for time-varying
19 exposures. The study authors concluded that steady-state solutions can reproduce or closely
20 approximate the solutions using a full PBPK model. Section 5.2.2 addresses the applicability of
21 using this model to conduct a route-to-route extrapolation in this assessment.
12 DRAFT - DO NOT CITE OR QUOTE
-------
1 4. HAZARD IDENTIFICATION
2
3
4 4.1. STUDIES IN HUMANS—EPIDEMIOLOGY, CASE REPORTS, CLINICAL
5 CONTROLS
6 4.1.1. Oral Exposure
7 A number of case reports provide information on effects of intentional acute exposure to
8 lethal oral doses of 1,1,2,2-tetrachloroethane (Mant, 1953; Lilliman, 1949; Forbes, 1943; Elliot,
9 1933; Hepple, 1927). Subjects usually lost consciousness within approximately 1 hour and died
10 3-20 hours postingestion, depending on the amount of food in the stomach. Postmortem
11 examinations showed gross congestion in the esophagus, stomach, kidneys, spleen, and trachea,
12 gross congestion and edema in the lungs, and histological effects of congestion and cloudy
13 swelling in the lungs, liver, and/or kidneys (Mant, 1953; Hepple, 1927). Amounts of
14 1,1,2,2-tetrachloroethane recovered from the stomach and intestines of the deceased subjects
15 included 12 mL (Hepple, 1927), 25 g (Lilliman, 1949), 48.5 mL (Mant, 1953), and 425 mL
16 (Mant, 1953). Assuming a density of 1.594 g/mL and an average body weight of 70 kg, the
17 approximate minimum doses consumed in these cases are estimated to be approximately 273,
18 357, 1,100, and 9,700 mg/kg, respectively. No deaths occurred in eight patients (six men and
19 two women) who were accidentally given 3 mL of 1,1,2,2-tetrachloroethane (68 mg/kg, using
20 the above assumptions) or three patients (one young man, one young woman, and one 12-year-
21 old girl) who were accidentally given 2 or 3 mL (98-117 mg/kg, using the density and reported
22 body weights) as medicinal treatment for hookworm (Ward, 1955; Sherman, 1953). These
23 patients experienced loss of consciousness and other clinical signs of narcosis that included
24 shallow breathing, faint pulse, and pronounced lowering of blood pressure.
25
26 4.1.2. Inhalation Exposure
27 The symptoms of high-dose acute inhalation exposure to 1,1,2,2-tetrachloroethane
28 commonly include drowsiness, nausea, headache, constipation, decreased red blood cell (RBC)
29 count, weakness, and at extremely high concentrations, jaundice, unconsciousness, and
30 respiratory failure (Coyer, 1944; Hamilton, 1917).
31 An experimental study was conducted in which two volunteers self-inhaled various
32 concentrations of 1,1,2,2-tetrachloroethane for up to 30 minutes (Lehmann et al, 1936). The
33 results of this study suggest that 3 ppm (6.9 mg/m3) was the odor detection threshold; 13 ppm
34 (89 mg/m ) was tolerated without effect for 10 minutes, while 146 ppm (1,003 mg/m ) for
35 30 minutes or 336 ppm (2,308 mg/m3) for 10 minutes produced irritation of the mucous
36 membranes, pressure in the head, vertigo, and fatigue. No other relevant information was
37 reported.
13 DRAFT - DO NOT CITE OR QUOTE
-------
1 Minot and Smith (1921) reported that symptoms of industrial 1,1,2,2-tetrachloroethane
2 poisoning (concentrations not specified) included fatigue, perspiration, drowsiness, loss of
3 appetite, nausea, vomiting, constipation, headache, and jaundice. Hematological changes
4 included increased large mononuclear cells, elevated white blood cell (WBC) count, a slight but
5 progressive anemia, and a slight increase in platelet number. Similar symptoms were reported by
6 Parmenter (1921) and Wilcox et al. (1915). Horiguchi et al. (1964) reported that in 127 coating
7 workers employed in artificial pearl factories and exposed to 75-225 ppm (500-1,500 mg/m3)
8 1,1,2,2-tetrachloroethane (along with other solvents), observed effects included decreased
9 specific gravity of the whole blood, decreased RBC count, relative lymphocytosis, neurological
10 findings (not specified), and a positive urobilinogen test.
11 Lobo-Mendonca (1963) observed a number of adverse health effects in a mixed-gender
12 group of 380 workers at 23 Indian bangle manufacturing facilities (80% of workers employed at
13 these facilities were examined). In addition to the inhalation exposure, approximately 50% of
14 the examined workers had a substantial amount of dermal exposure to 1,1,2,2-tetrachloroethane.
15 Some of the workers were exposed to a mixture of equal parts acetone and 1,1,2,2-tetrachloro-
16 ethane. Air samples were collected at several work areas in seven facilities. Levels of
17 1,1,2,2-tetrachloroethane in the air ranged from 9.1 to 98 ppm (62.5-672 mg/m3). High
18 incidences of a number of effects were reported, including anemia (33.7%), loss of appetite
19 (22.6%), abdominal pain (23.7%), headaches (26.6%), vertigo (30.5%), and tremors (35%). The
20 significance of these effects cannot be determined because a control group of unexposed workers
21 was not examined and coexposure to acetone was possible. The study authors noted that the
22 incidence of tremors appeared to be directly related to 1,1,2,2-tetrachloroethane exposure
23 concentrations, as the percentage of workers handling tetrachloroethane and displaying tremors
24 increased as the air concentration of 1,1,2,2-tetrachloroethane increased.
25 Over a 3-year period, Jeney et al. (1957) examined 34-75 workers employed at a
26 penicillin production facility. 1,1,2,2-Tetrachloroethane was used as an emulsifier, and wide
27 fluctuations in atmospheric levels occurred throughout the day. The investigators noted that the
28 workers were only in the areas with high 1,1,2,2-tetrachloroethane concentrations for short
29 periods of time, and gauze masks with organic solvent filters were worn in these areas. During
30 the first year of the study, 1,1,2,2-tetrachloroethane levels ranged from 0.016 to 1.7 mg/L (16-
31 1,700 mg/m3; 2-248 ppm). In the second year of the study, ventilation in the work room was
32 improved and 1,1,2,2-tetrachloroethane levels ranged from 0.01 to 0.85 mg/L (10-850 mg/m ;
33 1-124 ppm). In the third year of the study, the workers were transferred to a newly built facility
34 and 1,1,2,2-tetrachloroethane levels in the new facility ranged from 0.01 to 0.25 mg/L (10-
35 250 mg/m3; 1-36 ppm). At 2-month intervals, the workers received general physical
36 examinations, and blood was drawn for measurement of hematological parameters, serum
37 bilirubin levels, and liver function tests; urinary hippuric acid levels were measured every
38 6 months. It appears that workers with positive signs of liver damage, including palpability of
14 DRAFT - DO NOT CITE OR QUOTE
-------
1 the liver, rise in bilirubin levels, positive liver function tests, and urobilinogenuria, were
2 transferred to other areas of the facility and were not examined further.
3 In the first year of the study, 31% of the examined workers had "general or gastro-
4 intestinal symptoms." Loss of appetite, bad taste in the mouth, epigastric pain, and a "dull
5 straining pressure feeling in the area of the liver" were reported by 66% of the workers
6 experiencing gastrointestinal symptoms. Other symptoms included headaches, general weakness,
7 and fatigue in 29%, severe weight loss in 4%, and "tormenting itching" in 1%. Enlargement of
8 the liver was observed in 38% of the screened workers. Urobilinogenuria was detected in 50%
9 of the workers, most often following more than 6 months of employment, and 31 % of the
10 workers with urobilinogenuria also had palpable livers.
11 In the second year of the study, there was a decline in the number of symptomatic
12 workers (13% of examined workers) and in workers with positive urobilinogenuria findings
13 (24%). Liver enlargement was observed in 20% of the examined workers. In the third year, the
14 number of workers reporting symptoms decreased to 2%, and positive urobilinogen findings
15 were found in 12%. The investigators noted that the increased urobilinogen levels during the
16 third year of observation may have been secondary to excessive alcohol consumption or dietary
17 excess. Enlarged livers were found in 5% of the examined workers.
18 During the course of the study, no alterations in erythrocyte or hemoglobin (Hb) levels
19 were found. Leukopenia (defined as leukocyte levels of <5,800 cells/mL) was found in 20% of
20 the workers, but no relationship between the number of cases and duration of 1,1,2,2-tetrachloro-
21 ethane exposure was found. A positive relationship between duration of exposure and frequency
22 of abnormal liver function test results was observed, as statistically significant correlations were
23 found on the thymol and Takata-Ucko liver function tests, but not the gold sol reaction test. The
24 thymol liver function test measures the direct precipitation of both lipids and abnormal lipid
25 protein complexes appearing in liver disease by the addition of a thymol solution (Kunkel and
26 Hoagland, 1947). The Takata-Ucko (or Takata-Ara) test detects an increase in the amounts of
27 the globulin components of the serum, signifying liver disease (Kunkel and Hoagland, 1947).
28 Abnormal hippuric acid levels were only detected in 1% of the examined workers during the first
29 2 years, and no abnormalities were observed during the third year. Increased serum bilirubin
30 levels (>1 mg/dL) were observed in 20, 18.7, and 7.6% of the workers during the first, second,
31 and third years, respectively. The prevalence of hepatitis was assessed using sickness benefit
32 files. In the 1-year period prior to the study, 21 cases of hepatitis were found (total number of
33 workers not reported). Three cases of hepatitis were found in the first year of the study, eight
34 cases in the second year, and four cases in the third year. The lack of a control group and poor
35 reporting of study design and results precludes using this study for quantitative dose-response
36 analysis.
37 Norman et al. (1981) examined the mortality of the employees of 39 chemical processing
38 plants used by the Army during World War II. Ten plants used 1,1,2,2-tetrachloroethane to help
15 DRAFT - DO NOT CITE OR QUOTE
-------
1 treat clothing, while the others plants used water in the same process. Estimates of exposure
2 levels were not reported, and coexposure to dry-cleaning chemicals was expected. At the time of
3 evaluation, 2,414 deaths were reported in the study cohort. No differences in standard mortality
4 ratios were seen between the tetrachloroethane and water groups for total mortality,
5 cardiovascular disease, cirrhosis of the liver, or cancer of the digestive and respiratory systems.
6 The mortality ratio for lymphatic cancers in the tetrachloroethane group was increased relative to
7 controls or the water group, although the number of deaths was small (4 cases, with an expected
8 number of 0.85). No other differences were seen between the groups.
9
10 4.2. SUBCHRONIC AND CHRONIC STUDIES AND CANCER BIOASSAYS IN
11 ANIMALS—ORAL AND INHALATION
12 4.2.1. Oral Exposure
13 4.2.1.1. Subchronic Studies
14 NTP (2004) fed groups of male and female F344 rats (10/sex/group) diets containing 0,
15 268, 589, 1,180, 2,300, or 4,600 ppm of microencapsulated 1,1,2,2-tetrachloroethane for
16 14 weeks. NTP (2004) reported that the microcapsules containing 1,1,2,2-tetrachloroethane
17 were specified to be no greater than 420 (im in diameter, and were not expected to have any
18 significant effect on the study. The reported average daily doses were 0, 20, 40, 80, 170, or
19 320 mg/kg-day, and vehicle control (feed with empty microcapsules) and untreated control
20 groups were used for both genders. Endpoints evaluated throughout the study included clinical
21 signs, body weight, and feed consumption. Hematology and clinical chemistry were assessed on
22 days 5 and 21 and at the end of the study; urinalyses were not performed. Necropsies were
23 performed on all animals, and selected organs (liver, heart, right kidney, lung, right testis, and
24 thymus) were weighed. Comprehensive histological examinations were performed on untreated
25 control, vehicle control, and high dose groups. Tissues examined in the lower dose groups were
26 limited to bone with marrow, clitoral gland, liver, ovary, prostate gland, spleen, testis with
27 epididymis and seminal vesicle, and uterus. A functional observational battery (FOB) was
28 performed on rats in the control groups and the 20, 40, and 80 mg/kg-day groups during weeks 4
29 and 13. Sperm motility, vaginal cytology, estrous cycle length, and percentage of time spent in
30 the various estrus stages were evaluated in control groups and the 40, 80, and 170 mg/kg-day
31 groups.
32 All animals survived to the end of the study, but clinical signs of thinness and pallor were
33 observed in all animals in the 170 and 320 mg/kg-day groups (NTP, 2004). Final body weights
34 (Table 4-1) were statistically significantly lower than vehicle controls in males at 80, 170, and
35 320 mg/kg-day (7, 29, and 65% lower, respectively) and females at 80, 170, and 320 mg/kg-day
36 (9, 29, and 56% lower, respectively), with both genders at 320 mg/kg-day losing weight over the
37 course of the study. However, feed consumption by the rats also decreased with increasing dose
38 level (NTP, 2004).
16 DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-1. Final body weights (g) and percent change compared to controls
in F344/N rats exposed to l,l?2,2-tetrachloroethane in feed for 14 weeks
Dose (mg/kg-d)
Vehicle control
20
40
80
170
320
n
10
10
10
10
10
10
Males
366 ±5a
354 ±9
353±6
341 ±6b
259 ± 9b
127 ± 9b
-
-3%
-4
-7
-29
-65
n
10
10
10
10
10
10
Females
195 ±4a
192 ±4
189±2
177 ± 2b
139 ±4b
85±3b
-
-2%
-3
-9
-29
-56
2
3
4
5
9
10
aMean ± standard error.
V<0.05.
Source: NTP (2004).
Statistically significant increases in absolute liver weights were observed in female rats
exposed to 80 mg/kg-day, and statistically significant decreases in absolute liver weight were
observed at >170 mg/kg-day in males and at 320 mg/kg-day in females (Table 4-2a).
Statistically significant increases in relative liver weights (Table 4-2b) were observed at
>40 mg/kg-day in males and females (NTP, 2004). Significant alterations in absolute and/or
relative weights were also observed in the thymus, kidney, heart, lung, and testes primarily at
170 and 320 mg/kg-day.
Table 4-2a. Absolute liver weights (g) and percent change compared to
controls in F344/N rats exposed to l,l?2,2-tetrachloroethane in feed for
14 weeks
Dose (mg/kg-d)
Vehicle control
20
40
80
170
320
n
10
10
10
10
10
10
Males
12.74 ±0.26a
12. 99 ±0.35
14. 47 ±0.44
15.54±0.39
11.60±0.44b
6.57±0.18b
-
2%
14
22
-9
-48
n
10
10
10
10
10
10
Females
6.84±0.17a
7.03 ±0.12
7.14±0.16
7.80±0.08b
6.66 ±0.21
4.94±0.12b
-
3%
4
14
-3
-28
aMean ± standard error.
V<0.05.
Source: NTP (2004).
11
17
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-2b. Relative liver weight (mg organ weight/g body weight) and
percent change compared to controls in F344/N rats exposed to l,l?2,2-tetra-
chloroethane in feed for 14 weeks
Dose (mg/kg-d)
Vehicle control
20
40
80
170
320
n
10
10
10
10
10
10
Males
34.79 ±0.42a
36.72 ±0.44
41.03 ±0.85b
45.61 ±0.52b
44.68 ±0.45b
52.23 ±1.42b
-
6%
18
31
28
50
n
10
10
10
10
10
10
Females
35.07±0.56a
36.69 ±0.36
37.84±0.51b
44.20 ±0.27b
48.03 ±0.89b
58.40 ±1.42b
-
5%
8
26
37
67
1
2
3
4
5
9
10
11
12
13
14
15
16
17
18
19
aMean ± standard error.
V<0.05.
Source: NTP (2004).
Results of the FOB showed no exposure-related findings of neurotoxicity. The
hematology evaluations indicated that 1,1,2,2-tetrachloroethane affected the circulating erythroid
mass in both genders (Table 4-3). There was evidence of a transient erythrocytosis, as shown by
increases in hematocrit values, Hb concentration, and erythrocyte counts on days 5 and 21 at
>170 mg/kg-day. The erythrocytosis was not considered clinically significant and disappeared
by week 14, at which time minimal to mild, dose-related anemia was evident, as shown by
decreases in hematocrit and Hb at >40 mg/kg-day. For example, although males exposed to
40 mg/kg-day showed a statistically significant decrease in Hb at week 14, the magnitude of the
change was small (3.8%). The anemia was characterized as microcytic based on evidence
suggesting that the circulating erythrocytes were smaller than expected, including decreases in
mean cell volumes, mean cell Hb values, and mean cell Hb concentration in both genders at
>80 mg/kg-day at various time points. At week 14, there were no changes in reticulocyte counts,
suggesting that there was no erythropoietic response to the anemia, which was in turn supported
by the bone marrow atrophy observed microscopically. As discussed by NTP (2004), the
erythrocytosis suggested a physiological response consistent with hemoconcentration due to
dehydration, as well as compromised nutritional status due to the reduced weight gain and food
consumption, both of which may have contributed to the development of the anemia.
18
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-3. Serum chemistry and hematology changes3 in rats exposed to
dietary l,l?2,2-tetrachloroethane for 14 weeks
Oral dose (mg/kg-d)
Vehicle
control
20
40
80
170
320
Males (10/group)
Serum total
protein (g/dL)
Serum cholesterol
(mg/dL)
ALT (IU/L)
ALP (IU/L)
SDH (IU/L)
Bile acids (umol/L)
Hematocrit (%)
(automated)
Hb (g/dL)
Mean cell volume (fL)
Mean cell Hb (pg)
Platelets (103/uL)
7.2±0.1
73 ±2
48 ±2
256 ±7
23 ±1
29. 2 ±2. 9
45.2±0.5
15.8±0.1
50.7±0.1
17.7±0.1
728.4 ±12. 3
7. 3 ±0.1
74 ±3
49 ±2
260 ±5
27±lb
27. 5 ±2.7
44. 9 ±0.4
15.6±0.1
51.8±0.3
18.1±0.1
707.0 ±5. 8
7.3 ±0.1
76 ±2
53±2
248 ±5
26 ±2
27.2 ±2.7
44.0 ±0.9
15.2±0.3b
52. 3 ±0.2
18.0±0.1
727.0 ±25. 2
7. 3 ±0.1
67 ±2
69±3b
245 ±6
31±lb
35.9±3.9
43. 3 ±0.7
14.9±0.1b
51. 3 ±0.2
17.7 ±0.2
716.3 ±9.7
6.7±0.1b
68 ±2
115±8b
353±12b
47±2b
92.0±16.6b
43.1±0.6b
14.6±0.1b
49.4 ±0.2
16.8±0.1b
692.8 ±12.6b
6.0±0.1b
65±2b
292±18b
432±24b
74±4b
332.4 ±47.4b
39.0±l.lb
13.6±0.3b
44.4±0.4b
15.5±0.2b
773.4±23.2b
Females (10/group)
Serum total
protein (g/dL)
Serum cholesterol
(mg/dL)
ALT (IU/L)
ALP (IU/L)
SDH (IU/L)
Bile acids (umol/L)
Hematocrit (%)
(automated)
Hb (g/dL)
Mean cell volume (fL)
Mean cell Hb (pg)
Platelets (103/uL)
7.2±0.1
104 ±4
46 ±2
227 ±5
27 ±1
37.0±7.1
42.8 ±0.4
15.2±0.1
55.4±0.1
19.7±0.1
742.1 ±20.4
7. 3 ±0.0
105 ±3
42 ±1
216±4
27 ±1
46.6 ±6.5
43.2±0.4
15.3±0.1
56.1 ±0.1
19.8±0.1
725. 9 ±12.7
7.3 ±0.1
98 ±1
41±2
220 ±3
28 ±2
39.1 ±5.6
42.1 ±0.4
14.9±0.1
55.8±0.1
19.7±0.1
733. 9 ±8. 8
6.9±0.1
81±2b
49 ±2
225 ±11
25 ±1
36. 3 ±3. 9
40.1±0.5b
14.2±0.2b
53.3±0.2b
18.9±0.1b
727.4 ±14.2
6.4±0.1b
64±3b
112±7b
341 ±7b
45±3b
39.3 ±7. 9
42.8 ±0.7
14.5±0.2b
49.0±0.2b
16.6±0.2b
639.4 ±9.9b
5.6±0.1b
55±3b
339±18b
468 ± 22b
82±3b
321.5±50.6b
34.7±0.7b
12.5±0.2b
44.4±0.4b
16.0±0.2b
662.5 ±19.4b
1
2
3
4
5
"Mean ± standard error.
bStatistically significantly different from control value.
ALP = alkaline phosphatase; IU = international units; SDH = sorbitol dehydrogenase
Source: NTP (2004).
Changes in serum clinical chemistry parameters indicative of liver damage were observed
in both genders, occurring at all time points (day 5, day 21, and week 14) and increasing in
magnitude with increasing dose and time. At week 14 (Table 4-3), these effects included
statistically significant increases in ALT and sorbitol dehydrogenase (SDH) activity in males at
>80 mg/kg-day (41, 134, and 496%, and 15, 74, and 174%, respectively) and females at
19
DRAFT - DO NOT CITE OR QUOTE
-------
1 >170 mg/kg-day (167 and 707%, and 67 and 204%, respectively), increases in alkaline
2 phosphatase (ALP) activity in both genders at >170 mg/kg-day (36 and 66% in males and 58 and
3 117% in females), increases in bile acids levels in males at >170 mg/kg-day (233 and 1,110%)
4 and females at 320 mg/kg-day (590%), and decreases in serum cholesterol levels in females at
5 >80 mg/kg-day (23, 39, and 48%, respectively) and males at 320 mg/kg-day (12%). There were
6 no exposure-related changes in rat serum 5'-nucleotidase activity at week 14, although increases
7 occurred on day 5 in females at >20 mg/kg-day and on day 21 in males and females at 80, 170,
8 and/or 320 mg/kg-day.
9 A summary of histopathological alterations following 1,1,2,2-tetrachloroethane exposure
10 is presented in Table 4-4. Hepatic cytoplasmic vacuolization was noted in males exposed to
11 >20 mg/kg-day and in females exposed to >40 mg/kg-day. Although incidence of this alteration
12 was high in affected groups, severity was only minimal-to-mild and only increased with dose
13 from 20 to 40 mg/kg-day in males and 40 to 80 mg/kg-day in females. Females exposed to
14 >80 mg/kg-day showed an increase in the incidence of hepatocyte hypertrophy with an increase
15 in severity and incidence with increasing exposure level, and males showed similar results at
16 exposures >170 mg/kg-day. A statistically significant increase in the incidence of hepatocellular
17 necrosis was observed in male and female rats at 170 and 320 mg/kg-day, accompanied by an
18 increased severity with an increase in dose. At >170 mg/kg-day, additional effects in the liver in
19 both genders were hepatocyte pigmentation and mitotic alteration and mixed cell foci, with bile
20 duct hyperplasia observed in females only. Pigmentation of the spleen was statistically
21 significantly increased in male rats exposed to >80 mg/kg-day and in female rats exposed to
22 >170 mg/kg-day. Other histological effects included statistically significantly increased
23 incidences of atrophy (red pulp and lymphoid follicle) in the spleen of males at 170 and 320
24 mg/kg-day and the spleen of females at 320 mg/kg-day. A statistically significant increase in
25 atrophy of bone (metaphysis) and bone marrow, prostate gland, preputial gland, seminal vesicles,
26 testes (germinal epithelium), uterus, and clitoral gland, as well as an increase in ovarian
27 interstitial cell cytoplasmic alterations, was observed in females at >170 mg/kg-day and in males
28 at 320 mg/kg-day.
29
20 DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-4. Incidences of selected histopathological lesions in rats exposed to
dietary l,l?2,2-tetrachlorethane for 14 weeks
Dose (mg/kg-d)
Vehicle
control
20
40
80
170
320
Males (10/group)
Hepatocyte cytoplasmic
vacuolization
Hepatocyte hypertrophy
Hepatocyte necrosis
Hepatocyte pigmentation
Hepatocyte mitotic alteration
Mixed cell foci
Bile duct hyperplasia
Spleen pigmentation
Spleen red pulp atrophy
Spleen lymphoid follicle atrophy
Oa
0
0
0
0
0
0
0
0
0
7b(1.3)
0
0
0
0
0
0
0
0
0
9b (2.0)
0
0
0
0
0
0
1 (1.0)
0
0
10b(1.9)
1 (1.0)
0
0
0
0
0
9b(1.0)
0
0
8b(1.4)
9b(1.3)
8b(1.0)
7b(1.0)
0
3
0
9b(1.0)
5b(1.0)
0
0
10b(3.2)
10b(1.6)
10b(1.9)
6b (2.0)
5b
10b(1.7)
9b(1.6)
9b(1.4)
5b(1.0)
Females (10/group)
Hepatocyte cytoplasmic
vacuolization
Hepatocyte hypertrophy
Hepatocyte necrosis
Hepatocyte pigmentation
Hepatocyte mitotic alteration
Mixed cell foci
Bile duct hyperplasia
Spleen pigmentation
Spleen, red pulp atrophy
Spleen lymphoid follicle atrophy
oa
0
0
0
0
0
0
1 (1.0)
0
0
0
0
0
0
0
0
0
0
0
0
10b(1.7)
0
0
0
0
0
0
0
0
0
10b (2.2)
4b(1.0)
1 (1.0)
0
0
0
0
4(1.0)
0
0
4b(1.3)
10b(1.7)
7b(1.0)
10b(1.3)
3 (2.0)
8b
5b(1.0)
8b(l.l)
0
0
0
10b(2.8)
10b(l.l)
10b(2.0)
10b(1.9)
1
10b(1.9)
8b(1.3)
9b(1.6)
3(1.0)
aValues represent number of animals with the lesion, with the severity score in parenthesis; severity grades are as
follows: 1 = minimal, 2 = mild, 3 = moderate, 4 = severe.
bSignificantly different from vehicle control group.
Source: NTP (2004).
1
2 Epididymal spermatozoal motility was statistically significantly decreased at >40 mg/kg-
3 day, with statistically significant decreases in epididymis weight at >80 mg/kg-day and cauda
4 epididymis weight at 320 mg/kg-day. Exposed female rats spent more time in diestrus and less
5 time in proestrus, estrus, and metestrus than control rats (see Section 4.3.1).
6 In summary, the NTP (2004) 14-week rat study provides evidence that the liver is a
7 primary target of 1,1,2,2-tetrachloroethane toxicity. At the lowest dose tested, 20 mg/kg-day,
8 there was a significant increase in the incidence of hepatic cytoplasmic vacuolization in males.
9 At 40 mg/kg-day, significant increases in relative liver weights were observed in both males and
10 females. Hepatocellular hypertrophy and spleen pigmentation were observed at 80 mg/kg-day in
11 both males and females, although these changes were generally of minimal severity. Increases in
21
DRAFT - DO NOT CITE OR QUOTE
-------
1 serum ALT and SDH, were observed at 80 mg/kg-day in males and at 170 mg/kg-day in females.
2 Decreases in serum cholesterol levels were decreased in females at 80 mg/kg-day and at 320
3 mg/kg-day in males. A decrease in body weight (>10%) was observd at 170 mg/kg-day in both
4 males and females. Increases in serum ALP activity and bile acids levels, hepatocellular necrosis,
5 bile duct hyperplasia, hepatocellular mitotic alterations, foci of cellular alterations, and liver
6 pigmentation occurred at 170 and/or 320 mg/kg-day. A no-observed-adverse-effect level
7 (NOAEL) of 20 mg/kg-day and a lowest-observed-adverse-effect level (LOAEL) of 40 mg/kg-
8 day was identified by EPA for increased relative liver weight in male and female rats. NTP
9 (2004) identified a NOAEL of 20 mg/kg-day in rats based on survival and body weight changes
10 and increased lesion incidences. There were no clinical signs of neurotoxicity at doses as high as
11 320 mg/kg-day or exposure-related findings in the FOB at doses as high as 80 mg/kg-day
12 (highest tested dose in the FOB), indicating that the nervous system may be less sensitive than
13 the liver for subchronic dietary exposure.
14 NTP (2004) also exposed groups of male and female B6C3Fi mice (10/sex/group) to
15 diets containing 0, 589, 1,120, 2,300, 4,550, or 9,100 ppm of microencapsulated 1,1,2,2-tetra-
16 chloroethane for 14 weeks, with vehicle and untreated control groups for each gender. The
17 reported average daily doses were 0, 100, 200, 370, 700, or 1,360 mg/kg-day for males and 0, 80,
18 160, 300, 600, or 1,400 mg/kg-day for females. Endpoints evaluated throughout the study
19 included clinical signs, body weight, and feed consumption. Clinical chemistry was assessed at
20 the end of the study, but hematological evaluations and urinalyses were not performed.
21 Necropsies were conducted on all animals and selected organs (liver, heart, right kidney, lung,
22 right testis, and thymus) were weighed. Comprehensive histological examinations were
23 performed on untreated control, vehicle control, and high dose groups. Tissues examined in the
24 lower dose groups were limited to the liver, spleen, and thymus in both genders; preputial gland
25 in males; and lungs in females. An FOB (21 parameters) was performed on mice in both control
26 and 160/200, 300/370, and 600/700 mg/kg-day (1,120, 2,300, and 4,550 ppm, respectively) dose
27 groups during weeks 4 and 13. Sperm motility, vaginal cytology, estrous cycle length, and
28 percentage of time spent in the various estrus stages were evaluated in both control and 160/200,
29 600/700, and 1,360/1,400 mg/kg-day (1,120, 2,300, and 4,550 ppm, respectively) dose groups.
30 All mice survived to the end of the study (NTP, 2004). Thinness was observed clinically
31 in male mice (3/10, 9/10, 10/10) at 370, 700, and 1,400 mg/kg-day, respectively, and in female
32 mice (1/10, 2/10, 10/10) at 300, 600, and 1,360 mg/kg-day, respectively. Final body weights
33 were statistically significantly lower than vehicle controls in male mice at 370, 700, and
34 1,360 mg/kg-day (12, 16, and 23%, respectively) and female mice at 600 and 1,400 mg/kg-day
35 (11 and 12%, respectively) (Table 4-5). Feed consumption was less than controls in males at
36 >700 mg/kg-day, but similar to controls in females.
37
22 DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-5. Final body weights (g) and percent change compared to controls
in B6C3Fi mice exposed to 1,1^2,2-tetrachloroethane in feed for 14 weeks
Dose
(mg/kg-d)
Vehicle control
100
200
370
700
1,360
n
10
10
10
10
10
10
Vehicle control
80
160
300
600
1,400
10
10
10
10
10
10
Males
30.1±0.6a
30.6 ±0.6
30.0 ±0.3
26.5±0.4b
25.2±0.2b
23.1±0.5b
-
2%
0
-12
-16
-23
Females
24.3±0.5a
24.2 ±0.2
24. 3 ±0.6
23. 3 ±0.4
21.7±0.2b
21.5±0.6b
-
0%
0
-4
-11
-12
1
2
3
4
5
6
7
8
9
10
11
"Mean ± standard error.
V<0.05.
Source: NTP (2004).
Statistically significant increases in absolute liver weights were observed in the male
mice exposed to 200 and 370 mg/kg-day (16 and 10%, respectively), but not at higher doses, and
in female mice exposed to >80 mg/kg-day (11, 29, 27, 22, and 32%, respectively) (Table 4-6a).
Statistically significant increases in relative liver weights were observed in male mice at
>200 mg/kg-day (16, 24, 24, and 38%, respectively) and in female mice at >80 mg/kg-day (11,
28, 33, 36, and 49%, respectively) (Table 4-6b). Other organ weight changes (increased kidney
weights in males at >370 mg/kg-day and decreased thymus weights in both genders at 1,360/
1,400 mg/kg-day) were considered to be secondary to the body weight changes. Results of the
FOBs showed no exposure-related neurotoxicity.
23
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-6a. Absolute liver weights (g) and percent change compared to
controls in B6C3Fi mice exposed to l,l?2,2-tetrachloroethane in feed for
14 weeks
Dose
(mg/kg-d)
Vehicle control
100
200
370
700
1,360
n
10
10
10
10
10
10
Vehicle control
80
160
300
600
1,400
10
10
10
10
10
10
Males
1.467 ±0.020
1.557±0.039
1.701±0.020b
1.607 ±0.038b
1.531 ±0.052
1.558±0.045
-
6%
16
10
4
6
Females
1.048 ±0.028
1.160±0.022b
1.356±0.058b
1.336±0.037b
1.277 ±0.030b
1.386±0.047b
-
11%
29
27
22
32
"Mean ± standard error.
V<0.05.
Source: NTP (2004).
Table 4-6b. Relative liver weights (mg organ weight/g body weight) and
percent change compared to controls in B6C3Fi mice exposed to
l,l?2,2-tetrachloroethane in feed for 14 weeks
Dose
(mg/kg-d)
Vehicle control
100
200
370
700
1,360
n
10
10
10
10
10
10
Vehicle control
80
160
300
600
1,400
10
10
10
10
10
10
Males
48.84±1.17
50. 94 ±0.93
56.82±0.63b
60.63 ±1.20b
60.71±1.76b
67.43 ±1.83b
-
4%
16
24
24
38
Females
43.26 ±1.05
47.90±0.85b
55.54±1.17b
57.39±0.84b
58.73 ±1.23b
64.42 ±1.14b
-
11%
28
33
36
49
aMean ± standard error.
V<0.05.
Source: NTP (2004).
24
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
9
10
11
12
13
14
15
16
17
18
19
20
Clinical chemistry findings in the mice are summarized in Tables 4-7 and 4-8 and
included statistically significant decreases in total serum protein levels in males at >200 mg/kg-
day, total serum protein levels in females at >300 mg/kg-day, and serum albumin levels in
females at 1,400 mg/kg-day (NTP, 2004). Decreased serum albumin levels could not fully
account for the decreased total protein levels, suggesting that other factors (e.g., changes in other
protein fractions, hydration status, and/or hepatic function) contributed to the hypoproteinemia
(NTP, 2004). A statistically significant increase of serum SDH activity in females was observed
at>80 mg/kg-day (22, 111, 444, 575, and 1,181%, respectively) and in males at>200 mg/kg-day
(38, 424, 424, and 715%, respectively). A statistically significant decrease in serum cholesterol
levels was observed in females at >160 mg/kg-day (22, 38, 41, and 16%, respectively), and a
statistically significant increase in ALT activity was observed in females at >160 (30, 278, 294,
and 602%, respectively) and in males at >370 mg/kg-day (234, 177, and 377%, respectively).
Total bile acids levels increased statistically significantly in females at>160 mg/kg-day (18, 69,
97, and 290%, respectively) and in males at >370 mg/kg-day (148, 178, and 377%, respectively).
A statistically significant increase in ALP activity was observed in males (67, 83, and 136%,
respectively) and in females at 300 mg/kg-day (19, 28, 55%, respectively) at, and a statistically
significant increase in 5'-nucleotidase was observed in males at >370 mg/kg-day (88, 131, and
288%, respectively).
Table 4-7. Selected clinical chemistry changes in male mice exposed to
dietary I,l92,2-tetrachloroethane for 14 weeks
Dose (mg/kg-d)
Serum total protein
(g/dL)
Serum cholesterol
(mg/dL)
ALT (IU/L)
ALP (IU/L)
SDH (IU/L)
5'-Nucleotidase
(IU/L)
Bile acids (umol/L)
Vehicle
control
5.4±0.1a
131±7
66 ±8
85 ±2
55 ±3
18±1
25. 3 ±1.2
100
5.2±0.1
125 ±4
62 ±19
78 ±2
53 ±2
16±1
22.8 ±1.5
200
5.1±0.1b
94±3b
74 ±8
89 ±2
76±3b
18±0
24.8 ±0.6
370
5.1±0.1b
110±5
207±18b
130 ±3b
288 ± 20b
30±2b
56.5 ±5. lb
700
5.1±0.1b
112±4
172±18b
143 ±7b
288±29b
37±3b
63.3±7.5b
1,360
5.1±0.1b
126 ±5
296 ± 24b
184±llb
448±25b
62±7b
108.7 ±8.1b
aMean ± standard error.
Statistically significantly different from control value.
Source: NTP (2004).
21
25
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-8. Selected clinical chemistry changes in female mice exposed to
dietary l,l?2,2-tetrachloroethane for 14 weeks
Dose (mg/kg-d)
Serum total protein
(g/dL)
Serum cholesterol
(mg/dL)
ALT (IU/L)
ALP (IU/L)
SDH (IU/L)
5'-Nucleotidase
(IU/L)
Bile acids
(umol/L)
Vehicle control
5.6±0.1a
109 ±2
34 ±5
131±5
36 ±1
59 ±3
27.2 ±1.2
80
5.6±0.1
109 ±3
50 ±15
126 ±2
44±3b
71±2
26.1 ±1.9
160
5. 5 ±0.0
85±3b
65±5b
139±5
76±4b
84±5b
30.9±l.lb
300
5.4±0.1b
68±2b
189±33b
150 ±3b
197±15b
62 ±2
44.2±3.9b
600
5.4±0.0b
64±3b
197±21b
161 ±7b
243 ± 23b
62 ±3
51.5±3.6b
1,400
5.1±0.1b
92±4b
351±35b
195 ±6b
461±59b
83±4b
101.7 ±12.0b
1
2
3
4
5
6
7
8
9
10
11
12
13
14
"Mean ± standard error.
bStatistically significantly different from control value.
Source: NTP (2004).
The histopathological results in the B6C3Fi mice are summarized in Table 4-9. A
statistically significant increased incidence of minimal to moderate hepatocyte hypertrophy was
observed at >160 mg/kg-day in females and >200 mg/kg-day in males. The incidence of
hepatocellular necrosis was statistically significantly increased in male mice at >370 mg/kg-day
and in female mice at >300 mg/kg-day. A statistically significant increased incidence of
pigmentation and bile duct hyperplasia occurred at >300 mg/kg-day in females and >370 mg/kg-
day in males. Additionally, the histological findings included an increased incidence of preputial
gland atrophy in males in the 100, 700, and 1,360 mg/kg-day dose groups (Table 4-9), but this
effect did not appear dose-related. Based on the increase in serum SDH activity and increased
absolute and relative liver weights at 80 mg/kg-day in female mice, as well as serum chemistry
changes at > 160 mg/kg-day and clear evidence of histopathology at higher doses, a LOAEL of
80 mg/kg-day was identified based on liver toxicity.
26
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-9. Incidences of selected histopathological lesions in mice exposed to
dietary l,l?2,2-tetrachloroethane for 14 weeks
Males (10/group)
Oral dose (mg/kg-d)
Hepatocyte hypertrophy
Hepatocyte necrosis
Liver focal pigmentation
Bile duct hyperplasia
Preputial gland atrophy
Vehicle
control
Oa
0
0
0
0
100
0
0
0
0
4b (2.0)
200
7b(1.0)
1 (2.0)
0
0
2(1.0)
370
10b (2.2)
8b(l.l)
10b(1.2)
7b(1.4)
0
700
10b (2.8)
8b(1.0)
10b(1.4)
9b(1.3)
4b (2.5)
1,360
10b(3.1)
9b(1.0)
8b(1.3)
10b (2.0)
5b (2.2)
Females (10/group)
Oral dose (mg/kg-d)
Hepatocyte hypertrophy
Hepatocyte necrosis
Liver focal pigmentation
Bile duct hyperplasia
Vehicle
control
Oa
0
0
0
80
2(1.5)
0
0
0
160
9b(1.0)
0
2(1.0)
0
300
10b(1.9)
3 (1.0)
9b(1.0)
8b(1.0)
600
10b (2.5)
7b(1.0)
8b(1.0)
10b(1.4)
1,400
10b (3.0)
4b(1.0)
7b(l.l)
10b (2.0)
aValues represent number of animals with the lesion, with the severity score in parenthesis; severity grades are as
follows: 1 = minimal, 2 = mild, 3 = moderate, 4 = severe.
Significantly different from vehicle control group.
Source: NTP (2004).
1
2 4.2.1.2. Chronic Studies
3 Information on the chronic oral toxicity of 1,1,2,2-tetrachloroethane is available from a
4 bioassay in rats and mice. NCI (1978) exposed groups of 50 male and 50 female Osborne-
5 Mendel rats to 1,1,2,2-tetrachloroethane in corn oil via gavage 5 days/week for 78 weeks.
6 Vehicle and untreated control groups (20 animals/sex/species) were also used. The initial low
7 and high doses used for rats of both genders were 50 and 100 mg/kg-day. At week 15, the doses
8 were raised to 65 mg/kg-day for low-dose males and 130 mg/kg-day for high dose males. At
9 week 26, the doses were decreased to 40 mg/kg-day for the low-dose females and 80 mg/kg-day
10 for the high-dose females. Beginning at week 33, intubation of all high-dose rats was suspended
11 for 1 week followed by 4 weeks of dosing, and this cyclic pattern of dosing was maintained for
12 the remainder of the treatment period. Low-dose rats were not subject to this regimen. The
13 reported time-weighted average (TWA) doses were 62 and 108 mg/kg for male rats and 43 and
14 76 mg/kg for female rats. The exposure period was followed by a 32-week observation period in
15 which the rats were not exposed to 1,1,2,2-tetrachloroethane. Clinical signs, survival, body
16 weight, food consumption, gross pathology, and histology (32 major organs and tissues as well
17 as gross lesions) were evaluated.
18 There were no clear effects on survival in the male rats. In females, survival in the
19 vehicle control, low-dose, and high-dose groups at the end of the study was 70, 58, and 40%,
27
DRAFT - DO NOT CITE OR QUOTE
-------
1 respectively. Although there was a statistically significant association between increased
2 mortality and dose in the females, the increased mortality was affected by the deaths of 10 high-
3 dose females, 8 with pneumonia and 2 with no reported lesions, during the first 5 weeks of the
4 study. The study authors also stated that there was no evidence that the early deaths were tumor-
5 related. The male and female rats also demonstrated an increased incidence of endemic chronic
6 murine pneumonia. Incidences of chronic murine pneumonia in the vehicle control, low-, and
7 high-dose groups were 40, 68, and 76% in females and 55, 50, and 65% in males. Clinical
8 observations included squinted or reddened eyes in all control and treated groups of both genders,
9 but these effects occurred with greater frequency in the exposed rats. There was a low or
10 moderate incidence of labored breathing, wheezing, and/or nasal discharge in all control and
11 treated groups during the first year of the study, and near the end of the study these signs were
12 observed more frequently in the exposed animals.
13 Dose-related decreases in body weight gain were observed. However, as the study
14 approached termination (weeks 100-110), the differences in body weight across the dose groups
15 decreased.
16 Histopathological effects included a dose-related increased incidence of hepatic fatty
17 metamorphosis in high-dose males (2/20, 0/20, 2/50, and 9/49 in the untreated control, vehicle
18 control, low-dose, and high-dose groups, respectively). In addition, inflammation, focal cellular
19 changes, and angiectasis were observed in male and female rats but were not statistically
20 significant or biologically relevant. NCI (1978) stated that the inflammatory, degenerative, and
21 proliferative lesions observed in the control and dosed animals were similar in incidence and
22 type to those occurring in naturally aged rats.
23 A statistically significant increase in tumor incidence was not observed in the rats;
24 however, two hepatocellular carcinomas, which are rare tumors in male Osborne-Mendel rats
25 (NCI, 1978), as well as one neoplastic nodule, were observed in the high-dose males
26 (Table 4-10). A hepatocellular carcinoma was also observed in an untreated female control.
27 Although interpretation of this study is complicated by the chronic murine pneumonia, it is
28 unlikely to have contributed to the fatty metamorphosis observed in the liver of male rats.
29
28 DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-10. Incidence of neoplasms in male Osborne-Mendel rats exposed to
1,1^2,2-tetrachloroethane in feed for 78 weeks
Neoplasm
Papilloma, stomach
Squamous cell carcinoma, stomach
Neoplastic nodule/carcinoma, liver
Follicular-cell carcinoma, thyroid
Hemangiosarcoma, all sites
Adenocarcinoma, mammary gland
Fibroadenoma, mammary gland
Chromophode adenomas, pituitary
Islet-cell adenomas, pancreatic islets
Fibroma, subcutaneous tissue
Dose (mg/kg-d)
Control
Vehicle
control
62
108
Males
0/20
0/20
0/20
1/19
0/20
1/20
1/20
2/20
0/20
0/20
0/20
0/20
0/20
3/20
0/20
2/20
1/'20
5/14
2/20
1/20
0/50
0/50
0/50
0/49
2/50
2/50
1/50
5/48
2/49
2/50
1/48
1/48
3/49
2/48
3/49
0/49
0/49
5/48
2/49
2/49
Source: NCI (1978).
1
2 In addition, one papilloma of the stomach, one squamous-cell carcinoma of the stomach,
3 two follicular-cell carcinomas of the thyroid, and three hemangiosarcomas were each observed in
4 high-dose males (Table 4-10). In the low-dose males, two mammary gland adenocarcinomas
5 (2/20 in vehicle controls) and two hemangiosarcomas (0/20 in vehicle control) were observed.
6 Adenomas were observed as follows: pituitary chromophobe adenomas in the vehicle control
7 (5/14) and low- and high-dose males (5/48 and 5/48, respectively); pancreatic islet-cell
8 adenomas in the vehicle control (2/20) and low- and high-dose males (2/49 and 2/49,
9 respectively); mammary gland fibroadenomas in the vehicle control (1/20) and low-dose males
10 (1/50); and subcutaneous tissue fibromas in the vehicle control (1/20) and low- and high-dose
11 females (2/50 and 2/49, respectively). In male rats, the incidence of chromophobe adenomas,
12 islet-cell adenomas, and follicular-cell carcinomas in the vehicle controls was significantly
13 increased over the incidence in historical controls (NCI, 1978).
14 In the female rats (Table 4-11), one follicular-cell carcinoma was observed in both the
15 low- and high-dose groups. One mammary gland adenocarcinoma was observed in a low-dose
16 female, and two were observed in the high-dose group. One hemangiosarcoma was observed in
17 a low-dose female. Adenomas were observed as follows: pituitary chromophobe adenomas in
18 the vehicle control (3/20) and low- and high-dose females (11/49 and 6/48, respectively); one
19 pancreatic islet-cell adenoma in a low-dose female; mammary gland fibroadenomas in the
20 vehicle control (9/20) and low- and high-dose females (13/50 and 11/50, respectively); and
21 subcutaneous tissue fibromas in the vehicle control (1/20) and low- and high-dose females
22 (2/50 and 1/50, respectively). The incidence of fibroadenomas of the mammary gland in the
29
DRAFT - DO NOT CITE OR QUOTE
-------
1 vehicle control group was statistically significantly increased over the incidence in historical
2 controls (NCI, 1978).
3
Table 4-11. Incidence of neoplasms in female Osborne-Mendel rats exposed
to l,l?2,2-tetrachloroethane in feed for 78 weeks
Neoplasm
Adenocarcinoma, mammary gland
Fibroadenoma, mammary gland
Hemangiosarcomas, uterus
Chromophode adenomas, pituitary
Islet-cell adenomas, pancreatic islets
Follicular-cell carcinoma, thyroid
Fibroma, subcutaneous tissue
Dose (mg/kg-d)
Control
Vehicle
control
43
76
Females
2/20
2/20
0/20
6/19
1/20
0/20
0/20
0/20
9/20
0/20
3/20
0/20
0/20
1/20
1/50
13/50
1/50
11/49
1/50
1/49
2/50
2/50
11/50
0/50
6/48
0/50
1/50
1/50
4
5
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Source: NCI (1978).
NCI (1978) also exposed groups of 50 male and 50 female B6C3Fi mice to 1,1,2,2-tetra-
chloroethane in corn oil via gavage 5 days/week for 78 weeks. Initial dose levels were 100 and
200 mg/kg-day in both genders. In week 19, the doses were increased to 150 and 300 mg/kg-day,
respectively. Three weeks later, the doses were increased to 200 and 400 mg/kg-day,
respectively. In week 27, the doses were decreased to 150 and 300 mg/kg-day, respectively.
The reported TWA doses were 142 and 284 mg/kg for male and female mice. The exposure
period was followed by a 12-week observation period in which the mice were not exposed to
1,1,2,2-tetrachloroethane. Vehicle and untreated control groups (20 animals/sex) and a pooled
vehicle control were also used. The pooled vehicle control group comprised the vehicle controls
from the studies of 1,1,2,2-tetrachloroethane and chloropicrin. Clinical signs, survival, body
weight, food consumption, gross pathology, and histology (32 major organs and tissues as well
as gross lesions) were evaluated.
A statistically significant association between mortality and dose was observed, as
survival was markedly decreased in the high-dose male and female mice. Terminal survival data
were not reported for the males, although acute toxic tubular nephrosis was determined to be the
apparent cause of death in 33 high-dose males dying between weeks 69 and 70. Survival in the
vehicle control, low-dose, and high-dose females at the end of the study was 75, 74, and 34%,
respectively, but the cause of death in the high-dose females was not reported. The male and
female mice also demonstrated an increased incidence of endemic chronic murine pneumonia.
Incidences of chronic murine pneumonia in the vehicle control, low-, and high-dose groups were
11, 0, and 2% in males and 5, 13, and 18% in females.
30
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
9
10
A high incidence (approximately 95%) of pronounced abdominal distension, possibly
resulting from liver tumors, was observed in the high-dose females beginning in week 60 and
continuing throughout the recovery period. Nodular hyperplasia and organized thrombus were
observed in male and female mice, but the incidences were not statistically significant.
Nonneoplastic lesions observed included hydronephrosis (16/46) and chronic inflammation in
the kidneys (5/46) in high-dose females and chronic inflammation in the low- (13/39) and high-
dose (10/47) males (Table 4-12). In addition, acute toxic tubular nephrosis was observed, and
was the apparent cause of death as identified by the study authors, in high-dose male mice that
died during weeks 69 and 70.
Table 4-12. Incidence of nonneoplastic kidney lesions observed in male and
female B6C3Fi mice exposed to l,l?2,2-tetrachloroethane in feed for 78
weeks
Lesion
Chronic inflammation - kidney
Hydro nephro sis
Chronic inflammation
Dose (mg/kg-d)
Control
Vehicle
control
142
284
Males
7/19
5/18
13/39
10/47
Females
0/19
0/19
0/20
0/20
0/46
0/46
16/46
5/46
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Source: NCI (1978).
Statistically significant increases in the incidences of hepatocellular carcinomas occurred
in both sexes and at both dose levels (Table 4-13). The incidences in the vehicle control, pooled
vehicle control, 142, and 284 mg/kg-day groups were 1/18, 3/36, 13/50, and 44/49, respectively,
in males and 0/20, 1/40, 30/48, and 43/47, respectively, in females. Information on the
progression from preneoplastic pathology to hepatocellular carcinoma is not available due to the
lack of interim sacrifices. The hepatocellular carcinomas varied in microscopic appearance, with
some tumors composed of well-differentiated cells and a relatively uniform rearrangement of
cords, while other tumors were composed of anaplastic cells with large hyperchromatic nuclei
with eosinophilic inclusion bodies and/or vacuolated pale cytoplasm. In addition, a decrease in
the time to tumor for the hepatocellular carcinomas was also evident in both genderss of mice.
The spontaneous tumor rate for hepatocellular carcinoma in the historical vehicle controls at the
testing laboratory was 74/612 (12%) for male B6C3Fi mice and 8/560 for female B6C3Fi mice.
31
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-13. Incidence of hepatocelluar carcinomas in male and female
B6C3Fi mice exposed to l,l?2,2-tetrachloroethane in feed for 78 weeks
Hepatocellular carcinoma
Incidence
Time to first tumor
Incidence
Time to first tumor
Dose (mg/kg-d)
Vehicle control
Pooled vehicle
control
142
284
Males
1/18
72
3/36
NA
13/503
84
44/49a
52
Females
0/20
NA
1/40
NA
30/483
58
43/47a
53
1
2
3
4
5
"Significantly different from control groups.
Source: NCI (1978).
In addition to the liver tumors, alveolar/bronchiolar adenomas in the lung were observed
in the male matched vehicle controls (1/18), male and female pooled-vehicle controls (1/36 and
1/40, respectively), low-dose males and females (2/39 and 1/46, respectively), and high-dose
males and females (2/47 and 1/44, respectively) (Table 4-14). Lymphomas were observed in
low- and high-dose males (4/50 and 3/49, respectively), and in female pooled vehicle controls
(2/40) and low- and high-dose females (7/48 and 3/47, respectively).
Table 4-14. Incidence of additional neoplasms in male and female B6C3Fi
mice exposed to 1,1^2,2-tetrachloroethane in feed for 78 weeks
Neoplasm
Alveolar/bronchiolar adenomas, lung
Lymphomas, multiple organ
Alveolar/bronchiolar adenomas, lung
Lymphomas, multiple organ
Dose (mg/kg-d)
Matched control
Pooled vehicle control
142
284
Males
1/18
0/18
1/36
0/36
2/39
4/50
2/47
3/49
Females
0/20
0/20
1/40
2/40
1/46
7/48
1/44
3/47
9
10
11
12
13
14
15
Source: NCI (1978).
For chronic inflammation in the kidneys of male mice, a LOAEL of 142 mg/kg-day was
selected. A NOAEL was not identified. For hydronephrosis and chronic inflammation in the
kidneys in females, a NOAEL of 142 mg/kg-day and a LOAEL of 284 mg/kg-day were selected.
4.2.2. Inhalation Exposure
4.2.2.1. Subchronic Studies
32
DRAFT - DO NOT CITE OR QUOTE
-------
1 Truffert et al. (1977) exposed groups of female Sprague-Dawley rats (55/dose) to
2 1,1,2,2-tetrachloroethane vapor at reported calculated atmospheric concentrations of 0 or
3 560 mL/m3 5 days/week for 15 weeks (78 exposures). The daily exposure duration was 6 hours
4 for the first 8 exposures and 5 hours for the remaining 70 exposures. There is uncertainty
5 regarding the actual concentration employed due to the unusual unit of exposure (i.e., mL/m ). It
6 is assumed that mL/m3 is a volume/volume vapor concentration, so the reported concentration is
7 equivalent to 560 ppm (3,909 mg/m3). Interim sacrifices were conducted after 2, 4, 9, 19, 39,
8 and 63 exposures, although the number of animals killed at each time period was not reported.
9 This study is limited by poor reporting quality and minimal quantitative data.
10 Pronounced prostration was observed "after the first exposures to 1,1,2,2-tetrachloroethane,
11 followed by recovery". Body weight gain was decreased at the end of the study, but the
12 magnitude of the change was not reported. Increases in relative liver weights were observed
13 beginning 15 days after exposure initiation, but were not quantified. Hematological alterations
14 consisting of a decrease in hematocrit "confirmed by the joint RBC and WBC counts" were
15 observed at the end of the study, but were not quantified. A marked increase (313%) in
16 thymidine uptake in hepatic DNA was observed after four exposures, but by the ninth exposure
17 the thymidine uptake had decreased to levels similar to controls. Histological alterations were
18 observed in the liver after nine exposures and included granular appearance, cytoplasmic
19 vacuolization, and evidence of hyperplasia (increase in the number of binucleated cells and the
20 appearance of mitosis), but the alterations regressed after 19 exposures and were no longer
21 observed after 39 exposures. Incidences and severity of the liver lesions were not reported.
22 Considering the lack of incidence and severity data and other inadequately reported results, lack
23 of information on dose-response due to the use of a single exposure level, and uncertainty
24 regarding the exposure concentration, a NOAEL or LOAEL cannot be identified from this study.
25 Horiuchi et al. (1962) exposed one adult male monkey (Macaca cynomolga Linne] to
26 1,1,2,2-tetrachloroethane for 2 hours/day, 6 days/week for a total of 190 exposures in 9 months.
27 The exposure level was 2,000-4,000 ppm (13,700-27,500 mg/m3) for the first 20 exposures,
28 1,000-2,000 ppm (6,870-13,700 mg/m3) for the next 140 exposures, and 3,000^,000 ppm
29 (20,600-27,500 mg/m ) for the last 30 exposures. The TWA concentration was 1,974 ppm
30 (13,560 mg/m3). The authors noted that the monkey was weak after approximately seven
31 exposures and had diarrhea and anorexia between the 12th and 15th exposures. Beginning at the
32 15th exposure, the monkey was "almost completely unconscious falling upon his side" for 20-
33 60 minutes after each exposure. The authors noted a gradual increase in body weight during
34 months 3-5 followed by a gradual decrease until the study was terminated. Hematological
35 parameters demonstrated sporadic changes in hematocrit and RBC and WBC counts, but the
36 significance of these findings cannot be determined because there were no clear trends, only one
37 monkey was tested, and there was no control group. Histological alterations consisted of fatty
38 degeneration in the liver and splenic congestion, and no effects were observed in the heart, lung,
33 DRAFT - DO NOT CITE OR QUOTE
-------
1 kidneys, pancreas, or testes. This study cannot be used to identify a NOAEL or LOAEL for
2 subchronic exposure due to the use of a single animal without a control.
3 A 6-month inhalation study in rats was performed by the Mellon Institute of Industrial
4 Research (1947). Groups of 12 male and 12 female albino rats were exposed to 0 or 167 ppm
5 (1,150 mg/m3) of 1,1,2,2-tetrachloroethane for 7 hours/day on alternate days for the 6-month
6 study period. A statistically significant increase (15%) in kidney weight was observed in the
7 1,1,2,2-tetrachloroethane-exposed rats. The rats also appeared to develop lung lesions following
8 exposure to tetrachloroethane; however, the study authors stated that the pathology reported for
9 tetrachloroethane must be discounted due to approximately 50% of the control animals
10 demonstrating major pathology of the kidneys, liver, or lung. Meaningful interpretation of these
11 results is precluded by the observed endemic lung infection, which resulted in significant early
12 mortality in all of the rats (57 and 69% mortality in the control and tetrachloroethane-exposed
13 groups, respectively). This study also included one mongrel dog that followed the same study
14 design and evaluation as the rats. Serum phosphatase activity levels, mean of 33 units/100 mL,
15 and blood urea nitrogen levels, mean of 20.66%, were increased in the treated dog compared to
16 control values of 5.72/100 mL and 14.94%, respectively. The dog survived the 6-month
17 exposure with effects that included cloudy swelling of the liver and of the convoluted tubules of
18 the kidneys, and light congestion of the lungs. Identification of a LOAEL or NOAEL is
19 precluded by poor study reporting, high mortality in the rats, and the use of a single treated
20 animal in the dog study.
21 Kulinskaya and Verlinskaya (1972) examined effects of 1,1,2,2-tetrachloroethane on the
22 blood acetylcholine system in Chinchilla rabbits exposed to 0 or 10 mg/m3 (0 or 1.5 ppm)
23 3 hours/day, 6 days/week for 7-8.5 months. The animals were immunized twice, at 1.5-2 and
24 4 months, subcutaneously with a 1.2 and 1.5 billion microbe dose of typhoid vaccine in an
25 attempt to reveal changes in the immunological reactivity following 1,1,2,2-tetrachloroethane
26 exposures. The exposed group contained six animals, and the size of the control group was not
27 specified. In comparison with both initial and control levels, serum acetylcholine levels were
28 decreased after 1.5 months, significantly increased after 4.5 months, and significantly decreased
29 at the end of the study. The concentration of acetylcholine in the blood was increased following
30 the first immunization. No changes in serum acetylcholinesterase activity were reported,
31 although serum butyrylcholinesterase activity was reduced after 5-6 months of exposure. This is
32 a poorly reported study that did not examine any other relevant endpoints. A NOAEL or
33 LOAEL could not be identified because the changes in acetylcholine levels were inconsistent
34 across time and incompletely quantified, and the biological significance of the change is unclear.
35
36 4.2.2.2. Chronic Studies
37 In a chronic inhalation study by Schmidt et al. (1972), groups of 105 male rats were
38 exposed to 0 or 0.0133 mg/L (13.3 mg/m3) 1,1,2,2-tetrachloroethane for 4 hours daily for up to
34 DRAFT - DO NOT CITE OR QUOTE
-------
1 265 days. Subgroups of seven treated and seven control rats were killed after 110 or 265 days of
2 exposure and 60 days after exposure termination, with the remaining animals observed until
3 natural death. There were no significant alterations in survival. Weight gain in exposed rats was
4 2.1, 11.6, and 12.2% less than controls on study days 110, 260, and 324, although the only
5 statistically significant decreases in body weight gain occurred between days 90 and 170. Other
6 statistically significant changes included increased leukocyte (89%) and Pi-globulin (12%) levels
7 compared to controls after 110 days, and an increased percentage of segmented nucleated
8 neutrophils (36%), decreased percentage of lymphocytes (17%), and increased percentage of
9 liver total fat content (34%) after 265 days. There was a statistically significant decrease in
10 y-globulin levels (32%) at 60 days postexposure and a decrease in adrenal ascorbic acid content
11 (a measure of pituitary adrenocorticotropic hormone [ACTH] activity) at all three time periods
12 (64, 21, and 13%, respectively). This study is insufficient for identification of a NOAEL or
13 LOAEL for systemic toxicity because the experimental design and results were poorly reported,
14 and histological examinations were not conducted.
15
16 4.3. REPRODUCTIVE/DEVELOPMENTAL STUDIES—ORAL AND INHALATION
17 4.3.1. Oral Exposure
18 Gulati et al. (199la) exposed timed-pregnant CD Sprague-Dawley rats (8-9 animals/
19 group) to diets containing 0, 0.045, 0.135, 0.27, 0.405, or 0.54% microencapsulated
20 1,1,2,2-tetrachloroethane from gestation days (GDs) 4 through 20. Based on body weight and
21 food consumption data, the reported estimated doses of 1,1,2,2-tetrachloroethane were 0, 34, 98,
22 180, 278, or 330 mg/kg-day. Dams were sacrificed and litters were evaluated on GD 20.
23 Evaluations included maternal body weight, feed consumption and clinical signs, uterine weight,
24 and numbers of implantations, early and late resorptions, live fetuses, and dead fetuses.
25 Necropsies were performed on the maternal animals, but fetuses were not examined for
26 malformations.
27 All dams survived to study termination on GD 20. Maternal body weight was
28 statistically significantly decreased 9, 11, 14, and 24% at 98, 108, 278, and 330 mg/kg-day,
29 respectively, compared to controls, and demonstrated a dose-dependent and time-dependent
30 decrease in all dose groups. However, an increase in maternal body weight on day 20, compared
31 to body weight on day 4, was apparent for all dose groups. Daily food consumption was
32 significantly decreased in all dose groups, and this may have contributed to the decreased body
33 weights observed in the study. Four out of nine rats in the 278 mg/kg-day dose group had
34 slightly rough fur beginning on GD 10, while rough fur was present in all animals in the
35 330 mg/kg-day dose group. No statistically significant changes were observed in the numbers of
36 live fetuses/litter, dead fetuses/litter, resorptions/litter, or implants/litter. One dam in the
37 98 mg/kg-day group and four of nine dams in the 330 mg/kg-day group completely resorbed
38 their litters. At scheduled sacrifice, average fetal weights were statistically significantly
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
decreased 3.9, 12.7, 10.5, and 20.6% in the 98, 108, 278, and 330 mg/kg-day dose groups,
respectively (Table 4-15). Gravid uterine weight was statistically significantly reduced only in
the 330 mg/kg-day animals. Small, but statistically significant, decreases were seen in maternal
body weight and average fetal weight at >98 mg/kg-day. Using statistical significance and a
10% change as the criterion for an adverse change in maternal body weight, a NOAEL of 34
mg/kg-day and LOAEL of 98 mg/kg-day were selected for changes in maternal body weight. A
NOAEL of 34 mg/kg-day and LOAEL of 98 mg/kg-day were selected for developmental toxicity
based on the lowest dose that produced a statistically significant decrease in fetal body weight.
Table 4-15. Fetal body weight in CD Sprague-Dawley rats exposed
to microencapsulated l,!92,2-tetrachloroethane on gestation days
(CDs) 4-20
Dose (mg/kd-day)
0
34
98
180
278
330
N
9
8
8
9
9
5
Mean
2.28
2.17
2.19
1.99
2.04
1.81
SD
0.12
0.11
0.08
0.15
0.42
0.26
% change
4.8
3.9
12.7
10.5
20.6
Source: Gulati et al. (1991)
Gulati et al. (1991b) exposed timed-pregnant Swiss CD-I mice (n = 5-11) to diets
containing 0, 0.5, 1, 1.5, 2, or 3% microencapsulated 1,1,2,2-tetrachloroethane from GDs 4
through 17. Based on body weight and food consumption data, the reported estimated doses of
1,1,2,2-tetrachloroethane were 0, 987, 2,120, 2,216, or 4,575 mg/kg-day; an average dose could
not be calculated for the 3% group due to early mortality. Dams were sacrificed and litters were
evaluated on GD 17. Evaluations included maternal body weight, feed consumption and clinical
signs, uterine weight, and numbers of implantations, early and late resorptions, live fetuses, and
dead fetuses. Necropsies were performed on the maternal animals, but fetuses were not
examined for malformations.
All animals (9/9) in the 3% group died prior to the end of the study. Mortality was 0/11,
0/9, 2/10, 4/5, and 5/7 in the 0, 987, 2,120, 2,216, or 4,575 mg/kg-day groups, respectively, and
the mortality in the higher dose groups affected the statistical power of the study for those groups.
Maternal body weights were statistically significantly decreased compared to controls at
>2,120 mg/kg-day beginning on study day 9, although the day 17 data were not statistically
significantly different from controls for any treatment group. Average daily feed consumption
was statistically significantly decreased in all treated groups except in the 987 mg/kg-day
animals. Gross hepatic effects were reported in dams from all groups except the 987 mg/kg-day
36
DRAFT - DO NOT CITE OR QUOTE
-------
1 group and included pale or grey and/or enlarged livers and a prominent lobulated pattern.
2 Complete litter resorption occurred in 1/11, 0/9, 2/8, 1/1, and 1/2 dams in the 0, 987, 2,120,
3 2,216, and 4,575 mg/kg-day groups, respectively. No changes in developmental endpoints were
4 noted in the 987 or 2,120 mg/kg-day groups. The 2,120 and 4,575 mg/kg-day groups had too
5 few litters, due to maternal toxicity, to permit statistical analysis of the findings. The high
6 mortality in the exposed mice precluded the identification of a NOAEL or LOAEL for this study.
7 NTP (2004) conducted a 14-week study in which groups of 10 male and 10 female
8 F344 rats were fed diets containing microencapsulated 1,1,2,2-tetrachloroethane at reported
9 average daily doses of 0, 20, 40, 80, 170, or 320 mg/kg-day. The main part of this study is
10 summarized in Section 4.2.1.1. Reproductive function (fertility) was not evaluated. Endpoints
11 relevant to reproductive toxicity included histology (testis with epididymis and seminal vesicle,
12 preputial gland, prostate gland, clitoral gland, ovary, and uterus) and weights (left cauda
13 epididymis, left epididymis, and left testis) of selected reproductive tissues in all control and
14 treated groups. Sperm evaluations and vaginal cytology evaluations were performed in animals
15 in the 0, 40, 80, and 170 mg/kg-day dose groups. The sperm evaluations consisted of spermatid
16 heads per testis and per gram testis, spermatid counts, and epididymal spermatozoal motility and
17 concentration. The vaginal cytology evaluations consisted of measures of estrous cycle length.
18 Sperm motility was 17.1, 14.9, and 24.0% lower than in vehicle controls at 40, 80, and
19 170 mg/kg-day, respectively. Other statistically significant effects in the males included
20 reductions in absolute epididymis weight at >80 mg/kg-day and absolute left cauda epididymis
21 weight at 170 mg/kg-day, and statistically significant increases in the incidences (90-100%) of
22 minimal to moderate atrophy of the preputial and prostate gland, seminal vesicle, and testicular
23 germinal epithelium at 320 mg/kg-day. Effects in the females included statistically significant
24 increases in incidences of minimal to mild uterine atrophy (70-90%) at >170 mg/kg-day and
25 clitoral gland atrophy (70%) and ovarian interstitial cell cytoplasmic alterations (100%) at
26 320 mg/kg-day. The vaginal cytology evaluations indicated that the females in the 170 mg/kg-
27 day group spent more time in diestrus and less time in proestrus, estrus, and metestrus than did
28 the vehicle controls. Body weight loss and reduced body weight gain at the lower dose levels
29 may have contributed to the atrophy and other effects observed in both genders (NTP, 2004).
30 NTP (2004) also tested groups of 10 male and 10 female B6C3Fi mice that were
31 similarly exposed to 1,1,2,2-tetrachloroethane for 14 weeks at reported average daily dietary
32 doses of 0, 100, 200, 370, 700, or 1,360 mg/kg-day (males) or 0, 80, 160, 300, 600, or
33 1,400 mg/kg-day (females). The main part of this study is summarized in Section 4.2.1.1.
34 Reproductive function (fertility) was not evaluated, and toxicity endpoints in reproductive organs
35 are the same as those evaluated in the rat part of the study summarized above. The sperm and
36 vaginal cytology evaluations were performed in the 0, 1,120, 4,550, or 9,100 mg/kg-day dose
37 groups.
37 DRAFT - DO NOT CITE OR QUOTE
-------
1 Effects observed in the male mice included statistically significant increases in the
2 incidence of preputial gland atrophy at 100, 700, and 1,360 mg/kg-day (incidences in the control
3 to high dose groups were 0/10, 4/10, 2/10, 0/10, 4/10, and 5/10, respectively), decreased absolute
4 testis weight at >700 mg/kg-day and absolute epididymis and cauda epididymis weights at
5 1,360 mg/kg-day, and decreased epididymal spermatozoal motility at 1,360 mg/kg-day (3.1%
6 less than vehicle controls). In female mice, the length of the estrous cycle was significantly
7 increased at 9,100 pm (1,400 mg/kg-day) (8.7% longer than vehicle controls). The pronounced
8 decreases in body weight gain or body weight loss were similar to those observed in rats.
9
10 4.3.2. Inhalation Exposure
11 Male rats were exposed to 0 or 15 mg/m3 (2.2 ppm) 1,1,2,2-tetrachloroethane 4 hours/day
12 for up to 8 days in a 10-day period (Gohlke and Schmidt, 1972; Schmidt et al, 1972).
13 Reproductive function was not tested, but evaluations included histological examinations of the
14 testes in groups of seven control and seven treated males following the second, fourth, and eighth
15 exposures, as detailed in Schmidt et al. (1972) in Section 4.2.2.2. This study is limited by
16 imprecise and incomplete reporting of results. It was noted that testicular histopathology,
17 described as atrophy of the seminal tubules with strongly restricted or absent spermatogenesis,
18 was observed in five exposed animals following the fourth exposure; data for the other time
19 periods and the control group were not reported.
20 The Schmidt et al. (1972) chronic inhalation study, summarized in Section 4.2.2.2,
21 included a limited reproductive function/developmental toxicity assessment. Male rats were
22 exposed to 0 or 13.3 mg/m3 (1.9 ppm) 1,1,2,2-tetrachloroethane 4 hours/day for 265 days, as well
23 as during the mating period. One week before the end of the exposure period, seven control and
24 seven exposed males were each mated with five unexposed virgin females. Dams were
25 permitted to deliver and the offspring were observed for 84 days and were examined
26 macroscopically for malformations. The percentage of mated females having offspring, littering
27 interval, time to 50% littered, total number of pups, pups/litter, average birth weight, postnatal
28 survival on days 1, 2, 7, 14, 21, and 84, sex ratio, and average body weight on postnatal day 84
29 were also measured. No macroscopic malformations or significant group differences in the other
30 indices were found, indicating that 13.3 mg/m3 was a NOAEL for male reproductive toxicity.
31 No effects attributable to 1,1,2,2-tetrachloroethane were reported in rats exposed to 5 or
32 50 ppm (34.3 or 343 mg/m3, respectively) 7 hours/day for 5 days in a dominant lethal test
33 (McGregor, 1980). A viral infection may have resulted in increased numbers of early deaths in
34 all groups, including the control group, possibly affecting study sensitivity. The frequency of
35 sperm with hook abnormalities was statistically significantly increased in the 343 mg/m3 group,
36 but not at 34.3 mg/m3.
37
38 4.4. OTHER DURATION-OR ENDPOINT-SPECIFIC STUDIES
3 8 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
4.4.1. Acute Studies (Oral and Inhalation)
4.4.1.1. Oral Studies
Oral (single-dose gavage) median lethal dose (LDso) values of 250-800 mg/kg have been
reported in rats (NTP, 2004; Schmidt et al., 1980b; Gohlke et al, 1977; Smyth et al, 1969).
Cottalasso et al. (1998) described a series of experiments evaluating the effect of a single gavage
dose of 1,1,2,2-tetrachloroethane on the liver of exposed rats. In the first experiment, male
Sprague-Dawley rats (5/group) were given a single gavage dose of 0, 143.5, 287, 574, or
1,148 mg/kg in mineral oil and five animals from each group were sacrificed 5, 15, 30, or
60 minutes later. Sixty minutes after treatment, statistically significant, dose-related increases in
serum activity levels of AST (66, 129, and 201%, respectively) and ALT (54, 88, and 146%,
respectively) were observed at >287 mg/kg. The increase in rat serum activities of AST and
ALT were also increased in a time-dependent manner. Serum AST increased 13-130% from 5
to 60 minutes in rats at 574 mg/kg-day and serum ALT increased 8-88% from 5 to 60 minutes.
A statistically significant decrease in hepatic microsomal G6Pase activity (19, 36, and 47%,
respectively) was observed at >287 mg/kg. A statistically significant decrease in levels of
dolichol, a polyisoprenoid compound believed to be important in protein glycosylation reactions,
in the liver (41 and 56%, respectively) and a statistically significant increase in triglyceride
levels in liver homogenate (60 and 83%, respectively) were observed at >574 mg/kg. A
statistically significant increase in the trigylceride levels in liver microsomes (46, 65, and 97%,
respectively) was observed at >287 mg/kg. See Table 4-16 for a summary of these acute liver
toxicity results. A time-dependent effect was observed in the decrease in G6Pase, in the increase
in triglyceride levels, and in the decrease in levels of dolichol in the liver at 574 mg/kg-day from
5 to 60 minutes.
Table 4-16. Liver function and other effects observed following acute (60
minutes) exposure to 1,1^2,2-tetrachloroethane
Dose
(mg/kg)
0
143.5
287
574
1,148
Serum AST
(IU/L)
62 ±9
80 ±10
103±21a
143±13a
187±24a
Serum
ALT
(IU/L)
26 ±4
32 ±6
40±7a
49±6a
64±9a
Microsomal G6Pase
(nmol/min/mg
protein)
361 ±29
342 ±43
291±39a
230±18a
191±31a
Homogenate
triglycerides
(mg/g liver)
14. 5 ±2.0
15.9±2.3
19.7 ±3.2
23.2±2.8a
26.5±3.4a
Microsomal
triglycerides
(mg/g liver)
1.61 ±0.12
1.95 ±0.21
2.35±0.30a
2.65±0.35a
3.17±0.42a
Homogenate total
dolichol levels
(ng/mg protein)
335 ±0.28
302 ±53
268 ± 45
197±25a
147±21a
25
26
27
Significantly different from control.
Source: Cottalasso et al. (1998).
Schmidt et al. (1980b) administered 0 or 100 mg/kg doses of 1,1,2,2-tetrachloroethane in
corn oil by gavage to groups of 10 male Wistar rats, followed immediately by increased
39
DRAFT - DO NOT CITE OR QUOTE
-------
1 environmental temperatures, and evaluated hepatic effects 20-22 hours post administration.
2 Statistically significant increases in serum leucine aminopeptidase activity, hepatic ascorbic acid,
3 and hepatic triglyceride levels (10.5, 22.3, and 125% greater than control levels, respectively)
4 were observed, but changes in body weight, liver weight, hepatic N-demethylation of
5 aminopyrine, and serum ALT activity were not observed. The report includes a general
6 statement that all chemicals tested in this study led to necrosis and fatty degeneration, which
7 suggests that 100 mg/kg was a hepatotoxic dose of 1,1,2,2-tetrachloroethane. However, the
8 significance of the histology results cannot be assessed due to a lack of incidence and severity
9 measures. No other 1,1,2,2-tetrachloroethane-related histological data were reported in this
10 study.
11 Wolff (1978) exposed 8- to 10-week-old, female Wistar rats in groups of 8-10 animals,
12 to a single gavage dose of 0, 25, or 50 mg/kg of 1,1,2,2-tetrachloroethane 30 minutes prior to
13 testing for passive avoidance (shock level of 0.4 milliamperes [mA]). Passive avoidance was
14 measured by allowing the test rats to explore the test apparatus, which consisted of a larger, lit
15 box and a smaller, dark box. After 180 seconds, the darkened box received an electrical shock
16 through the grid floor. During the 180 seconds, the rats remained in the darkened box
17 approximately 80% of the time. The test was repeated 24 hours later. No differences in
18 avoidance were observed between the control and 25 mg/kg groups, but decreased passive
19 avoidance behavior was reported following exposure to 50 mg/kg. In the second test series, the
20 shock level was increased to 0.8 mA and the 1,1,2,2-tetrachloroethane dose was increased to
21 50 mg/kg. The 1,1,2,2-tetrachloroethane doses were then increased to 80 mg/kg and then to
22 100 mg/kg. Increasing the shock level to 0.8 mA resulted in no significant differences in
23 avoidance between the controls and the 50 mg/kg-day dose group (n = 10). Passive avoidance
24 was altered at 80 mg/kg (n = 10), and at 100 mg/kg, the animals (n = 10) were ataxic and did not
25 learn to avoid the shock. The authors stated that the treatment with 1,1,2,2-tetrachloroethane
26 may have affected the threshold of perception of the shock, rather than memory (Wolff, 1978).
27 This conclusion would be consistent with the high-dose anesthetic effects characteristic of
28 volatile organic compounds in general.
29
30 4.4.1.2. Inhalation Studies
31 Schmidt et al. (1980a) established a 24-hour median lethal concentration (LCso) of
32 8,600 mg/m3 (1,256 ppm) for 1,1,2,2-tetrachloroethane in rats for a single 4-hour exposure.
33 Carpenter et al. (1949) found that a 4-hour exposure to 1,000 ppm 1,1,2,2-tetrachloroethane
34 (6,870 mg/m ) was lethal in Sherman rats, with mortality in "2/6, 3/6, or 4/6" animals.
35 Price et al. (1978) exposed rats and guinea pigs to 576, 5,050, and 6,310 ppm
36 1,1,2,2-tetrachloroethane for 30 minutes. Rats exposed to 576 ppm (3,950 mg/m3) for
37 30 minutes showed a slight reduction in activity and alertness, while increasing the concentration
38 to 5,050 or 6,310 ppm (34,700 or 43,350 mg/m3) caused lacrimation, ataxia, narcosis, labored
40 DRAFT - DO NOT CITE OR QUOTE
-------
1 respiration, and 30-50% mortality (Price et al., 1978). Eye closure, squinting, lacrimation, and
2 decreased activity were observed in guinea pigs exposed to 576 ppm for 30 minutes; exposure to
3 5,050 ppm resulted in tremors, narcosis, and labored breathing, and exposure to 6,310 ppm
4 produced 30% mortality (Price et al., 1978). Organ weight measurements and gross pathology
5 and histology evaluations performed 14 days following the 30-minute exposures did not result in
6 chemical-related effects in the lungs, liver, kidneys, heart, brain, adrenals, testes, epididymides,
7 ovaries, or uterus in either species.
8 Pantelitsch (1933) exposed groups of three mice to 1,1,2,2-tetrachloroethane concent-
9 rations of 7,000, 8,000-10,000, 17,000, 29,000, or 34,000 mg/m3 (1,022, 1,168-1,460, 3,060,
10 5,220, or 6,120 ppm, respectively) for approximately 1.5-2 hours and examined changes in
11 clinical status of the animals. All concentrations resulted in disturbed equilibrium, prostration,
12 and loss of reflexes, with deaths occurring at >8,000-10,000 mg/m3; increasing the concentration
13 resulted in a more rapid onset of symptoms.
14 Horvath and Frantik (1973) determined that effective concentrations of 1,1,2,2-tetra-
15 chloroethane following a single 6-hour exposure in rats were 360 ppm (2,470 mg/m3) for a 50%
16 decrease in spontaneous motor activity and 200 ppm (1,370 mg/m3) for a 50% increase in
17 pentobarbital sleep time. No additional relevant information was reported.
18 Schmidt et al. (1980a) exposed groups of 10 male Wistar rats to 0, 410, 700, 1,030, 2,100,
19 or 4,200 mg/m3 (0, 60, 102, 150, 307, or 613 ppm, respectively) 1,1,2,2-tetrachloroethane (mean
20 concentrations) for 4 hours and evaluated the animals immediately (within 15-100 minutes), at
21 24 hours, or at 120 hours following exposure. The purpose of this study was to determine a
22 threshold concentration for effects on the liver following inhalation exposure. Evaluation of this
23 study is complicated by imprecise and incomplete reporting of results, exposure levels, and
24 observation durations. For example, results for endpoints other than liver histology, ascorbic
25 acid content, and histochemistry were not reported for the lowest concentration (410 mg/m3), and
26 liver ascorbic acid content and serum and liver triglyceride levels were the only results reported
27 quantitatively. Histological effects included diffuse fine droplet fatty degeneration in the liver at
28 410 and 700 mg/m3 (24 hours postexposure), nonspecific inflammation and Councilman bodies
29 (eosinophilic globules derived from necrosis of single hepatocytes) in the liver at 4,200 mg/m3
30 (24 hours postexposure), and interstitial nephritis in the kidneys at 700 mg/m3 (120 hours
31 postexposure). Additional information on these findings, including incidences and results for
32 other exposure concentrations, was not reported.
33 Hepatic ascorbic acid levels were statistically significantly increased in groups exposed
34 to >700 mg/m immediately after exposure (2, 64, 29, 167, and 182% higher than controls at 410,
35 700, 1,030, 2,100, and 4,200 mg/m3, respectively), but returned to control levels within 24 hours.
36 Serum triglyceride concentrations were statistically significantly decreased at >700 mg/m3 after
37 24 hours (35, 23, 29, and 56% at 700, 1,030, 2,100, and 4,200 mg/m3, respectively) and at
38 2,100 and 4,200 mg/m3 (39 and 42%, respectively) after 120 hours. Hepatic triglyceride levels
41 DRAFT - DO NOT CITE OR QUOTE
-------
1 were significantly increased at 2,100 and 4,200 mg/m3 (92 and 76%, respectively) at 24 hours
2 postexposure. Hexobarbital sleep time was increased at 2,100 and 4,200 mg/m3 (not quantified).
3 Assessing the biological significance and adversity of the effects in this study is complicated by
4 factors that include the lack of liver lesion incidence data, the paucity of other quantitative data,
5 and other reporting insufficiencies. The authors concluded that the threshold for effects on the
6 liver was between 410 and 700 mg/m3 because the fine droplet fatty degeneration was not
7 considered to be biologically significant in the absence of accompanying serum and liver
8 biochemical changes.
9 Hepatic effects were also reported by Tomokuni (1969), who administered a single
10 3-hour exposure of 600 ppm (4,120 mg/m3) 1,1,2,2-tetrachloroethane to female Cb mice. Total
11 hepatic lipids and triglycerides were statistically significantly increased following exposure and
12 continued to increase for 8 hours postexposure. Hepatic triglyceride levels increased more than
13 total lipid levels for 8 hours postexposure. Total hepatic adenosine triphosphate (ATP) levels
14 were decreased immediately following exposure and continued to decrease over the next 8 hours.
15 A later study by the same investigator (Tomokuni, 1970) evaluated female Cb mice (5-8/group)
16 exposed to 800 ppm (5,490 mg/m3) 1,1,2,2-tetrachloroethane for 3 hours and then followed the
17 time-course of the changes in hepatic lipids and phospholipids over the next 90 hours. Increased
18 tricglyceride and decreased phospholipid levels were seen for the first 30-45 hours postexposure,
19 but the effects generally resolved by 90 hours postexposure, demonstrating that hepatic effects
20 resolved after exposure was terminated.
21 Horiuchi et al. (1962) exposed 10 male mice for a single 3-hour period to an atmosphere
22 containing 5,900 ppm (-40,500 mg/m3) or 6,600 ppm (-45,300 mg/m3) 1,1,2,2-tetrachloroethane
23 and then observed the animals for 1 week following exposure. Tissues were obtained for
24 histologic evaluation from animals at sacrifice or when discovered dead. Three mice exposed to
25 5,900 ppm and four mice exposed to 6,600 ppm died prior to the end of the study. The
26 histological results reported by Horiuchi et al. (1962) are similar to the repeated vapor exposure
27 study in mice, described in Section 4.4.2.2, with slight to moderatie congestion and fatty
28 degeneration of the liver and congestion of the other mail tissues.
29 Deguchi (1972) administered a single 6-hour exposure of 0, 10, 100, or 1,000 ppm (0, 69,
30 690, or 6,900 mg/m3, respectively) of 1,1,2,2-tetrachloroethane to male rats and evaluated serum
31 AST activity and ALT activity levels up to 72 hours postexposure. This study was reported in
32 Japanese and included an English translation of the abstract. Based on information in the
33 English abstract and data graphs in this Japanese study, there was a minimal increase in serum
34 AST at all exposure concentrations 72 hours postexposure.
35
36 4.4.2. Short-term Studies (Oral and Inhalation)
37 4.4.2.1. Oral Studies
42 DRAFT - DO NOT CITE OR QUOTE
-------
1 Dow Chemical Company (1988) exposed groups of male Osborne-Mendel rats (n = 5) to
2 daily gavage doses of 0, 25, 75, 150, or 300 mg/kg-day 1,1,2,2-tetrachloroethane every 24 hours
3 for 4 days, followed by an injection of [3H]-thymidine, for DNA incorporation studies, 24 hours
4 following the last 1,1,2,2-tetrachloroethane dose. The fourth dose was not administered to the
5 300 mg/kg-day group due to signs of central nervous system (CNS) depression and debilitation,
6 and one animal in this group died before [ H]-thymidine injection. Terminal body weights of the
7 300 mg/kg-day animals were statistically significantly decreased 17% compared to controls.
8 Absolute liver weights at the highest dose were decreased and relative liver weights were
9 statistically significantly increased 14% in the 150 mg/kg-day dose group.
10 Histological examinations of the livers showed increased numbers of hepatocytes in
11 mitosis in the 75, 150, and 300 mg/kg-day groups, although this response was variable in high-
12 dose rats due, possibly, to the increased toxicity observed in this group (Dow Chemical
13 Company, 1988). Increased numbers of reticuloendothelial cells were seen at 300 mg/kg-day.
14 Increased hepatic glycogen content was found in hepatocytes of 75 and 150 mg/kg-day animals,
15 although this could be an outcome of altered feeding patterns resulting from sedative effects of
16 dosing (Dow Chemical Company, 1988).
17 Hepatic DNA synthesis ([3H]-thymidine incorporation) was increased 2.8-, 4.8-, and
18 2.5-fold at 75, 150, and 300 mg/kg-day, respectively; the decline at 300 mg/kg-day may have
19 been due to the poor clinical status of the rats in this group (Dow Chemical Company, 1988).
20 Total hepatic DNA content was not increased. Other endpoints were not evaluated. The 300
21 mg/kg-day dose is a frank effect level (FEL) based on the CNS depression and mortality. The 75
22 mg/kg dose may represent a NOAEL for increased relative liver weight in rats. However, the
23 increase in DNA synthesis and mitosis are not necessarily indicative of hepatotoxicity, and the
24 histological examinations showed no accompanying degeneration or other adverse liver lesions.
25 Dow Chemical Company (1988) similarly exposed groups of male B6C3Fi mice (n = 5)
26 to daily gavage doses of 0, 25, 75, 150, or 300 mg/kg-day 1,1,2,2-tetrachloroethane for 4 days,
27 followed by [3H]-thymidine injection for the DNA incorporation studies. All animals survived
28 treatment, and changes in body weight were not observed at any dose level. Absolute and
29 relative liver weights were increased 13 and 11%, respectively, at 150 mg/kg-day and 19 and
30 72%, respectively, at 300 mg/kg-day, although only the increase in relative liver weight at 300
31 mg/kg-day was statistically significantly.
32 Histopathologic examination of the liver revealed centrilobular swelling, with a
33 corresponding decrease in hepatocyte size in the periportal region due to decreased glycogen
34 content, in mice at >75 mg/kg-day. Increased hepatocyte mitosis was also observed in mice at
35 300 mg/kg-day. Hepatic DNA synthesis was increased 1.7-fold at 150 mg/kg-day and 4.4-fold at
36 300 mg/kg-day, although total hepatic DNA content was not increased. Other endpoints were
37 not evaluated.
43 DRAFT - DO NOT CITE OR QUOTE
-------
1 TSI Mason Laboratories (1993a, unpublished) administered 1,1,2,2-tetrachloroethane in
2 corn oil to groups of male and female (n = 5) F344/N rats at 0, 135, 270, or 540 mg/kg for
3 12 days over a 16-day period. Rats were weighed prior to dosing, after 7 days, and prior to
4 euthanasia, and all surviving rats were euthanized and subject to necropsy. Study endpoints
5 included clinical observations, body weight, necropsy, selected organ weights (liver, kidneys,
6 thymus, lung, heart, and testes), and histology of gross lesions. All of the high-dose rats died by
7 day 5 of the study. Male rats exposed to 270 mg/kg displayed an increase in body weight from
8 day 1 through day 17 of 37%, compared to an increase of 64% in controls. Female rats exposed
9 to 270 mg/kg displayed a decrease in body weight from day 1 through day 17 of 3%, compared
10 with an increase of 30% in controls. The automatic watering system for the low- and high-dose
11 males failed prior to the administration of 1,1,2,2-tetrachloroethane, and the low and high doses
12 of the study were repeated in a subsequent study by TSI Mason Laboratories (1993b,
13 unpublished).
14 Clinical signs were absent in the 135 mg/kg animals, but animals exposed to 270 or
15 540 mg/kg were lethargic following treatment. Absolute liver weights were statistically
16 significantly increased (19%) in the 135 mg/kg-day female rats, while relative liver weights were
17 statistically significantly increased at both 135 and 270 mg/kg-day (16 and 34%, respectively).
18 No changes in absolute or relative liver weights were seen in exposed male rats. Absolute right
19 kidney weight was significantly increased 9 and 37% in females at 135 and 270 mg/kg-day,
20 respectively. Absolute thymus weight was statistically significantly decreased in the mid-dose
21 group of male rats (33% at 270 mg/kg-day) while absolute (45%) and relative (32%) thymus
22 weights were statistically significantly decreased in only the mid-dose females. Relative right
23 testis weight was statistically significantly increased (10% at 270 mg/kg-day) in male rats.
24 Absolute, but not relative, lung weights were statistically significantly decreased in 270 mg/kg-
25 day females (17%), while relative heart weights were statistically significantly increased (14%)
26 in females.
27 Gross and microscopic lesions were observed in the liver (i.e., hepatodiaphragmatic
28 nodules) of one control, one mid-dose, and one high-dose rat, but these were common
29 spontaneous lesions.
30 In another study, TSI Mason Laboratories (1993b, unpublished) exposed groups of male
31 F344/N rats (n = 5) to 0, 135, 270, or 540 mg/kg-day 1,1,2,2-tetrachloroethane by gavage in corn
32 oil on 12 days in a 16-day period. Study endpoints included clinical observations, body weight,
33 necropsy, selected organ weights (liver, kidneys, thymus, lung, heart, and testes), and histology
34 of gross lesions. All animals exposed to 540 mg/kg-day died by day 3 of the study. Rats in the
35 270 and 540 mg/kg-day groups were extremely lethargic following administration of the test
36 article, with recovery observed only in the 270 mg/kg-day rats.
37 The weight gain observed in the low- and mid-dose rats was 55.2 and 28%, respectively.
38 At 135 mg/kg, statistically significant increases of 17 and 13% in absolute and relative liver
44 DRAFT - DO NOT CITE OR QUOTE
-------
1 weights, respectively, were observed compared to controls. In the mid-dose group, statistically
2 significant decreases in absolute testes weight (7%), absolute kidney weight (9%), absolute and
3 relative heart weight (10 and 6%, respectively), and absolute and relative thymus weight (33 and
4 21%, respectively) were observed. Statistically significant increases in relative thymus (10%),
5 liver (16%), and kidney weights (7%) were observed at 270 mg/kg compared to controls.
6 Gross and microscopic lesions were observed in the liver of one 270 mg/kg-day male and
7 in the glandular stomach of one 540 mg/kg-day male, but these were diagnosed as spontaneous
8 lesions commonly observed in F344/N rats. The lesion observed in the liver was a dark nodule
9 on the median lobe and corresponded histomorphologically to a hepatodiaphragmatic nodule,
10 and the lesion observed in the glandular stomach was a pale foci.
11 TSI Mason Laboratories (1993c, unpublished) exposed groups of five male and five
12 female B6C3Fi mice to 0, 337.5, 675, or 1,350 mg/kg-day 1,1,2,2-tetrachloroethane by gavage in
13 corn oil on 12 days during a 16-day period. Study endpoints included clinical observations, body
14 weight, necropsy, selected organ weights (liver, kidneys, thymus, lung, heart, and testes), and
15 histology of gross lesions. All mice of both genders in the 1,350 mg/kg-day groups were found
16 dead or euthanized by day 3 of the study. Additionally, one 675 mg/kg-day female died and one
17 337.5 mg/kg-day female was euthanized prior to the end of the study.
18 No significant changes in body weight were reported in treated groups. Animals in the
19 675 and 1,350 mg/kg-day groups appeared lethargic within 15 minutes of dosing, and the
20 1,350 mg/kg-day mice failed to recover after the third treatment. Lethargy also occurred in the
21 337.5 mg/kg-day female that was sacrificed, but not in other animals in that exposure group. In
22 male mice, relative liver weight was statistically significantly increased 9% at 337.5 mg/kg, and
23 absolute and relative liver weights were statistically significantly increased 28 and 37%,
24 respectively, at 675 mg/kg-day. In female mice, absolute and relative liver weights were
25 statistically significantly increased by 50 and 42%, respectively, at 675 mg/kg.
26 Gross hepatic changes, described as pale livers, were noted in one male and three females
27 at 337.5 mg/kg-day and in four males and three females at 675 mg/kg-day. Histological
28 examination of the gross lesions showed that they correlated with centrilobular hepatocellular
29 degeneration characterized by hepatocellular swelling, cytoplasmic rarefaction, and
30 hepatocellular necrosis in the 675 and 1,350 mg/kg-day males and the 337.5, 675, and
31 1,350 mg/kg-day females. Hepatocellular necrosis was the most common lesion observed at
32 675 mg/kg-day.
33 In a study examining the potential renal toxicity of orally administered halogenated
34 ethanes, groups of five male F344/N rats received 0, 0.62, or 1.24 mmol/kg-day 1,1,2,2-tetra-
35 chloroethane by gavage in corn oil (0, 104, or 208 mg/kg-day, respectively) for 21 consecutive
36 days (NTP, 1996). All rats in the high-dose group died or were killed moribund on days 13-14
37 and were not evaluated further. Evaluations of the 0 and 104 mg/kg-day animals included
38 weekly body weights, end-of-study urinalysis (volume, specific gravity, creatinine, glucose, total
45 DRAFT - DO NOT CITE OR QUOTE
-------
1 protein, AST, y-glutamyl transpeptidase, and N-acetyl-p-D-glucosaminidase), gross necropsy,
2 selected organ weights (right kidney, liver, and right testis), selected histopathology (right kidney,
3 left liver lobe, and gross lesions), and kidney cell proliferation analysis (proliferating cell nuclear
4 antigen [PCNA] labeling index for proximal and distal tubule epithelial cells in S phase).
5 Clinical signs in the high-dose animals included thinness and lethargy (5/5 rats), diarrhea,
6 abnormal breathing, and ruffled fur (3/5 rats). In the low-dose group, no effects on survival,
7 body weight gain, urinalysis parameters, absolute or relative kidney weights, renal or testicular
8 histopathology, or kidney cell PCNA labeling index were observed.
9 Hepatic effects in the low-dose group included increased absolute and relative liver
10 weights (24 and 29% greater than controls, respectively) and cytoplasmic vacuolization of
11 hepatocytes. The vacuolation occurred in hepatocytes of all low-dose rats and consisted of
12 multifocal areas with clear droplets within the cytoplasm. Changes in the kidneys of the male
13 rats were not observed.
14 In a range-finding study, the NTP (NTP, 2004; TSI Mason Laboratories, 1993d) exposed
15 male and female F344/N rats (5/sex/group) to 0, 3,325, 6,650, 13,300, 26,600, or 53,200 ppm
16 1,1,2,2-tetrachloroethane in the diet (microcapsules) for 15 days. Unexposed and vehicle control
17 groups were also evaluated, with the latter being given feed with empty microcapsules. Study
18 endpoints included clinical observations, body weight, food consumption, necropsy, selected
19 organ weights (liver, kidneys, thymus, lung, heart, and testes), and histology of gross lesions;
20 histology was not evaluated in animals without gross lesions. The study authors reported that
21 average daily doses for the three lowest concentrations were 300, 400, or 500 mg/kg-day for both
22 genders. All rats exposed to 26,600 or 53,200 ppm were killed moribund on day 11. The
23 average daily doses for these groups were not reported.
24 Female rats exposed to 400 mg/kg-day and both genders exposed to 500 mg/kg-day were
25 thin and displayed ruffled fur. Body weight at study termination was statistically significantly
26 lower than controls in both genders of all treated groups. Male rats exposed to 300 mg/kg-day
27 showed decreased weight gain compared to controls and those exposed to higher doses lost
28 weight, with final body weights in male rats 28, 46, and 53% less than vehicle controls at 300,
29 400, and 500 mg/kg-day, respectively. Females lost weight at doses of >300 mg/kg-day, with
30 final body weights in female rats 25, 38, and 47% less than vehicle controls at 300, 400, and
31 500 mg/kg-day, respectively. Decreased feed consumption likely contributed to the decreased
32 weight gains because consumption was reduced in a dose-related manner in both genders of all
33 treated groups (NTP, 1996).
34 Absolute thymus weights were decreased 24, 69, and 84% in male rats and 37, 61, and
35 81% in female rats at doses of >300 mg/kg-day and relative thymus weights were decreased
36 42 and 65% in male rats and 38 and 65% in female rats at >400 mg/kg-day (NTP, 2004; TSI
37 Mason Laboratories, 1993d). In male rats, absolute liver weights were decreased 22, 49, and
38 60% compared to controls at 300, 400, and 500 mg/kg-day, respectively. Relative liver weight
46 DRAFT - DO NOT CITE OR QUOTE
-------
1 was increased 7% compared to controls at 300 mg/kg-day and decreased 14% compared to
2 controls at 500 mg/kg-day. In female rats, absolute liver weight was decreased 25 and 34%
3 compared to controls at 400 and 500 mg/kg-day, respectively, and relative liver weight was
4 increased 34 and 23% compared to controls at 300 and 500 mg/kg-day, respectively. Relative
5 kidney weights were increased 14, 26, and 18% in male rats at 300, 400, and 500 mg/kg-day,
6 respectively, and 17 and 36% in female rats at 400 and 500 mg/kg-day, respectively. Absolute
7 kidney weights were decreased 17, 32, and 45% in males and 16, 27, and 27% in females at 300,
8 400, and 500 mg/kg-day, respectively. Other organ weight decreases were considered a
9 reflection of the decreased body weights.
10 Focal areas of alopecia occurred on the skin of four female rats in the 500 mg/kg-day
11 group, and these lesions correlated with minimal to moderate acanthosis, which is an abnormal
12 benign increase in the thickness of the stratum spinosum, a layer of cells that is capable of
13 undergoing mitotic cell division, of the epidermis. In the liver, mild or moderate centrilobular
14 degeneration was observed microscopically in the exposed male and female rats.
15 Groups of five male and five female B6C3Fi mice were exposed to 0, 3,325, 6,650,
16 13,300, 26,600, or 53,200 ppm of encapsulated 1,1,2,2-tetrachloroethane in the diet for 15 days
17 (NTP, 2004; TSI Mason Laboratories, 1993d). Organ weights, gross necropsy, and histology of
18 gross lesions were evaluated in surviving mice at the termination of the study. Average daily
19 doses were not determined by the study authors because feed consumption could not be
20 measured accurately due to excessive scattering of feed. All male and female mice exposed to
21 53,200 ppm, all males exposed to 26,600 ppm, and two males exposed to 13,300 ppm were
22 sacrificed in extremis before the end of the study. Final body weights were decreased 16, 24,
23 and 22%, in comparison to vehicle controls, in males at 3,325, 6,650, and 13,300 ppm,
24 respectively. In females, final body weights were decreased 9, 20, 31, and 34% at 3,325, 6,650,
25 13,300, and 26,600 ppm, respectively.
26 Clinical findings included hyperactivity in males and females exposed to 3,325, 6,650, or
27 13,300 ppm and in females in the 26,600 ppm group. Males in the 26,600 and 53,200 ppm
28 groups were lethargic. Males exposed to >6,650 ppm and females exposed to 26,600 and
29 53,200 ppm were thin and had ruffled fur. A statistically significant decrease in absolute (31, 47,
30 82, and 81%, respectively) and relative (22, 33, 74, and 72%, respectively) thymus weights
31 compared to controls was observed in all exposed female mice. Relative liver weights were
32 statistically significantly increased 22, 31, and 34% in male mice at 3,325, 6,650, and
33 13,300 ppm, respectively. Absolute liver weights were statistically significantly decreased 11,9,
34 and 5% in female mice at 6,650, 13,300, and 26,600 ppm, respectively, and relative liver weight
35 increased 30 and 44% at 13,300 and 26,600 ppm, respectively. Other organ weight changes
36 were associated with changes in body weight. Pale or mottled livers were noted in all exposed
37 groups of male and female mice and correlated microscopically with hepatocellular degeneration,
38 which was characterized by hepatocellular swelling, cytoplasmic rarefaction, single paranuclear
47 DRAFT - DO NOT CITE OR QUOTE
-------
1 vacuoles, hepatocellular necrosis, and infrequent mononuclear infiltrates. The severity of the
2 hepatic changes increased with increasing exposure concentration.
3 The histological examinations in the surviving mice showed hepatocellular degeneration
4 in 3/3, 4/4, 4/4, 1/1, and 1/1 males, and 4/4, 4/4, 3/3, 3/3, and 3/3 females, at 3,325, 6,650,
5 13,300, 26,600, and 53,200 ppm, respectively (TSI Mason Laboratories, 1993d). For both
6 genders, the lesions tended to be minimal to mild at 3,325 and 6,650 ppm, with more moderate to
7 marked severity observed at the higher doses.
8 The National Cancer Institute (NCI, 1978) conducted a range-finding study in rats and
9 mice in order to estimate the maximum tolerated dose for administration in the chronic bioassay.
10 In this study, Osborne-Mendel rats (5/sex/group) received gavage doses of 0 (vehicle control
11 group), 56, 100, 178, 316, or 562 mg/kg 1,1,2,2-tetrachloroethane in corn oil 5 days/week for
12 6 weeks, followed by a 2-week observation period. B6C3Fi mice (5/sex/group) were similarly
13 exposed to 0, 32, 56, 100, 178, or 316 mg/kg 1,1,2,2-tetrachloroethane. It appears that mortality
14 and body weight gain were the only endpoints used to assess toxicity and determine the high-
15 dose levels for the NCI (1978) chronic bioassays in rats and mice. In the rats, one male exposed
16 to 100 mg/kg and all five females exposed to 316 mg/kg died (mortality rates in the 562 mg/kg
17 groups were not reported). Body weight gain was reduced 3, 9, and 38% in male rats and 9, 24,
18 and 41% in female rats at 56, 100, and 178 mg/kg-day, respectively. No deaths or significant
19 alterations in body weight gain were observed in the mice. In male rats, 100 and 178 mg/kg-day,
20 were selected as the NOAEL and LOAEL, respectively, for the observed decrease in body
21 weight, while in female rats the NOAEL and LOAEL were 56 and 100 mg/kg-day, respectively,
22 for the same endpoint. The highest dose in mice, 316 mg/kg-day, was selected as the NOAEL
23 for body weight changes and mortality.
24
25 4.4.2.2. Short-term Inhalation Studies
26 Rats (n = 84) were exposed to 0 or 15 mg/m3 (2.2 ppm) 1,1,2,2-tetrachloroethane
27 4 hours/day for up to 8 days in a 10-day period (Gohlke and Schmidt, 1972; Schmidt et al., 1972).
28 Following the first, third, and seventh exposures, seven control and exposed rats were given an
29 unknown amount of ethanol. Evaluations were performed on seven males from the control and
30 treated groups, with and without ethanol, following the second, fourth, and eighth exposures.
31 Statistically significant changes included increased serum total protein and decreased
32 serum ar and (X2-globulin fractions compared to controls after the eighth exposure (day 10),
33 although the difference was not quantified (Schmidt et al., 1972). Histological effects included a
34 fine to medium droplet fatty degeneration of the liver that involved increasing numbers of
35 animals with increasing duration of exposure, although the incidences and severity were not
36 reported (Gohlke and Schmidt, 1972). The results of the serum and histochemical evaluations
37 were illegible in the best copy of the translated reference available. Testicular atrophy in the
38 seminal tubules was observed in five treated animals following the fourth exposure (Gohlke and
48 DRAFT - DO NOT CITE OR QUOTE
-------
1 Schmidt, 1972). This study is limited by imprecise and incomplete reporting of results.
2 Assessment of the adversity of liver and other effects in this study is complicated by the
3 reporting insufficiencies, particularly the paucity of incidence and other quantitative data, as well
4 as effects that were not consistently observed in the three time periods and a lack of information
5 on dose-response due to the use of a single exposure level.
6 Horiuchi et al. (1962) exposed nine male mice to an average concentration of
7 approximately 7,000 ppm (48,000 mg/m3) 1,1,2,2-tetrachloroethane for 2 hours once/week for a
8 total of five exposures over 29 days. All animals died during the study with none of the deaths
9 occurring during exposure, and most (5/9) of the mice died within 5 days of the first exposure.
10 The only other reported findings in the exposed animals were slight to moderate congestion and
11 fatty degeneration of the liver and congestion of "other main tissues."
12 Horiuchi et al. (1962) exposed six male rats to an average concentration of 9,000 ppm
13 (62,000 mg/m ) 1,1,2,2-tetrachloroethane 2 hours/day, 2-3 times a week for 11 exposures in
14 29 days. All rats died during the study. No changes in body weight were reported. Exposed
15 animals generally showed hypermotility within the first few minutes of exposure, followed by
16 atactic gait within approximately 20 minutes and eventual near-complete loss of consciousness
17 1-1.5 hours after the onset of exposure. Hematology was assessed in three rats that survived
18 beyond 2 weeks, and two of these animals showed a decrease in RBC count and Hb content.
19 Exposed animals generally showed moderate congestion and fatty degeneration of the liver and
20 congestion of "other main tissues."
21 As discussed in Section 4.2.2.1, one monkey was exposed to varying concentrations
22 (2,000-4,000 ppm for the first 20 exposures, 1,000-2,000 ppm for the 20th-160th exposure, and
23 3,000^,000 ppm for the remaining exposures) of 1,1,2,2-tetrachloroethane for 2 hours/day,
24 6 days/week for 9 months (Horiuchi et al., 1962). Effects of short-term exposure included
25 weakness after seven exposures, diarrhea and anorexia between the 12th and 15th exposures, and
26 beginning at the 15th exposure, near-complete unconsciousness for 20-60 minutes after each
27 exposure.
28
29 4.4.3. Acute Injection Studies
30 Paolini et al. (1992) exposed groups of male and female Swiss Albino mice to a single i.p.
31 dose of 0, 300, or 600 mg/kg 1,1,2,2-tetrachloroethane and sacrificed the animals 24 hours after
32 dosing to assess hepatotoxicity. An LD50 of 1,476 mg/kg for 1,1,2,2-tetrachloroethane was
33 calculated using six animals/dose and eight dose groups. At 600 mg/kg, absolute and relative
34 liver weights were statistically significantly decreased 16 and 37%, respectively, in female mice.
35 No changes in total microsomal protein were noted. Statistically significant decreases (37-74%)
36 in hepatic cytochrome P450 enzymes of numerous classes were reported at both dose levels in
37 male and female mice (see Section 3.3). Other hepatic enzymes with statistically significantly
38 decreased activity included NADPH-cytochrome c-reductase, 5-aminolevulinic acid-synthetase,
49 DRAFT - DO NOT CITE OR QUOTE
-------
1 ethoxyresorufin-O-deethylase, pentoxyresorufin O-depentylase, GST (600 mg/kg only), and
2 epoxide hydrolase. Total hepatic heme was reduced at both doses, and heme oxygenase activity
3 was increased in a dose-related manner, but was statistically significant only in high-dose males
4 and females.
5 Wolff (1978) exposed groups of female Wistar rats to a single i.p. dose of 0, 20, or
6 50 mg/kg 30 minutes prior to testing for passive avoidance of a 0.4 mA electric shock. No
7 differences between the control and 25 mg/kg groups were reported, but doses of 50 mg/kg
8 resulted in decreased passive avoidance behavior. Similarly, no differences were seen in the
9 open-field test at any dose level. In male ICR-mice, a single i.p. dose of 20 mg/kg resulted in a
10 significant reduction in spontaneous locomotor activity, and 50-60 mg/kg resulted in a 50%
11 reduction (Wolff, 1978).
12 In an abstract, Andrews et al. (2002) described the exposure of a rat whole embryo
13 culture system to 1,1,2,2-tetrachloroethane. Gestational day 9 embryos were exposed to
14 concentrations between 0.5 and 2.9 mM 1,1,2,2-tetrachloroethane for 48 hours and then
15 evaluated for morphological changes. At concentrations >1.4 mM, 1,1,2,2-tetrachloroethane
16 resulted in rotational defects and anomalies of the heart and eye. Embryo lethality was observed
17 at>2.4mM.
18
19 4.4.4. Immunotoxicological Studies
20 Shmuter (1977) exposed groups of 12 Chinchilla rabbits to 0, 2, 10, or 100 mg/m3 (0, 0.3,
21 1.5, or 14.6 ppm, respectively) 1,1,2,2-tetrachloroethane 3 hours/day, 6 days/week for 8-
22 10 months. Animals were vaccinated with 1 mL of a 1.5 x 10 suspension of heated typhoid
23 vaccine 1.5, 4.5-5, and 7.5-8 months after the initiation of 1,1,2,2-tetrachloroethane exposure.
24 Significant increases and decreases in total antibody levels were observed in the 2 and
25 100 mg/m3 groups, respectively. No significant changes in 7S-typhoid antibody levels were
26 observed. Significant alterations in the levels of "normal" hemolysins to the Forsman's antigen
27 of sheep erythrocytes were observed in the 10 and 100 mg/m3 groups, as levels were increased in
28 the 10 mg/m group after 1.5, 2, and 2.5 months of exposure, decreased after 4 months, and
29 absent at 5 months of exposure. Levels of these hemolysins were decreased in the 100 mg/m
30 group during the first 6 months of exposure. Increases in the electrophoretic mobility of specific
31 antibodies following 1,1,2,2-tetrachloroethane were also reported. Exposure to 100 mg/m3
32 1,1,2,2-tetrachloroethane resulted in a decrease in the relative content of antibodies in the
33 y-globulin fraction and an increase in the T and (3 fractions. This is a poorly reported study that
34 provides inadequate quantitative data. The inconsistent dose-response patterns preclude
35 assessing biological significance and identification of a NOAEL or LOAEL.
36
37 4.5. MECHANISTIC DATA AND OTHER STUDIES IN SUPPORT OF THE MODE OF
38 ACTION
50 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
4.5.1. Genotoxicity
As discussed in Section 3.4, radiolabeled 1,1,2,2-tetrachloroethane may covalently bind
to DNA and RNA (Colacci et al, 1987), suggesting the potential for mutagenicity. A summary
of the results of genotoxicity studies of 1,1,2,2-tetrachloroethane is presented in Table 4-17.
Table 4-17. Results of in vitro and in vivo genotoxicity studies of
l,l?2,2-tetrachloroethane
In vitro gene mutation assays
Test system
Endpoint
Cells/strain
Concentrations
Results
-S9
+S9
Reference
(a) Bacterial assays
Salmonella
typhimurium
(Ames test)
Escherichia coli
Saccharomyces
cerevisisae
Aspergillus
nidulans
Reverse
mutation
Forward
mutation
DNA damage
Gene
conversion
Gene
reversion
Gene
recombina-
tion
Mitotic
crossover
TA100, 1535,
1537, 1538,98
TA1530, 1535,
1538
TA1535, 1537,
98
TA1535
TA97, 98, 100,
1535, 1537
TA98, 100,
1535, 1537
TA98, 100,
1535, 1537
TA100
BA13
pol A+/pol Af
WP2S(X)
D7
D7
D7
PI
NA
10 uL/plate
10 uL/plate
NA
10-3,333 uL/plate
NA
5-l,OOOuL/plate
NA
0.06-2,979 nmol/
plate
10 uL/plate
15-23 6 mM
3. 1-7.3 mM
NA
3. 1-7.3 mM
NA
3. 1-7.3 mM
0.01-0.04%v:v
-
NP
-
-
-
-
-
-
-
NP
+
NP
NP
NP
NP
NP
NP
-
+
-
-
-
-
-
-
-
+
-
+
-
+
-
+
+
Nestmann et al., 1980
Rosenkranz, 1977;
Bremetal, 1974
Mitomaetal, 1984
Onoetal, 1996
NTP, 2004
Milmanetal, 1988
Haworthetal., 1983
Warner et al., 1988
Roldan-Arjona et al.,
1991
Rosenkranz, 1977;
Bremetal., 1974
DeMarini and Brooks,
1992
Callenetal, 1980
Nestmann and Lee, 1 983
Callenetal., 1980
Nestmann and Lee, 1 983
Callenetal., 1980
Crebellietal, 1988
(b) Mammalian cell assays
Mouse Lymphoma
Hepatocytes
(primary)
Gene
mutation
DNA repair
L5178Y
Osborne
Mendel rats
B6C3F] mice
25-500 nL/mL
NA
NA
-
NP
NP
-
-
-
NTP, 2004
Milmanetal., 1988;
Williams, 1983
51
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-17. Results of in vitro and in vivo genotoxicity studies of
1,1^2,2-tetrachloroethane
In vitro chromosomal damage assays
Test system
Cells/organs
Concentrations
Results
Reference
Mammalian Cells
Chromosomal
Aberrations
Sister chromatid
exchanges (SCE)
UDS
CHO cells
CHO cells
BALB/C-3T3 cells
Human embryonic
intestinal fibroblasts
453-804 ug/mL
16.8-558 ug/mL
500-1,000 ug/mL
<15,869 ug/mL
-
+
+
-
-
+
+
NP
NTP, 2004;
Galloway et al, 1987
NTP, 2004;
Galloway et al., 1987
Colaccietal, 1992
McGregor (1980)
Other in vitro assays:
Cell transformation
(initiation)
Cell transformation
(promotion)
BALB/C-3T3 cells
1-250 ug/mL
1-250 ug/mL
125-1,000 ug/mL
NA
0. 1-1,000 ng/mL
NP
NP
+
-
NP
-
-
+
-
-
Arthur Little, Inc., 1983
Tuetal, 1985
Colaccietal., 1990
Milmanetal, 1988
Colaccietal., 1996
In vivo bioassays
Test system
Cells/organs
Doses
Results
Reference
Chromosomal damage: mammalian
Chromosomal
aberrations
Micronucleus
UDS
DNA alkylation
Rat bone marrow cells,
male
Rat bone marrow cells,
female
Mouse peripheral blood
erythrocytes
Mouse hepatocytes
Mouse hepatocytes, male
Mouse hepatocytes, female
Mouse hepatocytes
50 ppm
50 ppm
589-9, 100 ppm
200 mg/kg
50-1, 000 (mg/kg)
50-1,000 mg/kg
1 50 mg/kg
-
+
+
+
-
-
+
McGregor, 1980
NTP, 2004
Miyagawa et al., 1995
Mirsalis et al., 1989
Dow Chemical Co., 1988
Other in vivo assays
S-phase DNA
synthesis
Mitotic recombination
Recessive lethal
mutation
Mouse hepatocytes, male
Mouse hepatocytes, female
Drosophila melanogaster
D. melanogaster
200-700 mg/kg
200-700 mg/kg
500-1, 000 ppm
800 ppm (injected)
1,500 (feed)
-
+/-
-
-
Mirsalis et al., 1989
Vogel and Nivard, 1993
Woodruff etal, 1985
1
2
3
4
5
+ = positive; - = negative/no change; CHO = Chinese hamster ovary; NA = not available; NP = assay not
performed; UDS = unscheduled DNA synthesis
1,1,2,2-Tetrachloroethane has been shown to be predominantly inactive in reverse
mutation assays in Salmonella typhimurium (strains TA97, TA98, TA100, TA1530, TA1535,
TA1537, and TA1538), either with or without the addition of S9 metabolic activating mixture,
even at concentrations that lead to cytotoxicity (NTP, 2004; Ono et al., 1996; Milman et al.,
52
DRAFT - DO NOT CITE OR QUOTE
-------
1 1988; Warner et al., 1988; Mitoma et al, 1984; Haworth et al, 1983; Nestmann et al, 1980).
2 Two studies reported reverse mutation activity in S. typhimurium (Rosenkranz, 1977; Brem et al.,
3 1974). Results of studies employing methods to prevent volatilization were not notably different
4 from those that did not use those methods. 1,1,2,2-Tetrachloroethane also did not induce
5 forward mutations (L-arabinose resistance) in S. typhimurium strain BA13 (Roldan-Arjona et al.,
6 1991). Assays with Escherichia coli indicated that 1,1,2,2-tetrachloroethane induced DNA
7 damage, as shown by growth inhibition in DNA polymerase deficient E. coli (Rosenkranz, 1977;
8 Brem et al., 1974) and induction of prophage lambda (DeMarini and Brooks, 1992). In
9 Saccharomyces cerevisiae, 1,1,2,2-tetrachloroethane induced gene conversion, reversion, and
10 recombination in one study (Callen et al., 1980), whereas another study found no conversion or
11 reversion (Nestmann and Lee, 1983). In Aspergillus nidulans, 1,1,2,2-tetrachloroethane induced
12 aneuploidy, but no crossing over (Crebelli et al., 1988).
13 1,1,2,2-Tetrachloroethane did not induce trifluorothymidine resistance in L5178Y mouse
14 lymphoma cells, with or without S9, at concentrations up to those producing lethality (NTP,
15 2004). Primary hepatocytes from rats and mice exposed in vitro to 1,1,2,2-tetrachloroethane did
16 not show altered DNA repair at concentrations that were not cytotoxic (Milman et al., 1988;
17 Williams, 1983). McGregor (1980) reported no increase in unscheduled DNA synthesis (UDS)
18 in human embryonic intestinal fibroblasts exposed to 1,1,2,2-tetrachloroethane. Treatment of
19 Chinese hamster ovary (CHO) cells with up to 653 ug/mL (which was cytotoxic) did not result in
20 increased induction of chromosomal aberrations (NTP, 2004; Galloway et al., 1987) but did
21 produce an increased frequency of sister chromatid exchanges (SCEs) at concentrations of
22 >55.8 ug/mL (NTP, 2004; Galloway et al., 1987). SCEs were also induced in BALB/C-3T3 cells
23 treated in vitro with high concentrations (>500 ug/mL) of 1,1,2,2-tetrachloroethane, either with
24 or without S9 activating mixture (Colacci et al., 1992).
25 In BALB/C-3T3 cells, 1,1,2,2-tetrachloroethane exposure of up to 250 ug/mL in the
26 absence of exogenous metabolic activation did not result in increased numbers of transformed
27 cells (Colacci et al., 1992; Milman et al., 1988; Tu et al., 1985; Arthur Little, Inc., 1983);
28 survival was generally >70%. Higher concentrations (>500 ug/mL) were capable of
29 transforming the cells, but also showed higher levels of cytotoxicity (Colacci et al., 1990).
30 However, even relatively low levels (31.25 ug/mL) of 1,1,2,2-tetrachloroethane used as an
31 initiating agent, followed by promotion with 12-O-tetradecanoylphorbol-13 -acetate, resulted in
32 increased numbers of transformed cells (Colacci et al., 1992). 1,1,2,2-Tetrachloroethane did not
33 act as a promoter in BALB/C-3T3 cells in vitro without metabolic activation (Colacci et al.,
34 1996).
35 1,1,2,2-Tetrachloroethane tested negative for sex-linked recessive lethal mutations and
36 mitotic recombination inD. melanogaster (NTP, 2004; Vogel and Nivard, 1993; Woodruff et al.,
37 1985; McGregor, 1980). Replicative DNA synthesis was increased in hepatocytes isolated from
38 male B6C3Fi mice exposed to a single gavage dose of 200 mg/kg (24 and 48 hours
53 DRAFT - DO NOT CITE OR QUOTE
-------
1 postexposure) or 400 mg/kg (24, 39, and 48 hours postexposure) relative to hepatocytes from
2 unexposed mice (Miyagawa et al., 1995). Hepatocytes isolated from mice following a single
3 gavage dose of up to 1,000 mg/kg did not show an increase in UDS or S-phase DNA synthesis
4 (Mirsalis et al., 1989). Hepatocytes isolated from B6C3Fi mice 6 hours after a single gavage
5 dose of 150 mg/kg in corn oil demonstrated irreversible alkylation of hepatic DNA (Dow
6 Chemical Co., 1988). Inhalation exposure to 5 or 50 ppm (34.3 or 343 mg/m3) for 7 hours/day,
7 5 days/week did not result in increased frequency of chromosomal aberrations in bone marrow
8 cells isolated from male rats (McGregor, 1980); female rats exposed to 50 ppm (343 mg/m ), but
9 not to 5 ppm (34.3 mg/m3), showed an increase in bone marrow cell aberrations other than gaps
10 (McGregor, 1980).
11 In summary, genotoxicity studies provide limited evidence of a mutagenic mode of action.
12 1,1,2,2-Tetrachloroethane has some genotoxic activity, but in vitro genotoxicity tests generally
13 reported non-positive results. Similarly, in vivo studies had mostly non-positive results with the
14 exception of chromosomal aberrations in female rat bone marrow cells and micronucleus
15 formation in mouse bone marrow peripheral erythrocytes. The results of rat liver preneoplastic
16 foci and mouse BALB/C-3T3 cell neoplastic transformation assays suggest that 1,1,2,2-tetra-
17 chloroethane may have initiating and promoting activity. Overall, results of genotoxicity studies
18 for 1,1,2,2-tetrachloroethane are mixed and insufficient for establishing a mutagenic mode of
19 action.
20
21 4.5.2. Short-Term Tests of Carcinogenicity
22 Treatment of partially hepatectomized male Osborne-Mendel rats with a single
23 100 mg/kg gavage dose of 1,1,2,2-tetrachloroethane, followed by 7 weeks of promotion with
24 phenobarbital in the diet, did not result in increased numbers of preneoplastic (GGT-positive)
25 foci in the liver (Milman et al., 1988; Story et al., 1986). Exposure of partially hepatectomized
26 male Osborne-Mendel rats to a single i.p. dose of diethylnitrosamine (DEN) as an initiating agent
27 followed by promotion with 100 mg/kg-day of 1,1,2,2-tetrachloroethane by gavage 5 days/week
28 for 7 weeks produced a significantly increased number of GGT-positive foci in the liver (Milman
29 et al., 1988; Story et al., 1986). 1,1,2,2-Tetrachloroethane also significantly increased the
30 number of GGT-positive foci in rats administered the promotion protocol in the absence of the
31 DEN initiator. The study authors concluded that 1,1,2,2-tetrachloroethane induces
32 hepatocarcinogenesis primarily through a promoting mechanism (Story et al., 1986).
33 Using a mouse strain that had been shown to be susceptible to pulmonary adenomas
34 when exposed to organic chemicals, Theiss et al. (1977) administered i.p. injections of 80, 200,
35 or 400 mg/kg 1,1,2,2-tetrachloroethane in Tricaprylin 5-18 times to groups of 20 male A/St mice
36 for 8 weeks. There was a dose-related increase in number of lung tumors/mouse (Table 4-18),
37 and the dose-response was nearly statistically significant (Theiss et al., 1977).
38
54 DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-18. Pulmonary adenomas from l,l?2,2-tetrachloroethane exposure
in mice
Dose/injection (mg/kg)
Number of i.p. injections
Total dose (mg/kg)
Number of surviving animals
Number of lung tumors/mouse
0
24
0
15/20
0.27±0.15
80
5
400
10/20
0.30±0.21
200
18
3,600
15/20
0.50 ±0.14
400
16
6,400
5/20
1.00 ±0.45
Source: Thiess et al. (1977).
1
2 Maronpot et al. (1986) tested 65 chemicals at three doses in 6- to 8-week-old male and
3 female strain A/St or A/J mice housed 10/cage. Doses were set based on the highest dose
4 exhibiting a lack of overt toxicity from a preliminary dose-setting study, with the mid and low
5 dose as half the higher dose. Mice were injected i.p. 3 times/week for 8 weeks. Lungs were
6 examined histologically. The data for 1,1,2,2-tetrachloroethane-exposed male and female strain
7 A/St are presented in Table 4-19.
8
Table 4-19. Pulmonary adenomas from l,l?2,2-tetrachloroethane exposure
in A/St mice
Compound
Dose/injection
(mg/kg)
Vehicle
Untreated
control
—
-
Saline
vehicle
control
—
-
Tricaprylin
vehicle
control
—
-
Urethan
positive
control
1,000
-
1,1,2,2-Tetrachloroethane
62.5
Tricaprylin
99
Tricaprylin
187.5
Tricaprylin
Male A/St mice
Number of
surviving animals3
Percent survivors
with tumors
Tumors per mouseb
119/120
2
0.017
45/50
9
0.089
54/60
13
0.167
47/50
96
11.9
10/10
10
0.1
8/10
0
0
5/10
0
0
Female A/St mice
Number of
surviving animals3
Percent survivors
with tumors
Tumors per mouse
79/80
8
0.076
44/50
14
0.186
54/60
11
0.11
47/50
96
10.3
9/10
0
0
5/10
20
0.2
3/10
0
0
3Numerator is number of mice alive at study termination; denominator is number of mice started on study.
Based on all surviving mice at study termination.
Source: Maronpot et al. (1986).
55
DRAFT - DO NOT CITE OR QUOTE
-------
1 4.6. SYNTHESIS OF MAJOR NONCANCER EFFECTS
2 4.6.1. Oral
3 4.6.1.1. Human Data
4 Information on the acute oral toxicity of 1,1,2,2-tetrachloroethane in humans is available
5 from several case reports. Based on amounts of 1,1,2,2-tetrachloroethane recovered from the
6 gastrointestinal tract of deceased subjects following intentional ingestion (Mant, 1953; Sherman,
7 1953; Lilliman, 1949; Forbes, 1943; Elliot, 1933; Hepple, 1927), estimated lethal doses ranged
8 from 273 to 9,700 mg/kg. Patients who accidentally consumed a known volume of 1,1,2,2-tetra-
9 chloroethane, corresponding to single doses ranging from 68 to 117 mg/kg, as medicinal
10 treatment for hookworm experienced loss of consciousness and other clinical signs of narcosis
11 (Ward, 1955; Sherman, 1953). Chronic oral effects of 1,1,2,2-tetrachloroethane in humans have
12 not been reported in the literature.
13
14 4.6.1.2. Animal Data
15 Few studies have evaluated acute oral toxicity in animals, and the endpoints assessed
16 consist of data on lethality and neurological and liver effects (Table 4-20). Oral LDso values
17 ranged from 250 to 800 mg/kg in rats (NTP, 2004; Schmidt et al, 1980a; Gohlke et al, 1977;
18 Smyth et al., 1969). Neurological effects of acute, oral 1,1,2,2-tetrachloroethane administration
19 revealed ataxic effects and decreased passive avoidance behavior (Wolff, 1978). Hepatic
20 changes were noted in two separate acute oral toxicity studies. Male Sprague-Dawley rats
21 administered between 287 and 1,148 mg/kg 1,1,2,2-tetrachloroethane had dose-dependent
22 increases in the serum activity levels of AST and ALT as well as a decrease in hepatic
23 microsomal G6Pase activity (Cottalasso et al., 1998). Male Wistar rats were administered 100
24 mg/kg 1,1,2,2-tetrachloroethane and had increases in hepatic ascorbic acid levels and serum
25 leucine aminopeptidase activity, but no changes in serum ALT activity (Schmidt et al., 1980a, b).
26 Both studies noted increases in triglyceride levels in the liver.
56 DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-20. Summary of noncancer results of major studies for oral exposure of animals to l,l?2,2-tetrachloroethane
Species
Sex
Average daily
dose
(mg/kg-d)
Exposure
duration
NOAEL
(mg/kg-d)
LOAEL
(mg/kg-d)
Response
Comments
Reference
Acute exposure
Rat
(Wistar)
Rat
(Sprague-
Dawley)
Rat (Wistar)
F
M
M
0, 25, 50, 80, 100
(gavage)
0, 143.5,287,
574, or 1,148
(gavage)
0 or 100
Single dose
Single dose
Single dose
25
143.5
100
50
287
ND
Increased electric shock
perception threshold.
Increased serum AST
activity and ALT activity,
increased liver triglycerides
levies; decreased liver
dolichol levels.
Increased hepatic ascorbic
acid levels and serum
leucine aminopeptidase
activity
Results suggestive of a subtle
anesthetic effect. Ataxia observed
at 100 mg/kg.
Evaluations performed 1 hr
postexposure. Approximately
twofold increases in AST and ALT
at >574 mg/kg. Liver histology
and neurotoxicity not assessed.
No changes in serium ALT
Wolff, 1978
Cottalasso et al.,
1998
Schmidt et al.,
1980a,b
Short-term exposure
Rat
(Osborne-
Mendel)
Mouse
(B6C3FO
Rat (F344/N)
Rat (F344/N)
Mouse
(B6C3FO
Rat (F344/N)
M
M
M,F
M
M,F
M
0,25,75, 150, or
300
(gavage)
0,25,75, 150, or
300
(gavage)
0, 135,270, or
540
(gavage)
0, 135,270, or
540
(gavage)
0,337.5, 675, or
1,350
(gavage)
0, 104, or 208
(gavage)
3-4 d
4d
12 doses in
16 d
12 doses in
16 d
12 doses in
16 d
13-21 d
150
300
135
135
ND
ND
300 (PEL)
ND
270
270
337.5
104 (PEL)
CNS depression and
mortality. No
histopathological changes
in liver.
Decreased body weight in
females, plus lethargy and
increased organ weights.
Lethargy, decreased body
weight gain.
Hepatocellular
degeneration (females).
Hepatic cytoplasmic
vacuolization at low dose,
mortality at high dose.
Increased hepatocellular DNA
synthesis and mitosis at
>75 mg/kg-d; increased liver
weight at > 150 mg/kg-d. No
nonhepatic endpoints evaluated.
Centrilobular swelling at
>75 mg/kg-d and increased
hepatocellular DNA synthesis and
mitosis at > 150 mg/kg-d. No
nonhepatic endpoints evaluated.
The highest dose caused 100%
mortality. Limited histology3.
Mortality at 540 mg/kg-d. Limited
histology3.
Lethargy, increased liver weight,
and mortality at higher doses.
Limited histology3.
No changes in body weight, kidney
weights, kidney histology, or
urinalysis.
Dow Chemical
Company, 1988
Dow Chemical
Company, 1988
TSI Mason
Laboratories,
1993a, unpubl.
TSI Mason
Laboratories,
1993b, unpubl.
TSI Mason
Laboratories,
1993c, unpubl.
NTP, 1996;
57
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-20. Summary of noncancer results of major studies for oral exposure of animals to l,l?2,2-tetrachloroethane
Species
Rat (F344/N)
Mouse
(B6C3FO
Sex
M,F
M F
Average daily
dose
(mg/kg-d)
0, 300, 400, or
500
(diet)
3,325,6,650,
13,300,26,600,
or 53,200 ppm
Exposure
duration
15 d
15 d
NOAEL
(mg/kg-d)
ND
ND
LOAEL
(mg/kg-d)
300
ND
Response
Decreased body weight
gain.
Decreased body weight,
hyperactivity, decreased
absolute and relative
thymus weight, increased
relative liver weight, pale
or mottled livers,
hepatocellular degeneration
Comments
Changes in liver and kidney
weights and clinical signs at higher
doses. Limited histology3.
feed consumption could not be
measured accurately
Reference
NTP, 2004
NTP, 2004; TSI
Mason
Laboratories,
1993d
Subchronic exposure
Rat (F344)
Mouse
(B6C3FO
M,F
M F
0, 20, 40, 80,
170, or 320
(diet)
0, 100, 200, 370,
700, or 1,360
(male); 0, 80,
160, 300, 600, or
1,400 (female)
(diet)
14wks
14wks
20
40
80
40
80
160
Increased liver weight, as
well as decreased sperm
motility.
Increased serum ALT
activity, SDH activity, and
cholesterol levels, reduced
epididymis weight.
Increased liver weight,
increased ALT activity,
ALP activity, SDH activity,
and bile acids levels.
Comprehensive study. More
serious hepatic effects, including
hepatocyte necrosis and bile duct
hyperplasia, as well as effects on
other organs, at >170 mg/kg-d.
Comprehensive study. Wide array
of endpoints evaluated, including
histopathology. More serious
hepatic effects, including
hepatocyte necrosis and bile duct
hyperplasia, as well as effects on
other organs, at >300 mg/kg-d.
NTP, 2004
NTP, 2004
Chronic exposure
Rat
(Osborne-
Mendel)
Mouse
(B6C3FO
M,F
M,F
0, 62, or 108
(male)
0,43, or 76
(female)
(gavage)
0, 142, or 284
(gavage)
78wks
78wks
62 (M)
76 (F)?
ND
142
108 (M)
ND(F)
142 (M)
284 (F)
Fatty changes in liver.
Reduced survival. Acute
toxic tubular nephrosis,
hydronephrosis, and
chronic inflammation in the
kidneys.
Study is confounded by endemic
chronic murine pneumonia, but
this is unlikely to have contributed
to the liver pathology.
High incidences of hepatocellular
tumors in all dose groups
precluded evaluation of noncancer
effects in the liver.
NCI, 1978
NCI, 1978
58
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-20. Summary of noncancer results of major studies for oral exposure of animals to l,l?2,2-tetrachloroethane
Species
Sex
Average daily
dose
(mg/kg-d)
Exposure
duration
NOAEL
(mg/kg-d)
LOAEL
(mg/kg-d)
Response
Comments
Reference
Developmental exposure
Rat
(Sprague-
Dawley)
Mouse
(CD-I)
F
F
0, 34, 98, 180,
278, or 330
(diet)
0,987,2,120,
2,216, or 4,575
(diet)
GDs 4-20
GDs 4-17
34
ND
98
ND
Decreased maternal and
fetal body weights.
Maternal mortality and
litter resorptions.
Effects were more pronounced at
higher doses.
high mortality in the exposed mice
precluded the identification of a
NOAEL or LOAEL.
Gulati et al.,
1991a
Gulati et al.,
1991b
aHistology only evaluated in animals with gross lesions.
59
DRAFT - DO NOT CITE OR QUOTE
-------
1 Short-term oral exposure (Table 4-18) to 1,1,2,2-tetrachloroethane produced clinical
2 signs of neurotoxicity and mortality at doses as low as 208-300 mg/kg-day by gavage in rats
3 (NTP, 1996; TSI Mason Laboratories, 1993a, b, unpublished; Dow Chemical Company, 1988).
4 Body weight gain was decreased at similar dose levels in rats exposed by gavage or diet (NTP,
5 2004; TSI Mason Laboratories, 1993a, b, unpublished; Dow Chemical Company, 1988; NCI,
6 1978). Hepatic effects consisted of increased DNA synthesis and centrilobular swelling in mice
7 exposed to 75 mg/kg-day in the diet (Dow Chemical Company, 1988) and hepatocellular
8 cytoplasmic vacuolation in rats exposed to 104 mg/kg-day (NTP, 1996). At higher doses (337.5
9 mg/kg-day), hepatocellular degeneration was observed in mice (TSI Mason Laboratories, 1993c,
10 unpublished).
11 Subchronic and chronic oral administration studies (Table 4-18) with 1,1,2,2-tetrachloro-
12 ethane in animals indicated that the liver is the most sensitive organ for toxicity. Oral toxicity
13 studies in F344 and Osborne-Mendel rats and B6C3Fi mice were evaluated (NTP, 2004, NCI,
14 1978). The 14-week subchronic study by the National Toxicology Program (NTP, 2004) in both
15 F344 rats and B6C3Fi mice was the most comprehensive evaluation of 1,1,2,2-tetrachloroethane-
16 mediated toxicity through an orally administered route. NCI (1978) conducted a chronic study
17 on Osborne Mendel rats and B6C3Fi mice in which dosing regimens were modified during the
18 course of the study.
19 In F344 rats, an increased incidence of hepatocellular cytoplasmic vacuolization was
20 observed at 20 mg/kg-day in males and 40 mg/kg-day in females, increased relative liver weights
21 were observed at 40 mg/kg-day, and hepatocellular hypertrophy was observed at 80 mg/kg-day
22 in the subchronic NTP (2004) study. Additional hepatic effects included increases in serum ALT
23 and SDH activity at 80 mg/kg-day, decreases in serum cholesterol levels at 80 mg/kg-day, and
24 increases in serum ALP activity and bile acids levels, hepatocellular necrosis, bile duct
25 hyperplasia, hepatocellular mitotic alterations, foci of cellular alterations, and hepatocyte
26 pigmentation at 170 and 320 mg/kg-day. A NOAEL of 20 mg/kg-day and a LOAEL of 40
27 mg/kg-day was selected based on the increase in relative liver weight; however, it should be
28 noted that an increased incidence of hepatocellular cytoplasmic vacuolization was observed at 20
29 and 40 mg/kg-day in male and female rats, respectively. In the Osborne-Mendel rats, significant
30 increases in hepatic fatty metamorphosis were observed in male rats following a chronic
31 exposure to 108 mg/kg-day (TWA, based on changes in dosing regimen) (NCI, 1978). Mortality
32 was significantly increased in female rats dosed at a TWA dose of 43 and 76 mg/kg-day;
33 however, the increased mortality was affected by the deaths of 10 high-dose females, 8 with
34 pneumonia and 2 with no reported lesions, during the first 5 weeks of the study. A NOAEL of
35 62 mg/kg-day and a LOAEL of 108 mg/kg-day were identified in male rats based on an
36 increased incidence of hepatic fatty metamorphosis (NCI, 1978).
37 Mice appear to be less sensitive than rats to noncancer effects mediated by orally
38 administered 1,1,2,2-tetrachloroethane. Relative liver weight was statistically significantly
60 DRAFT - DO NOT CITE OR QUOTE
-------
1 increased in female and male B6C3Fi mice at 80 and 200 mg/kg-day, respectively. Effects in the
2 mice also included minimal hepatocellular hypertrophy, increased serum SDH activity, ALT
3 activity, and bile acids levels, and decreased serum cholesterol levels at 160-200 mg/kg-day, and
4 increased serum ALP and 5'-nucleotidase activities, necrosis, pigmentation, and bile duct
5 hyperplasia at 300-370 mg/kg-day. Based on the increase in relative liver weight observed in
6 the NTP (2004) study, a NOAEL of 100 mg/kg-day and a LOAEL of 200 mg/kg-day in male
7 mice and a LOAEL of 80 mg/kg-day in female mice was identified . In addition, male and
8 female B6C3Fi mice were evaluated for chronic oral toxicity by NCI (1978). For this study, a
9 LOAEL of 142 mg/kg-day was selected for chronic inflammation in the kidneys of male mice,
10 while a NOAEL of 142 mg/kg-day and a LOAEL of 284 mg/kg-day were selected for
11 hydronephrosis and chronic inflammation in the kidneys of female mice.
12 Comprehensive neurobehavioral testing showed no evidence of neurotoxicity in either
13 species at doses equal to or higher than the LOAELs based on liver effects (NTP, 2004),
14 indicating that the liver is more sensitive than the nervous system to subchronic dietary exposure
15 to 1,1,2,2-tetrachloroethane.
16 Developmental parameters were significantly affected by oral administration of
17 1,1,2,2-tetrachloroethane in rats and mice. Significant decreases in rat maternal and fetal body
18 weights were noted at doses of >98 mg/kg-day (Gulati et al, 1991a). Using statistical
19 significance and a 10% change as the criteria for establishing an adverse effect in maternal body
20 weight, a NOAEL of 34 mg/kg-day and LOAEL of 98 mg/kg-day were selected. A NOAEL of
21 34 mg/kg-day and LOAEL of 98 mg/kg-day were selected for developmental toxicity based on
22 the lowest dose that produced a statistically significant decrease in fetal body weight. In mice,
23 the FEL based on maternal toxicity and resorption of litters is 2,120 mg/kg-day (Gulati et al.,
24 1991 b). The high mortality in the exposed mice precluded the identification of a NOAEL or
25 LOAEL from this study.
26 Toxicity to reproductive tissues following 1,1,2,2-tetrachloroethane exposure to adult rats
27 and mice was observed at dose levels as low as 40 mg/kg-day (NTP, 2004). In male rats, sperm
28 motility was decreased at >40 mg/kg-day. Higher doses resulted in decreased epididymal
29 absolute weight and increased atrophy of the preputial and prostate gland, seminal vesicle, and
30 testicular germinal epithelium. In female rats, minimal to mild uterine atrophy was increased at
31 >170 mg/kg-day and clitoral gland atrophy and ovarian interstitial cell cytoplasmic alterations
32 were increased at 320 mg/kg-day. Female F344 rats in the 170 mg/kg-day group spent more
33 time in diestrus than did the vehicle controls.
34 Male B6C3Fi mice had increased incidences of preputial gland atrophy at >100 mg/kg-
35 day. Less sensitive effects included decreases in absolute testis weight (>700 mg/kg-day) and
36 absolute epididymis and cauda epididymis weights (1,360 mg/kg-day) and a decrease in
37 epididymal spermatozoal motility (1,360 mg/kg-day). The only noted reproductive toxicity
61 DRAFT - DO NOT CITE OR QUOTE
-------
1 parameter in female mice affected was a significant increase in the length of the estrous cycle at
2 a dose of 1,400 mg/kg-day (NTP, 2004).
3
4 4.6.2. Inhalation
5 4.6.2.1. Human Data
6 Limited information is available on the acute inhalation toxicity of 1,1,2,2-tetrachloro-
7 ethane in humans (Table 4-21). The results of an early, poorly reported experimental study with
8 two volunteers suggest that 3 ppm (6.9 mg/m3) was the odor detection threshold. Irritation of the
9 mucous membranes, pressure in the head, vertigo, and fatigue were observed at 146 ppm (1,003
10 mg/m3) for 30 minutes or 336 ppm (2,308 mg/m3) for 10 minutes. Common reported symptoms
11 of high-level acute inhalation exposure to 1,1,2,2-tetrachloroethane in humans include
12 drowsiness, nausea, headache, and weakness, and at extremely high concentrations, jaundice,
13 unconsciousness, and respiratory failure (Coyer, 1944; Hamilton, 1917).
62 DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-21. Summary of noncancer results of major human studies of inhalation exposure to 1,1^2,2-tetrachloroethane
Study
population
Sex
Exposure
level (mg/m3)
Exposure NOAEL
duration (mg/m3)
LOAEL
(mg/m3)
Response
Comments
Reference
Acute exposure
Two volunteers
NS
6.9-2,308
30 min
ND
ND
Irritation, vertigo, head
pressure, fatigue.
Effect levels could not be
determined due to limited
analysis.
Lehmann et al., 1936
Occupational exposure
127 coating
workers
Workers from
3 9 chemical
processing
plants
380 workers
from 23
factories
34-75 workers
in penicillin
production
NS
NS
M,F
NS
500-1,500
NS
62.5-672
10-1,700
NS
NS
Generally <1 yr
Up to 3 yrs
ND
ND
ND
ND
ND
ND
ND
ND
Decreased whole blood
specific gravity,
decreased RBC count,
lymphocytosis,
unspecified neurological
findings.
Increased mortality for
lymphatic cancers.
Anemia, loss of appetite,
abdominal pain,
headache, vertigo, and
tremors.
Loss of appetite,
epigastric pain, hepatic
enlargement,
urobilinogenuria,
weakness, fatigue, weight
loss, and itching.
Effect levels could not be
determined due to limited
analysis.
Mortality from cardiovascular
disease, cirrhosis of the liver, and
digestive or respiratory cancers
was not elevated.
Effect levels could not be
determined due to a lack of a
control population and possible
coexposure.
Effect levels could not be
determined due to a lack of a
control population, limited
reporting, and possible
coexposure.
Horiguchi et al., 1964
Norman etal, 1981
Lobo-Mendonca, 1963
Jeney et al., 1957
ND = not determined; NS = not stated
1
63
DRAFT - DO NOT CITE OR QUOTE
-------
1 Chronic toxicity of inhaled 1,1,2,2-tetrachloroethane in humans (Table 4-19) resulted in
2 neurological symptoms including headache, weakness, fatigue, and hematological changes such
3 as anemia and elevated WBC count (Norman et al, 1981; Lobo-Mendonca, 1963; Jeney et al,
4 1957; Minot and Smith, 1921). Most occupational exposure studies failed to evaluate hepatic
5 endpoints, other than an urobilinogen test. Jeney et al. (1957) reported a positive relationship
6 between duration of exposure and frequency of abnormal liver function test results, loss of
7 appetite, bad taste in the mouth, epigastric pain, and a "dull straining pressure feeling in the area
8 of the liver".
9
10 4.6.2.2. Animal Data
11 Acute inhalation exposures in animals (Table 4-22) resulted in near-lethal or lethal effects
12 at levels >1,000 ppm (Schmidt et al., 1980a; Price et al., 1978; Horiuchi et al., 1962; Carpenter et
13 al., 1949; Pantelitsch, 1933). Death was typically preceded by signs of CNS toxicity (e.g.,
14 incoordination, loss of reflexes, labored respiration, prostration, and loss of consciousness) and
15 was often accompanied by congestion and fatty degeneration of the liver. Nonlethal exposures
16 increased lipid and triglyceride levels in the liver in mice following exposure to 600-800 ppm
17 (4,120-5,490 mg/m3)for 3 hours (Tomokuni, 1970, 1969). Nonlethal exposures also reduced
18 motor activity in rats following exposure to 576 ppm (3,950 mg/m3) for 30 minutes (Price et al.,
19 1978) and 360 ppm (2,470 mg/m ) for 6 hours (Horvath and Frantik, 1973) and in guinea pigs
20 following exposure to 576 ppm (3,950 mg/m3) (Price et al., 1978).
64 DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-22. Summary of noncancer results of major studies for inhalation exposure of animals to
l,l?2,2-tetrachloroethane.
Species
Sex
Exposure
level (mg/m3)
Exposure
duration
NOAEL
(mg/m3)
LOAEL
(mg/m3)
Response
Comments
Reference
Acute exposure
Rat
Rat (Wistar)
Rat (Sherman)
Rat
Guinea pig
Rat
(NR)
Mouse (Cb)
Mouse (Cb)
Mouse
Mouse
Rat
NR
M
NR
NR
NR
NR
F
F
NS
M
M
NR
0,410,700,
1,030,2,100,
or 4,200
6870
3,950,
34,700, or
43,350
3,950,
34,700, or
43,350
1,370 or
2,470
4,120
5,490
7,000, 8,000-
10,000,
17,000,
29,000, or
34,000
40,500 or
45,300
0, 69, 690, or
6,900
4Hrs
4Hrs
4Hrs
30 mins
30 mins
6Hrs
3Hrs
3Hrs
1.5-2Hrs
3Hrs
6Hrs
NR
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
8,600
ND
ND
3,950
3,950
2,470
4,120
ND
7,000
ND
69
LC50
24-Hr observation.
Hepatic effects included histological alterations and
increases in serum enzymes and liver triglycerides.
Identification of a NOAEL or LOAEL precluded by
reporting inadequacies.
Mortality
slight reduction in activity and alertness; lacrimation,
ataxia, narcosis, labored respiration, and 30-50%
mortality when concentration increased
Eye closure, squinting, lacrimation, and decreased
activity; tremors, narcosis, and labored breathing and
mortality when concentration increased
Effective concentration
for a 50% decrease in
spontaneous motor
activity.
Increased hepatic lipid
and triglyceride levels,
decreased hepatic
ATP.
Increased tricglyceride
and decreased
phospholipid levels
Disturbed equilibrium,
prostration, and loss of
reflexes.
Effective concentration for a
50% increase in pentobarbital
sleep time was 1,370 mg/m3.
A limited number of
endpoints were evaluated.
effects generally resolved by
90 hours postexposure
Limited number of endpoints
and poor reporting. Mortality
at >8,000 mg/m3.
Mortality: 3/10 and 4/10, respectively
minimal increase in serum AST at all exposure
concentrations 72 hours postexposure
Schmidt et al, 1980a
Schmidt et al, 1980a
Carpenter etal, 1949
Price etal, 1978
Price etal., 1978
Horvath and Frantik, 1973
Tomokuni, 1969
Tomokuni, 1970
Pantelitsch, 1933
Horiuchietal, 1962
Deguchi, 1970
65
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-22. Summary of noncancer results of major studies for inhalation exposure of animals to
l,l?2,2-tetrachloroethane.
Species
Sex
Exposure
level (mg/m3)
Exposure
duration
NOAEL
(mg/m3)
LOAEL
(mg/m3)
Response
Short-term exposure
Rat
Rat
Mouse
M
M
M
Oor 15
62,000
48,000
4 Hrs/d for up to
eight exposures
inlOd
2 Hrs/d, 2-3
times a week for
1 1 exposures in
29 d
2 Hrs/d for 5
exposures in 29
d
ND
ND
ND
ND
ND
ND
Comments
Reference
Increases in serum proteins and histological
alterations in the liver. Identification of a NOAEL or
LOAEL precluded by reporting inadequacies.
All rats died during the study. No changes in body
weight were reported. Exposed animals generally
showed moderate congestion and fatty degeneration
of the liver
Moderate congestion
and fatty degeneration
of the liver
Subchronic exposure
Rat (Osborne-
Mendel)
Mouse
(B6C3F1)
Rat
(Sprague-
Dawley)
Monkey
(Macaca sp.)
Rats
Mongrel dog
Rabbits
M, F
M, F
F
M
M,F
M
NS
0, 56, 100,
178,316,or
562
0,32,56,
100, 178, or
316
0 or 3,909
13,560
Oor 1,150
Oor 1,150
Oor 10
5 d/wk for 6 wks
5 d/wk for 6 wks
5-6 Hrs/d,
5 d/wk for
15 wks
2 hrs/d, 6 d/wk
for total of
190 exposures in
9 mo
7 hrs/d for 6 mo
7 hrs/d for 6 mo
3 hrs/d, 6 d/wk
for 7-8.5 mo
100 (male)
56 (female)
316
ND
ND
ND
ND
ND
178 (male)
100 (female)
ND
ND
ND
ND
ND
ND
Decreased body
weight gain
Body weight changes
and mortality
Most (5/9) of the mice died
within 5 days of the first
exposure
Gohlke and Schmidt,
1972; Schmidt etal, 1972
Horiuchi et al, 1962
Horiuchi et al, 1962
Mortality and body weight
gain were the only endpoints
used to assess toxicity
Mortality and body weight
gain were the only endpoints
used to assess toxicity
Increased liver weight, transient liver cytoplasmic
vacuolization. Identification of a NOAEL or LOAEL
precluded by reporting inadequacies.
Fatty degeneration and splenic congestion.
Identification of a LOAEL or NOAEL is precluded
by the use of a single animal and lack of control.
Pathological effects in the liver, kidney, and lung,
precluded by an endemic lung infection.
Increased serum phosphatase and blood urea nitrogen
levels, cloudy swelling of the liver and convoluted
tubule of the kidney, and light congestion of the
lungs. A NOAEL or LOAEL was not identified due
to single treated dog
Altered serum acetylcholine levels. A NOAEL or
LOAEL can not be identified due to incomplete
NCI, 1978
NCI, 1978
Truffertetal, 1977
Horiuchi etal., 1962
Mellon Institute of
Industrial Research, 1947
Mellon Institute of
Industrial Research, 1947
Kulinskaya and
Verlinskaya, 1972
66
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-22. Summary of noncancer results of major studies for inhalation exposure of animals to
l,l?2,2-tetrachloroethane.
Species
Rabbits
Sex
NS
Exposure
level (mg/m3)
0,2, 10, or
100
Exposure
duration
3 hrs/d, 6 d/wk
for 8- 10 mo
NOAEL
(mg/m3)
ND
LOAEL
(mg/m3)
ND
Response
quantitation.
Comments
Increase and decrease in total antibody levels,
increase in the mobility of specific antibodies,
decrease in the relative content of y-globulin
antibodies and an increase in the T and |3 fractions.
Poorly reported study that provides inadequate
quantitative data.
Reference
Shmuter, 1977
Chronic exposure
Rats
M
0 or 13.3
4 hrs/d, 110 or
265 d
ND
ND
Increased leukocyte and Pi-globulin levels, increased
percentage of segmented nucleated neutrophils,
decreased percentage of lymphocytes, increased liver
total fat content. Experimental design and results
were poorly reported and histological examinations
do not appear to have been conducted.
Schmidt et al, 1972
ND = not determined
67
DRAFT - DO NOT CITE OR QUOTE
-------
1 Acute and short-term inhalation exposure (Table 4-22) to high concentrations (>7,000
2 ppm) of 1,1,2,2-tetrachloroethane produced mortality and neurological and liver effects in
3 animals. Mortality occurred in mice exposed to 7,000 ppm (48,000 mg/m ) for 2 hours
4 once/week for 4 exposures in 29 days and in rats exposed to 9,000 ppm (62,000 mg/m3) for 2
5 hours/day, 2-3 times/week for 11 exposures in 29 days. Congestion and fatty degeneration in
6 the liver (mice and rats), as well as a biphasic change in neurological motor activity
7 (hyperactivity followed by ataxia, rats only), were also reported (Horiuchi et al., 1962). At the
8 lowest inhalation exposure of 2.2 ppm (15 mg/m ) for 4 hours/day (8-10 days), rats had fine
9 droplet fatty degeneration in the liver and changes in levels of serum proteins, but no
10 neurological changes were reported (Gohlke and Schmidt, 1972; Schmidt et al., 1972).
11 There are a few subchronic inhalation exposure studies and one chronic exposure study
12 with 1,1,2,2-tetrachloroethane (Table 4-20). Overall these studies either had poor study designs,
13 one exposure concentration, low number of animals, or a combination of the above. The
14 available subchronic and chronic inhalation studies indicate that the liver was the most sensitive
15 organ to 1,1,2,2-tetrachloroethane exposure. Increased relative liver weights were reported at
16 exposures of 560 ppm (3,909 mg/m ) for 15 weeks (Truffert et al., 1977). Other transient hepatic
17 changes (e.g., histological alterations and cytoplasmic vacuolation) were observed, but these
18 effects did not persist (Truffert et al., 1977). In the chronic exposure study, rats exposed to 13.3
19 mg/m3 (1.9 ppm) 1,1,2,2-tetrachloroethane 4 hours/day for 265 days exhibited increased liver fat
20 content (Schmidt et al., 1972). In the third rat study (Mellon Institute of Industrial Research,
21 1947), none of the effects noted from 1,1,2,2-tetrachloroethane exposure could be evaluated
22 since the control animals experienced a high degree of pathological effects in the kidneys, liver,
23 and lung. Hepatic effects from long-term exposure to 1,1,2,2-tetrachloroethane were also
24 reported in a study with one mongrel dog with cloudy swelling of the liver at 167 ppm (1,150
25 mg/m3) for 6 months (Mellon Institute of Industrial Research, 1947) and one male monkey with
26 fatty degeneration of the liver at 1,974 ppm (13,560 mg/m3) for 9 months (Horiuchi et al., 1962).
27 Other endpoints that were observed following subchronic and chronic inhalation
28 exposure are described below. Hematological alterations, including increased leukocyte and
29 Pi-globulin levels, increased percentage of segmented nucleated neutrophils and decreased
30 percentage of lymphocytes, decreased y-globulin, and decreased adrenal ascorbic acid levels,
31 were observed in rats exposed to 1.9 ppm (13.3 mg/m3) for 265 days (Schmidt et al., 1972), and
32 splenic congestion was noted in a study of a single monkey (Horiuchi et al., 1962). In the
33 mongrel dog study noted above, cloudy swelling of the convoluted tubules of the kidneys and
34 light congestion of the lungs were observed (Mellon Institute of Industrial Research, 1947).
35 Kulinskaya and Verlinskaya (1972) observed alterations in serum acetylcholine levels in rabbits
36 exposed to 10 mg/m3 (1.5 ppm) 3 hours/day, 6 days/week for 7-8.5 months. Shmuter (1977)
37 observed immunological alterations (changes in antibody levels) in rabbits exposed to 2-100
38 mg/m3 (0.3-14.6 ppm) 3 hours/day, 6 days/week for 8-10 months.
68 DRAFT - DO NOT CITE OR QUOTE
-------
1 A reproductive toxicity assessment was conducted on seven male rats exposed to
2 13.3 mg/m3 1,1,2,2-tetrachloroethane for 258 days. No significant changes in reproductive
3 parameters were observed, indicating that 13.3 mg/m3 (1.9 ppm) was a NOAEL for male
4 reproductive effects in the rat (Schmidt et al, 1972).
5
6 4.6.3. Mode of action Information
7 1,1,2,2-Tetrachloroethane is rapidly and extensively absorbed following both oral and
8 inhalation exposures, with absorption of 70-100% following oral exposure in animals (Dow
9 Chemical Company, 1988; Mitoma et al., 1985) and 40-97% following inhalation exposures in
10 humans (Morgan et al., 1970; Lehmann et al., 1936). Following absorption, the chemical is
11 distributed throughout the body, although the high tissue:air partition coefficient for fat (Gargas
12 et al., 1989) suggests that it may accumulate more in lipid-rich tissues. Metabolism is extensive,
13 with >68% of a total administered dose generally found as metabolites (Dow Chemical Company,
14 1988; Mitoma et al., 1985; Yllner, 1971), and is believed to occur mostly in the liver. Urinary
15 elimination occurs mainly as metabolites, including dichloroacetic acid, glyoxalic acid, formic
16 acid, trichloroethanol, and trichloroacetic acid, while a fraction of an absorbed dose may be
17 eliminated in expired air as parent compound or carbon dioxide.
18 Metabolism of 1,1,2,2-tetrachloroethane to reactive products is likely to play a key role in
19 its toxicity. Both nuclear and microsomal cytochrome P450 enzymes have been implicated in
20 the metabolism of the compound, possibly forming a number of biologically active compounds
21 including aldehydes, alkenes, acids, and free radicals (see Figure 3-1 in Section 3.3), which may
22 react with biological tissues. Evidence for metabolism to reactive compounds comes from
23 studies of radiolabel incorporation following single doses of radiolabeled 1,1,2,2-tetrachloro-
24 ethane in which incorporated radiolabel was enhanced by pretreatment with phenobarbital,
25 xylene, or ethanol, and the variety of inducers capable of influencing this effect suggest that
26 multiple P450 isozymes may be involved (Casciola and Ivanetich, 1984; Halpert, 1982; Sato et
27 al., 1980), including members of the CYP2A, CYP2B, CYP2E, and CYP3A subfamilies
28 (Omiecinski et al., 1999; Nebert et al., 1987). Additionally, mice are known to metabolize
29 1,1,2,2-tetrachloroethylene at a 1.1 -3.5 -fold greater rate than rats and have been demonstrated to
30 have approximately a twofold greater binding of radiolabel to tissues, further implicating
31 metabolic activation as a possible step in the mode of action. However, there is uncertainty as to
32 whether the presence of radiolabel in proteins, DNA, and RNA may be radiolabeled carbon that
33 has been incorporated into biomolecules through normal biochemical processes. Studies
34 describing the mechanism of 1,1,2,2-tetrachloroethane-induced noncancer toxicological effects
35 are not available.
36
69 DRAFT - DO NOT CITE OR QUOTE
-------
1 4.7. EVALUATION OF CARCINOGENICITY
2 4.7.1. Summary of Overall Weight of Evidence
3 Under the Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005a) 1,1,2,2-tetra-
4 chloroethane is "likely to be carcinogenic to humans" based on data from an oral cancer bioassay
5 in male and female Osborne-Mendel rats and B6C3Fi mice (NCI, 1978). In B6C3Fi mice, a
6 statistically significant increase in the incidence of hepatocellular carcinomas in both genders
7 was observed at doses of 142 and 284 mg/kg-day. A decrease in the time to tumor in both
8 genders of mice was also observed. In this same bioassay, male Osborne-Mendel rats exhibited
9 an increased incidence of hepatocellular carcinomas, a rare tumor in this strain (NCI, 1978), at
10 the high dose only, although this increased incidence was not statistically significant. An
11 untreated female control rat also developed a hepatocellular carcinoma. Limitations in the study
12 included increased mortality in male and female mice and the variable doses given to the mice
13 over the course of the 78-week exposure period. In the high-dose male mice, acute toxic tubular
14 nephrosis was characterized as the cause of death in the mice that died prior to study termination,
15 although hepatocellular carcinomas were observed in most of these mice.
16 The predominant proposed metabolic pathway for 1,1,2,2-tetrachloroethane involves
17 production of dichloroacetic acid (Casciola and Ivanetich, 1984; Halpert and Neal, 1981; Yllner,
18 1971). Dichloroacetic acid was identified as the major urinary metabolite in mice treated with
19 1,1,2,2-tetrachloroethane by i.p. injection (Yllner et al, 1971) and in in vitro systems with rat
20 liver microsomal and nuclear cytochrome P450 (Casciola and Ivanetich, 1984; Halpert, 1982;
21 Halpert and Neal, 1981). Other pathways involve the formation of trichloroethylene, via
22 dehydrochlorination, or tetrachloroethylene, via oxidation, as initial metabolites (Mitoma et al.,
23 1985; Ikeda and Ohtsuji, 1972; Yllner et al., 1971). 1,1,2,2-Tetrachloroethane may also form
24 free radicals by undergoing reductive dechlorination (ATSDR, 1996).
25 Dichloroacetic acid induces hepatocellular carcinomas in both genders of F344 rats and
26 B6C3Fi mice (DeAngelo et al., 1999; DeAngelo et al., 1996; Pereira, 1996; Pereira and Phelps,
27 1996; Ferreira-Gonzalez et al., 1995; Richmond et al., 1995; Daniel et al., 1992; DeAngelo et al.,
28 1991; U.S. EPA, 1991b; Bull et al., 1990; Herren-Freund et al., 1987). Trichloroethylene, also a
29 metabolite of 1,1,2,2-tetrachloroethane, has been shown to produce hepatocellular carcinomas
30 and hepatocellular adenomas in male and female B6C3Fi mice, respectively, but did not
31 demonstrate carcinogenicity in Osborne-Mendel or Sprague-Dawley rats (NTP, 1990; NCI,
32 1976). Tetrachloroethylene, another metabolite of 1,1,2,2-tetrachloroethane, was characterized
33 by NCI (1977) as a liver carcinogen in B6C3Fi mice, but an evaluation of carcinogenicity in
34 Osborne-Mendel rats was inadequate due to early mortality. In a study by NTP (1986),
35 tetrachloroethylene demonstrated evidence of carcinogenicity in F344 rats, as shown by
36 increased incidences of mononuclear cell leukemia, and in B6C3Fi mice, as shown by increased
37 incidences of hepatocellular adenomas and carcinomas in males and carcinomas in females.
70 DRAFT - DO NOT CITE OR QUOTE
-------
1 Additional information on the carcinogenic potential comes from studies on the tumor
2 initiating and promoting activity in mammalian cells (Colacci et al, 1996, 1992). The results of
3 the in vivo and in vitro genotoxicity studies for 1,1,2,2-tetrachloroethane, which were generally
4 non-positive, provide limited evidence of a mutagenic mode of action and are insufficient for
5 establishing a mutagenic mode of action.
6 No animal cancer bioassay data following inhalation exposure to 1,1,2,2-tetrachloro-
7 ethane are available. However, U.S. EPA's Guidelines for Carcinogen Risk Assessment (2005a)
8 indicates that for tumors occurring at a site other than the initial point of contact the cancer
9 descriptor generally applies to all routes of exposure that have not been adequately studied unless
10 there is convincing information to indicate otherwise. No additional information is available for
11 1,1,2,2-tetrachloroethane. Thus, 1,1,2,2-tetrachloroethane is considered "likely to be
12 carcinogenic to humans" by any route of exposure.
13 The weight of evidence for the carcinogenicity of 1,1,2,2-tetrachloroethane could be
14 strengthened by additional cancer bioassays demonstrating tumor development. Currently, the
15 NCI (1978) bioassay is the only study available demonstrating 1,1,2,2-tetrachloroethane
16 tumorgenicity. The NCI (1978) study was a 78-week study, compared to a 104-week bioassay,
17 and the limitations of the study included increased mortality in male and female mice, the
18 variable doses given to the mice over the course of the 78-week exposure period, and the acute
19 toxic tubular nephrosis, characterized as the cause of death, in the high-dose male mice that died
20 prior to study termination (although hepatocellular carcinomas were observed in most of these
21 mice).
22
23 4.7.2. Synthesis of Human, Animal, and Other Supporting Evidence
24 Only one study in humans evaluated the possible carcinogenic effects of 1,1,2,2-tetra-
25 chloroethane. Norman et al. (1981) evaluated groups of clothing-treatment workers employed
26 during World War II in which some workers used 1,1,2,2-tetrachloroethane and some used water.
27 Inhalation exposure concentrations and durations were not reported and dermal exposures were
28 likely. In addition, coexposures to dry-cleaning chemicals occurred. No differences in standard
29 mortality ratios were seen between the 1,1,2,2-tetrachloroethane and water groups for total
30 mortality, cardiovascular disease, cirrhosis of the liver, or cancer of the digestive and respiratory
31 systems. The mortality ratio for lymphatic cancers in the 1,1,2,2-tetrachloroethane group was
32 increased relative to controls and the water group, although the number of deaths was small
33 (4 cases observed compared to 0.85 cases expected). No other information was located
34 regarding the carcinogenicity of 1,1,2,2-tetrachloroethane in humans.
35 The only comprehensive animal study that evaluated the carcinogenicity of 1,1,2,2-tetra-
36 chloroethane was performed by the NCI (1978). Male and female Osborne-Mendel rats were
37 exposed to TWA doses of 0, 62, or 108 mg/kg-day (males) or 0, 43, or 76 mg/kg-day (females)
38 5 days/week for 78 weeks, followed by a 32-week observation period during which the rats were
71 DRAFT - DO NOT CITE OR QUOTE
-------
1 not exposed. No statistically significant increases in tumor incidences were observed in rats.
2 However, two hepatocellular carcinomas, which were characterized by NCI (1978) as rare in
3 Osbourne-Mendel rats, and one neoplastic nodule were observed in the high-dose male rats. A
4 hepatocellular carcinoma was also observed in a female rat in the control group. NCI (1978)
5 characterized the carcinogenic results in male rats as "equivocal." Male and female B6C3Fi
6 mice were exposed to TWA doses of 0, 142, or 284 mg/kg-day 5 days/week for 78 weeks,
7 followed by a 12-week observation period during which the mice were not exposed. Statistically
8 significant, dose-related increases in the incidence of hepatocellular carcinoma were observed in
9 males (3/36, 13/50, and 44/49 in the control, low-, and high-dose groups, respectively) and
10 females (1/40, 30/48, and 43/47, respectively). In addition, a decrease in the time to tumor for
11 the hepatocellular carcinomas was also evident in both genders of mice. Lymphomas were also
12 seen in the male and female mice, but the incidences were not found to be statistically significant.
13 The only other available study observed pulmonary adenomas in female Strain A/St mice given
14 99 mg/kg injections i.p. 3 times/week for 8 weeks (Maronpot et al., 1986).
15 In vitro studies of the genotoxicity of 1,1,2,2-tetrachloroethane have yielded mixed,
16 though mainly nonpositive, results. Mutagenicity studies in S. typhimurium were predominantly
17 negative, with only 2 of 10 available studies reporting activity (NTP, 2004; Ono et al., 1996;
18 Roldan-Arjona et al., 1991; Milman et al., 1988; Warner et al., 1988; Mitoma et al., 1984;
19 Haworth et al., 1983; Nestmann et al., 1980; Rosenkranz, 1977; Brem et al., 1974). Mixed
20 results were reported for gene conversion, reversion, and recombination in S. cerevisiae
21 (Nestmann and Lee, 1983; Callen et al., 1980), and aneuploidy, but not mitotic cross over, was
22 induced in A. nidulans (Crebelli et al., 1988). Tests for DNA damage inE. coli were positive
23 (DeMarini and Brooks, 1992; Rosenkranz, 1977; Brem et al., 1974). 1,1,2,2-Tetrachloroethane
24 was not mutagenic in mouse L5178Y lymphoma cells (NTP, 2004) and was negative in tests for
25 DNA damage in other mammalian cells, including induction of DNA repair in primary rat or
26 mouse hepatocytes (Milman et al., 1988; Williams, 1983), induction of chromosomal aberrations
27 in CHO cells (NTP, 2004; Galloway et al., 1987), and induction of cell transformation in
28 BALB/C-3T3 cells (Colacci et al., 1992; Milman et al., 1988; Tu et al., 1985; Arthur Little, Inc.,
29 1983). 1,1,2,2-Tetrachloroethane was positive for induction of SCEs in both BALB/C-3T3
30 (Colacci et al., 1992) and CHO cells (NTP, 2004; Galloway et al., 1987) and for induction of cell
31 transformation in BALB/C-3T3 cells at high (cytotoxic) doses (Colacci et al., 1990).
32 1,1,2,2-Tetrachloroethane also had mixed results for genotoxicity following in vivo
33 exposure. Tests for sex-linked recessive lethal mutations and mitotic recombination in
34 Drosophila were negative (NTP, 2004; Vogel and Nivard, 1993; Woodruff et al., 1985;
35 McGregor, 1980). Both positive (Miyagawa et al., 1995) and negative results (Mirsalis et al.,
36 1989) have been reported in mouse hepatocytes tested for UDS, and tests for S-phase DNA
37 induction in hepatocytes were negative in male mice and equivocal in female mice (Mirsalis et
72 DRAFT - DO NOT CITE OR QUOTE
-------
1 al., 1989). Rat bone marrow cells were negative for chromosomal aberrations in male rats, but
2 positive in female rats (McGregor, 1980).
3 1,1,2,2-Tetrachloroethane showed promoting activity, but limited initiating activity, in rat
4 liver preneoplastic (GGT-positive) foci assays (Milman et al., 1988; Story et al., 1986).
5 1,1,2,2-Tetrachloroethane initiated, but did not promote, neoplastic transformation in mouse
6 BALB/c-3t3 cells (Colacci et al., 1996, 1992).
7
8 4.7.3. Mode of action Information
9 The mode of action of the carcinogenic effects of 1,1,2,2-tetrachloroethane is unknown.
10 Colacci et al. (1987) reported possible covalent binding of radiolabeled 1,1,2,2-tetrachloroethane
11 to DNA, RNA, and protein in the liver, kidneys, lung, and stomach of rats and mice exposed to a
12 single intravenous dose and analyzed 22 hours postexposure. However, the conclusion of
13 covalent binding may be influenced by the presence of radiolabel in the DNA, RNA, and protein
14 that was the result of incorporated radiolabeled carbon into the biomolecules through normal
15 biochemical processes.
16 The mutagenicity data for 1,1,2,2-tetrachloroethane are inconclusive, with in vitro
17 genotoxicity tests generally reporting negative results except for assays of SCE and cell
18 transformation, and in vivo tests of genotoxicity showing a similar pattern. Several studies have
19 reported increases in the number of hepatocytes in mitosis, but the possible role these effects
20 may have on the carcinogenicity of 1,1,2,2-tetrachloroethane has not been evaluated. The results
21 of rat liver preneoplastic foci and mouse BALB/C-3T3 cell neoplastic transformation assays
22 suggest that 1,1,2,2-tetrachloroethane may have initiating and promoting activity (Colacci, 1996,
23 1992; Milman et al., 1988; Story et al., 1986), but tumor initiation and promotion studies have
24 not been conducted.
25 Tumor formation by 1,1,2,2-tetrachloroethane may involve metabolism to one or more
26 active compounds, with the predominant pathway leading to the production of dichloroacetic
27 acid (Casciola and Ivanetich, 1984; Halpert and Neal, 1981; Yllner, 1971). 1,1,2,2-Tetrachloro-
28 ethane is metabolized extensively following absorption, at least in part, by cytochrome P450
29 enzymes from the members of the CYP2A, CYP2B, CYP2E, and CYP3A subfamilies (see
30 Section 3.3). Mice are known to metabolize 1,1,2,2-tetrachloroethane to a greater extent than
31 rats, which may, in part, account for the fact that liver tumors occurred in mice at statistically
32 significant levels, but not in rats, following chronic oral exposure.
33 Dichloroacetic acid, which appears to be the main metabolite of 1,1,2,2-tetrachloroethane,
34 induces hepatocellular carcinomas in both genders of F344 rats and B6C3Fi mice (DeAngelo et
35 al., 1999; DeAngelo et al., 1996; Pereira, 1996; Pereira and Phelps, 1996; Ferreira-Gonzalez et al.,
36 1995; Richmond et al., 1995; Daniel et al., 1992; DeAngelo et al., 1991; U.S. EPA, 1991b; Bull et al.,
37 1990; Herren-Freund et al., 1987). Dichloroacetic acid is recognized as hepatocarcinogenic in
38 both genders of two rodent species
73 DRAFT - DO NOT CITE OR QUOTE
-------
1 1,1,2,2-tetrachloroethane may be metabolized to form free radicals, which may, in turn,
2 covalently bind to macromolecules, including DNA. Formation of free radicals during
3 1,1,2,2-tetrachloroethane metabolism has been demonstrated in spin-trapping experiments
4 (Tomasi et al, 1984). Both nuclear and microsomal forms of cytochrome P450 enzymes have
5 been implicated in this process, as increased metabolism and covalent binding of metabolites
6 following pretreatment with phenobarbital (Casciola and Ivanetich, 1984; Halpert, 1982), xylene
7 (Halpert, 1982), or ethanol (Sato et al., 1980) have been reported. The presence of covalently
8 bound label has been reported following inhalation (Dow Chemical Company, 1988), oral
9 (Mitoma et al., 1985), and intravenous (Eriksson and Brittebo, 1991) administration of
10 radiolabeled 1,1,2,2-tetrachloroethane.
11 In summary, only limited data are available regarding the possible mode(s) of action of
12 1,1,2,2-tetrachloroethane carcinogenicity. Metabolism to one or more active compounds may
13 play a role in tumor development. Results of genotoxicity studies of 1,1,2,2-tetrachloroethane
14 are mixed and provide inconclusive evidence for establishing a mutagenic mode of action.
15 There is some evidence to indicate that the mode of carcinogenic action may involve
16 tumor promotion. Milman et al. (1988) and Story et al., (1986) concluded that 1,1,2,2-tetra-
17 chloroethane induces hepatocarcinogenesis primarily through a promoting mechanism following
18 treatment of partially hepatectomized male Osborne-Mendel rats with a single 100 mg/kg gavage
19 dose of 1,1,2,2-tetrachloroethane, followed by 7 weeks of promotion with phenobarbital in the
20 diet. This regimen failed to result in increased numbers of preneoplastic (GGT-positive) foci in
21 the liver; whereas an exposure of partially hepatectomized male Osborne-Mendel rats to a single
22 i.p. dose of diethylnitrosamine (DEN) as an initiating agent followed by promotion with 100
23 mg/kg-day of 1,1,2,2-tetrachloroethane by gavage 5 days/week for 7 weeks produced a
24 significantly increased number of GGT-positive foci in the liver..
25
26 4.8. SUSCEPTIBLE POPULATIONS AND LIFE STAGES
27 4.8.1. Possible Childhood Susceptibility
28 Studies in humans and laboratory animals have not thoroughly examined the effect of
29 1,1,2,2-tetrachloroethane exposure on the immature organism. The Gulati rat study (Gulati et al.,
30 1991b) demonstrated that fetuses exposed in utero can be adversely affected. At scheduled
31 sacrifice, average fetal weights were statistically significantly decreased in all dose groups
32 except the 34 mg/kg-day group. In the Gulati mouse study (Gulati et al., 199la), complete litter
33 resorption occurred in mice in 1/11, 0/9, 2/8, 1/1, and 1/2 dams in the 0, 987, 2,120, 2,216, and
34 4,575 mg/kg-day dose groups, respectively. The limited data evaluating the effect of
35 1,1,2,2-tetrachloroethane on the developing organism have not indicated effects on the offspring
36 at levels that did not also produce maternal effects.
37
38 4.8.2. Possible Gender Differences
74 DRAFT - DO NOT CITE OR QUOTE
-------
1 Studies evaluating the differences in potency of 1,1,2,2-tetrachloroethane in male and
2 female rodents are not available. Some toxicity studies which evaluated both genders in the
3 same study showed close concordance between genders with often no more than one dose
4 distinguishing between response levels for a given effect. Men normally have a smaller volume
5 of body fat than women, even accounting for average size differences, contributing to differential
6 disposition of organic solvents between genders (Sato and Nakajima, 1987). Rats have
7 pronounced sex-specific differences in CYPs, primarily involving the CYP2C family which is
8 not found in humans, but humans have not demonstrated sex-specific isoforms of CYP450
9 (Mugford and Kedderis, 1998). Humans have differences in CYP 3A4 activity related to
10 estrogen and progesterone, but these differences are regulated by the hormones at the level of
11 gene expression (Harris et al, 1995). Other differences may occur at the Phase 2 level
12 attributable to conjugation. Overall, no consistent differences have been reported between
13 women and men in the handling of xenobiotics such as 1,1,2,2-tetrachloroethane by CYP
14 isoforms (Shimada et al., 1994). These distinctions make it difficult to predict from the animal
15 data gender-relevant differences for human exposure to 1,1,2,2-tetrachloroethane.
16
17 4.8.3. Other Susceptible Populations
18 As metabolism is believed to play an important role in the toxicity of 1,1,2,2-tetrachloro-
19 ethane, particularly in the liver, individuals with elevated levels of cytochrome P450 enzymes
20 may have an increased susceptibility to the compound. Halpert (1982) reported an increase in in
21 vitro metabolite formation and in covalently bound metabolites following pretreatment with
22 xylene or phenobarbital, both of which increased cytochrome P450 activity. Sato et al. (1980)
23 similarly reported an increased metabolism of 1,1,2,2-tetrachloroethane in rats following ethanol
24 pretreatment. Since 1,1,2,2-tetrachloroethane has been demonstrated to inhibit cytochrome P450
25 enzymes (Paolini et al., 1992; Halpert, 1982), presumably through a suicide inhibition
26 mechanism, it is also possible that people coexposed to chemicals that are inactivated by
27 cytochrome P450 enzymes will be more susceptible to those compounds.
28 In addition, studies of human GST-zeta polymorphic variants show different enzymatic
29 activities toward and inhibition by dichloroacetic acid that could affect the metabolism of
30 1,1,2,2-tetrachloroethane (Lantum et al., 2002; Blackburn et al., 2001, 2000; Tzeng et al., 2000).
31 Dichloroacetic acid may covalently bind to GST-zeta (Anderson et al., 1999), irreversibly
32 inhibiting one of two stereochemically different conjugates, thus inhibiting its own metabolism
33 and leading to an increase in unmetabolized dichloroacetic acid as the dose and duration of
34 exposure increases (U.S. EPA, 2003). GST zeta is a hepatic enzyme that also functions in the
35 pathway for tyrosine catabolism. Populations, or single individuals, may be more sensitive to
36 1,1,2,2-tetrachloroethane toxicity depending on which GST-zeta variant they possess.
37
75 DRAFT - DO NOT CITE OR QUOTE
-------
1 5. DOSE-RESPONSE ASSESSMENTS
2
3
4 5.1. ORAL REFERENCE DOSE (RfD)
5 5.1.1. Subchronic Oral RfD
6 5.1.1.1. Choice of Principal Study and Critical Effect
1 The data available on subchronic oral exposure to 1,1,2,2-tetrachloroethane are limited to
8 experimental studies in animals. Though a number of case reports provide information on
9 effects of intentional acute oral exposure to lethal oral doses of 1,1,2,2-tetrachloroethane (Mant,
10 1953; Lilliman, 1949; Forbes, 1943; Elliot, 1933; Hepple, 1927), no subchronic studies of oral
11 exposure to 1,1,2,2-tetrachloroethane in humans exist. A single, well-designed 14-week
12 subchronic study in rats and mice that tested multiple dose levels and examined an array of
13 endpoints and tissues in rats is available (NTP, 2004). Furthermore, a developmental toxicity
14 study in rats and mice exists (Gulati et al, 1991a, b). These studies in laboratory animals
15 provide evidence suggesting that the liver and the developing fetus may be targets of toxicity
16 following subchronic oral exposure to 1,1,2,2-tetrachloroethane.
17 NTP reported multiple effects on the livers of both male and female rats and mice
18 following subchronic oral exposure to 1,1,2,2-tetrachloroethane. Specifically, NTP (2004)
19 exposed F344 rats (10/sex/group) to 0, 20, 40, 80, 170, or 320 mg/kg-day (both males and
20 females) and B6C3Fi mice (10/sex/group) to 0, 100, 200, 370, 700, or 1,360 mg/kg-day for
21 males and 0, 80, 160, 300, 600, or 1,400 mg/kg-day for females in the diet for 14 weeks. A
22 statistically significant decrease in body weight gain (<10%) in both male and female rats at
23 >80 mg/kg-day was observed. Low dose effects observed in the liver included statistically
24 significantly increased relative liver weights in both male and female rats at >40 mg/kg-day. In
25 addition, hepatocyte vacuolization was observed at >20 mg/kg-day in male rats and >40 mg/kg-
26 day in female rats. The severity of vacuolization was reported to be minimal to mild. Serum
27 enzyme activity levels of both male and female rats were also affected. For example, increases
28 in serum ALT and SDH activity were observed at >80 mg/kg-day in male rats and >170 mg/kg-
29 day in female rats. In addition, increased cholesterol levels and ALP activity were observed in
30 female rats at >80 and 170 mg/kg-day, respectively. Additional histopathology observed in the
31 liver included a statistically significantly increased incidence of minimal to moderate hepatocyte
32 hypertrophy at >170 mg/kg-day in females and >200 mg/kg-day in males. Also, increased
33 incidence of necrosis and pigmentation were observed at >80 mg/kg-day and hepatocellular
34 mitotic alterations and foci of cellular alterations were observed at >80 and >170 mg/kg-day in
35 male rats, respectively. In females, increased incidence of hepatocellular hypertrophy was
36 observed at >80 mg/kg-day and necrosis, pigmentation, and foci of cellular alterations were
37 reported at >170 mg/kg-day. Bile duct hyperplasia, increased bile acids, spleen pigmentation,
38 and spleen atrophy were also observed in both male and female rats at the two highest doses.
76 DRAFT - DO NOT CITE OR QUOTE
-------
1 Evidence of liver effects were also observed in mice by NTP (2004). A statistically
2 significant increase in relative liver weights was observed in both male and female mice at
3 >200 and 80 mg/kg-day, respectively. Increases in serum ALT and ALP activity, bile acids
4 levels, and hepatic 5'-nucleotidase activity (males only) were observed in males and females at
5 >370 and 160 mg/kg-day, respectively. The study authors also reported an increase in SDH
6 activity at >200 and 80 mg/kg-day in male and female mice, respectively. Serum cholesterol
7 levels were statistically significantly increased in female mice at >160 mg/kg-day. The
8 incidence of hepatocellular necrosis was statistically significantly increased in male mice at >370
9 mg/kg-day and in female mice at >700 mg/kg-day. Hepatocellular hypertrophy was also
10 reported in both genders at >160-200 mg/kg-day. A statistically significant increase in incidence
11 of liver pigmentation and bile duct hyperplasia occurred at >300 mg/kg-day in females and
12 >370 mg/kg-day in males.
13 In addition to effects on the liver, NTP (2004) also observed effects associated with
14 reproduction in adult rats and mice following subchronic exposure to 1,1,2,2-tetrachloroethane at
15 dose levels as low as 40 mg/kg-day. In male rats, sperm motility was decreased at >40 mg/kg-
16 day, and higher doses resulted in decreased epididymis weight and increased atrophy of the
17 preputial and prostate gland, seminal vesicle, and testicular germinal epithelium. In female rats,
18 minimal to mild uterine atrophy was increased at > 170 mg/kg-day and clitoral gland atrophy and
19 ovarian interstitial cell cytoplasmic alterations were increased at 320 mg/kg-day. Female F344
20 rats in the 170 mg/kg-day group also spent more time in diestrus compared to controls. Male
21 mice had increased incidences of preputial gland atrophy at >100 mg/kg-day. Less sensitive
22 effects included decreases in absolute testes weight (>700 mg/kg-day), absolute epididymis, and
23 cauda epididymis weights (1,360 mg/kg-day), and a decrease in epididymal spermatozoal
24 motility (1,360 mg/kg-day). The only noted reproductive toxicity parameter in female mice
25 affected was a significant increase in the length of the estrous cycle at a dose of 1,400 mg/kg-day.
26 A developmental toxicity study by Gulati et al. (1991 a) demonstrated that the developing
27 fetus may be sensitive to 1,1,2,2-tetrachloroethane exposure. Gulati et al. (1991 a) exposed
28 pregnant CD Sprague-Dawley rats to 0, 34, 98, 180, 278, or 330 mg/kg-day
29 1,1,2,2-tetrachloroethane from GDs 4 through 20. Small, but statistically significant, decreases
30 were observed in maternal body weight and average fetal weight at >98 mg/kg-day. No other
31 maternal or fetal effects were reported by the study authors. In a second study, Gulati et al.
32 (1991b) exposed pregnant Swiss CD-I mice to 0, 987, 2,120, 2,216, or 4,575 mg/kg-day
33 1,1,2,2-tetrachloroethane from GDs 4 through 17. All animals (9/9) in the high-dose group died
34 prior to the end of the study, precluding calculation of the average dose in this exposure group.
35 Maternal body weights were statistically significantly decreased compared to controls at
36 >2,120 mg/kg-day beginning on study day 9. Gross hepatic effects such as pale or grey and/or
37 enlarged livers and a prominent lobulated pattern were also reported in dams from all groups
38 except at the low dose. Complete litter resorption occurred in 1/11, 0/9, 2/8, 1/1, and 1/2 dams in
77 DRAFT - DO NOT CITE OR QUOTE
-------
1 the 0, 987, 2,120, 2,216, and 4,575 mg/kg-day groups, respectively. No other developmental
2 effects were reported. Gulati et al. (1991a, b) suggested that the developing fetus may be a target
3 of 1,1,2,2-tetrachloroethane-induced toxicity. However, these developmental studies were
4 conducted at doses higher than the subchronic NTP (2004) study, which demonstrated liver
5 effects at lower doses. Therefore, Gulati et al. (1991a, b) was not selected as the principal study
6 and the observed reproductive effects were not selected as the critical effect following
7 subchronic exposure to 1,1,2,2-tetrachloroethane. Nevertheless, potential points of departure
8 (PODs) based on the observed developmental effects from Gulati et al. (1991a) were provided
9 for comparison (see Section 5.1.2 and Appendix B).
10 In consideration of the available studies reporting effects of subchronic oral exposure to
11 1,1,2,2-tetrachloroethane in animals, NTP (2004) was chosen as the principal study for the
12 derivation of the subchronic RfD. This study was conducted in both genders of two species,
13 used five dose levels and a concurrent control group, measured a wide-range of endpoints and
14 tissues, and provides data that were transparently and completely reported. NTP (2004)
15 identified the liver as the most sensitive target organ of 1,1,2,2-tetrachloroethane-induced
16 toxicity. Specifically, NTP (2004) identified effects on the liver, including increased liver
17 weight and increased incidence of hepatocellular vacuolization, at low dose levels. Other liver
18 effects observed in rats and mice at higher doses included increased liver weight, increased ALT,
19 ALP, and SDH serum activity levels, increased bile acid levels, and an increased incidence of
20 hepatocellular vacuolization and necrosis.
21 Based on the available data from the NTP (2004) study, the liver appears to be the most
22 sensitive target organ for 1,1,2,2-tetrachloroethane-induced toxicity. Thus, the observed effects
23 in the liver were considered in the selection of the critical effect for the derivation of the
24 subchronic RfD. Specifically, liver effects including increased liver weight, increased ALT,
25 ALP, and SDH serum levels, increased bile acid levels, and an increased incidence of
26 hepatocellular vacuolization were taken into consideration and modeled for the determination of
27 the critical effect and POD (Section 5.1.1.2 and Appendix B). EPA selected increased liver
28 weight as the critical effect because this effect may represent a sensitive endpoint that occurs
29 early in the process leading to hepatocellular necrosis associated with subchronic oral exposure
30 to 1,1,2,2-tetrachloroethane. The increase in relative liver weight was selected as the basis for
31 the selection of the POD because this analysis takes into account the substantive, dose-dependent
32 decreases in body weight that were observed in both genders of rats. Rats were selected as the
33 representative species because they appeared to be more sensitive than mice to the hepatotoxic
34 effects of 1,1,2,2-tetrachloroethane. EPA recognizes that the POD for the increased incidence of
35 hepatocellular vacuolization is approximately an order of magnitude lower than the POD for
36 increased relative liver weight, and would result in a lower RfD than that derived for increased
37 relative liver weight (See Sections 5.1.1.2 and 5.1.3 for more information). However, the
78 DRAFT - DO NOT CITE OR QUOTE
-------
1 biological significance of this effect following 1,1,2,2-tetrachloroethane exposure is unclear
2 based on the following considerations.
3 Vacuoles are defined as cavities bound by a single membrane that serve several
4 functions; usually providing storage areas for fat, glycogen, secretion precursors, liquid, or debris
5 (Osol, 1972). Vacuolization is defined as the process of accumulating vacuoles in a cell or the
6 state of accumulated vacuoles (Grasso, 2002). This process can be classified as either a normal
7 physiological response or may reflect an early toxicological process. As a normal physiological
8 response, vacuolization is associated with the sequestration of materials and fluids taken up by
9 cells, and also with secretion and digestion of cellular products (Henics and Wheatley, 1999). In
10 addition, Robbins et al. (1976) characterized vacuolization (i.e., intracellular autophagy) as a
11 normal cellular functional, homeostatic, and adaptive response.
12 Vacuolization is not only a normal physiological response. Vacuolization has been
13 identified as one of four principal types of chemical-induced injury (the other three being cloudy
14 swelling, hydropic change, and fatty change) (Grasso, 2002). It is one of the most common
15 responses of the liver following a chemical exposure, typically in the accumulation of fat in
16 parenchymal cells, most often in the periportal zone (Plaa and Hewitt, 1998). The ability to
17 detect subtle ultrastructural defects, such as vacuolization, early in the course of toxicity often
18 permits identification of the initial site of the lesion and thus can provide clues to possible
19 biochemical mechanisms involved in the pathogenesis of liver injury (Hayes, 2001).
20 The hepatocellular vacuolization reported by NTP (2004) was not observed consistently
21 across species (i.e., reported only in male and female rats); whereas the other observed liver
22 effects were reported in both sexes of both species. In addition, NTP (2004) did not characterize
23 the vacuole content following exposure to 1,1,2,2-tetrachloroethane. The study authors indicated
24 that the severity of the hepatocellular vacuolization was minimal to mild and was concentration
25 independent, but NTP (2004) did not report the localization of the vacuolization in the liver. The
26 observed vacuolization in the liver at low doses appeared to diminish as dose increased.
27 Specifically, hepatocellular vacuolization increased in a dose dependant manner from 20 to
28 80 mg/kg-day in male rats. At 80 mg/kg-day, 100% of male rats were affected, and at doses of
29 >80 mg/kg-day, the incidence of vacuolization began to decrease. Concurrent with this decrease
30 in incidence of vacuolization, an increased incidence of hepatocyte hypertrophy, necrosis, and
31 pigmentation were observed. In female rats, the incidence of vacuolization was 100% at 40 and
32 80 mg/kg-day followed by a diminished response at the two highest doses. Necrosis and
33 pigmentation were observed in the females at the two high doses. Thus, the qualitative and
34 quantitative biological relationship between the observed hepatocellular toxicity (i.e., hepato-
35 cellular necrosis) and the increased incidence of hepatocellular cytoplasmic vacuolization in
36 NTP (2004) is unknown.
37
38 5.1.1.2. Methods of Analysis—Including Models (PBPK, BMD, etc.)
79 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
9
10
11
12
13
14
15
Benchmark dose (BMD) modeling was conducted using the EPA's benchmark dose
software (BMDS, version 2.1.1.) to analyze the hepatotoxic effects associated with subchronic
exposure to 1,1,2,2-tetrachloroethane (see Appendix B for modeling details). The software was
used to calculate potential PODs for deriving the subchronic RfD by estimating the effective
dose at a specified level of response (BMDX) and its 95% lower bound (BMDLX). For all
continuous endpoints, a BMR of 1SD of the control mean was considered appropriate for
derivation of the RfD under the assumption that it represents a minimally biologically significant
response level. A BMR of 1 standard deviation (SD) of the control mean was also included for
comparative purposes. For the dichotomous data, i.e., the incidence of hepatocellular
cytoplasmic vacuolization, a BMR of 10% extra risk was considered appropriate for derivation
of the RfD under the assumption that it represents a minimally biologically significant response
level. The effects modeled include liver weight changes, serum ALT and SDH, bile acids,
hepatocellular cytoplasmic vacuolization, and rat fetal body weights. Table 5-1 summarizes the
BMD modeling results for the selected toxicological endpoints.
Table 5-1. Summary of BMD model results for rats exposed to 1,1^2,2-tetra-
chloroethane
Endpoint
Model
BMR
BMD
(mg/kg-d)
BMDL
(mg/kg-d)
Males
Cytoplasmic vacuol.
Relative liver weight
Absolute live weight
ALT
SDH
Bile acids
Polynomial
None
Polynomial
Polynomial
None
Power
10% extra risk
NA
1 SD
1 SD
NA
1 SD
13
NA
30
41
NA
72
11
NA
23
26
NA
57
Females
Cytoplasmic vacuol.
Relative liver weight
Absolute liver weight
ALT
SDH
Bile acids
Weibull
Polynomial
Polynomial
Hill
Power
Polynomial
10% extra risk
1 SD
1 SD
1 SD
1 SD
1 SD
31
22
36
82
157
188
19
15
26
69
113
170
Developmental
Rat fetal weight
Linear
1 SD
83
60
16
17
18
19
20
21
Changes in hepatocellular cytoplasmic vacuolization, ALT, SDH, ALP, and bile acids
serum levels from NTP (2004), as well as mean rat fetal weights from Gulati et al. (1991a), were
modeled for comparison in identifying a POD. A BMD of 31 mg/kg-day and BMDL of 19
mg/kg-day were derived from the multistage model for the increased incidence of hepatocellular
cytoplasmic vacuolization in female rats. For serum ALT levels in female rats, a BMD of 82
80
DRAFT - DO NOT CITE OR QUOTE
-------
1 mg/kg-day and a BMDL of 69 mg/kg-day was derived from the Hill model. For serum SDH in
2 female rats, a BMD of 157 mg/kg-day and a BMDL of 113 mg/kg-day was derived from the
3 power model. The serum ALP data were not amenable to BMD modeling; a LOAEL of 160
4 mg/kg-day was identified. For bile acid levels in female rats, a BMD of 188 mg/kg-day and a
5 BMDL of 170 mg/kg-day were derived from the polynomial model. BMD modeling derived a
6 BMD of 83 mg/kg-day and a BMDL of 60 mg/kg-day from a linear model with a BMR of 1 SD
7 for decreased rat fetal weight.
8 The BMDiso of 22 mg/kg-day and BMDLiso of 15 mg/kg-day based on increased
9 relative liver weight in the female rat was selected as the POD for the subchronic RfD. The
10 observed changes in liver weights, serum liver enzyme levels, and hepatocellular necrosis
11 combine to support hepatotoxicity as the major toxic effect following 1,1,2,2-tetrachloroethane
12 exposure.
13
14 5.1.1.3. RfD Derivation—Including Application of Uncertainty Factors (UFs)
15 To derive the subchronic RfD, the 15 mg/kg-day BMDLiso for increased relative liver
16 weight in female rats is divided by a total UF of 300. The UF of 300 comprises component
17 factors of 10 for interspecies extrapolation, 10 for interhuman variability, and 3 for database
18 deficiencies.
19 A default UF of 10 was selected to account for the interspecies variability in
20 extrapolating from laboratory animals (rats) to humans (i.e., interspecies variability), because
21 information was not available to quantitatively assess toxicokinetic or toxicodynamic differences
22 between animals and humans for 1,1,2,2-tetrachloroethane.
23 A default UF of 10 was selected to account for inter-individual variability (UFH) to
24 account for human-to-human variability in susceptibility in the absence of quantitative
25 information to assess the toxicokinetics and toxicodynamics of 1,1,2,2-tetrachloroethane in
26 humans. However, studies of human GST-zeta polymorphic variants demonstrate different
27 enzymatic activities toward and inhibition by dichloroacetic acid that could affect the
28 metabolism of 1,1,2,2-tetrachloroethane (Lantum et al, 2002; Blackburn et al, 2001, 2000;
29 Tzeng et al., 2000). Populations, or single individuals, may be more sensitive to 1,1,2,2-tetra-
30 chloroethane toxicity depending on which GST-zeta variant they possess. Animal toxicity
31 studies did not show consistent sex-related differences.
32 An UF of 3 was selected to account for deficiencies in the database. The NTP (2004)
33 14-week study provides comprehensive evaluations of systemic toxicity and neurotoxicity in two
34 species. The NTP (2004) study provides information of effects on sperm, estrous cycle, and
35 male and female reproductive tissues in rats and mice, but the database lacks a two-generation
36 reproductive toxicity study. Available developmental toxicity studies provide information on
37 embryo or fetotoxicity in orally exposed rats and mice (Gulati et al., 1991a, b), but the studies
38 did not include skeletal and visceral examinations.
81 DRAFT - DO NOT CITE OR QUOTE
-------
1 An UF for LOAEL-to-NOAEL extrapolation was not used because the current approach
2 is to address this factor as one of the considerations in selecting a BMR for benchmark dose
3 modeling. In this case, a BMR associated with a change of 1 SD from the control mean was
4 selected under an assumption that it represents a minimal biologically significant change.
5 The subchronic RfD for 1,1,2,2-tetrachloroethane is calculated as follows:
6
7 Subchronic RfD = BMDLiso-UF
8 = 15 mg/kg-day-3 00
9 =0.05 mg/kg-day (or 5 x 10"2 mg/kg-day)
10
11 5.1.2. Chronic Oral RfD
12 5.1.2.1. Choice of Principal Study and Critical Effect - with Rationale and Justification
13 Information on the chronic oral toxicity of 1,1,2,2-tetrachloroethane is limited to a
14 78-week cancer bioassay in rats and mice that were exposed by gavage (NCI, 1978).
15 Interpretation of the rat study may be confounded by high incidences of endemic chronic murine
16 pneumonia, although it is unlikely that this contributed to effects observed in the liver. Based on
17 an increased incidence of hepatic fatty changes, the NOAEL and LOAEL for liver effects were
18 62 and 108 mg/kg-day, respectively. In the mouse study, a LOAEL of 142 mg/kg-day was
19 selected for chronic inflammation in the kidneys of males and a NOAEL of 142 mg/kg-day and a
20 LOAEL of 284 mg/kg-day were selected for hydronephrosis and chronic inflammation in the
21 kidneys of females, respectively.
22 The 14-week dietary study in rats and mice (NTP, 2004), used to derive the subchronic
23 RfD, was also considered for the derivation of the chronic RfD. The subchronic NTP (2004)
24 study appears to be a more sensitive assay than the chronic NCI (1978) bioassay. The NTP
25 (2004) study also uses lower dose levels and a wider dose range than the NCI (1978) study, and
26 thereby provides a better characterization of the dose-response curve in the low-dose region.
27 Additionally, dietary exposure is a more relevant route of exposure for the general population
28 exposed to 1,1,2,2-tetrachloroethane in the environment than is gavage exposure. For these
29 reasons, the NTP (2004) subchronic study was selected as the principal study.
30 EPA selected increased liver weight as the critical effect because this effect may
31 represent a potential sensitive endpoint that may occur early in the process leading to
32 hepatocellular necrosis associated with subchronic oral exposure to 1,1,2,2-tetrachloroethane.
33 The increase in relative liver weight was selected as the basis for the selection of the POD
34 because this analysis takes into account the substantive, dose-dependent decreases in body
35 weight that were observed in both sexes of rats. Additional liver effects observed included
36 increased liver weight, increased ALT, ALP, and SDH serum levels, increased serum bile acid
37 levels, and increased incidence of hepatocellular vacuolization and necrosis.
38
39 5.1.2.2. Methodsof Analysis—Including Models (PBPK, BMD, etc.)
82 DRAFT - DO NOT CITE OR QUOTE
-------
1 The subchronic BMDLiso of 15 mg/kg-day based on the increased relative liver weight
2 in female rats was used as the POD for the chronic RfD. The observed increases in liver weights,
3 serum liver enzyme levels, and incidence of hepatocellular necrosis combine to support
4 hepatotoxicity as the critical effect of toxicity of 1,1,2,2-tetrachloroethane.
5
6 5.1.2.3. RfD Derivation—Including Application of UFs
1 To derive the chronic RfD, the subchronic BMDLiso of 15 mg/kg-day, based on
8 increased relative liver weights in female rats, was divided by a UF of 1,000. The UF of 1,000
9 comprises component factors of 10 for interspecies extrapolation, 10 for interhuman variability,
10 3 for subchronic to chronic duration extrapolation, and 3 for database deficiencies, as explained
11 below.
12 A default UF of 10 was selected to account for the interspecies variability in
13 extrapolating from laboratory animals (rats) to humans (i.e., interspecies variability), because
14 information was not available to quantitatively assess toxicokinetic or toxicodynamic differences
15 between animals and humans for 1,1,2,2-tetrachloroethane.
16 A default UF of 10 was selected to account for inter-individual variability (UFn) to
17 account for human-to-human variability in susceptibility in the absence of quantitative
18 information to assess the toxicokinetics and toxicodynamics of 1,1,2,2-tetrachloroethane in
19 humans. However, studies of human GST-zeta polymorphic variants demonstrate different
20 enzymatic activities toward and inhibition by dichloroacetic acid that could affect the
21 metabolism of 1,1,2,2-tetrachloroethane (Lantum et al, 2002; Blackburn et al, 2001, 2000;
22 Tzeng et al., 2000). Populations, or single individuals, may be more sensitive to 1,1,2,2-tetra-
23 chloroethane toxicity depending on which GST-zeta variant they possess. Animal toxicity
24 studies which evaluated both sexes in the same study did not show consistent sex-related
25 differences. Developmental toxicity studies in animals are limited in scope, but have not
26 indicated effects on the offspring at levels that did not also cause maternal effects.
27 An UF of 3 was selected to account for extrapolation from a subchronic exposure
28 duration study to a chronic RfD. The study selected as the principal study was a 14-week study
29 by NTP (2004), a study duration that is minimally past the standard subchronic (90 day) study
30 and falls well short of a standard lifetime study. In addition, some data are available to inform
31 the nature and extent of effects that would be observed with a longer duration of exposure to
32 1,1,2,2-tetrachloroethane. Specifically, the available chronic cancer bioassay data (NCI, 1978)
33 suggest that liver damage observed in F344 rats following subchronic exposure to 1,1,2,2-tetra-
34 chloroethane (NTP, 2004), e.g., increased liver weight and incidence of necrosis, and altered
35 serum enzyme and bile levels, may not progress to more severe effects following chronic
36 exposures. The chronic cancer bioassay was conducted in Osborne-Mendel rats and did not
37 measure liver enzyme levels. However, NCI (1978) observed minimal alterations in liver
38 pathology, including inflammation, fatty metamorphosis, focal cellular change, and angiectasis
83 DRAFT - DO NOT CITE OR QUOTE
-------
1 in rats, and organized thrombus and nodular hyperplasia in mice. NCI (1978) reported that the
2 study authors performed complete histological analysis on the liver, but specific endpoints
3 assessed were not included. The available database does not abrogate all concern associated
4 with using a subchronic study as the basis of the RfD. For these reasons, a threefold UF was
5 used to account for the extrapolation from subchronic to chronic exposure duration for the
6 derivation of the chronic RfD.
7 An UF of 3 was selected to account for deficiencies in the database. The NTP (2004)
8 14-week study provides comprehensive evaluations of systemic toxicity and neurotoxicity in
9 both rats and mice. However, the database is limited by the lack of a two-generation
10 reproductive toxicity study. The NTP (2004) study provides information on effects on sperm,
11 estrous cycle, and male and female reproductive tissues in rats and mice, but the database lacks a
12 two-generation reproductive toxicity study. Available developmental toxicity studies provide
13 information on embryo or fetotoxicity in orally exposed rats and mice (Gulati et al., 199la, b),
14 but the studies did not include skeletal and visceral examinations.
15 An UF for LOAEL-to-NOAEL extrapolation was not used because the current approach
16 is to address this factor as one of the considerations in selecting a BMR for benchmark dose
17 modeling. In this case, a BMR associated with a change of 1 SD from the control mean was
18 selected under an assumption that it represents a minimal biologically significant change.
19 The chronic RfD for 1,1,2,2-tetrachloroethane is calculated as follows:
20
21 Chronic RfD = BMDLiSD-UF
22 =15 mg/kg-day - 1,000
23 = 0.015 mg/kg-day (or 1.5 x 10"2 mg/kg-day)
24
25 5.1.3. RfD Comparison Information
26 Figure 5-1 is an exposure-response array that presents NOAELs, LOAELs, and the dose
27 range tested corresponding to selected health effects. The effects observed in the subchronic and
28 chronic studies were considered candidates for the derivation of the sample subchronic and
29 chronic RfDs.
30 In addition to the increase in relative liver weight and the increased incidence of
31 hepatocellular cytoplasmic vacuolization, changes in absolute liver weight and serum levels of
32 ALT and SDH, bile acid levels, and serum cholesterol levels were considered for comparison.
33 Mean rat fetal weights observed following subchronic or chronic exposure to 1,1,2,2-tetrachloro-
34 ethane were also considered for comparison. Table 5-3 provides a tabular summary of sample
35 PODs and resulting subchronic sample RfDs for these endpoints in female rats. Additionally,
36 Figure 5-2 provides a graphical representation of this information. This figure should be
37 interpreted with caution since the PODs across studies are not necessarily comparable, nor is the
38 confidence the same in the data sets from which the PODs were derived. Figure 5-3 provides a
84 DRAFT - DO NOT CITE OR QUOTE
-------
1 graphical representation of the derivation of sample chronic RfDs for sample PODs from the
2 subchronic data.
3
85 DRAFT - DO NOT CITE OR QUOTE
-------
350
0>
(A
O
O
1
<
>
>
1
I
» LOAEL
• NOAEL
1
TU., ,,„,+;„„!
lines represent
the range of
doses tested in
a given study.
f
I
T T T T T T T T I
1-^f Iss s|I -- SQ: -~- ifs llo.- Is|^ l|o'
= E .0 CM CD ^S = 05 CM <^'-- 051——. T=> — o CD co i _ o ' ~— <-, ^^^CD ^--cor^
o Q_.b! |— — _£= 0- .>i|— coJ«C3 CD^,-, to-o" cOcoQ- o««coo ca^-- OO"±;CD
ro o o Z co °>H- ro^Z ^SP^ rorooi ro^l- 2^!~ ro^roOi CD ^^ r -^ E 2 ^
o-^^— -^^5, ^ra^ 2 S^ eroz ""5- 0.°^ oo-ro roS^o
CDOOCO rofe C- — CO 0 ^X3 >— ^-. o ^X3- — - £= CO CD 0 o+ CD -Q 2 Q- CD
-^^ro ? ra .c " s= ' -<=a> -o a^c'
1
2
Figure 5-1. Exposure response array for subchronic and chronic oral exposure to 1,1^2,2-tetrachloroethane.
86
DRAFT - DO NOT CITE OR QUOTE
-------
Table 5-3. Potential PODs with applied UFs and resulting subchronic RfDs
Effect
Hepatocellular
cytoplasmic
vacuolization
Relative liver weight
Absolute liver
weight
ALT
SDH
Bile acids
Fetal body weight
POD (mg/kg-d)
l.lb
15C
23C
26C
113C
57C
60d
Gender
and
Species
Male Rat
Female
Rat
Male
Rat
Male
Rat
Female
Rat
Male
Rat
Rat
UFsa
A
10
10
10
10
10
10
10
H
10
10
10
10
10
10
10
L
-
-
-
-
-
-
s
-
-
-
-
-
-
D
3
3
3
3
3
3
3
Total
300
300
300
300
300
300
300
Subchronic
RfD
4 x 10"3
5 x 10'2
8 x 10'2
9 x 10"2
0.38
0.20
0.20
1
2
aUFs: A = animal to human (interspecies); H = interindividual (intraspecies); L = LOAEL to NOAEL;
S = subchronic-to-chronic duration; D = database deficiency.
bPOD based on BMDL determined through BMD modeling of a 10% response; source: NTP (2004).
°POD based on BMDL determined through BMD modeling of a 1 SD response; source: NTP (2004).
dPOD based on BMDL determined through BMD modeling of a 5% response; source: Gulati et al. (1991a).
87
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
1000
100
10
(O
T3
0.1
0.01
0.001
if
If
If
POD
UFA
UFD
RfD
hepatocellular relative liver weight- absolute liver weight increased ALT-o increased SDH-? increased bile acids - decreased fetal body
cytoplasnic ? rats (HIP, 2004) - o rats (NIP, 2004) rats (NTP, 2004) rats (NTP, 2004) o rats (NIP, 2004) weight-rats (Gulati
vacuolization - o rats etal., 1991 a)
(NTP, 2004)
Figure 5-2. PODs for selected endpoints (with critical effect circled) from Table 5-3 with corresponding applied
UFs and derived sample subchronic oral reference values (RfVs).
DRAFT - DO NOT CITE OR QUOTE
-------
1000
100
10
ra
_§> 1
"TO
E
0.1
0.01
0.001
%\\\\V
%\\\\V
POD
UFA
UFH
UFD
UFsc
RfD
\\\\v
\\\\v
hepatocellular
cytoplasmic
vacuolization - <$
rats (NTP, 2004)
relative liver
weight- ? rats
(NTP, 2004)
absolute liver
w eight - (5 rats
(NTP, 2004)
increased ALT-
S rats (NTP,
2004)
increased SDH -
2004)
increased bile decreased fetal
acids - (5 rats body w eight -
(NTP, 2004) rats (Gulati et al.,
1991 a)
1
2
3
Figure 5-3. PODs for selected endpoints (with critical effect circled) from Table 5-3 with corresponding applied UFs
and derived sample chronic oral reference values (RfVs).
89
DRAFT - DO NOT CITE OR QUOTE
-------
1 5.1.4. Previous RfD Assessment
2 An oral assessment for 1,1,2,2-tetrachloroethane was not previously available on IRIS.
3
4 5.2. INHALATION REFERENCE CONCENTRATION (RfC)
5 5.2.1. Choice of Principal Study and Critical Effect—with Rationale and Justification
6 Information on the inhalation toxicity of 1,1,2,2-tetrachloroethane is limited. In Truffert
7 et al. (1977), rats were exposed to a presumed concentration of 560 ppm (3,909 mg/m3) for a
8 TWA duration of 5.1 hours/day, 5 days/week for 15 weeks. Findings included transient
9 histological alterations in the liver, including granular appearance and cytoplasmic vacuolation,
10 which were observed after 9 exposures and were no longer evident after 39 exposures. Because
11 of the uncertainty regarding the actual exposure concentration for the single dose, and a lack of
12 incidence and severity data, this report cannot be used to identify a NOAEL or LOAEL or for
13 possible derivation of an RfC.
14 Horiuchi et al. (1962) observed fatty degeneration of the liver and splenic congestion in a
15 single monkey exposed to a TWA of 1,974 ppm (15,560 mg/m3) 1,1,2,2-tetrachloroethane for
16 2 hours/day, 6 days/week for 9 months. The monkey was weak after approximately seven
17 exposures and had diarrhea and anorexia between the 12th and 15th exposures. Beginning at the
18 15th exposure, the monkey was "almost completely unconscious falling upon his side" for 20-
19 60 minutes after each exposure. Also, hematological parameters demonstrated sporadic changes
20 in hematocrit and RBC and WBC counts, but the significance of these findings cannot be
21 determined. This study cannot be utilized to identify a NOAEL or LOAEL due to the use of a
22 single test animal with no control group.
23 Mellon Institute of Industrial Research (1947) observed an increased incidence of lung
24 lesions and an increase in kidney weight in rats following a 6-month exposure to 200 ppm
25 1,1,2,2-tetrachloroethane, but these results were not evaluated because the control animals
26 experienced a high degree of pathological effects in the kidney, liver, and lung, and because of
27 the presence of an endemic lung infection in both controls and treated groups. MIIR (1947) also
28 observed increased serum phosphatase levels and blood urea nitrogen levels in a dog exposed to
29 200 ppm 1,1,2,2-tetrachloroethane, compared to control values, along with cloudy swelling of
30 the liver and the convoluted tubules of the kidney, and light congestion of the lungs. However,
31 identification of a LOAEL or NOAEL is precluded by poor study reporting, high mortality and
32 lung infection in the rats, and the use of a single treated animal in the dog study.
33 Kulinskaya and Verlinskaya (1972) observed inconsistent changes in acetylcholine levels
34 in Chinchilla rabbits exposed to 10 mg/m3 (1.5 ppm) 1,1,2,2-tetrachloroethane for 3 hours/day,
35 6 days/week for 7-8.5 months. A NOAEL or LOAEL was not identified because the changes in
36 acetylcholine were not consistent across time and incompletely quantified, and the biological
37 significance of the change is unclear.
90 DRAFT - DO NOT CITE OR QUOTE
-------
1 Shmuter (1977) observed increases in antibody levels in Chinchilla rabbits at 2 mg/m3
2 1,1,2,2-tetrachloroethane and decreases in antibody levels at 100 mg/m3. Exposure to
3 100 mg/m3 1,1,2,2-tetrachloroethane also resulted in a decrease in the relative content of
4 antibodies in the y-globulin fraction and an increase in the T and P fractions. This is a poorly
5 reported study that provides inadequate data, including reporting limitations, toxicological
6 uncertainty in the endpoints, and inconsistent patterns of response, which preclude the
7 identification of a NO AEL or LO AEL.
8 Effects following the chronic inhalation toxicity of 1,1,2,2-tetrachloroethane included
9 hematological alterations and increased liver fat content in rats exposed to 1.9 ppm (13.3 mg/m3)
10 4 hours/day for 265 days (Schmidt et al, 1972). Statistically significant changes included
11 increased leukocyte (89%) and pVglobulin (12%) levels compared to controls after 110 days,
12 and an increased percentage of segmented nucleated neutrophils (36%), decreased percentage of
13 lymphocytes (17%), and increased liver total fat content (34%) after 265 days. A statistically
14 significant decrease in y-globulin levels (32%) at 60 days postexposure and a decrease in adrenal
15 ascorbic acid content (a measure of pituitary ACTH activity) were observed at all three time
16 periods (64, 21, and 13%, respectively). This study is insufficient for identification of a NO AEL
17 or LOAEL for systemic toxicity because most of the observed effects occurred at a single dose or
18 time point, or there was a reversal of the effect at the next dose or time point. A reproductive
19 assessment in the Schmidt et al. (1972) study was sufficient for identification of a NO AEL for
20 the single dose tested, 1.9 ppm (13.3 mg/m ), for reproductive effects in male rats, including
21 percentage of mated females having offspring, littering interval, time to 50% littered, total
22 number of pups, pups per litter, average birth weight, postnatal survival on days 1, 2, 7, 14, 21,
23 and 84, sex ratio, and average body weight on postnatal day 84. However, macroscopic
24 malformations or significant group differences in the other indices were not observed at
25 13.3 mg/m3. The lack of information on the reproductive toxicity precludes utilizing the selected
26 NO AEL in the derivation of the RfC.
27 In addition, effects of chronic exposure to 1,1,2,2-tetrachloroethane included alterations
28 in serum acetylcholinesterase activity in rabbits exposed to 1.5 ppm (10 mg/m ) 1,1,2,2-tetra-
29 chloroethane 3 hours/day, 6 days/week for 7-8.5 months (Kulinskaya and Verlinskaya, 1972)
30 and immunological alterations in rabbits exposed to 0.3-14.6 ppm (2-100 mg/m3) 3 hours/day,
31 6 days/week, for 8-10 months (Shmuter, 1977). These studies are inadequate for identification
32 of NOAELs or LOAELs for systemic toxicity due to inadequate study reporting.
33 The inhalation toxicity database lacks a well-conducted study that demonstrates a dose-
34 related toxicological effect following subchronic and/or chronic exposure to 1,1,2,2-tetrachloro-
35 ethane. Therefore, an inhalation RfC was not derived.
36
37 5.2.2. Methods of Analysis—Including Models (PBPK, BMD, etc.)
91 DRAFT - DO NOT CITE OR QUOTE
-------
1 A route-to-route extrapolation using the computational technique of Chiu and White
2 (2006), as described in Section 3.5, was considered. However, U.S. EPA (1994b) recommends
3 not conducting a route-to-route extrapolation from oral data when a first-pass effect by the liver
4 or respiratory tract is expected, or a potential for a portal-of-entry effect in the respiratory tract is
5 indicated following analysis of short-term inhalation, dermal irritation, in vitro studies, or
6 evaluation of the physical/chemical properties. In the case of 1,1,2,2-tetrachloroethane, a first-
7 pass effect by the liver is expected. In addition, the presence of tissue-bound metabolites in the
8 epithelial linings in the upper respiratory tract may demonstrate a first-pass effect by the
9 respiratory tract (Eriksson and Brittebo, 1991). Lehmann et al. (1936) observed irritation of the
10 mucous membranes of two humans following inhalation of 146 ppm (1,003 mg/m3) for
11 30 minutes or 336 ppm (2,308 mg/m ) for 10 minutes, indicating the potential for portal-of-entry
12 effects in the respiratory system.
13
14 5.2.3. Previous RfC Assessment
15 An inhalation assessment for 1,1,2,2-tetrachloroethane was not previously available on
16 IRIS.
17
18 5.3. UNCERTAINTIES IN THE ORAL REFERENCE DOSE (RfD) AND INHALATION
19 REFERENCE CONCENTRATION (RfC)
20 The following discussion identifies some uncertainties associated with the RfD for
21 1,1,2,2-tetrachloroethane. As presented earlier (Sections 5.1.2 and 5.1.3; 5.2.2 and 5.2.3), EPA
22 standard practices and RfC and RfD guidance (U.S. EPA, 1994b) were followed in applying an
23 UF approach to a POD, a BMDLiso for the subchronic and chronic RfDs. Factors accounting
24 for uncertainties associated with a number of steps in the analyses were adopted to account for
25 extrapolating from an animal bioassay to human exposure, a diverse human population of
26 varying susceptibilities, and to account for database deficiencies. These extrapolations are
27 carried out with standard approaches given the lack of extensive experimental and human data on
28 1,1,2,2-tetrachloroethane to inform individual steps.
29 An adequate range of animal toxicology data is available for the hazard assessment of
30 1,1,2,2-tetrachloroethane, as described in Section 4. Included in these studies are short-term and
31 long-term bioassays and a developmental toxicity bioassay in rats and mice, as well as numerous
32 supporting genotoxicity and metabolism studies. Toxicity associated with oral exposure to
33 1,1,2,2-tetrachloroethane is observed in the liver, kidney, and developing organism, including
34 decreased fetal body weight and increased number of litter resorptions.
35 Consideration of the available dose-response data to determine an estimate of oral
36 exposure that is likely to be without an appreciable risk of adverse health effects over a lifetime
37 led to the selection of the 14-week oral dietary study in rats (NTP, 2004) and increased relative
38 liver weight in females as the principal study and critical effect, respectively, for deriving the
92 DRAFT - DO NOT CITE OR QUOTE
-------
1 subchronic and chronic RfDs for 1,1,2,2-tetrachloroethane. The NTP (2004) data demonstrate
2 hepatocellular damage, including increased liver weight, increased serum liver enzyme levels,
3 and increased incidence of hepatic necrosis. Increased liver weight was chosen as the critical
4 effect because it may represent a sensitive indicator of 1,1,2,2,-tetrachloroethane-induced
5 hepatoxicity and occurs at a dose lower than the observed overt liver necrosis. The increase in
6 relative liver weight was selected as the basis for the selection of the POD because this analysis
7 takes into account the substantive, dose-dependent decreases in body weight that were observed
8 in both sexes of rats. The dose-response relationships between oral exposure to 1,1,2,2-tetra-
9 chloroethane and fetal body weight in rats and mice are also suitable for deriving an RfD, but are
10 associated with BMDLs that are less sensitive than the selected critical effect and corresponding
11 BMDL.
12 For comparison purposes, Figure 5-2 presents potential PODs, applied UFs, and derived
13 potential RfDs for the additional endpoints that were modeled using the EPA's BMDS, version
14 2.1.1. The additional endpoints included increased absolute liver weight, changes in serum ALT
15 and SDH, increased bile acids, and increased incidence of hepatocellular necrosis, all of which
16 support the liver as the target of 1,1,2,2-tetrachloroethane-induced toxicity following oral
17 exposure. A decrease in rat fetal weight was also modeled. The change in serum ALP was
18 modeled, but a model with adequate fit was not available.
19 The selection of the BMD model for the quantitation of the RfD does not lead to
20 significant uncertainty in estimating the POD, since benchmark effect levels were within the
21 range of experimental data. However, the selected model, the polynomial model, does not
22 represent all possible models one might fit, and other models could be selected to yield more
23 extreme results, both higher and lower than those included in this assessment.
24 Extrapolating from animals to humans embodies further issues and uncertainties. An
25 effect and its magnitude associated with the concentration at the POD in rodents are extrapolated
26 to human response. Pharmacokinetic models are useful in examining species differences in
27 pharmacokinetic processing, however, dosimetric adjustment using pharmacokinetic modeling
28 was not possible for the toxicity observed following oral and inhalation exposure to 1,1,2,2-tetra-
29 chloroethane. Additional interspecies uncertainty may arise from the rate of metabolism across
30 species, as it has been demonstrated that mice have greater metabolic capacity following
31 exposure to tetrachloroethylene than rats and humans (Reitz et al, 1996). Reitz et al. (1996)
32 demonstrated that mice possessed a greater relative ability to metabolize tetrachloroethylene than
33 rats and humans, and, although data are not available, a similar situation may exist for 1,1,2,2-
34 tetrachloroethane.
35 Heterogeneity among humans is another uncertainty associated with extrapolating from
36 animals to humans. Uncertainty related to human variation needs to be considered; also,
37 uncertainties in extrapolating from a subpopulation, say of one sex or a narrow range of life
38 stages typical of occupational epidemiologic studies, to a larger, more diverse population need to
93 DRAFT - DO NOT CITE OR QUOTE
-------
1 be addressed. In the absence of 1,1,2,2-tetrachloroethane-specific data on human variation, a
2 factor of 10 was used to account for uncertainty associated with human variation in the
3 derivation of the RfD. Human variation may be larger or smaller; however, 1,1,2,2-tetrachloro-
4 ethane-specific data to examine the potential magnitude of over- or under-estimation are
5 unavailable.
6 Extrapolating from subchronic PODs to derive chronic reference values is also an
7 uncertainty encountered in this assessment. A threefold UF was selected to account for
8 extrapolation from a subchronic exposure duration study to a chronic RfD. Based on the
9 available data for 1,1,2,2-tetrachloroethane, the toxicity observed in the liver does not appear to
10 increase over time. The use of data from a subchronic study to derive a chronic RfD becomes a
11 concern when the damage, in this case hepatoxicity, has the potential to accumulate; however, if
12 the progression of the effect is not apparent, a reduced UF may be considered (U.S. EPA, 1994b).
13 Specifically, liver damage observed in F344 rats following subchronic exposure to 1,1,2,2-tetra-
14 chloroethane (NTP, 2004), e.g., increased incidence of necrosis or altered serum enzyme and bile
15 levels, did not progress to more severe effects such as cirrhosis or major liver disease following
16 chronic exposures (NCI, 1978). NCI (1978) observed minimal alterations in liver pathology,
17 including inflammation, fatty metamorphosis, focal cellular change, and angiectasis in rats, and
18 organized thrombus and nodular hyperplasia in mice. Therefore, the available database does not
19 abrogate all concern associated with using a subchronic study as the basis of the RfD, but
20 supports the utilization of a database UF of 3.
21 Data gaps have been identified that are associated with uncertainties in database
22 deficiencies specific to the developmental and reproductive toxicity of 1,1,2,2-tetrachloroethane
23 following oral exposure. The developing fetus may be a target of toxicity, and the absence of a
24 study specifically evaluating the full range of developmental toxicity endpoints represents an
25 area of uncertainty or gap in the database. The database of inhalation studies is of particular
26 concern due to the paucity of studies, especially subchronic and chronic studies, a multi-
27 generational reproductive study, and a developmental toxicity study.
28
29 5.4. CANCER ASSESSMENT
30 As discussed in Section 4.7, under U.S. EPA's Guidelines for Carcinogen Risk
31 Assessment (U.S. EPA, 2005a), 1,1,2,2-tetrachloroethane is "likely to be carcinogenic to
32 humans" based on data from an oral cancer bioassay in male and female Osborne-Mendel rats
33 and B6C3Fi mice (NCI, 1978) demonstrating an increase in the incidence of hepatocellular
34 carcinomas in both sexes of mice. In this study, the incidence of hepatocellular carcinomas was
35 statistically significantly increased in both sexes of B6C3Fi mice at 142 (13/50 males; 30/48
36 females) and 284 mg/kg-day (44/49 males; 43/47 females), with incidences in the male and
37 female controls of 3/36 and 1/40, respectively. NCI (1978) also demonstrated a decrease in the
38 time to tumor in both sexes of mice. Male rats exhibited an increased incidence in hepatocellular
94 DRAFT - DO NOT CITE OR QUOTE
-------
1 carcinomas, characterized as rare tumors, but the increased incidence was not statistically
2 significantly different from controls. NCI (1978) has characterized the carcinogenic results in
3 male rats as "equivocal."
4 The epidemiological human data available are inadequate for evaluation for cancer risk
5 (IARC, 1999). There are a limited number of positive results from genotoxicity studies which
6 suggest that 1,1,2,2-tetrachloroethane treatment in animals can result in UDS (Miyagawa et al.,
7 1995), chromosomal aberrations (McGregor, 1980), SCE (NTP, 2004; Colacci et al., 1992), and
8 micronucleus formation (NTP, 2004). The ability of 1,1,2,2,-tetrachloroethane to alkylate
9 enzymatically purified hepatic DNA was observed following a single oral dose of 150 mg/kg of
10 1,1,2,2-tetrachloroethane in B6C3Fi mice (Dow Chemical Company, 1988). 1,1,2,2-Tetra-
11 chloroethane may have tumor initiating and promoting activity in mammalian cells (Colacci et
12 al., 1996, 1992; Milrnan et al., 1988; Story et al., 1986).
13
14 5.4.1. Choice of Study/Data—with Rationale and Justification
15 The only carcinogenicity bioassay for 1,1,2,2-tetrachloroethane is a chronic gavage study
16 in Osborne-Mendel rats and B6C3Fi mice performed by NCI (1978). This study was conducted
17 in both sexes in two species with an adequate number of animals per dose group, with
18 examination of appropriate toxicological endpoints in both sexes of rats and mice. Selection of
19 doses was aided by range-finding toxicity tests. The rat study did not identify statistically
20 significant increases in tumor incidences in males or females. Three rare liver tumors in high-
21 dose male rats were noted.
22 The mouse study identified statistically significant, dose-related increases in the
23 incidences of hepatocellular carcinomas in both sexes. Based on these increases in
24 hepatocellular carcinomas, NCI (1978) concluded that orally administered 1,1,2,2-tetrachloro-
25 ethane is a liver carcinogen in male and female B6C3Fi mice. NCI (1978) stated that there was
26 no evidence for carcinogenicity of 1,1,2,2-tetrachloroethane in Osborne-Mendel rats (NCI, 1978).
27 The tumor data in mice from the NCI study was used for dose-response analysis for oral
28 exposure.
29
30 5.4.2. Dose-response Data
31 Data on the incidences of hepatocellular carcinomas in male and female mice from the
32 NCI (1978) study were used for cancer dose-response assessment. These data are shown in
33 Table 5-4. The control data were pooled from vehicle control groups. The cancer bioassay for
34 1,1,2,2-tetrachloroethane demonstrated evidence of increased incidence of tumors in both sexes
35 of one species.
36
95 DRAFT - DO NOT CITE OR QUOTE
-------
Table 5-4. Incidences of hepatocellular carcinomas in B6C3Fi mice used for
dose-response assessment of 1,1^2,2-tetrachloroethane
Sex
Male
Female
Dose (mg/kg-d)a
0
3/36b
l/40b
142
13/50
30/48
284
44/49
43/47
aTWA dose administered by gavage on 5 d/wk for 78 wks.
Pooled vehicle (corn oil) control groups from this and another, concurrent, bioassay. Pooling based on identical
housing and care, similar spontaneous tumor rates, placed on test at about the same time, and examined by the same
pathologists.
Source: NCI (1978).
1
2 5.4.3. Dose Adjustments and Extrapolation Method(s)
3 Conversion of the doses in the NCI (1978) mouse study to human equivalent doses
4 (HEDs) to be used for dose-response modeling was accomplished in three steps. The mice were
5 treated with 1,1,2,2-tetrachloroethane by gavage 5 days/week for 78 weeks and then observed
6 untreated for 12 weeks for a total study duration of 90 weeks. Because the reported TWA doses
7 were for a 5 day/week, 78 week exposure, they were duration-adjusted to account for the partial
8 week exposure (by multiplying by 5 days/7 days) and untreated observation period (by
9 multiplying by 78 weeks/90 weeks). These duration-adjusted animal doses were then converted
10 to HEDs by adjusting for differences in body weight and lifespan between humans and mice. In
11 accordance with the U.S. EPA (2005a) Guidelines for Carcinogen Risk Assessment, a factor of
12 BW3/4 was used for cross-species scaling. Because the study duration (90 weeks) was less than
13 the animal lifespan (104 weeks), the scaled dose was then multiplied by the cubed ratio of
14 experimental duration to animal lifespan to complete the extrapolation to a lifetime exposure in
15 humans. The equation and data used to calculate the HEDs are presented below, and the
16 calculated HEDs are presented in Table 5-5.
17
18 HED = Dose* x (W/70 kg)1/4 x (Le/L)3
19 Where:
20 Dose = average daily animal dose (* TWA converted for 5/7 days, 78/90 weeks)
21 W = average animal body weight (0.030 kg for male and female B6C3Fi mice [U.S. EPA,
22 1988]).
23 70 kg = reference human body weight (U. S. EPA, 1988)
24 Le = duration of experiment (90 weeks)
25 L = reference mouse lifespan (104 weeks) (U.S. EPA, 1988)
26
96 DRAFT - DO NOT CITE OR QUOTE
-------
Table 5-5. HEDs corresponding to duration-adjusted TWA doses in mice
Duration-adjusted dose in male and female mice (mg/kg-d)
HED for use with both male and female mouse incidence data (mg/kg-d)
Dose (mg/kg-d)
0
0
87.9
8.22
175.8
16.5
1
2 The mode of action of 1,1,2,2-tetrachloroethane carcinogenicity is unknown. It appears
3 that metabolism to one or more active compounds is likely to play a role in the development of
4 the observed liver tumors, but insufficient data preclude proposing a specific mode of action.
5 Dichloroacetic acid, a metabolite of 1,1,2,2-tetrachloroethane, induces hepatocellular carcinomas
6 in male and female B6C3Fi mice and F344 rats. Trichloroethylene (NTP, 1990; NCI, 1976) and
7 tetrachloroethylene (NTP, 1996; NCI, 1977), also metabolites of 1,1,2,2-tetrachloroethane, have
8 also been shown to be hepatocarcinogens in rodents.
9 Results of genotoxicity and mutagenicity studies of 1,1,2,2-tetrachloroethane are mixed
10 and insufficient for informing whether 1,1,2,2-tetrachloroethane carcinogenicity is associated
11 with a mutagenic mode of action. Given that the mechanistic and other information available on
12 cancer risk from exposure to 1,1,2,2-tetrachloroethane is sparse and that the existing data do not
13 inform the mode of action of carcinogenicity, a linear low-dose extrapolation was conducted as a
14 default option for the derivation of the oral slope factor.
15 Dose-response modeling was performed to obtain a POD for quantitative assessment of
16 cancer risk. The data sets for hepatocellular carcinoma in both sexes of mice were modeled for
17 determination of the POD. In accordance with the U.S. EPA (2005a) cancer guidelines, the
18 BMDLio (lower bound on dose estimated to produce a 10% increase in tumor incidence over
19 background) was estimated by applying the multistage cancer model in the EPA's BMDS
20 (version 2.1.1.) for the dichotomous incidence data, and selecting the results of the model that
21 best characterizes the cancer incidences. The BMD modeling of the male mouse data did not
22 achieve adequate model fit for any of the dichotomous models; thus, a cancer slope factor was
23 not derived from the male data. The 1° multistage model was selected for the derivation of the
24 cancer slope factor from the female data because this model provided adequate model fit and the
25 lowest Akaike's Information Criterion (AIC) when compared to the results of the 2° multistage
26 model. In addition, the 2° multistage model had insufficient degrees of freedom to test the
27 goodness-of-fit. The BMDL of 0.65 mg/kg-day from the modeling of the tumor incidence data
28 in female mice is selected as the POD for use in calculation of an oral slope factor (Table 5-6).
29 Details of the BMD modeling are presented in Appendix C.
30
97 DRAFT - DO NOT CITE OR QUOTE
-------
Table 5-6. Summary of human equivalent BMDs and BMDLs based on
hepatocellular carcinoma incidence data in female B6C3Fi mice
Female mice
BMR
(% extra risk)
10
HMD
(mg/kg-d)a
0.81
BMDL10
(mg/kg-d)a
0.65
aHED.
1
2 5.4.4. Oral Slope Factor and Inhalation Unit Risk
3 The oral slope factor was derived from the BMDLio (the lower bound on the exposure
4 associated with a 10% extra cancer risk) by dividing the BMR by the BMDLio, and represents an
5 upper bound on cancer risk associated with a continuous lifetime exposure to 1,1,2,2-tetrachloro-
6 ethane. In accordance with the U.S. EPA (2005a) guidelines, an oral slope factor (mg/kg-day)"1
7 was calculated by dividing the human equivalent BMDLio into 0.1 (10%) (Appendix C). The
8 BMDLio, the lower 95% bound on exposure at 10% extra risk, is 0.65 mg/kg-day, and the cancer
9 slope factor, the slope of the linear extrapolation from the BMDLio to 0, is 0.10/0.65 = 0.15 per
10 mg/kg-day. The slope of the linear extrapolation from the central estimate (i.e., BMD) is
11 0.1/0.81 mg/kg-day or 0.12 (mg/kg-day)"1.
12 In the absence of any suitable data on the carcinogenicity of 1,1,2,2-tetrachloroethane via
13 the inhalation route, an inhalation unit risk has not been derived in this evaluation.
14
15 5.4.5. Uncertainties in Cancer Risk Values
16 Extrapolation of data from animals to estimate potential cancer risks to human
17 populations from exposure to 1,1,2,2-tetrachloroethane yields uncertainty. Several types of
18 uncertainties may be considered quantitatively, but other important uncertainties cannot be
19 considered quantitatively. Thus, an overall integrated quantitative uncertainty analysis is not
20 presented. This section and Table 5-7 summarize the principal uncertainties.
21
98 DRAFT - DO NOT CITE OR QUOTE
-------
Table 5-7. Summary of uncertainty in the l,l?2,2-tetrachloroethane cancer
risk assessment
Consideration/
approach
Low-dose
extrapolation
procedure
Dose metric
Cross-species
scaling
Statistical
uncertainty at POD
Bioassay
Species/gender
combination
Human relevance of
mouse tumor data
Human population
variability in
metabolism and
response/sensitive
subpopulations
Impact on oral slope
factor
Departure from U.S.
EPA's Guidelines for
Carcinogen Risk
Assessment POD
paradigm, if justified,
could J, or 1 slope
factor an unknown
extent
Alternatives could f
or I slope factor by an
unknown extent
Alternatives could J,
or 1 slope factor (e.g.,
3. 5 -fold J, [scaling by
BW] or t twofold
(scaling by BW2/3])
1 slope factor if MLE
used rather than lower
bound on POD
Alternatives could f
or J, slope factor by an
unknown extent
Human risk could J, or
1, depending on
relative sensitivity
Human relevance of
mouse tumor data
could I slope factor
Low-dose risk f or J,
to an unknown extent
Decision
Multistage cancer
model to determine
POD, linear low-
dose extrapolation
from POD
Used administered
exposure
BW3/4
LEC (method for
calculating
reasonable upper
bound slope factor)
NCI study
Female mice liver
cancer
Liver tumors in
mice are relevant
to human exposure
Considered
qualitatively
Justification
Available mode of action data do not inform
selection of dose-response model; linear approach
used in absence of an alternative as per U.S.
EPA's Guidelines for Carcinogen Risk
Assessment.
Experimental evidence supports a role for
metabolism in toxicity, but actual responsible
metabolites are not clearly identified.
There are no data to support alternatives. Because
the dose metric was not an AUC, BW3 4 scaling
was used to calculate equivalent cumulative
exposures for estimating equivalent human risks.
Limited size of bioassay results in sampling
variability; lower bound is 95% confidence
interval on administered exposure.
Alternative bioassays were unavailable.
There are no mode of action data to guide
extrapolation approach for any choice. It was
assumed that humans are as sensitive as the most
sensitive rodent gender/species tested; true
correspondence is unknown. The carcinogenic
response occurs across species. Generally, direct
site concordance is not assumed; consistent with
this view, some human tumor types are not found
in rodents and rat and mouse tumor types also
differ.
1,1,2,2-tetrachloroethane is carcinogenic through
an unknown mode of action.
No data to support range of human
variability/sensitivity, including whether children
are more sensitive. Metabolic activation mode of
action (if fully established) could indicate | or {
early-life susceptibility.
1
2
3
4
5
6
Choice of low-dose extrapolation approach. The mode of action is a key consideration in
clarifying how risks at low-dose exposures should be estimated. A linear low-dose extrapolation
approach was used to estimate human carcinogenic risk associated with 1,1,2,2-tetrachloroethane
exposure due to the unavailability of data that supports any specific mode of carcinogenic action
for 1,1,2,2-tetrachloroethane.
99
DRAFT - DO NOT CITE OR QUOTE
-------
1 The extent to which the overall uncertainty in low-dose risk estimation could be reduced
2 if the mode of action for 1,1,2,2-tetrachloroethane were known is of interest, but data on the
3 mode of action of 1,1,2,2-tetrachloroethane are not available.
4 Dose metric. 1,1,2,2-Tetrachloroethane is metabolized to intermediates with
5 carcinogenic potential. Dichloroacetic acid is recognized as hepatocarcinogenic in male B6C3Fi
6 mice and F344 rats (U.S. EPA, 2003). However, it is unknown whether a metabolite or some
7 combination of parent compound and metabolites is responsible for the observed toxicity. If the
8 actual carcinogenic moiety is proportional to administered exposure, then use of administered
9 exposure as the dose metric is the least biased choice. On the other hand, if this is not the correct
10 dose metric, then the impact on the slope factor is unknown.
11 Cross-species scaling. An adjustment for cross-species scaling (BW3/4) was applied to
12 address toxicological equivalence of internal doses between the rodent species and humans,
13 consistent with the 2005 Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005a). It is
14 assumed that equal risks result from equivalent constant lifetime exposures.
15 Statistical uncertainty at the POD. Parameter, or probabilistic, uncertainty can be
16 assessed through confidence intervals. Each description of parameter uncertainty assumes that
17 the underlying model and associated assumptions are valid. For the multistage cancer model
18 applied to the female mice data, there is a reasonably small degree of uncertainty at a 10%
19 increase in tumor incidence (the POD for linear low-dose extrapolation).
20 Bioassay selection. The study by NCI (1978) was used for development of an oral slope
21 factor. This study was conducted in both sexes in two species with an adequate number of
22 animals per dose group, with examination of appropriate toxicological endpoints in both sexes of
23 rats and mice. Alternative bioassays were unavailable. Both genders of mice exhibited liver
24 tumors. Uncertainties associated with the use of this study in the derivation of the oral slope
25 factor arise, primarily, from the study design. The dose levels used in the study were poorly
26 selected and were modified over the exposure duration, and the exposure duration of the study
27 (78 weeks) was less then the standard 104 week chronic exposure duration. In addition, the bolus
28 nature of the 1,1,2,2-tetrachloroethane gavage exposures in NCI (1978) may lead to more
29 pronounced irritation, inflammation, cell death, and an eventual increase in tumor incidence at
30 portals of entry because of direct contact of the test chemical with the gastroinstestinal tissues. There
31 was also an increased incidence of endemic chronic murine pneumonia in male and female rats and
32 mice, and while interpretation of this study is complicated by the chronic murine pneumonia, it is
33 unlikely to have contributed to the carcinogenicity results observed in male and female rats.
34 Choice of species/gender. The oral slope factor for 1,1,2,2-tetrachloroethane was
35 quantified using the tumor incidence data for female mice. The hepatocelluar carcinoma data in
36 male mice demonstrated tumorigenicity, but the data in male mice did not achieve adequate
37 model fit for any of the dichotomous models when BMD modeled. The male and female rat
38 tumor incidence data were not suitable for deriving low-dose quantitative risk estimates, and NCI
100 DRAFT - DO NOT CITE OR QUOTE
-------
1 described the rat strain as relatively insensitive to the carcinogenic effects of chlorinated organic
2 compounds.
3 Relevance to humans. The oral slope factor is derived from the incidence of
4 hepatocellular carcinomas in female mice. Using liver tumors in B6C3Fi mice as the model for
5 human carcinogenesis is a concern because of the prevalence of and susceptibility to developing
6 liver tumors in this strain of mice. Hasemen et al. (1998) reported an increased liver carcinoma rate
7 of 17.9 and 8.4% for male and female B6C3F1 mice, respectively, from NTP carcinogenicity feeding
8 bioassays, and a combined adenoma and carcinoma rate of 42 and 24% for male and female B6C3F1
9 mice, respectively. The B6C3F1 mouse was also used in the NCI (1978) study and may be
10 excessively sensitive to the development of hepatocellular tumors.
11 Additional interspecies uncertainty may arise from the rate of metabolism across species,
12 as it has demonstrated that mice have greater metabolic capacity following exposure to
13 tetrachloroethylene than rats and humans (Reitz et al., 1996). Reitz et al. (1996) demonstrated
14 that mice possessed a greater relative ability to metabolize tetrachloroethylene than rats and
15 humans, and, although data are not available, a similar situation may exist for 1,1,2,2-
16 tetrachloroethane.
17 In addition, the genotoxicity and mutagenicity studies provide limited evidence of a
18 mutagenic mode of action, with 1,1,2,2-tetrachloroethane displaying equivocal results of
19 mutagenic activity. In addition, there are inadequate data to support any mode of action
20 hypothesis.
21 Human population variability. The extent of inter-individual variability in animals for
22 1,1,2,2-tetrachloroethane metabolism has not been characterized. A separate issue is that the
23 human variability in response to 1,1,2,2-tetrachloroethane is also unknown. This lack of
24 understanding about potential differences in metabolism and susceptibility across exposed
25 animal and human populations thus represents a source of uncertainty.
26
27 5.4.6. Previous Cancer Assessment
28 In the previous IRIS assessment, posted to the IRIS database in 1987, 1,1,2,2-tetrachloro-
29 ethane was characterized as "Classification — C; possible human carcinogen" based on the
30 increased incidence of hepatocellular carcinomas in mice observed in the NCI (1978) bioassay
31 (U.S. EPA, 1987). An oral slope factor of 0.2 (mg/kg-day)"1 was derived using the increased
32 incidence of hepatocellular carcinomas in female mice (NCI, 1978) and a linear multistage
33 extrapolation method.
34
101 DRAFT - DO NOT CITE OR QUOTE
-------
1 6. MAJOR CONCLUSIONS IN THE CHARACTERIZATION OF HAZARD AND DOSE
2 RESPONSE
3
4
5 6.1. HUMAN HAZARD POTENTIAL
6 1,1,2,2-Tetrachloroethane (CAS No. 79-34-5) has been used as an insecticide, fumigant,
7 and weed killer (Hawley, 1981), although it presently is not registered for any of these purposes.
8 It was once used as an ingredient in an insect repellent, but registration was canceled in the late
9 1970s. In the past, the major use for 1,1,2,2-tetrachloroethane was in the production of
10 trichloroethylene, tetrachloroethylene, and 1,2-dichloroethylene (Archer, 1979). It was also used
11 as a solvent, in cleaning and degreasing metals, in paint removers, varnishes, and lacquers, in
12 photographic films, and as an extractant for oils and fats (Hawley, 1981). With the development
13 of new processes for manufacturing chlorinated ethylenes, the production of 1,1,2,2-tetrachloro-
14 ethane as a commercial end-product in the United States and Canada had steadily declined since
15 the late 1960s and had ceased by the early 1990s (HSDB, 2009; Environment Canada and Health
16 Canada, 1993). 1,1,2,2-Tetrachloroethane may still appear as a chemical intermediate in the
17 production of a variety of other common chemicals.
18 1,1,2,2-Tetrachloroethane is well absorbed from the respiratory and gastrointestinal
19 tracts, is rapidly and extensively metabolized, and is eliminated mainly as metabolites in the
20 urine and breath. Both reductive and oxidative metabolisms occur, producing reactive radical
21 and organochlorine intermediates, respectively. Trichloroethanol, trichloroacetic acid, and
22 dichloroacetic acid are initial metabolites that subsequently yield glyoxalic acid, oxalic acid, and
23 carbon dioxide.
24 A limited amount of information is available addressing the toxicity of 1,1,2,2-tetra-
25 chloroethane in humans. CNS depression was the predominant effect of high-dose acute oral
26 and inhalation exposures, although acute inhalation also caused irritation of the mucous
27 membranes. Occupational studies suggest that repeated exposure to 1,1,2,2-tetrachloroethane
28 can affect the liver and the nervous system.
29 Animal studies have established that the CNS and liver are the main targets of toxicity at
30 high levels of oral and inhalation exposures. Death in laboratory animals typically was preceded
31 by signs of CNS depression (e.g., lethargy, incoordination, loss of reflexes, depressed
32 respiration, prostration, and loss of consciousness), and postmortem examinations mainly
33 showed fatty degeneration in the liver. The most sensitive target of sublethal ingestion and
34 inhalation appears to be the liver, and short-term and subchronic exposures caused hepatic
35 effects that included serum chemistry changes, hepatocellular degeneration, and other
36 histopathological alterations. Comprehensive neurobehavioral testing in 14-week feeding studies
37 showed no effects in rats or mice, indicating that the liver was more sensitive than the nervous
38 system for subchronic oral exposure (Chan, 2004). A limited amount of information is available
102 DRAFT - DO NOT CITE OR QUOTE
-------
1 on other effects of 1,1,2,2-tetrachloroethane. Reduced body weight gain and weight loss were
2 effects of repeated oral exposures in rats and mice that generally occurred at high doses and may
3 have contributed to mild anemia and atrophy in the spleen, bone, bone marrow, and reproductive
4 tissues in these animals. Kidney lesions (acute toxic tubular necrosis and chronic inflammation)
5 occurred in mice that were chronically exposed to oral doses that also caused reduced survival.
6 Adequate immunological testing of 1,1,2,2-tetrachloroethane has not been performed.
7 The reproductive and developmental toxicity of 1,1,2,2-tetrachloroethane has not been
8 adequately evaluated. Significant decreases in maternal and fetal body weights were observed in
9 rats. In mice, litter resorption was observed along with high maternal mortality. Toxicity to
10 reproductive tissues following 1,1,2,2-tetrachloroethane exposure to adult rats and mice was
11 observed in F344 rats and B6C3Fi mice. Effects observed in rats and/or mice include:
12 decreased sperm and spermatozoal motility; decreased testis and epididymal weight; increased
13 atrophy of the preputial and prostate gland, seminal vesicle, testicular germinal epithelium,
14 uterus, and clitoral gland; ovarian interstitial cell cytoplasmic alterations; and lengthened estrus
15 cycle. Chronic low-level inhalation caused no effects on reproductive function in male mice, but
16 multigeneration or other tests of reproductive function in females have not been conducted for
17 any route of exposure. Developmental toxicity was assessed in rats and mice that were
18 gestationally exposed to 1,1,2,2-tetrachloroethane in the diet. These studies did not include
19 examinations for skeletal or visceral abnormalities, although effects that included reduced fetal
20 body weight gain in rats and litter resorptions in mice occurred at doses that were maternally
21 toxic.
22 The carcinogenicity of 1,1,2,2-tetrachloroethane was evaluated in a chronic gavage study
23 in rats and mice conducted by NCI (1978). Hepatocellular carcinomas were induced in male and
24 female mice, but there were no statistically significant increases in tumor incidences in the rats.
25 Three rare tumors in high dose male rats were noted. Thus, 1,1,2,2-tetrachloroethane is "likely
26 to be carcinogenic to humans" by any route of exposure, according to the Guidelines for
27 Carcinogen Risk Assessment (U. S. EPA, 2005a).
28
29 6.2. DOSE RESPONSE
30 6.2.1. Noncancer/Oral
31 The NTP (2004) study was selected as the principal study because it was a well-designed
32 subchronic dietary study, conducted in both sexes in two rodent species with a sufficient number
33 of animals per dose group. The number of test animals allocated among three dose levels and an
34 untreated control group was acceptable, with examination of appropriate toxicological endpoints
35 in both sexes of rats and mice. The liver was the most sensitive target in both species and the
36 rats were more sensitive than the mice. In addition to the observed liver weight increases, there
37 is evidence of hepatocellular effects, including increased serum liver enzyme levels and an
38 increased incidence of both hepatocellular cytoplasmic vacuolization and necrosis, from the
103 DRAFT - DO NOT CITE OR QUOTE
-------
1 subchronic NTP (2004) study. EPA selected increased liver weight as the critical effect because
2 this effect may represent an indicator of liver toxicity that occurs early in the process leading to
3 hepatocellular necrosis associated with subchronic oral exposure to 1,1,2,2-tetrachloroethane.
4 Potential PODs for a subchronic RfD were derived by BMD modeling of dose-response
5 data for increases in liver weight, increases in serum levels of ALT, SDH, and ALP, increased
6 levels of bile acids, and increased incidence of hepatocellular cytoplasmic vacuolization in rats.
7 All available dichotomous models in the EPA's BMDS (version 2.1.1) were fit to the incidence
8 data for hepatocellular cytoplasmic vacuolization, and all available continuous models in the
9 software were applied to the data for liver weight and serum enzyme levels, as well as the data
10 for rat fetal body weight. A BMR of 10% (10% extra risk above control) was selected for
11 derivation of the BMDL for hepatocellular cytoplasmic vacuolization in female rats, and a BMR
12 of 1 SD (a change in the mean equal to 1 SD from the control mean) was selected for the
13 derivation of the BMDL for the continuous female rat liver weight and rat fetal body weight data.
14 The BMDiso of 22 mg/kg-day and BMDLiso of 15 mg/kg-day based on the relative liver
15 weight effects seen in the female rat represents a reasonable POD for the derivation of the RfD.
16 To derive the subchronic RfD, the 15 mg/kg-day BMDLiso based on female rat relative liver
17 weight was divided by a total UF of 300, yielding a subchronic RfD of 0.05 mg/kg-day. The UF
18 of 300 comprises component factors of 10 for interspecies extrapolation, 10 for interhuman
19 variability, and 3 for database deficiencies.
20 The choice of BMD model is not expected to introduce a considerable amount of
21 uncertainty in the risk assessment since the chosen response rate of 1 SD is within the observable
22 range of the data. Additional BMD modeling for other amenable data sets, including serum liver
23 enzyme levels, liver lesions, and fetal body weight, were also conducted to provide other PODs
24 for comparison purposes (see Appendix B). A graphical representation of these potential PODs
25 and resulting subchronic reference values is shown below in Figure 6-1.
104 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
1000
100
(O
T3
0.1
0.01
0.001
If
If
POD
UFA
UFD
RID
hepatocellular relative liver weight- absolute liver weight increased ALT-o increased SDH-? increased bile acids - decreased fetal body
cytoplasnic ? rats (HIP, 2004) - o rats (HIP, 2004) rats (NTP,2004) rats (NTP, 2004) o rats (NTP, 2004) weight-rats (Gulati
vacuolization - o rats etal., 1991a)
(NTP, 2004)
Figure 6-1. PODs for selected endpoints (with critical effect circled) with corresponding applied UFs and
derived sample subchronic oral RfVs.
105
DRAFT - DO NOT CITE OR QUOTE
-------
1 The default UF of 10 for the extrapolation from animals and humans is a composite of
2 uncertainty to account for toxicokinetic differences and toxicodynamic differences between the
3 animal species in which the POD was derived and humans.
4 PBTK models can be useful for the evaluation of interspecies toxicokinetics; however,
5 information was unavailable to quantitatively assess toxicokinetic or toxicodynamic differences
6 between animals and humans and the potential variability in human susceptibility; thus, the
7 interspecies and intraspecies UFs of 10 were applied for a total UF of 100. Human variation may
8 be larger or smaller; however, 1,1,2,2-tetrachloroethane-specific data to examine the potential
9 magnitude of human variability of response are unknown.
10 In addition, a threefold database UF was applied due to the lack of information
11 addressing the potential reproductive toxicity associated with 1,1,2,2-tetrachloroethane.
12 Uncertainties associated with data gaps in the 1,1,2,2-tetrachloroethane database have been
13 identified, specifically, uncertainties associated with database deficiencies characterizing
14 reproductive toxicity associated with oral exposure to 1,1,2,2-tetrachloroethane. The developing
15 fetus may be a target of toxicity (Gulati et al, 1991a), and the absence of a study specifically
16 evaluating the full range of developmental toxicity represents an additional area of uncertainty or
17 gap in the database.
18 The overall confidence in this subchronic RfD assessment is medium. Confidence in the
19 principal study (NTP, 2004) is high. Confidence in the database is medium. Reflecting high
20 confidence in the principal study and medium confidence in the database, confidence in the
21 subchronic RfD is medium.
22 Information on the chronic oral toxicity of 1,1,2,2-tetrachloroethane consists of a limited
23 78-week gavage study in rats and mice (NCI, 1978). The high incidences of hepatocellular
24 tumors in all treated groups of mice precluded evaluation of noncancer effects in the liver and
25 identification of a NOAEL or LOAEL. Additionally, the NCI (1978) study performed
26 histological examinations on the animals when they died or at the termination of the study, which
27 was beyond the point at which more sensitive hepatotoxic effects, including nonneoplastic
28 effects, would be expected. The 14-week dietary study (NTP, 2004) was used to derive the
29 subchronic oral RfD. The NTP (2004) study also utilized a more relevant type of exposure (i.e.,
30 oral feeding) for the general population exposed to 1,1,2,2-tetrachloroethane in the environment.
31 The chronic RfD of 0.015 mg/kg-day was calculated by dividing the subchronic
32 BMDLiso of 15 mg/kg-day for increased relative liver weight by a total UF of 1,000: 10 for
33 interspecies extrapolation, 10 for interhuman variability, 3 for subchronic to chronic duration
34 extrapolation, and 3 for database deficiencies.
35 The choice of BMD model is not expected to introduce a considerable amount of
36 uncertainty in the risk assessment since the chosen BMR of 1 SD from the control mean is within
37 the observable range of the data. Additional BMD modeling for other amenable data sets,
38 including serum liver enzyme levels, liver lesions, and fetal body weight, were also conducted to
106 DRAFT - DO NOT CITE OR QUOTE
-------
1 provide other PODs for comparison purposes (see Appendix B). A graphical representation of
2 these potential PODs and resulting chronic reference values is shown below in Figure 6-2.
107 DRAFT - DO NOT CITE OR QUOTE
-------
1000
100
10
co
0.1
0.01
0.001
*\\\vo
A\\\V
— w 1 1 1 1 1 1
hepatocellular relative liver weight-absolute liver weight increased ALT -$ increased SDH-? increased bile acids - decreased fetal body
cytoplasnic ? rats (NTP, 2004) - $ rats (NTP, 2004) rats (NTP, 2004) rats (NTP, 2004) $ rats (NTP, 2004) weight-rats (Gulati
vacuolization - o rats etal., 1991 a)
(NTP, 2004)
POD
UFA
UFH
UFD
\\\\v
\\\\v
RfD
I
2
3
Figure 6-2. PODs for selected endpoints (with critical effect circled) from Table 5-3 with corresponding applied UFs
and derived sample subchronic oral RfVs.
108
DRAFT - DO NOT CITE OR QUOTE
-------
1 The default UF of 10 for the extrapolation from animals and humans is a composite of
2 uncertainty to account for toxicokinetic differences and toxicodynamic differences between the
3 animal species in which the POD was derived and humans.
4 PBTK models can be useful for the evaluation of interspecies toxicokinetics; however,
5 information was unavailable to quantitatively assess toxicokinetic or toxicodynamic differences
6 between animals and humans and the potential variability in human susceptibility, thus, the
7 interspecies and intraspecies UFs of 10 were applied for a total UF of 100. Human variation may
8 be larger or smaller; however, 1,1,2,2-tetrachloroethane-specific data to examine the potential
9 magnitude of human variability of response are unknown.
10 A threefold UF was applied for extrapolation from a subchronic exposure duration study
11 to a chronic RfD. Based on the available data for 1,1,2,2-tetrachloroethane, the toxicity observed
12 in the liver does not appear to increase over time. Specifically, liver damage observed in
13 F344 rats following subchronic exposure to 1,1,2,2-tetrachloroethane (NTP, 2004), e.g.,
14 increased incidence of necrosis or altered serum enzyme and bile levels, did not progress to more
15 severe effects such as cirrhosis or major liver disease following chronic exposures (NCI, 1978).
16 Therefore, the available database does not abrogate all concern associated with using a
17 subchronic study as the basis of the RfD but supports the utilization of a database UF of 3.
18 In addition, a threefold database UF was applied due to the lack of information
19 addressing the potential reproductive toxicity associated with 1,1,2,2-tetrachloroethane.
20 Uncertainties associated with data gaps in the 1,1,2,2-tetrachloroethane database have been
21 identified, specifically, uncertainties associated with database deficiencies characterizing
22 reproductive toxicity associated with oral exposure to 1,1,2,2-tetrachloroethane. The developing
23 fetus may be a target of toxicity (Gulati et al, 199la), and the absence of a study specifically
24 evaluating the full range of developmental toxicity represents an additional area of uncertainty or
25 gap in the database.
26 The overall confidence in this chronic RfD assessment is medium. Confidence in the
27 principal study (NTP, 2004) is high. Confidence in the database is medium. Reflecting high
28 confidence in the principal study and medium confidence in the database, confidence in the
29 chronic RfD is medium.
30
31 6.2.2. Noncancer/Inhalation
32 An RfC was not calculated due to insufficient data. Information on the subchronic and
33 chronic inhalation toxicity of 1,1,2,2-tetrachloroethane is limited to the results of one study in
34 rats that found transient liver effects (Truffert et al., 1977). Reporting inadequacies preclude
35 identification of a NOAEL or LOAEL and derivation of an RfC in the usual manner.
36 A route-to-route extrapolation using the computational technique of Chiu and White
37 (2006), as described in Section 3.5, was considered. However, U.S. EPA (1994b) recommends
38 not conducting a route-to-route extrapolation from oral data when a first-pass effect by the liver
109 DRAFT - DO NOT CITE OR QUOTE
-------
1 or respiratory tract is expected, or a potential for portal-of-entry effects in the respiratory tract is
2 indicated following analysis of short-term inhalation, dermal irritation, in vitro studies, or
3 evaluation of the physical properties of the chemical. In the case of 1,1,2,2-tetrachloroethane, a
4 first-pass effect by the liver is expected. In addition, the presence of tissue-bound metabolites in
5 the epithelial linings in the upper respiratory tract may demonstrate a first-pass effect by the
6 respiratory tract (Eriksson and Brittebo, 1991). Lehmann et al. (1936) observed irritation of the
7 mucous membranes of two humans following exposure to 1,1,2,2-tetrachloroethane air
8 concentrations of 146 ppm (1,003 mg/m ) for 30 minutes or 336 ppm (2,308 mg/m ) for
9 10 minutes, indicating the potential for portal-of-entry effects in the respiratory system.
10 Information regarding the chronic inhalation toxicity of 1,1,2,2-tetrachloroethane is
11 available from four animal studies that include limited data on liver effects and serum
12 acetylcholinesterase, hematological, and immunological alterations (Shmuter, 1977; Kulinskaya
13 and Verlinskaya, 1972; Schmidt et al., 1972; Mellon Institute of Industrial Research, 1947).
14 However, the reporting of results from these chronic bioassays is inadequate for identification of
15 NOAELs or LOAELs for systemic toxicity. A chronic NOAEL was identified for reproductive
16 effects in male rats (Schmidt et al., 1972); however, macroscopic malformations or significant
17 group differences in the other indices were not observed at 13.3 mg/m3. This lack of information
18 on reproductive toxicity precludes utilizing this selected NOAEL in the derivation of an RfC.
19
20 6.2.3. Cancer/Oral and Inhalation
21 Under the Guidelines for Carcinogen Risk Assessment (U. S. EPA, 2005a), 1,1,2,2-tetra-
22 chloroethane is characterized as "likely to be carcinogenic to humans", based on the existence of
23 evidence of the compound's tumorigenicity in a single study in a single animal species (NCI,
24 1978) and the induction of hepatocellular carcinomas in both rats and mice by the main
25 metabolite, 1,2-dichloroacetic acid (U.S. EPA, 2003). The epidemiological human data available
26 are inadequate for evaluation of cancer risk (IARC, 1999). The NCI (1978) provided evidence
27 that 1,1,2,2-tetrachloroethane causes hepatocellular tumors in male and female mice. A few,
28 statistically nonsignificant, rare tumors were seen in high-dose male rats (NCI, 1978). The NCI
29 concluded that 1,1,2,2-tetrachloroethane causes cancer in mice.
30 The only carcinogenicity bioassay for 1,1,2,2-tetrachloroethane was a chronic gavage
31 study in Osborne-Mendel rats and B6C3Fi mice performed by NCI (1978). This was a well-
32 designed study, conducted in both sexes in two rodent species with an adequate number of
33 animals per dose group and with examination of appropriate toxicological endpoints in both
34 sexes of rats and mice. The rat study found no statistically significant increases in tumor
35 incidences in males or females. Three rare hepatocellular tumors in high-dose male rats were
36 noted and NCI (1978) characterized the carcinogenic results in male rats as "equivocal." The
37 mouse study found significant, dose-related increases in the incidences of hepatocellular
38 carcinomas in both sexes. Based on the increased incidences of hepatocellular carcinomas, NCI
110 DRAFT - DO NOT CITE OR QUOTE
-------
1 (1978) concluded that orally administered 1,1,2,2-tetrachloroethane is a liver carcinogen in male
2 and female B6C3Fi mice. This NCI study was used for dose-response analysis for oral exposure.
3 Data on the incidences of hepatocellular carcinomas in male and female mice from the
4 NCI (1978) study were used for cancer dose-response assessment. Conversion of the doses in
5 the NCI (1978) mouse study to HEDs to be used for dose-response modeling was accomplished
6 in two steps. The mice were treated with 1,1,2,2-tetrachloroethane by gavage 5 days/week for
7 78 weeks, and then observed untreated for 12 weeks for a total study duration of 90 weeks.
8 Because the reported TWA doses were doses for 5 days/week for 78 weeks, they were duration-
9 adjusted to account for the partial week exposure (by multiplying by 5 days/7 days) and
10 untreated observation period (by multiplying by 78 weeks/90 weeks). The duration-adjusted
11 animal doses were converted to HEDs by adjusting for differences in body weight and lifespan
12 between humans and mice. In accordance with U.S. EPA (2005a) Guidelines for Carcinogen
13 Risk Assessment, a factor of BW3 4 was used for cross-species scaling. Because the study
14 duration (90 weeks) was less than the animal lifespan (104 weeks), the scaled dose was then
15 multiplied by the cubed ratio of experimental duration to animal lifespan to complete the
16 extrapolation to a lifetime exposure in humans.
17 The mode of action of 1,1,2,2-tetrachloroethane carcinogenicity is unknown. It appears
18 that metabolism to one or more active compounds is likely to play a role in the development of
19 the observed liver tumors, but insufficient data preclude proposing this as a mode of action.
20 Results of genotoxicity and mutagenicity studies of 1,1,2,2-tetrachloroethane are mixed and
21 insufficient for informing the mode of action. Given that the mechanistic and other information
22 available on cancer risk from exposure to 1,1,2,2-tetrachloroethane is sparse and that the data
23 that does exist is equivocal, there is inadequate information to inform the low dose extrapolation.
24 Dose-response modeling was performed to obtain a POD for quantitative assessment of
25 cancer risk. The incidences of hepatocellular carcinomas in both sexes of mice were modeled for
26 determination of the POD. In accordance with the U.S. EPA (2005a) cancer guidelines, the
27 BMDLio (lower bound on dose estimated to produce a 10% increase in tumor incidence over
28 background) was estimated by applying the multistage cancer model in the the EPA's BMDS
29 (version 2.1.1) for the dichotomous incidence data and selecting the results for the model that
30 best fits the data. The BMD modeling of the male mouse data did not achieve adequate fit for
31 any of the dichotomous models; thus, a cancer slope factor was not derived from the male data.
32 The 1° multistage model was selected for the derivation of the cancer slope factor from the
33 female data because this model provided adequate model fit and the lowest AIC when compared
34 to the results of the 2° multistage model. In addition, the 2° multistage model had insufficient
35 degrees of freedom to test the goodness-of-fit. The BMDLio of 0.65 mg/kg-day from the
36 modeling of the tumor incidence data in female mice is selected as the POD for use in
37 calculation of an oral slope factor. Details of the BMD modeling are presented in Appendix C.
111 DRAFT - DO NOT CITE OR QUOTE
-------
1 In accordance with the U. S. EPA (2005a) guidelines, an oral slope factor of 0.15 (mg/kg-
2 day)"1 is calculated by dividing the human equivalent BMDLio of 0.65 mg/kg-day into 0.1 (10%)
3 (Appendix C).
4 In the absence of any data on the carcinogenicity of 1,1,2,2-tetrachloroethane via the
5 inhalation route, an inhalation unit risk has not been derived in this evaluation.
6
7
112 DRAFT - DO NOT CITE OR QUOTE
-------
7. REFERENCES
Amoore, JE; Hautala E. (1983) Odor as an aid to chemical safety: odor thresholds compared with threshold limit
values and volatilities for 214 industrial chemicals in air and water dilution. J Appl Toxicol 3(6):272-290.
Anderson, WB; Board, PG; Gargano, B; et al. (1999) Inactivation of glutathione transferase zeta by dichloroacetic
acid and other fluorine-lacking a-haloalkanoic acids. Chem Res Toxicol 12:1144-1149.
Andrews, JE; Nichols, H; Hunter, ES. (2002) Developmental toxicity of di- and tetrachloroethane and
dichloropropane in the rat whole embryo culture system. Toxicologist 66(1-S):23.
Archer, WL. (1979) Other chloroethanes. In: Grayson, H; Eckroth, D; eds. Kirk-Othmer encyclopedia of chemical
technology. New York, NY: John Wiley & Sons, 3rd edition, Vol 5, pp. 722-742.
Arthur Little, Inc. (1983) Cell transformation assays of 11 chlorinated hydrocarbon analogs. Final report. ICAIR
Work Assignment No. 10. EPA Document No. 40-8324457; NTIS No. OTS0509392.
ASTER (Assessment Tools for the Evaluation of Risk). (1995) ASTER ecotoxicity profile. U.S. Environmental
Protection Agency, Environmental Research Laboratory, Duluth, MN.
AT SDR (Agency for Toxic Substances and Disease Registry). (1996) Toxicological profile for
1,1,2,2-tetrachloroethane (update). U.S. Department of Health and Human Services, Public Health Service.
Blackburn, AC; Coggan, M; Tzeng, HF; et al. (2001) GSTZld: a new allele of glutathione transferase zeta and
maleylacetoacetate isomerase. Pharmacogenetics 11:671-678.
Blackburn, AC; Tzeng, HF; Anders, MW; et al. (2000) Discovery of a functional polymorphism in human
glutathione transferase zeta by expressed sequence tag database analysis. Pharmacogenetics 10:49-57.
Brem, H; Smith, AB; Rosenkranz, HS. (1974) The mutagenicity and DNB-modifying effect of haloalkanes. Cancer
Res 34:2576-2579.
Bull, RJ; Sanchez, IM; Nelson, MA; et al. (1990) Liver tumor induction in B6C3F1 mice by dichloroacetate and
trichloroacetate. Toxicology 63:341-359.
Callen, DF; Wolf, CR; Richard, MP. (1980) Cytochrome P-450 mediated genetic activity and cytotoxicity of seven
halogenated aliphatic hydrocarbons in Saccharomyces cerevisiae. Mutat Res 77:55-63.
Carpenter, CP; Smyth, HF, Pozzani, UC. (1949) The assay of acute vapor toxicity, and the grading and
interpretation of results on 96 chemical compounds. J Ind Hyg Toxicol 31:343-346.
CAS. (Chemical Abstracts Service). (1994). CA Registry File. Computer printout.
Casciola, LAF; Ivanetich, KM. (1984) Metabolism of chloroethanes by rat liver nuclear cytochrome P-450.
Carcinogenesis 5:543-548.
Chan, PC. (2004) NTP technical report on the toxicity studies of 1,1,2,2-tetrachloroethane (CAS No. 79-34-5)
administered in microcapsules in feed to F344/N rats and B6C3F1 mice. Toxic Rep Ser (49):6-Fl 1.
Chiou, CT; Peters, LJ; Freed, VH. (1979) A physical concept of soil-water equilibria for nonionic organic
compounds. Science 206:831-832.
Chiu, WA; 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.
Colacci, A; Grilli, S; Lattanzi, G; et al. (1987) The covalent binding of 1,1,2,2-tetrachloroethane to macromolecules
of rat and mouse organs. Teratog Carcinog Mutagen 7:465-474.
113 DRAFT - DO NOT CITE OR QUOTE
-------
Colacci, A; Perocco, P; Vaccari, M; et al. (1990) In vitro transformation of BALB/c 3T3 cells by
1,1,2,2-tetrachloroethane. Jpn J Cancer Res 81:786-792.
Colacci, A; Perocco, P; Bartoli, S; et al. (1992) Initiating activity of 1,1,2,2-tetrachloroethane in two-stage BALB/c
3T3 cell transformation. Cancer Lett 64:145-153.
Colacci, A; Vaccari, M; Perocco, P; et al. (1996) Enhancement of BALB/c 3T3 cells transformation by
1,2-dibromoethane promoting effect. Carcinogenesis 17:225-231.
Cottalasso, D; Bellocchio, A; Domenicotti, C; et al. (1998) 1,1,2,2-Tetrachloroethane-induced early decrease of
dolichol levels in rat liver microsomes and Golgi apparatus. J Toxicol Environ Health 54:133-144.
Coyer, HA. (1944) Tetrachloroethane poisoning. Ind Med 13:230-233.
Crebelli, R; Franekic, J; Conti, G; et al. (1988) Induction of chromosome malsegregation by halogenated organic
solvents in Aspergillus nidulans: unspecific or specific mechanism? Mutat Res 201:401-411.
Daniel, FB; DeAngelo, AB; Stober, JA; et al. (1992) Hepatocarcinogenicity of chloral hydrate,
2-chloroacetaldehyde, and dichloroacetic acid in male B6C3F1 mouse. Fundam Appl Toxicol 19:159-168.
DeAngelo, AB; Daniel, FB; Stober, JA; et al. (1991) The carcinogenicity of dichloroacetic acid in the male B6C3F1
mouse. Fundam Appl Toxicol 16:337-347.
DeAngelo, AB; Daniel, FB; Most, BM; et al. (1996) The carcinogenicity of dichloroacetic acid in the male Fischer
344 rat. Toxicology 114:207-221.
DeAngelo, AB; George, MH; House, DE. (1999) Hepatocarcinogenicity in the male B6C3F1 mouse following a
life-time exposure to dichloroacetic acid in the drinking water: dose-response determination and modes of action.
J Toxicol Environ Health 58:485-507.
Deguchi, T. (1972) A fundamental study of the threshold limit values for solvent mixtures in the air. OsakB-shiritsu
Daigaku Igaku Zasshi 21:187-209.
DeMarini, DM; Brooks, HG. (1992) Induction of prophage lambda by chlorinated organics: detection of some
single-species/single-site carcinogens. Environ Mol Mutagen 19:98-111.
Dow Chemical Company. (1988) The metabolism and hepatic micromolecular interactions of
1,1,2,2-tetrachloroethane (TCE) in mice and rats. Document D002628.
Elliott, JM. (1933) Report of a fatal case of poisoning by tetrachloroethane. JR Army Med Corps 60:373-374
Environment Canada and Health Canada. (1993). Canada environmental protection act. Priority substances list
assessment report. 1,1,2,2,-Tetrachlorethane. Ottawa, Canada.
Eriksson, C; Brittebo, EB. (1991) Epithelial binding of 1,1,2,2-tetrachloroethane in the respiratory and upper
alimentary tract. Arch Toxicol 65:10-14.
Ferreira-Gonzalez, A; DeAngelo, AB; Nasim, S; et al. (1995) Ras oncogene activation during hepatocarcinogenesis
in B6C3F male mice by dichloroacetic and trichloroacetic acid. Carcinogenesis 16:495-500.
Flick, EW. (1985) Industrial solvents handbook. Park Ridge, NJ: Noyes Publication, p. 134.
Forbes, G. (1943) Tetrachloroethane poisoning. Br Med J 1:348-350.
Galloway, SM; Armstrong, MJ; Reuben, C; et al. (1987) Chromosome aberrations and sister chromatid exchanges in
Chinese hamster ovary cells: evaluations of 108 chemicals. Environ Mol Mutagen 10:1-175.
114 DRAFT - DO NOT CITE OR QUOTE
-------
Gargas, ML; Anderson, ME. (1989) Determining kinetic constants of chlorinated metabolism in the rat from rates of
exhalation. Toxicol Appl Pharmacol 99:344-353.
Gargas, ML; Burgess, RJ; Voisard, DE; et al. (1989) Partition coefficients of low-molecular-weight volatile
chemicals in various liquids and tissues. Toxicol Appl Pharmacol 98:87-99.
Gohlke, R; Schmidt, P. (1972) Subacute action of low concentrations of chlorinated ethane with and without
additional ethanol treatment in the rat. Int Arch Arbeitsmed 30:299-312.
Gohlke, R; Schmidt, P; Bahmann, H. (1977) 1,1,2,2-Tetrachloroethane and heat stress in animal experiment.
Morphological results. Z Gesamte Hyg 20: 278-282. (German)
Grasso, P. (2002) Essentials of pathology for toxicologists. Taylor and Francis, London.
Gulati, DK; Grimes, LK; Smith, MR; et al. (199la) Range finding studies: developmental toxicity.
1,1,2,2-Tetrachloroethane (repeat) then administered via feed in CD Sprague-Dawley rats. Study No: NTP-91-
RF/DT-017.
Gulati, DK; Grimes, LK; Smith, MR; et al. (1991b) Range finding studies: developmental toxicity.
1,1,2,2-Tetrachloroethane (repeat) then administered via feed in Swiss CD-I mice. Study No: NTP-91-RF/DT-020.
Halpert, J. (1982) Cytochrome P-450 dependent covalent binding of 1,1,2,2-tetrachloroethane in vitro. Drug Metab
Dispos 10:465-468.
Halpert, J; Neal, RA. (1981) Cytochrome P-450-dependent metabolism of 1,1,2,2-tetrachloroethane in
dichloroacetic acid in vitro. Biochem Pharmacol 30:1366-1368.
Halpert, JR; Balfour, C; Miller, NE; et al. (1986) Dichloromethyl compounds as mechanism-based inactivators of rat
liver cyrochromes P-450 in vitro. Mol Pharmacol 30:19-24.
Hamilton, A. (1917) Military medicine and surgery. JAMA 69:2037-2039.
Hansch, C; Leo, AJ. (1985) Medchem project. Issue no. 26. Claremont, CA: Pomona College.
Harris, RZ; Benet, LZ; Schwartz, JB. (1995) Gender effects in pharmacokinetics and pharmacodynamics. Drugs
50(2):222-239.
Haseman, JK; Hailey, JR; Morris, RW. (1998) Spontaneous neoplasm incidences in Fischer 344 rats and B6C3F1
mice in two-year carcinogenicity studies: a National Toxicology Program update. Toxicol Pathol 26(3):428-441.
Hawley, GG. (1981) Condensed chemical dictionary. New York, NY: VanNostrand Reinhold, p. 1003.
Haworth, S; Lawlor, T; Mortelmans, K; et al. (1983) Salmonella mutagenicity test results for 250 chemicals.
Environ Mutag 5(Suppl 1):3-142.
Hayes, W. (2001) Principles and methods of toxicology. 4th edition. Philadelphia, PA: Taylor and Francis.
Henics, T; Wheatley, DN. (1999) Cytoplasmic vacuolation, adaptation and cell death: a view on new perspectives
and features. Biol Cell. 91:485-498.
Hepple, RA. (1927) An unusual case of poisoning. J R Army Med Corps 49:442-445.
Herren-Freund, SL; Pereira, MA; Khoury, DK; et al. (1987) The carcinogenicity of trichloroethylene and its
metabolites, trichloroacetic acid and dichloroacetic acid, in mouse liver. Toxicol Appl Pharm 90:183-189.
Horiguchi, S; Morioka, S; Utsunomiya, T; et al. (1964) A survey of the actual conditions of artificial pearl factories
with special reference to work using tetrachloroethane. Jpn J Ind Health 6:251-256.
115 DRAFT - DO NOT CITE OR QUOTE
-------
Horiuchi, K; Horiguchi, S; Hashimoto, K; et al. (1962) Studies on the industrial tetrachloroethane poisoning. Osaka
City Med 18:29-38.
Horvath, M; Frantik, E. (1973) To the relative sensitivity of nervous functions and behavior to nonspecific effects of
foreign substances. Act Nerv Super (Praha) 15:25-27.
HSDB (Hazardous substance data bank). (2009) National Library of Medicine, Bethesda, MD.
http://toxnet.nlm.nih.gov/cgi-bin/sis/search (August 10, 2009).
IARC (International Agency for Research on Cancer). (1999) 1,1,2,2-Tetrachloroethane. IARC Monogr Eval
Carcmog Risks Hum 71(Pt 2):817-827.
Ikeda, M; Ohtsuji, H. (1972) A comparative study of the excretion of Fujiwara reaction-positive substances in urine
of humans and rodents given trichloro- or tetrachloro-derivatives of ethane and ethylene. Br J Ind Med 29:99-104.
Jeney, E; Bartha, F; Kondor, L; et al. (1957) Prevention of industrial tetrachloroethane intoxication—Part III.
Egeszsegtudomany 1:142-164. (Hungarian)
Koizumi, A; Kumai, M; Ikeda, M. (1982) Enzymatic formation of an olefin in the metabolism of
1,1,2,2-tetrachloroethane: an in vitro study. Bull Environ Contam Toxicol 29:562-565.
Kulinskaya, IL; Verlinskaya, RV. (1972) Comparative effect of low concentrations of di-, tetra-, and
pentachloroethane on the blood acetylcholine system. Gig Tr Prof Zabol 16:56-58. (Russian)
Kunkel, GH; Hoagland, CL. (1947) Mechanism and significance of the thymol turbidity test for liver disease. J Clin
Invest 26(6): 1060-1071.
Lantum, HB; Board, PG; Anders, MW. (2002) Kinetics of the biotransformation of maleylacetone and
chlorofluoroacetic acid by polymorphic variants of human glutathione transferase zeta (hGSTZl-1). Chem Res
Toxicol 15:957-963.
Lehmann, KB; Schmidt-Kehl, L; Ruf, H; et al. (1936) The thirteen most important chlorinated aliphatic
hydrocarbons from the standpoint of industrial hygiene. Arch Hyg 116:132-200.
Lide, DR, ed. (1993) CRC handbook of chemistry and physics, Boca Raton, FL: CRC Press Inc., pp. 4-39.
Lilliman, B. (1949) Suggested mechanism of poisoning by liquid tetrachloroethane. Analyst 74:510-511.
Lobo-Mendonca, R. (1963) Tetrachloroethane -a survey. Br J Ind Med 20:51-56
Mackay, D; Shiu, WY. (1981) A critical review of Henry's Law constants for chemicals of environmental interest. J
Phys Chem Ref Data 10:1175-1199.
Mant, AK. (1953) Acute tetrachloroethane poisoning: a report on two fatal cases. Br Med J 1:655-656.
Maronpot, RR; Shimkin, MB; Witschi, HP; et al. (1986) Strain a mouse pulmonary tumor test results for chemicals
previously tested in the national cancer institute carcinogenicity tests. J Natl Cancer Inst 76:1101-1112.
McGregor, DB. (1980) Tier II mutagenic screening of 13 NIOSH priority compounds, individual compound report,
1,1,2,2-tetrachloroethane . Inveresk Research International Limited, Musselburgh EH21 7UB Scotland NIOSH,
Cincinnati, OH. , Report No 26.
McKim, JM; Lien, GJ; Hoffman, AD; et al. (1999) Respiratory-cardiovascular physiology and xenobiotic gill flux in
the lake trout (Salvelinus namaycush). Comp Biochem Physiol A Mol Integr Physiol 123:69-81.
Mellon Institute of Industrial Research. (1947) Repeated exposure of rats and dogs to vapors of eight chlorinated
hydrocarbons. Union Carbide Corp. Submitted under TSCA Section 8D; EPA Document No. 86-870001397; NTIS
No. OTS0515559.
116 DRAFT - DO NOT CITE OR QUOTE
-------
Meulenberg, CJW; Vijverberg, HPM. (2000) Empirical relations predicting human and rat tissue:air partition
coefficients of volatile organic compounds. Toxicol Appl Pharmacol 165:206-216.
Meulenberg, CJ; Wijnker, AG; Vijverberg, HPM. (2003) Relationship between olive oil:air, saline:air, and rat
brain:air partition coefficients of organic solvents in vitro. J Toxicol Environ Health, Part A 66:1985-1998.
Milman, HA; Story, DL; Riccio, ES; et al. (1988) Rat liver foci and in vitro assays to detect initiating and promoting
effects of chlorinated ethanes and ethylenes. Ann NY Acad Sci 534:521-530.
Minot, GR; Smith, LW. (1921) The blood in tetrachloroethane poisoning. Arch Intern Med 28:687-702.
Mirsalis, JC; Tyson, CK; Steinmetz, KL; et al. (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.
Mitoma, C; Tyson, CA; Riccio, ES. (1984) Investigations of the species sensitivity and mechanism of
carcinogenicity of halogenated hydrocarbons. Standford Research Institute International, Menlo Park, CA;
Submitted under TSCA Section 4; EPA Document No. 40-8424225; NTIS No. OTS0509408.
Mitoma, C; Steeger, T; Jackson, SE; et al. (1985) Metabolic disposition study of chlorinated hydrocarbons in rats
and mice. Drug Chem Toxicol 8:183-194.
Miyagawa, M; Takasawa, H; Sugiyama, A; et al. (1995) The in vivo-in vitro replicative DNA synthesis (RDS) test
with hepatocytes prepared from male B6C3F1 mice as an early prediction assay for putative nongenotoxic (Ames-
negative) mouse hepatocarcinogens. Mutat Res 343:157-183.
Morgan, A; Black, A; Belcher, DR. (1970) The excretion in breath of some aliphatic halogenated hydrocarbons
following administration by inhalation. Ann Occup Hyg 13:219-233.
Mugford, CA; Kedderis, GL. (1998) Sex-dependent metabolism of xenobiotics. Drug Metab Rev 30:441-498.
NCI (National Cancer Institute). (1976) Carcinogenesis bioassay of trichloroethylene. Natl Cancer Inst Carcinogen
Tech Rep SerNo. 2; NCI-CG-TR-2. NIH Publication No. 76-802.
NCI (National Cancer Institute). (1977) Bioassay of tetrachloroethylene for possible carcinogenicity. Natl Cancer
Inst Tech Rep SerNo. 13; NCI-CG-TR-13. NIH Publication No. 77-813.
NCI (National Cancer Institute). (1978) Bioassay of 1,1,2,2-tetrachloroethane for possible carcinogenicity. Natl
Cancer Inst Carcmog Tech Rep Ser 27:1-86.:l-86; NIH Publication No. 78-827. PB2774537GA.
NRC (National Research Council). (1983) Risk assessment in the federal government: managing the process.
Washington, DC: National Academy Press.
NTP (National Toxicology Program). (1986) NTP technical report on the toxicology and carcinogenesis
tetrachloroethylene (perchloroethylene) (CAS No. 1127-18-4) in F344/N and B6C3F1 mice (inhalation studies).
U.S. DHHS, Public Health Service, National Institute of Health. Technical Report Series, Number 311; NIH
Publication No. 86-2567.
NTP (National Toxicology Program). (1990) Carcinogenesis studies of trichloroethylene (without epichlorohydrin)
(CAS No. 79-01-6) in F344/N rats and B6C3F1 mice (gavage studies). U.S. DHHS, Public Health Service, National
Institute of Health. Technical Report Series No. 243; NIH Publication No. 90-1779.
NTP. (National Toxicology Program). (1996) NTP technical report on renal toxicity studies of selected halogenated
ethanes administered by gavage to F344/Nrats. U.S. DHHS, Public Health Service, National Institute of Health.
Toxicity Report Series, Number 45.
NTP. (National Toxicology Program). (2004) NTP technical report on the toxicity studies of
1,1,2,2-tetrachloroethane administered in microcapsules in feed to F344/N rats and B6C3F1 mice. U.S. DHHS,
Public Health Service, National Institute of Health. Toxicity Report Series, Number 49.
117 DRAFT - DO NOT CITE OR QUOTE
-------
Nebert, DW; Adesnik, M; Coon, MJ; et al. (1987) The P450 gene superfamily: recommended nomenclature. DNA
6:1-11.
Nestmann, ER; Lee, EGH. (1983) Mutagenicity of constituents of pulp and paper mill effluent in growing cells of
Saccharomyces cerevisiae. Mutat Res 119:273-280.
Nestmann, ER; Lee, EG-H; Matula, TI; et al. (1980) Mutagenicity of constituents identified in pulp and paper mill
effluents using the salmonella/mammalian-microsome assay. Mutat Res 79:203-212.
Nichols, JW; Mckim, JM; Lien, GJ; et al. (1993) Physiologically-based toxicokinetic modeling of three waterborne
chloroethanes in channel catfish, Ictaluruspunctatus. Aqua Toxicol 27:83-111.
Norman, JE; Robinette, CD; Fraumeni, IF. (1981) The mortality experience of army World War II chemical
processing companies. J Occup Med 23:818-822.
Omiecinski, CJ; Remmel, RP; Hosagrahara, VP. (1999) Concise review of the cytochrome P450s and their roles in
toxicology. Toxicol Sci 48:151-156.
Ono, Y; Kobayashi, U; Somiya, I; et al. (1996) Evaluation of DNA damage by active oxygen induced by
organochlorine compounds and nitroarenes. Mizu Kankyo Gakkaishi 19:871-877.
Osol, A. (1972) Blakiston's gould medical dictionary. 3rd edition. New York, NY: McGraw-Hill, Inc., p. 1828.
Pantelitsch, M. (1933) Experiments concerning the effect of chlorinated methane and ethane on mice: the relative
sensitivity of mice and cats to poisons. Wurzburg: Julius Maximilian University, pp. 1-13.
Paolini, M; Sapigni, E; Mesirca, R; et al. (1992) On the hepatotoxicity of 1,1,2,2-tetrachloroethane. Toxicology
73:101-115.
Parmenter, DC. (1921) Tetrachloroethane poisoning and its prevention. J Ind Hyg 2:456-465.
Pereira, MA. (1996) Carcinogenic activity of dichloroacetic acid and trichloroacetic acid in the liver of female
B6C3F1 mice. Fundam Appl Toxicol 31:192-199.
Pereira, MA; Phelps, JB. (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.
Plaa, GL; Hewitt, WR. (1998) Toxicology of the liver. 2nd edition. Washington, DC: Taylor and Francis, p. 431.
Price, NH; Allen, SD; Daniels, AU; et al. (1978) Toxicity data for establishing "immediately dangerous to life or
health" (IDLH) values. Cincinnati, OH: National Institute for Occupational Safety andHealth; PB87163531.
Reitz, RH; Gargas, ML; Mendrala, AL; et al. (1996) In vivo and in vitro studies of perchloroethylene metabolism
for physiologically based pharmacokinetic modeling in rats, mice, and humans. Toxicol and App Pharm 136:289-
306.
Richmond, RE; Carter, JH; Carter, HW; et al. (1995) Immunohistochemical analysis of dichloroacetic acid (DCA)-
induced hepatocarcinogenesis in male Fischer (F344) rats. Cancer Lett 92:67-76.
Riddick, JA; Bunger, WB; Sakano, TK. (1986) 1,1,1-Trichloroethane. In: Organic solvents. Physical properties and
methods of purification. New York, NY: John Wiley and Sons; pp. 358-359.
Robbins, SL; Angell M. (1976) Disease at the cellular level. In: Basic pathology, 2nd edition. Philadelphia, PA:
Saunders; pp. 3-30.
Roldan-Arjona, T; GarciB-Pedrajas, MD; Luque-Romero, FL. (1991) An association between mutagenicity of the
Ara test of Salmonella typhimurium and carcinogenicity in rodents for 16 halogenated aliphatic hydrocarbons.
Mutagenesis 6:199-205.
118 DRAFT - DO NOT CITE OR QUOTE
-------
Rosenkranz, HS. (1977) Mutagenicity of halogenated alkanes and their derivatives. Environ Health Perspect 21:79-
84.
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.
Sato, A; Nakajima, T. (1987) Pharmacokinetics of organic solvent vapors in relation to their toxicity. Scand J Work
Environ Health 13:81-93.
Schmidt, P; Binnewies, S; Gohlke, R; et al. (1972) Subacute action of low concentrations of chlorinated ethanes on
rats with and without additional ethanol treatment. 1. Biochemical and toxicometric aspects, especially results in
subacute and chronic toxicity studies with 1,1,2,2-tetrachloroethane. Int Arch Arbeitsmed 30:283-298. (German)
Schmidt, P; Burck, D; Buerger, A; et al. (1980a) On the hepatotoxicity of benzene, 1,1,2,2-tetrachloroethane and
carbon tetrachloride. Z Gesamte Hyg 26:167-172. (German)
Schmidt, P; Gohlke, R; Just, A; et al. (1980b) Combined action of hepatotoxic substances and increased
environmental temperature on the liver of rats. J Hyg Epidemiol Microbiol Immunol 24:271-277.
Sherman, JB. (1953) Eight cases of acute tetrachloroethane poisoning. J Trop Med Hyg 56:139-140.
Shimada, T; Yamazaki, H; Mimura, M; et al. (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.
Shmuter, LM. (1977) The effect of chronic exposure to low concentrations of ethane series chlorinated
hydrocarbons on specific and nonspecific immunological reactivity in animal experiments. Gig Tr Prof Zabol
20:38-43.
Smyth, HF; Carpenter, CP; Weil, CS; et al. (1969) Range-finding toxicity data: list VII. Am Ind Hyg Assoc J
30:470-476
Story, DL; Meierhenry, EF; Tyson, CA; et al. (1986) Differences in rat liver enzyme-altered foci produced by
chlorinated aliphatics and phenobarbital. Toxicol Ind Health 2:351-362.
Theiss, JC; Stoner, GD; Shimkin, MB; et al. (1977) Test for carcinogenicity of organic contaminants of United
States drinking waters by pulmonary tumor response in strain A mice. Cancer Res 37(8):2717-2720.
Tomasi, A; Albano, E; Bini, A; et al. (1984) Free radical intermediates under hypoxic conditions in the metabolism
of halogenated carcinogens. Toxicol Pathol 12:240-246.
Tomokuni, K. (1969) Studies on hepatotoxicity induced by chlorinated hydrocarbons. Lipid and ATP metabolisms
in the liver of mice exposed to 1,1,2,2-tetrachloroethane. Acta Med Okayama 23:273-282.
Tomokuni, K. (1970) Hepatotoxicity induced by chlorinated hydrocarbons. II. Lipid metabolism and absorption
spectrum of microsomal lipids in the mice exposed to 1,1,2,2-tetrachloroethane. Acta Med Okayama 24:315-322.
TOXNET (Toxicology Data Network). (2009). National Library of Medicine, National Institutes of Health, U.S.
Department of Health and Human Services, Bethesda, MD. Available online at http://toxnet.nlm.nih.gov (accessed
August 4, 2009).
Truffert, L; Girard-Wallon, C; Emmerich, E; et al. (1977) Early experimental demonstration of the hepatotoxicity of
some chlorinated solvents by the study of the synthesis of hepatic DNA. Arch Mai Prof 38:261-263.
TSI Mason Laboratories. (1993a) 14 Day pilot gavage toxicity study of 1,1,2,2-tetrachloroethane in male F344/N
rats. Contract NO1-ES-15326. MLI-NTP-1-93-1. Submitted to NTP.
119 DRAFT - DO NOT CITE OR QUOTE
-------
TSI Mason Laboratories. (1993b) 14 Day pilot gavage toxicity study of 1,1,2,2-tetrachloroethane in male F344/N
rats. Contract NO1-ES-15326. MLI-NTP-13-93-13. Submitted to NTP.
TSI Mason Laboratories. (1993c) 14 Day pilot gavage toxicity study of 1,1,2,2-tetrachloroethane in B6C3F1 mice.
Contract NO1-ES-15326. MLI-NTP-10-93-10. Submitted to NTP.
TSI Mason Laboratories. (1993d) 14 Day pilot dosed feed toxicity study of microencapsulated 1,1,2,2-tetrachloro-
ethane in B6C3F1 mice. Contract NO1-ES-15326. MLI-NTP-9-93-9. Submitted to NTP.
Tu, AS; Murray, TA; Hatch, KM; et al. (1985) In vitro transformations of BALB/C-3T3 cells by chlorinated ethanes
and ethylenes. Cancer Lett 28:85-92.
Tzeng, HF; Blackburn, AC; Board, PG; et al. (2000) Polymorphism- and species-dependent inactivation of
glutathione transferase zeta by dichloroacetate. Chem Res Toxicol 13:231-236.
U.S. EPA (Environmental Protection Agency). (1986) Guidelines for mutagenicity risk assessment. Federal Register
51(185):34006-34012.
U.S. EPA (Environmental Protection Agency). (1987) 1,1,2,2-tetrachloroethane (CASRN 79-34-5). Available
online at http:// www.epa.gov/iris (accessed August 4, 2009).
U.S. EPA (Environmental Protection Agency). (1988) Recommendations for and documentation of biological values
for use in risk assessment. EPA 600/6-87/008. Available from: National Technical Information Service, Springfield,
VA; PB88-179874/AS.
U.S. EPA (Environmental Protection Agency). (1991a) Guidelines for developmental toxicity risk assessment.
Federal Register 56(234):63798-63826.
U.S. EPA (Environmental Protection Agency). (1991b) Toxicology of the chloroacetic acids by-products of the
drinking water disinfection process. II. The comparative carcinogenicity of dichloroacetic and trichloroacetic acid:
implication for risk assessment. Document No. HERL-0820. Research Triangle Park, NC: Health Effects Research
Laboratory.
U.S. EPA (Environmental Protection Agency). (1994a) Interim policy for particle size and limit concentration issues
in inhalation toxicity studies. Federal Register 59(206):53799.
U.S. EPA (Environmental Protection Agency). (1994b) Methods for derivation of inhalation reference
concentrations and application of inhalation dosimetry. EPA/600/8-90/066F. Available from: National Technical
Information Service, Springfield, VA; PB2000-500023. Available online at http://www.epa.gov/iris/backgr-d.htm
(accessed August 4, 2009).
U.S. EPA (Environmental Protection Agency). (1995) Use of the benchmark dose approach in health risk
assessment. EPA/630/R-94/007. Available from: National Technical Information Service (NTIS) , Springfield, VA;
PB95-213765. Available online at http://www.epa.gov/iris/backgr-d.htm (accessed August 4, 2009).
U.S. EPA (Environmental Protection Agency). (1996) Guidelines for reproductive toxicity risk assessment. Federal
Register 61(212):56274-56322.
U.S. EPA (Environmental Protection Agency). (1998a) Guidelines for neurotoxicity risk assessment. Federal
Register 63(93):26926-26954.
U.S. EPA (Environmental Protection Agency). (2000a) Science policy council handbook: risk characterization.
Prepared by the Office of Science Policy, Office of Research and Development, Washington, DC. EPA 100-B-OO-
002. Available online at http://www.epa.gov/iris/backgr-d.htm (accessed August 4, 2009).
U.S. EPA (Environmental Protection Agency). (2000b) Benchmark dose technical guidance document [external
review draft]. EPA/630/R-00/001. Available online at http://www.epa.gov/iris/backgr-d.htm (accessed August 4,
2009).
120 DRAFT - DO NOT CITE OR QUOTE
-------
U.S. EPA (Environmental Protection Agency). (2000c) Supplemental guidance for conducting health risk
assessment of chemical mixtures. EPA/630/R-00/002. Available online at http://www.epa.gov/iris/backgr-d.htm
(accessed August 4, 2009).
U.S. EPA (Environmental Protection Agency). (2003) Toxicological review of dichloroacetic acid. In support of the
Integrated Risk Information System (IRIS). August 2003. EPA 635/R-03/007.
U.S. EPA (Environmental Protection Agency). (2005a) Guidelines for carcinogen risk assessment. Risk Assessment
Forum, Washington, DC; EPA/630/P-03/00IB. Available online at http://www.epa.gov/iris/backgrd.htm (accessed
January 15, 2009).
U.S. EPA (Environmental Protection Agency). (2005b) Supplemental guidance for assessing susceptibility from
early-life exposure to carcinogens. Risk Assessment Forum, Washington, DC; EPA/630/R-03/003F. Available
online at http://www.epa.gov/iris/backgr-d.htm (accessed January 15, 2009).
U.S. EPA (Environmental Protection Agency). (2006a) Peer review handbook. 3rd edition. Review draft. Science
Policy Council, Washington, DC. Available online at http://www.epa.gov/peerreview (accessed August 4, 2009).
U. S. EPA (Environmental Protection Agency). (2006b) A framework for assessing health risk of environmental
exposures to children. National Center for Environmental Assessment, Washington, DC, EPA/600/R-05/093F.
Available online at http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=l58363 (accessed August 4, 2009).
Vogel, EW; Nivard, MJM. (1993) Performance of 181 chemicals in a Drosophila assay predominantly monitoring
interchromosomal mitotic recombination. Mutagenesis 8:57-81.
Ward, JM. (1955) Accidental poisoning with tetrachloroethane. Br Med J 1:1136.
Warner, JR; Hughes, TJ; Claxton, LD. (1988) Mutagenicity of 16 volatile organic chemicals in a vaporization
technique with Salmonella typhimurium TA100. Environ Mol Mutagen 11:111-112.
Willcox, WH; Spilsbury, BH; Legge, TM. (1915) An outbreak of toxic jaundice of a new type amongst aeroplane
workers-its clinical and toxological aspects. Trans Med Soc London 38:129-156.
Williams, GM. (1983) DNA repair tests of 11 chlorinated hydrocarbon analogs. Final report. U.S. Environmental
Protection Agency. NTIS No. OTS408324292.
Wolff, L. (1978) The effect of 1,1,2,2,-tetrachloroethane on passive avoidance learning and spontaneous locomotor
activity. Act Nerv Super (Praha) 20:14-16.
Woodruff, RC; Mason, JM; Valencia, R; et al. (1985) Chemical mutagenesis testing in drosophila. V. Results of 53
coded compounds tested for the National Toxicology Program. Environ Mutagen 7:677-702.
Yllner S. (1971) Metabolism of l,l,2,2-tetrachloroethane-14C in the mouse. Acta Pharmacol Toxicol (Copenh)
29:499-512.
121 DRAFT - DO NOT CITE OR QUOTE
-------
1 APPENDIX A: SUMMARY OF EXTERNAL PEER REVIEW AND PUBLIC
2 COMMENTS AND DISPOSITION
3
4 The Toxicological Review of 1,1,2,2-tetrachloroethane (dated August, 2009) has
5 undergone a formal external peer review performed by scientists in accordance with EPA
6 guidance on peer review (U.S. EPA, 2006a, 2000a). The external peer reviewers were tasked
7 with providing written answers to general questions on the overall assessment and on chemical-
8 specific questions in areas of scientific controversy or uncertainty. A summary of significant
9 comments made by the external reviewers and EPA's responses to these comments arranged by
10 charge question follow. In many cases, the comments of the individual reviewers have been
11 synthesized and paraphrased in development of Appendix A. An external peer-review workshop
12 was held January 27, 2010. EPA did not receive any scientific comments from the public.
13
14 EXTERNAL PEER REVIEW PANEL COMMENTS
15 The reviewers made several editorial suggestions to clarify specific portions of the text.
16 These changes were incorporated in the document as appropriate and are not discussed further.
17 In addition, the reviewers provided comments specific to particular decisions and
18 analyses presented in the Toxicological Review under multiple charge questions. These
19 comments were organized and responded to under the most appropriate charge question.
20
21 A. General Comments
22
23 1. Is the Toxicological Review logical, clear and concise? Has EPA clearly synthesized the
24 scientific evidence for noncancer and cancer hazard?
25
26 Comments: The reviewers, generally, commented that the Toxicological Review was
27 logically written. One reviewer recommended an improvement to the clarity of the document
28 by reducing the text describing the available studies and presenting the individual study data
29 in a bulleted format, and this was echoed by another reviewer who recommended condensing
30 the study summaries and discussions.
31
32 Response: The content of the Toxicological Review is consistent with the current outline for
33 IRIS Toxicological Reviews, although an effort has been made to streamline the document and
34 reduce the redundancy. The general structure of a Toxicological Review is to present a factual
35 summary of toxicity studies in Section 4 and critical interpretation/synthesis in Section 5.
36
37 2. Please identify any additional studies that should be considered in the assessment of the
38 noncancer and cancer health effects of l,!92,2-tetrachloroethane.
39
A-l B-l DRAF1
-------
1 Comments: One reviewer identified the following studies:
2
3 Matsuoka A, Yamakage K, Kusakabe H, Wakuri S, Asakura M, Noguchi T, Sugiyama T,
4 Shimada H, Nakayama S, Kasahara Y, Takahashi Y, Miura KF, Hatanaka M, Ishidate M
5 Jr, Monta T, Watanabe K, Hara M, Odawara K, Tanaka N, Hayashi M, Sofuni T. Re-
6 evaluation of chromosomal aberration induction on nine mouse lymphoma assay "unique
7 positive'NTP carcinogens. 1996. Mutat Res. Aug 12;369(3-4):243-52.
8
9 Sofuni T, Honma M, Hayashi M, Shimada H, Tanaka N, Wakuri S, Awogi T, Yamamoto
10 KI, Nishi Y, Nakadate M. Detection of in vitro clastogens and spindle poisons by the
11 mouse lymphoma assay using the microwell method: interim report of an international
12 collaborative study. Mutagenesis. 1996 Jul;ll(4):349-55.
13
14 Ashley DL, Bonin MA, Cardinali FL, McCraw JM, Wooten JV. Blood concentrations of
15 volatile organic compounds in a nonoccupationally exposed US population and in groups
16 with suspected exposure. Clin Chem. 1994 Jul;40(7Pt 2):1401-4.
17
18 Response: The references [Matsuoka et al. (1996), Sofuni et al. (1996), Ashley et al. (1994)]
19 were examined but have not been added to the Toxicological Review, as these references do not
20 contribute significant information to the discussion and analysis in the document.
21
22 B. Oral Reference Dose (RfD) for 1,1,2,2-tetrachloroethane
23
24 1. Subchronic and chronic RfDs for l,l?2,2-tetrachloroethane have been derived from a
25 13-week oral gavage study (NTP, 2004) in rats and mice. Please comment on whether
26 the selection of this study as the principal study has been scientifically justified. Please
27 identify and provide the rationale for any other studies that should be selected as the
28 principal study.
29
30 Comment: The reviewers generally agreed that the selection of the NTP (2004) report as the
31 principal study was scientifically justified.
32
33 Response: No response.
34
35 Comment: One reviewer commented that the Gulati et al. (1991a,b) is the only other study
36 that could be a candidate principal study and provides what may be a more significant
37 endpoint for human health protection; but also states that EPA has made a reasonable
38 selection in the NTP study.
A-2 B-l DRAF1
-------
1
2 Response: The Gulati et al. developmental studies were conducted at doses higher than the
3 subchronic NTP (2004) study, which demonstrated liver effects at lower doses. Therefore,
4 the Gulati et al. studies were not selected as the principal studies. However, potential points
5 of departure (PODs) based on the observed developmental effects from Gulati et al. (1991 a)
6 were provided in the document for comparison purposes.
7
8 Comment: One reviewer requested additional explanation regarding the statement that high
9 incidences of hepatocellular tumors in all mouse groups of the NCI (1978) study precluded
10 evaluation of noncancer effects in the liver.
11
12 Response: A LOAEL of 142 mg/kg-day was selected for chronic inflammation in the
13 kidneys of male mice, while a NOAEL of 142 mg/kg-day and a LOAEL of 284 mg/kg-day
14 were selected for hydronephrosis and chronic inflammation in the kidneys of female mice.
15 The text in Section 5.1.2.1., Choice of Principal Study and Critical Effect - with Rationale
16 and Justification, addressing the high incidence of hepatocellular tumors in all mouse dose
17 groups and the evaluation of noncancer effects in the liver was deleted.
18
19 2. Increased relative liver weight was selected as the critical effect for the derivation of the
20 subchronic and chronic RfDs. Please comment on whether the rationale for the
21 selection of this critical effect has been scientifically justified. Please provide a detailed
22 explanation. Please identify and provide the rationale for any other endpoints that
23 should be considered in the selection of the critical effect.
24
25 Comment: The reviewers generally agreed that the selection of increased relative liver
26 weight as the critical effect for the derivation of the subchronic and chronic RfDs was
27 justified. One reviewer commented that increased relative liver weight is a less
28 toxicologically significant index of liver change than increased absolute liver weight, due to
29 the treatment-induced loss of body weight; whereas another reviewer believed the change in
30 relative liver weight is more appropriate than absolute liver weight where body weights in
31 general are being affected. Another reviewer commented that increased serum enzyme
32 activity is an alternative critical effect and a true measure of hepatocellular damage, and the
33 most toxicologically-significant endpoint should be selected as the critical effect. A reviewer
34 commented that the only other endpoint that is a candidate critical effect is reduced fetal
35 body weight in the Gulati et al. studies, but also states that EPA's selection of the relative
36 liver weight as the critical effect is reasonable.
37 Two reviewers questioned the statement in the Toxicological Review that the critical
38 effect was selected "because this effect may represent a sensitive endpoint that occurs early
A-3 B-l DRAF1
-------
1 in the process leading to hepatocellular necrosis." The reviewers questioned whether
2 increases in liver weight reflect other, earlier changes that have been going on long enough to
3 cause the cell proliferation, inflammation, or other effects responsible for the observed
4 weight gain.
5
6 Response: The increase in relative liver weight was selected as the basis for the selection of
7 the POD because the relative liver weight analysis takes into account the substantive, dose-
8 dependent decreases in body weight that were observed in both sexes of rats.
9 The reduction in fetal body weight was observed at doses higher than the
10 demonstrated liver effects from the subchronic NTP (2004) study. Therefore, the decrease in
11 fetal body weight was not selected as the critical effect. However, potential points of
12 departure (PODs) based on the observed developmental effects from Gulati et al. (199la)
13 were provided in the document for comparison purposes.
14 EPA considered that, given the available data, increased liver weight represents the most
15 sensitive effect observed in the liver and that it may occur early in the process of liver toxicity
16 associated with oral exposure to 1,1,2,2-tetrachloroethane. In addition to increased liver weight
17 following subchronic exposure, the evidence of hepatocellular damage includes; increased serum
18 concentrations of hepatocellular enzymes (ALT and SDH), decreased serum cholesterol, and
19 increased incidences of hepatocellular necrosis, bile duct hyperplasia, hepatocelluar mitotic
20 alterations, and hepatic pigmentation. In addition, evidence of the 'earlier changes' reflected by
21 the increase in liver weight as suggested by two reviewers is unavailable. Thus, EPA concluded
22 that the observed increase in liver weight may represent the most sensitive effect that occurs early
23 in the process of 1,1,2,2-tetrachloroethane-induced hepatoxicity following subchronic oral
24 exposure.
25
26 3. Hepatocellular vacuolization was observed at the lowest dose in the principal study
27 (NTP, 2004). This effect was not selected as the critical effect for the determination of
28 the POD for derivation of the subchronic and chronic RfDs. Please comment on the
29 rationale and justification for not selecting this endpoint as the critical effect.
30
31 Comment: The reviewers generally considered the rationale and justification for not
32 selecting hepatocellular vacuolization as the critical effect as reasonable, justified, logical,
33 and comprehensive. One reviewer recommended slight refinements to the justification, and
34 questioned whether the comments that vacuolization was not observed across species and the
35 severity was not dose-dependent supported the conclusion. Another reviewer asked if NTP
36 (2004) specified the lobular distribution of the vacuoles.
37
A-4 B-l DRAF1
-------
1 Response: The decision to not select hepatocellular vacuolization as the critical effect
2 involved more than a consideration of cross species observations and severity (see Section
3 5.1.1.1., Choice of Principal Study and Critical Effect - with Rationale and Justification).
4 The biological significance of the hepatocellular vacuolization observed following
5 1,1,2,2-tetrachloroethane exposure was unclear based on the paucity of information provided
6 by NTP (2004).
7 NTP did not specify the lobular distribution of the observed vacuoles.
8
9 4. The subchronic and chronic RfDs have been derived utilizing benchmark dose (BMD)
10 modeling to define the point of departure (POD). All available models were fit to the
11 data in both rats and mice for increased absolute and relative liver weight, increased
12 incidence of hepatocellular cytoplasmic vacuolization (rats only), increased levels of
13 ALT, SDH, and bile acids, and decreased fetal body weight. Has the BMD modeling
14 been appropriately conducted? Is the benchmark response (BMR) selected for use in
15 deriving the POD (i.e., one standard deviation from the control mean) scientifically
16 justified? Please identify and provide the rationale for any alternative approaches
17 (including the selection of the BMR, model, etc.) for the determination of the POD and
18 discuss whether such approaches are preferred to EPA's approach.
19
20 Comment: Three reviewers stated that the BMD modeling was appropriate. One reviewer
21 disagreed with the reasoning provided in the document for eliminating the two highest dose
22 groups from the BMD modeling analysis for all of the endpoints, and stated that dropping
23 doses is typically only done when the issues of model fit are encountered. A second reviewer
24 commented that EPA should at least show earlier BMD modeling results with the highest
25 doses included and show the lack of model fit that led to the elimination of the two highest
26 doses.
27
28 Response: In agreement with the reviewers' comments, the current reasoning, provided in
29 Section 5.1.1.2 of the document, Methods of Analysis—Including Models (PBPK, BMD, etc.),
30 for dropping the two highest dose groups (exceeding the MTD) was removed. In its place, a
31 rationale for dropping dose groups based on adequacy of model fit was employed. In
32 addition, as requested by two of the external peer reviewers, the endpoints in Table 5-1 were
33 remodeled using the most recent version of BMDS (i.e., 2.1.1). Because of these changes,
34 Appendix B was essentially replaced with a new version showing BMD modeling results
35 (generated using version 2.1.1 of BMDS) with the highest dose groups included to
36 demonstrate that lack of model fit led to the elimination of one or more of these dose groups
37 in order to obtain adequate fit. As a result of this remodeling, a new critical effect was
A-5 B-l DRAF1
-------
1 selected, relative liver weight in female rats, where before, relative liver weight in male rats
2 had been selected.
3
4 5. Please comment on the selection of the uncertainty factors applied to the POD for the
5 derivation of the RfDs. For instance, are they scientifically justified? If changes to the
6 selected uncertainty factors are proposed, please identify and provide a rationale(s).
7
8 Please comment specifically on the following uncertainty factor:
9 • A database uncertainty factor of 3 was used to account for the lack of oral
10 reproductive and developmental toxicity data for 1,1,2,2-tetrachloroethane.
11 Please comment on whether the application of this uncertainty factor has
12 been scientifically justified.
13
14 Comment: The reviewers generally considered the applications of the uncertainty factors to
15 be adequate, acceptable, reasonable, and appropriate.
16
17 Response: No response.
18
19 Comment: One reviewer requested a comparison between the RfD derived from the
20 subchronic NTP study and an approximate RfD derived from the chronic NCI study.
21
22 Response: A comparison between the RfD derived from the subchronic NTP (2004) study
23 and an approximate RfD derived from the chronic NCI (1978) study was considered. The
24 RfD from the subchronic NTP study was based on a study that used lower dose levels and a
25 wider dose range than the NCI (1978) study, and thereby provided a better characterization
26 of the dose-response curve in the low-dose region. Additionally, the route of exposure used
27 in the NTP study, dietary exposure, is a more relevant route of exposure for the general
28 population exposed to 1,1,2,2-tetrachloroethane in the environment than the gavage exposure
29 used in the NCI study. However, if one were to use the observance of chronic inflammation
30 in the kidneys of male mice in the NCI study as a LOAEL, for purposes of comparison, the
31 POD of 142 mg/kg-day could be divided by a total UF of 3 00 to yield an RfD of 0.5 mg/kg-
32 day.
33
34 Comment: A reviewer recommended the addition of text addressing the major metabolites of
35 1,1,2,2-tetrachloroethane (dichloroacetic acid, trichloroethylene, perchloroethylene) and how
36 the results of these assessments compare to those derived for 1,1,2,2-tetrachloroethane.
37
A-6 B-l DRAF1
-------
1 Response: This comparison was considered outside of the scope of the IRIS assessment for
2 1,1,2,2-tetrachloroethane.
3
4 Comment: One reviewer commented that there is a considerable amount of information
5 about the toxicokinetics of related halocarbons [e.g., trichloroethylene (TCE),
6 perchloroethylene (PERC), chloroform, 1,1,1-trichloroethane] in rodents and humans, and
7 that the rank of metabolic activation of the compounds is: mice » rats > humans. Therefore,
8 the toxicokinetic component of the interspecies UF of 10 could be reduced, resulting in a
9 interspecies uncertainty factor of 3.
10
11 Response: The potential difference between animal and human toxicokinetics following 1,1,2,2-
12 tetrachloroethane exposure based on information from related halocarbons was added to Section
13 5.3, Uncertainties in the Oral Reference Dose (RfD) and Inhalation Reference Concentration
14 (RfC)- Upon further evaluation, this information was not considered sufficient to reduce the UF
15 for 1,1,2,2-tetrachloroethane and the UF of 10 was retained.
16
17 Comment: A reviewer commented that Section 5.3 is a restatement of the features that
18 contributed to the valuation of the standard uncertainty factors, and recommended a
19 consideration of what additional uncertainties are present that might impact the results.
20
21 Response: Additional text was added to this section in response to the reviewer's comment.
22
23 C. Inhalation Reference Concentration (RfC) for l,l?2,2-tetrachloroethane
24
25 1. An RfC for l,l?2,2-tetrachloroethane has not been derived. Has the scientific
26 justification for not deriving an RfC been described in the document? Please identify
27 and provide the rationale for any studies that should be selected as the principal study.
28 Please identify and provide the rationale for any endpoints that should be considered in
29 the selection of the critical effect.
30
31 Comment: The reviewers agreed with the decision not to derive an RfC. One reviewer
32 comment that a comparison to metabolically-related compounds is useful and recommended
33 including this information in the discussion of the uncertainties associated with not deriving
34 an RfC.
35
36 Response: Most reviewers were in agreement with the decision to not derive an RfC based
37 on the available data. Additional text related to uncertainties was added to Section 5.3.
38
A-7 B-l DRAF1
-------
1 D. Carcinogenicity of l,l?2,2-tetrachloroethane
2
3 1. Under EPA's 2005 Guidelines for carcinogen risk assessment (www.epa.gov/iris/backgr-
4 d.htm), the Agency concluded that l,l?2,2-tetrachloroethane is likely to be carcinogenic
5 to humans by all routes of exposure. Please comment on the cancer weight of the
6 evidence characterization. Is the cancer weight of evidence characterization
7 scientifically justified?
8
9 Comment: One reviewer commented that the conclusion that 1,1,2,2-tetrachloroethane is
10 likely to be carcinogenic to humans is one of the weakest likely to be carcinogenic to humans
11 characterizations demonstrated when the data is singularly considered; in addition, given the
12 prevalence of and susceptibility to developing liver tumors in B6C3Fi mice, the reviewer
13 questioned whether a slope factor should be derived from this study. A second reviewer did
14 not concur with the conclusion that 1,1,2,2-tetrachloroethane is likely to be carcinogenic to
15 humans, and thought it would be more accurate to characterize 1,1,2,2-tetrachloroethane as a
16 possible human carcinogen. Several reviewers recommended incorporating the carcinogenic
17 conclusions for related compounds/major metabolites (dichloroacetic acid, trichloroethylene,
18 and perchloroethylene) to make a stronger case for the likely to be carcinogenic to humans
19 determination.
20
21 Response: The cancer weight of evidence descriptor for 1,1,2,2-tetrachloroethane is based
22 on the statistically significant increase in the incidence of hepatocellular carcinomas in both
23 male and female B6C3F1 mice, and the rare hepatocellular tumors observed in the male
24 Osborne-Mendel rats (NCI, 1978). According to the Guidelines for Carcinogen Risk
25 Assessment (U.S. EPA, 2005a), the likely to be carcinogenic to humans descriptor is
26 supported when an agent has tested positive in animal experiments in more than one species,
27 sex, strain, site, or exposure route with or without evidence of carcinogenicity in humans, and
28 in the case of 1,1,2,2-tetrachloroethane, a positive tumor response was observed in both male
29 and female mice. This descriptor is also supported when a rare animal tumor is observed in a
30 single experiment that is assumed to be relevant to humans, and in the case of 1,1,2,2-
31 tetrachloroethane, NCI (1978) considered the liver tumors observed in male rats to be a rare
32 tumor response.
33 Additional text was added to the discussion of the potential susceptibility of B6C3F1
34 mice to developing hepatocellular carcinomas following 1,1,2,2-tetrachloroethane exposure
35 is included in Section 5.4.5, Uncertainties in Cancer Risk Values.
36 Section 4.7.1, Summary of Overall Weight of Evidence, presents the carcinogenicity
37 data available for 1,1,2,2-tetrachloroethane. This section also includes a discussion of the
A-8 B-l DRAF1
-------
1 carcinogenicity data available for dichloroacetic acid, trichloroethylene, and
2 perchloroethylene.
3
4 2. A two-year oral gavage cancer bioassay (NCI, 1978) was selected as the principal study
5 for the derivation of an oral slope factor. Please comment on the appropriateness of the
6 selection of the principal study.
7
8 Comment: The reviewers generally agreed with the selection of the NCI (1978) study as the
9 principal study for the development of an oral slope factor, although the reviewers highlighted
10 that this was the only study available for this purpose.
11
12 Response: No response.
13
14 Comment: One reviewer commented that the NCI study used poorly selected dose levels that
15 were adjusted during the course of the study, the exposure duration was 78 weeks as opposed
16 to the more standard 104 weeks, that there was also a concurrent disease (pneumonia)
17 observed, and that these deficiencies and resulting uncertainties need to be stated in the
18 document.
19
20 Response: Text was added to Section 5.4.5, Uncertainties in Cancer Risk Values, to address
21 the concern associated with the doses selection and modification and the increased incidence
22 of chronice murine pneumonia in the rats.
23
24 Comment: A reviewer expressed concerns that gavage dosing may deliver the chemical in a
25 short term bolus dose and may not provide the same results as a dietary or other oral dosing
26 method that delivers the chemical more gradually over time.
27
28 Response: The potential effect of the corn oil vehicle, as well as the bolus nature of the
29 gavage dose, on the effects observed in the liver following 1,2,3-trichloropropane exposure
30 has been added to Section 5.4.5, Uncertainties in Cancer Risk Values.
31
32 3. An increased incidence of hepatocellular carcinomas in B6C3F1 mice was used to
33 estimate the oral cancer slope factor. Please comment on the scientific justification of
34 this analysis. Has the BMD modeling been appropriately conducted?
35
36 Comment: Several reviewers considered the modeling of the increased incidence of
37 hepatocellular tumors in B6C3F1 mice to be justified and appropriate. One reviewer
38 commented that maybe an oral slope factor should not be derived given the prevalence of and
A-9 B-l DRAF1
-------
1 susceptibility to developing liver tumors in this strain of mice. A reviewer commented that
2 both sexes of B6C3F1 mice have a high spontaneous cancer incidence and referenced a study
3 by Haseman et al. (1998) which reported that male B6C3F1 control mice have a 42% liver
4 cancer incidence.
5
6 Response: The U.S. EPA considers liver tumors in mice to be relevant to humans unless
7 chemical-specific information is available to indicate otherwise. Text addressing this issue is
8 included in Section 5.4.5, Uncertainties in Cancer Risk Values.
9 Text was also added to Section 5.4.5, Uncertainties in Cancer Risk Values,
10 addressing the high spontaneous cancer incidence of liver cancer in male B6C3F1 mice. The
11 42% liver cancer rate for male B6C3F1 mice was for liver adenomas and carcinomas
12 combined, but the NCI (1978) study analysis was for hepatocellular carcinomas, only.
13 Haseman et al. (1998) reported a 17.9 and 8.4% hepatocellular carcinoma rate in feeding
14 studies for male and female B6C3F1 mice, respectively.
15 It should be noted, that even though the B6C3F1 strain may have a high
16 spontaneous cancer incidence, the incidence in the control mice in NCI (1978) was 1/18 in
17 the male vehicle controls and 0/20 in the female vehicle controls, and 3/36 and 1/40 in male
18 and female pooled-vehicle controls, respectively. Comparison of an experimental group is
19 with its concurrent controls was considered to be the most appropriate comparison, and in
20 this case, the control values were considered low (Haseman et al., 1992; Tarone et al., 1981;
21 Gart et al., 1979 referenced in Haseman et al., 1998).
22
23 Comment: One reviewer requested additional model output information, in Appendix C,
24 describing how the multi-stage model fit the data points, even if the reported goodness-of-fit
25 p-value was provided as "NA" because of too many model parameters.
26
27 Response: In response to this comment, the incidence of hepatocellular carcinomas in male
28 and female mice were remodeled using the most recent version of BMDS (version 2.1.1), and
29 the relevant information describing the fit of both the one- and two-stage multistage models
30 to these incidence data have now been included in Appendix C.
31
32 Comment: A reviewer requested additional analysis of the mode of action of carcinogenesis,
33 as the preponderance of genotoxicity data suggest that 1,1,2,2-tetrachloroethane is not
34 genotoxic and the data available indicate promotion potential. This reviewer recommended
35 an uncertainty factor approach for the cancer assessment. A second reviewer also
36 commented that it is more likely that 1,1,2,2-tetrachloroethane may act as a tumor promoter,
37 provided that the majority of the in vitro and in vivo genotoxicity and mutagenicity studies
38 yielded non-positive results.
A-10 B-l DRAF1
-------
1
2 Response: The two available studies providing some evidence to support the promotion
3 potential of 1,1,2,2-tetrachloroethane were added to Section 4.7.3, Mode of action
4 Information. However, the key events associated with any hypothesized mode of action of
5 carcinogenesis of 1,1,2,2-tetrachloroethane cannot be determined provided the information
6 available.
7
8 Comment: A reviewer commented that mice and other rodents metabolize a considerably
9 larger portion of high doses of halocarbons than humans, and, therefore, experience more
10 severe hepatocellular injury, greater formation of covalent adducts, and higher cancer
11 incidences. This reviewer also commented that male B6C3F1 mice have a very high
12 spontaneous liver cancer incidence as indicated by Haseman et al. (1998). The reviewer
13 recommended including a discussion addressing this in the uncertainty section.
14
15 Response: Text was added to Section 5.4.5, Uncertainties in Cancer Risk Values, addressing
16 the potential difference between animal and human toxicokinetics following 1,1,2,2-
17 tetrachloroethane exposure based on information from related halocarbons demonstrating
18 increased metabolic activation in mice compared with humans. In addition, text was also added
19 to Section 5.4.5, Uncertainties in Cancer Risk Values, addressing the high spontaneous
20 cancer incidence of liver cancer in male B6C3F1 mice.
21
22 Comment: A reviewer commented that the document should recognize that administration of
23 large quantities of corn oil promotes lipid accumulation and lipoperoxidative damage of
24 hepatocytes, and that corn oil is believed to be tumorigenic in rats and humans through
25 increased expression of protooncogenes, decreased apotosis, mitogenesis, etc. The reviewer
26 recommended including a discussion addressing this in the uncertainty section.
27
28 Response: EPA has included text in Section 5.4.5, Uncertainties in Cancer Risk Values, that
29 addresses that the bolus administration of 1,2,3-trichloropropane was in corn oil.
30
31
A-11 B-l DRAF1
-------
APPENDIX B. BENCHMARK DOSE MODELING RESULTS FOR THE DERIVATION
OF THE RfD
Dichotomous Endpoints
Incidence of hepatocellular cytoplasmic vacuolization in male and female rats (NTP, 2004)
Table B-l. Incidences of hepatocellular cytoplasmic vacuolization in
rats exposed to dietary I,l92,2-tetrachlorethane for 14 weeks
Nonneoplastic lesion
Dose (mg/kg-d)
Vehicle
control
20
40
80
170
320
Males3
Hepatocellular cytoplasmic
vacuolization
0/10
7/1 Ob
(1.3)
9/1 Ob
(2.0)
10/10b
(1.9)
8/1 Ob
(1.4)
0/10
Females3
Hepatocellular cytoplasmic
vacuolization
0/10
0/10
10/10b
(1.7)
10/10b
(2.2)
4/1 Ob
(1.3)
0/10
a Values represent proportion of animals with the lesion; for those dose groups in which lesions were found,
the average severity score is in parenthesis; severity grades were as follows: 1 = minimal, 2 = mild, 3 =
moderate, 4 = severe.
b Statistically significantly different from vehicle control group.
Source: NTP (2004).
All available dichotomous models (except the "quantal-linear" and "quantal-quadratic")
in the EPA's BMDS (version 2.1.1) were fit to the incidence of hepatocellular cytoplasmic
vacuolization in male and female rats administered 1,1,2,2-tetrachloroethane in the diet for 14
weeks. Table B-l displays the incidence data for this endpoint for both males and females.
BMDs and their associated 95 percent lower confidence limits (i.e., BMDLs) at an extra risk of
10% were estimated by each model. The results of this BMD modeling for male and female rats
are summarized in Tables B-2 and B-3, respectively, and the BMDS output from the selected
model are displayed following each table.
A-A
B-l
DRAF1
-------
Table B-2. Summary of BMD modeling results for the incidence of hepatocellular
cytoplasmic vacuolization in male rats
Model
DF
x2
5£2 Goodness of fit
/>-valuea
Scaled residuals
of interest1"
AIC
BMD10
(mg/kg-
day)
BMDL10
(mg/kg-
day)
All dose groups included
BMDS was unable to generate model outputs
Highest dose group dropped
Gamma0
Logistic
Log-logistic '
Log-probit
Multistage (1 -degree)6
Probit
Weibull0
4
3
4
4
4
3
4
57.61
22.78
6.78
36.46
57.61
20.45
57.61
O.001
O.001
0.15
O.001
O.001
O.001
O.001
0.00/1.66
-2.77/1.01
0.00/-0.06
0.00/0.85
0.00/1.66
3.00/0.94
0.00/1.66
47.97
57.05
36.14
41.77
47.97
58.24
47.97
3.64
10.59
0.91
4.70
3.64
13.29
3.64
2.60
6.70
0.40
3.03
2.60
8.99
2.60
Two highest dose groups dropped
Gamma0
Logistic
Log-logistic
Log-probit
Multistage (l-degree)e'g
Multistage (2 -degree)0
Multistage (3 -degree)0
Probit
Weibull0
2
2
2
2
3
2
2
2
2
0.10
2.50
0.25
0.18
0.10
0.08
0.06
2.56
0.10
0.95
0.29
0.88
0.92
0. 99
0.96
0.97
0.28
0.95
0.00/0.08
-0.82/0.81
0.00/0.09
0.00/0.10
0.00/-0.02
0.00/0.12
0.00/0.13
-0.81/1.03
0.00/0.10
22.87
25.51
23.09
22.98
20.89
22.83
22.80
25.71
22.86
2.47
6.78
6.16
5.49
1.73
1.99
1.89
6.45
2.32
1.12
3.67
0.31
1.80
1.12
1.12
1.13
3.73
1.12
AIC = Akaike Information Criterion; BMD = maximum likelihood estimate of the dose associated with the selected
benchmark response; BMDL = 95% lower confidence limit on the BMD; DF = degrees of freedom
aValues < 0.1 fail to meet conventional goodness-of-fit criteria.
Scaled residuals at doses immediately below and immediately above the benchmark dose.
°Power restricted to >1.
dSlope restricted to >1.
"Betas restricted to >0.
fAlthough the overall goodness of flip-value suggested adequate fit of this model to the data, the model was rejected
because the very high residual at the high dose (-2.32) suggested that fit of the model to the data would be improved by
dropping that dose.
8Selected model is displayed in boldface type. BMDLs for models with adequate fit differed by > threefold. However, the
results from the log-logistic model were rejected as unreliable due to the large spread between BMD and BMDL (20-fold)
and because the BMDL from this model was an outlier in relation to the results of the other models. After dropping this
model, the results of the other models were within approximately threefold. Among the remaining models, the 1-degree
polynomial had the lowest AIC and also produced the lowest BMDL and was therefore selected as the most suitable model
for this dataset.
A-B
B-l
-------
As shown in Table B-2, in attempting to model the incidence of hepatocellular
cytoplasmic vacuolization in male rats with all six dose groups included, the BMDS failed to
generate any output because response was not a monotonically increasing function of dose (i.e.,
the response in the penultimate dose group was 80%, while the response in the highest dose
group was 0). A key underlying assumption for the fitting of the dichotomous models in BMDS
is that response must be a monotonically non-decreasing function of dose. Therefore, the highest
dose group was dropped and the models were fit to the data again. In this instance, the chi-
square goodness-of-fit test found that all models exhibited inadequate fit (i.e., p < 0.1). Finally,
in an attempt to find a model that fit, the two highest dose groups were dropped and the models
were refit to these data. In this case, all of the models exhibited adequate fit (p > 0.10).
Of these models exhibiting adequate fit, a "best-fit" model was selected consistent with
the EPA's Benchmark Dose Technical Guidance Document (USEPA 2000), as follows. If the
BMDL estimates from the models exhibiting adequate fit were "sufficiently close," then the
model with the lowest AIC is to be used to estimate the BMDL from which the POD will be
derived. In this particular case, as explained in the footnote in Table B-2, BMDLs for models
with adequate fit differed by greater than threefold. However, the results from the log-logistic
model were rejected as unreliable due to the large spread between BMD and BMDL (20-fold)
and because the BMDL from this model was an outlier in relation to the results from the other
models. After dropping the log-logistic model, the BMDLs from the other models were within
approximately threefold. Among the remaining models, the one-stage multistage model had the
lowest AIC, and also produced the lowest BMDL, and was therefore selected as the most
suitable model for this dataset. The BMDLio from this model (i.e., 1.12 mg/kg-day) was then
selected as a possible POD. The standard BMDS output from the one-stage multistage model is
displayed below.
A-C B-l DRAF1
-------
Multistage Model with 0.95 Confidence Level
a
o
'•8
0.8
0.6
0.4
0.2
Multistage
3MDL BMD
0 10 20 30 40 50 60 70 80
dose
11:41 03/302010
Multistage Model. (Version: 3.0; Date: 05/16/2008)
Input Data File:
C:\USEPA\lRlS\TCE\NTP\hepcytvac\male\mst_hepcytvacM2HDD_MS_l.(d)
Gnuplot Plotting File:
C:\USEPA\lRlS\TCE\NTP\hepcytvac\male\mst_hepcytvacM2HDD_MS_l.pit
Tue Mar 30 12:41:48 2010
HMDS Model Run
The form of the probability function is:
P[response] = background + (1-background)* [1-EXP(
-betal*doseAl)]
The parameter betas are restricted to be positive
Dependent variable = incidence
Independent variable = dose
Total number of observations = 4
A-4
B-l
DRAF1
-------
Total number of records with missing values
Total number of parameters in model = 2
Total number of specified parameters = 0
Degree of polynomial = 1
= 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial Parameter Values
Background = 0
Beta(l) = 1.28571e+018
Asymptotic Correlation Matrix of Parameter Estimates
Beta(l)
( *** The model parameter(s) -Background
have been estimated at a boundary point, or have been specified by the user,
and do not appear in the correlation matrix )
Beta(l)
1
Parameter Estimates
Variable
Background
Beta(l)
Estimate
0
0. 0607678
Std. Err.
95.0% Wald Confidence Interval
Lower Conf. Limit Upper Conf. Limit
* - Indicates that this value is not calculated.
Model
Full model
Fitted model
Reduced model
Analysis of Deviance Table
Log(likelihood)
-9.35947
-9.44611
-25.8979
# Param's
4
1
1
Deviance Test d.f.
P-value
0 .173273
33.0768
0.9818
< .0001
AIC:
20.8922
Goodness of Fit
Scaled
0
20
40
80
Dose
.0000
.0000
.0000
.0000
Est
0.
0.
0.
0.
. Prob.
0000
7034
9120
9923
Expected
0.
7 .
9.
9.
.000
.034
.120
.923
Observed
0.
7.
9.
10.
.000
.000
.000
.000
Size
10
10
10
10
Residual
0.
-0.
-0.
0.
.000
.024
.134
.279
ChiA2 = 0.10
d.f. = 3
P-value = 0.9922
A-5
B-l
DRAF1
-------
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Extra risk
Confidence level = 0.95
HMD = 1.73382
BMDL = 1.11682
BMDU = 2.71595
Taken together, (1.11682, 2.71595) is a 90 % two-sided confidence
interval for the BMD
A-6 B-l DRAF1
-------
Table B-3. Summary of benchmark dose model results for the incidence of
hepatocellular cytoplasmic vacuolization in female rats
Model
DF
7.2
X2 Goodness
of fit
/>-valuea
Scaled
residuals of
interest1"
AIC
BMD10
(mg/kg-day)
BMDL10
(mg/kg-day)
All dose groups included
BMDS was unable to generate model outputs
Highest dose group dropped
Gamma0
Logistic
Log-logistic
Log-probit
Multistage (1 -degree polynomial)6
Probit
Weibull0
4
3
4
4
4
3
4
45.13
38.70
31.61
49.11
45.13
38.70
45.13
O.001
O.001
O.001
O.001
O.001
O.001
O.001
0.00/-1.66
-2.52/3.63
0.00/-2.36
0.00/-1.61
0.00/-1.66
-2.50/3.65
0.00/-1.66
61.33
69.75
53.57
58.57
61.33
69.79
61.33
8.65
30.61
3.99
12.62
8.65
31.28
8.65
6.18
18.21
2.24
8.86
6.18
19.39
6.18
Two highest dose groups dropped
Gamma0
Logistic
Log-logistic"1
Log-probitd
Multistage (1 -degree polynomial)6
Multistage (2 -degree polynomial)6
Multistage (3 -degree polynomial)6
Probit
WeibulF'1
3
2
3
3
3
3
3
2
3
1.56
0.00
0.04
0.00
13.83
7.48
4.41
0.00
0.00
0.67
1.00
1.00
1.00
0.003
0.06
0.22
1.00
1.00
-0.95/0.82
0.00/0.00
-0.14/0.14
0.00/0.00
0.00/-3.09
0.00/-2.24
0.00/-1.78
0.00/0.00
-0.02/0.01
5.00
4.00
2.08
2.00
22.89
14.54
9.85
4.00
2.00
20.59
29.46
25.03
26.36
3.14
10.17
14.53
28.77
30.68
17.05
19.38
19.51
19.56
2.05
5.95
9.15
19.85
19.16
AIC = Akaike Information Criterion; BMD = maximum likelihood estimate of the concentration associated with the selected
benchmark response; BMDL = 95% lower confidence limit on the BMD; DF = degrees of freedom
aValues < 0.1 fail to meet conventional goodness-of-fit criteria.
bScaled residuals at doses immediately below and immediately above the benchmark dose.
°Power restricted to >1.
Slope restricted to >1.
6Betas restricted to >0.
Selected model is displayed in boldface type. BMDLs for models with adequate fit differed by < threefold, so the models
with the lowest AIC (Log-probit and Weibull models) were initially selected as the best fitting. The Weibull model had a
slightly lower BMDL between the two models; thus the Weibull was selected.
1
B-l
DRAFT - DO NOT CITE OR QUOTE
-------
1 As shown in Table B-3, in attempting to model the incidence of hepatocellular
2 cytoplasmic vacuolization in female rats with all six dose groups included, the BMDS failed to
3 generate any output because response was not a monotonically increasing function of dose (i.e.,
4 the response in the penultimate dose group was 40%, while the response in the highest dose
5 group was 0). A key underlying assumption for the fitting of the dichotomous models in BMDS
6 is that response must be a monotonically non-decreasing function of dose. Therefore, the highest
7 dose group was dropped and the models were fit to the data again. In this instance, the chi-
8 square goodness-of-fit test showed that all models exhibited inadequate fit (i.e., p < 0.1). Finally,
9 in an attempt to find a model that fit, the two highest dose groups were dropped and the models
10 were refit to these data. In this case, all of the models exhibited adequate fit, except for the one-
11 and two-stage multistage models (p > 0.10).
12 Of the models exhibiting adequate fit, a "best-fit" model was selected consistent with the
13 EPA's Benchmark Dose Technical Guidance Document (USEPA 2000), as follows. If the
14 BMDL estimates from the models exhibiting adequate fit were "sufficiently close," then the
15 model with the lowest AIC is to be used to estimate the BMDL from which the POD will be
16 derived. In this particular case, as explained in the footnote in Table B-3, BMDLs for models
17 with adequate fit differed by less than threefold. Among these models, the log-probit and
18 Weibull models shared the lowest AIC, and thus the average BMDLio from these two models
19 (i.e., 19.36 mg/kg-day) was used to derive a possible POD. The standard BMDS outputs from
20 the log-probit and Weibull models are displayed below.
B-2 DRAFT - DO NOT CITE OR QUOTE
-------
LogProbit Model with 0.95 Confidence Level
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
C
o
•
0.8
0.6
0.4
0.2
LogProbit
BMDL
BMD
10
20
30
40
dose
50
60
70
80
11:5403/302010
Probit Model. (Version: 3.1; Date: 05/16/2008)
Input Data File:
C:\USEPA\lRlS\TCE\NTP\hepcytvac\female\lnp_hepcytvacF2HDD_logprobit.(d)
Gnuplot Plotting File:
C:\USEPA\lRlS\TCE\NTP\hepcytvac\female\lnp_hepcytvacF2HDD_logprobit.plt
Tue Mar 30 12:54:34 2010
HMDS Model Run
The form of the probability function is:
P[response] = Background
+ (1-Background) * CumNorm(Intercept+Slope*Log(Dose)
where CumNorm(.) is the cumulative normal distribution function
Dependent variable = incidence
Independent variable = dose
Slope parameter is restricted as slope >= 1
Total number of observations = 4
Total number of records with missing values
= 0
B-3
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
c
o
9
10
11
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
53
J -J
54
55
56
57
58
59
60
61
62
63
64
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
User has chosen the log transformed model
Default Initial (and Specified) Parameter Valu
background = 0
intercept = -8.43383
slope = 2.43905
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter (s) -background -slope
have been estimated at a boundary point, or have been specified
and do not appear in the correlation matrix )
intercept
intercept 1
Parameter Estimates
95.0% Wald Confidence I
Variable Estimate Std. Err. Lower Conf. Limit Upper
background 0 NA
intercept -60.1746 2420.13 -4803.54
slope 18 NA
NA - Indicates that this parameter has hit a bound
implied by some inequality constraint and thus
has no standard error.
Analysis of Deviance Table
Model Log (likelihood) # Param's Deviance Test d.f. P- value
Full model 0 4
Fitted model -4 .43789e-009 1 8.87578e-009 3 1
Reduced model -27.7259 1 55.4518 3 <.0001
AIC: 2
Goodness of Fit
Dose Est. Prob. Expected Observed Size
0.0000 0.0000 0.000 0.000 10
20.0000 0.0000 0.000 0.000 10 -o.ooo
40.0000 1.0000 10.000 lO.ooo 10 o.ooo
80.0000 1.0000 10.000 10.000 10
ChiA2 = 0.00 d.f. = 3 P-value = 1.0000
Benchmark Dose Computation
Specified effect = 0.1
es
by the user
nterval
Conf. Limit
4683 .19
Scaled
Residual
0.000
0.000
B-4
DRAFT - DO NOT CITE OR QUOTE
-------
1
2 Risk Type = Extra risk
3
4 Confidence level = 0.95
5
6 BMD = 26.3597
7
8 BMDL = 19.557
B-5 DRAFT - DO NOT CITE OR QUOTE
-------
Weibull Model with 0.95 Confidence Level
2
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
C
o
•
0.8
0.6
0.4
0.2
Weibull
BMDL
BMD
10
20
30
40
dose
50
60
70
80
11:5403/302010
Weibull Model using Weibull Model (Version: 2.12; Date: 05/16/2008)
Input Data File:
C:\USEPA\lRlS\TCE\NTP\hepcytvac\female\wei_hepcytvacF2HDD_weibull.(d)
Gnuplot Plotting File:
C:\USEPA\lRlS\TCE\NTP\hepcytvac\female\wei_hepcytvacF2HDD_weibull.plt
Tue Mar 30 12:54:37 2010
HMDS Model Run
The form of the probability function is:
P[response] = background + (1-background)*[1-EXP(-slope*doseApower)]
Dependent variable = incidence
Independent variable = dose
Power parameter is restricted as power >=1
Total number of observations = 4
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
B-6
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
9
10
11
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
T-O
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
Default Initial (and Specified) Parameter Valu
Background = 0.0454545
Slope = 0.00369372
Power = 1.53227
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter (s) -Background -Power
have been estimated at a boundary point, or have been specified
and do not appear in the correlation matrix )
Slope
Slope -1.$
Parameter Estimates
95.0% Wald Confidence I
Variable Estimate Std. Err. Lower Conf. Limit Upper
Background 0 NA
Slope 1.815596-028 1 . #QNAN 1 . #QNAN
Power 18 NA
NA - Indicates that this parameter has hit a bound
implied by some inequality constraint and thus
has no standard error.
Analysis of Deviance Table
Model Log (likelihood) # Param's Deviance Test d.f. P- value
Full model 0 4
Fitted model -0.000514093 1 0.00102819 3 1
Reduced model -27.7259 1 55.4518 3 <.0001
AIC: 2.00103
Goodness of Fit
Dose Est. Prob. Expected Observed Size
0.0000 0.0000 0.000 0.000 10
20.0000 0.0000 0.000 0.000 10
40.0000 1.0000 10.000 10.000 10
80.0000 1.0000 10.000 10.000 10
ChiA2 = 0.00 d.f. = 3 P-value = 1.0000
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Extra risk
es
by the user
nterval
Conf. Limit
1 . #QNAN
Scaled
Residual
0.000
-0.022
0.006
0.000
B-7
DRAFT - DO NOT CITE OR QUOTE
-------
1 Confidence level = 0.95
2
3 HMD = 30.681
4
5 BMDL = 19.1631
6
B-8 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
Continuous Endpoints
Organ weight and serum chemistry changes in male and female rats (NTP, 2004)
Table B-4. Selected organ weight and serum chemistry changes in male and female
F344 rats administered l,l?2,2-tetrachlroethane in the diet for 14 weeks
Endpoint
Absolute liver wt
(g)
Relative liver wt
(mg organ wt / g
body wt)
Serum ALT
activity (IU/L)
Serum SDH
activity (IU/L)
Serum bile acid
levels (umol/L)
Sex
M
F
M
F
M
F
M
F
M
F
Dose (mg/kg-day)
0
12.74 ±0.26a
6.84±0.17
34.79 ±0.42
35.07 ±0.56
48 ±2
46 ±2
23 ±1
27 ±1
29.2 ±2.9
37.0±7.1
20
12. 99 ±0.35
7.03 ±0.13
36.72 ±0.44
36.69 ±0.36
49 ±2
42 ±1
27 ±1
27 ±1
27.5 ±2.7
46.6 ±6.5
40
14.47 ±0.44
7.14±0.16
41.03 ±0.85
37.84±0.51
53±2
41±2
26 ±2
28 ±2
27.2 ±2.7
39.1 ±5.6
80
15. 54 ±0.40
7. 80 ±0.08
45.61 ±0.52
44.20 ±0.27
69 ±3
49 ±2
31±1
25 ±1
35.9±3.9
36. 3 ±3. 9
170
11. 60 ±0.44
6.66 ±0.22
44.68 ±0.45
48.03 ±0.89
115±8
112±7
47 ±2
45 ±3
92.0 ±16.6
39. 3 ±7.9
320
6.57±0.18
4.94±0.12
52.23 ±1.42
58.40 ±1.42
292 ±18
339±18
74 ±4
82 ±3
332.4 ±47.4
321. 5 ±50.6
9
10
11
12
13
14
15
16
17
18
19
20
21
22
"Values are means ± SE for 10 animals.
Source: NTP (2004).
All available continuous models in theEPA's BMDS (version 2.1.1) were fit to each of
the endpoints listed in Table B-4 for both male and female rats administered 1,1,2,2-tetrachloro-
ethane in the diet for 14 weeks. BMDs and their 95 percent lower confidence limits (i.e.,
BMDLs) associated with a change in the response of one standard deviation from the control
were estimated by each model. The results of this BMD modeling for male and female rats are
summarized in Tables B-5 through B-14. Following each table is the BMDS output for the
selected model.
The model fitting procedure for continuous data was as follows. The simplest model
(linear) is first applied to the data while assuming constant variance. If the data are consistent
with the assumption of constant variance (p > 0.1), then the fit of the linear model to the means is
evaluated and the polynomial, power, and Hill models are fit to the data while assuming constant
variance. In accordance with U.S. EPA (2000) guidance, BMDs and BMDLs are estimated
employing a BMR that represents a change of 1 standard deviation from the control. Adequate
model fit is judged primarily by the goodness-of-fit/>-value (p > 0.1), but visual inspection of the
dose-response curve and the examination of scaled residual at the data point (except the control)
closest to the predefined BMR also play a role. If the test for constant variance is negative, the
linear model is run again while applying the power model integrated into BMDS to account for
B-9
DRAFT - DO NOT CITE OR QUOTE
-------
1 nonhomogeneous variance. If the nonhomogeneous variance model provides an adequate fit (p >
2 0.1) to the variance data, then the fit of the linear model to the means is evaluated and the
3 polynomial, power, and Hill models are fit to the data and evaluated while the variance model is
4 applied. If no model provides adequate fit to the data based on these criteria, then the highest
5 dose is dropped, if appropriate, and the continuous modeling procedure is repeated.
6
7 Absolute liver weights in male and female rats (Tables B-5 andB-6)
8 No adequate fit to the data for absolute liver weight in males or females was achieved
9 until the two highest doses were dropped. After dropping the two highest doses, the assumption
10 of constant variance was met and all models provided adequate fit (except the Hill model, which
11 has too many parameters for the number of remaining data points). BMDL estimates across the
12 models with adequate fit differed by less than threefold. In accordance with U.S. EPA (2000),
13 the model with the lowest AIC (linear, for both males and females) was selected as the basis for
14 the BMDiso and BMDLiso estimates for these endpoints (respectively, 30 and 23 mg/kg-day for
15 males, and 36 and 26 mg/kg-day for females).
16
B-10 DRAFT - DO NOT CITE OR QUOTE
-------
Table B-5. Summary of benchmark dose modeling results for absolute liver weight
in male rats
Model
Test for
significant
difference
/>-valuea
Variance
/>-valueb
Means
/>-valueb
Scaled
residuals of
interest0
AIC
BMD1SD
(mg/kg-
day)
BMDL1SD
(mg/kg-
day)
All dose groups included
Constant variance
Lineard
O.0001
0.07
O.0001
NA
198.13
NA
3,925.92
Non-constant variance
Hille
Lineard
Polynomial (2 -degree/
Polynomial (3 -degree)
Polynomial (4-degree)d
Polynomial (5-degree)d
Power6
O.0001
O.0001
<0.0001
O.0001
<0.0001
<0.0001
<0.0001
0.39
0.39
0.39
0.39
0.39
0.39
0.39
O.0001
O.0001
O.0001
<0.0001
<0.0001
O.0001
<0.0001
-0.7/1.81
NA
NA
NA
NA
NA
-1.43/0.08
160.48
200.13
200.13
200.13
200.13
200.13
106.77
36.49
NA
NA
NA
NA
NA
173.92
NA
10.43
10.45
733.03
595.06
533.37
141.52
Highest dose group dropped
Constant variance
Hill6
Lineard
Polynomial (2-degree)d
Polynomial (3 -degree)
Polynomial (4-degree)d
Power6
<0.0001
O.0001
<0.0001
<0.0001
<0.0001
<0.0001
0.49
0.49
0.49
0.49
0.49
0.49
<0.0001
<0.0001
<0.0001
O.0001
<0.0001
<0.0001
3.3/0.00
NA
NA
NA
NA
3.3/0.00
100.95
112.67
112.67
112.67
112.67
98.95
165.58
NA
NA
NA
NA
166.09
94.36
606.09
416.42
326.66
282.11
145.65
Two highest dose groups dropped
Constant variance
Hill6
LinearAf
Power6
O.0001
<0.0001
<0.0001
0.41
0.41
0.41
NA
0.32
0.13
0.00/0.00
-1.07/0.97
-1.03/1.01
57.97
56.26
58.25
32.10
30.40
31.30
20.62
22.92
22.93
AIC = Akaike Information Criterion; BMD = maximum likelihood estimate of the dose associated with the selected
benchmark response; BMDL = 95% lower confidence limit on the BMD; NA = not applicable (BMD/BMDL computation
failed or insufficient degrees of freedom to fit model); SD = standard deviation
aValues >0.05 fail to meet conventional goodness-of-fit criteria.
Values <0.10 fail to meet conventional goodness-of-fit criteria.
°Scaled residuals at doses immediately below and immediately above the benchmark dose.
dCoefficients restricted to be positive.
ePower restricted to >1.
fBest-fitting model is displayed in boldface type. BMDLs for models providing adequate fit differed by < threefold, so the
model with the lowest AIC was selected.
B-ll
DRAFT - DO NOT CITE OR QUOTE
-------
Linear Model with 0.95 Confidence Level
c
o
Q_
(/)
0
ce
c
ro
0
16
15
14
13
12
Linear
BMDL BMD
10 20
30
40
dose
50 60 70 80
14:1203/262010
B-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
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
Polynomial Model. (Version: 2.13; Date: 04/08/2008)
Input Data File:
C:\USEPA\lRlS\TCE\NTP\abslivwt\male\lin_abslivwtM2HDD_linear.(d)
Gnuplot Plotting File:
C:\USEPA\IRIS\TCE\NTP\abslivwt\male\lin_abslivwtM2HDD_linear.plt
Fri Mar 26 15:12:39 2010
HMDS Model Run
The form of the response function is:
Y[dose] = beta 0 + beta l*dose + beta 2*doseA2 + ...
Dependent variable = mean
Independent variable = dose
rho is set to 0
The polynomial coefficients are restricted to be positive
A constant variance model is fit
Total number of dose groups = 4
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial Parameter Values
alpha = 1.35605
rho = 0 Specified
beta_0 = 12.626
beta 1 = 0.0374
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -rho
have been estimated at a boundary point, or have been
specified by the user,
and do not appear in the correlation matrix )
alpha
alpha 1
beta_0 -6.9e-010
beta 1 -4.8e-011
Variable
alpha
beta 0
Estimate
1.29235
12.626
beta_0 beta_l
-6.96-010 -4.88-011
1 -0.76
-0.76 1
Parameter Estimates
95.0% Wald Confidence Interval
Std. Err. Lower Conf. Limit Upper Conf. Limit
0.288979 0.725966 1.85874
0.278462 12.0802 13.1718
B-13
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
beta 1
0.0374
0.00607655
0.0254902
0.0493098
Table of Data and Estimated Values of Interest
Dose N Obs Mean Est Mean Obs Std Dev Est Std Dev Scaled Res.
0
20
40
80
10
10
10
10
12 .7
13
14 .5
15 .5
12 .6
13 .4
14 .1
15 .6
0.82
1.11
1.39
1.26
1.14
1 .14
1 .14
1 .14
0.317
-1 .07
0. 968
-0.217
Model Descriptions for likelihoods calculated
Model Al: Yij = Mu(i) + e(ij)
Var{e(ij) } = SigmaA2
Model A2: Yij = Mu(i) + e(ij)
Var{e(ij) } = Sigma(i) A2
Model A3: Yij = Mu(i) + e(ij)
Var{e(ij)} = SigmaA2
Model A3 uses any fixed variance parameters that
were specified by the user
Yi = Mu + e(i)
Var{e (i)} = SigmaA2
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
MO
we
Model
Test 1
Test 2
Test 3
Test 4
(Note:
Likelihoods of Interest
Model
Al
A2
A3
fitted
R
Log (likelihood)
-23.984311
-22.556035
-23.984311
-25.129323
-38 .455553
# Param' s
5
8
5
3
2
AIC
57.968622
61.112070
57.968622
56.258645
80.911106
Explanation of Tests
Do responses and/or variances differ among Dose levels?
(A2 vs. R)
Are Variances Homogeneous? (Al vs A2)
Are variances adequately modeled? (A2 vs. A3)
Does the Model for the Mean Fit? (A3 vs. fitted)
When rho=0 the results of Test 3 and Test 2 will be the same.)
B-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
36
37
38
Test
Test 1
Test 2
Test 3
Test 4
Tests of Interest
-2*log (Likelihood Ratio) Test df
31.799
2.85655
2.85655
2.29002
p-value
<.0001
0.4143
0.4143
0.3182
The p-value for Test 1 is less than .05. There appears to be a
difference between response and/or variances among the dose levels
It seems appropriate to model the data
The p-value for Test 2 is greater than .1. A homogeneous variance
model appears to be appropriate here
The p-value for Test 3 is greater than .1.
to be appropriate here
The p-value for Test 4 is greater than .1.
to adequately describe the data
The modeled variance appears
The model chosen seems
Benchmark Dose Computation
Specified effect = 1
Risk Type = Estimated standard deviations from the control mean
Confidence level = 0.95
HMD = 30.3962
BMDL =
22.9198
B-15
DRAFT - DO NOT CITE OR QUOTE
-------
Table B-6. Summary of benchmark dose modeling results for absolute liver weight
in female rats
Model
Test for
significant
difference
/>-valuea
Variance
/>-value
Means
/>-value
Scaled
residuals of
interest0
AIC
BMD1SD
(mg/kg-
day)
BMDL1SD
(mg/kg-
day)
All dose groups included
Constant variance
Linear
O.0001
0.05
O.0001
NA
62.98
NA
3,632.46
Non-constant variance
Linear
O.0001
0.02
O.0001
NA
64.98
NA
24.07
Highest dose group dropped
Constant variance
Lineard
O.0001
0.04
O.0001
NA
5.69
NA
377.10
Non-constant variance
Hille
Lineard
Polynomial (2 -degree)
Polynomial (3-degree)d
Polynomial (4-degree)d
Power6
O.0001
O.0001
O.0001
<0.0001
O.0001
<0.0001
0.84
0.84
0.84
0.84
0.84
0.84
<0.0001
<0.0001
O.0001
<0.0001
<0.0001
<0.0001
0.00f
NA
NA
NA
NA
0.00f
4.52
7.69
7.69
7.69
7.69
2.52
170.20
NA
NA
NA
NA
170.19
NA
397.23
343.87
290.54
67.91
153.95
Two highest dose groups dropped
Constant variance
Hill6
Linear11*
Polynomial (2-degree)d
Polynomial (3-degree)d
Power6
<0.0001
<0.0001
<0.0001
O.0001
<0.0001
0.11
0.11
0.11
0.11
0.11
NA
0.55
0.63
0.71
0.57
-0.30/0.05
0.05/-0.91
-0.28/0.05
-0.19/0.02
-0.30/0.05
-19.17
-22.27
-21.25
-21.35
-21.17
48.28
35.62
48.21
49.83
48.28
25.37
26.10
27.58
27.77
27.44
AIC = Akaike Information Criterion; BMD = maximum likelihood estimate of the dose associated with the selected
benchmark response; BMDL = 95% lower confidence limit on the BMD; NA = not applicable (BMD/BMDL computation
failed or insufficient degrees of freedom to fit model); SD = standard deviation
aValues >0.05 fail to meet conventional goodness-of-fit criteria.
Values <0.10 fail to meet conventional goodness-of-fit criteria.
°Scaled residuals at doses immediately below and immediately above the benchmark dose.
dCoefficients restricted to be positive.
ePower restricted to >1.
Residual at highest dose tested.
gBest-fitting model displayed in boldface type. BMDLs for models providing adequate fit differed by < threefold, so the
model with the lowest AIC was selected.
B-16
DRAFT - DO NOT CITE OR QUOTE
-------
Linear Model with 0.95 Confidence Level
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
c
o
Q_
(/)
0
ce
c
ro
0
7.8
7.6
7.4
7.2
6.8
6.6
6.4
14:5803/262010
70
80
Polynomial Model. (Version: 2.13; Date: 04/08/2008)
Input Data File:
C:\USEPA\lRlS\TCE\NTP\abslivwt\female\lin_abslivwtF2HDD_linear.(d)
Gnuplot Plotting File:
C:\USEPA\lRlS\TCE\NTP\abslivwt\female\lin_abslivwtF2HDD_linear.plt
Fri Mar 26 15:58:54 2010
HMDS Model Run
The form of the response function is:
Y[dose] = beta 0 + beta l*dose + beta 2*doseA2
Dependent variable = mean
Independent variable = dose
rho is set to 0
The polynomial coefficients are restricted to be positive
A constant variance model is fit
Total number of dose groups = 4
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
B-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
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
Parameter Convergence has been set to: le-008
Default Initial Parameter Values
alpha = 0.195575
rho = 0 Specified
beta_0 = 6.784
beta 1 = 0.0119571
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -rho
have been estimated at a boundary point, or have been specified by the user,
and do not appear in the correlation matrix )
alpha
beta_0
beta 1
alpha
1
-8e-009
8.2e-009
beta_0
-8e-009
1
-0.76
beta_l
8.2e-009
-0.76
1
Variable
alpha
beta_0
beta 1
Estimate
0 .181435
6.784
0. 0119571
Parameter Estimates
95.0% Wald Confidence Interval
Std. Err. Lower Conf. Limit Upper Conf. Limit
0.04057 0.101919 0.26095
0.104336 6.5795 6.9885
0.00227681 0.00749468 0.0164196
Table of Data and Estimated Values of Interest
Dose N Obs Mean Est Mean Obs Std Dev Est Std Dev Scaled Res.
0
20
40
80
10
10
10
10
6.84
7. 03
7.14
7 .8
6.78
7. 02
7.26
7.74
0.54
0.41
0.51
0.25
0.426
0.426
0.426
0.426
0.416
0 .0509
-0. 908
0.441
Model Descriptions for likelihoods calculated
Model Al: Yij = Mu(i) + e(ij)
Var{e(ij) } = SigmaA2
Model A2: Yij = Mu(i) + e(ij)
Var{e(ij) } = Sigma(i)A2
Model A3: Yij = Mu(i) + e(ij)
Var{e(ij)} = SigmaA2
Model A3 uses any fixed variance parameters that
were specified by the user
Model R:
Yi = Mu
B-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
Test 1
Test 2
Test 3
Test 4
(Note:
Test
Var{e (i) } = SigmaA2
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
Likelihoods of Interest
Log (likelihood)
14 .743437
17 .781442
14 .743437
14.137196
3.648385
# Param' s
5
8
5
3
2
AIC
-19.486874
-19.562884
-19.486874
-22.274391
-3.296770
Model
Al
A2
A3
fitted
R
Explanation of Tests
Do responses and/or variances differ among Dose levels?
(A2 vs. R)
Are Variances Homogeneous? (Al vs A2)
Are variances adequately modeled? (A2 vs. A3)
Does the Model for the Mean Fit? (A3 vs. fitted)
When rho=0 the results of Test 3 and Test 2 will be the same.)
Tests of Interest
-2*log (Likelihood Ratio) Test df
p-value
28 .2661
6.07601
6.07601
1.21248
6
3
3
2
<.0001
0.108
0.108
0.5454
Test 1
Test 2
Test 3
Test 4
The p-value for Test 1 is less than .05. There appears to be a
difference between response and/or variances among the dose levels
It seems appropriate to model the data
The p-value for Test 2 is greater than .1.
model appears to be appropriate here
The p-value for Test 3 is greater than .1.
to be appropriate here
The p-value for Test 4 is greater than .1.
to adequately describe the data
A homogeneous variance
The modeled variance appears
The model chosen seems
B-19
DRAFT - DO NOT CITE OR QUOTE
-------
1
2 Benchmark Dose Computation
3
4 Specified effect = 1
5
6 Risk Type = Estimated standard deviations from the control mean
7
8 Confidence level = 0.95
9
10 HMD = 35.6232
11
12
13 BMDL = 26.1046
14
15
16 Relative liver weights in male and female rats (Tables B-7 and B-8)
17 No model provided an adequate fit to the relative liver weight data in male rats even after
18 dropping the two highest dose groups. Therefore, these data are considered unsuitable for BMD
19 modeling. For the relative liver weight data in females, the assumption of constant variance was
20 satisfied and the power and 2- and 3-degree polynomial models provided adequate fit to the data
21 after the highest two dose groups were dropped. BMDL estimates across these models differed
22 by less than threefold. In accordance with U. S. EPA (2000), the model with the lowest AIC (3-
23 degree polynomial) was selected as the basis for the BMDiso and BMDLiso estimates of 22 and
24 15 mg/kg-day, respectively, for this endpoint.
B-20 DRAFT - DO NOT CITE OR QUOTE
-------
Table B-7. Summary of benchmark dose modeling results for relative liver weight
in male rats
Model
Test for
significant
difference
/>-valuea
Variance
/>-valueb
Means
/>-valueb
Scaled
residuals of
interest0
AIC
BMD1SD
(mg/kg-
day)
BMDL1SD
(mg/kg-
day)
All dose groups included
Constant variance
Linear
O.0001
<0.0001
O.0001
1.6/4.15
208.74
68.02
56.64
Non-constant variance
Lineard
O.0001
0.03
O.0001
1.93/4.36
208.89
55.05
37.77
Highest dose group dropped
Constant variance
Lineard
O.0001
0.09
<0.0001
1.84/4.25
165.27
51.62
40.95
Non-constant variance
Lineard
O.0001
0.06
<0.0001
-0.79/-0.95
157.11
12.93
8.10
Two highest dose groups dropped
Constant variance
Lineard
O.0001
0.07
0.15
0.25/-1.24
94.60
13.14
10.76
Non-constant variance
Linear
O.0001
0.08
0.09
0.35/-1.32
95.74
10.97
7.77
3 highest doses dropped
Constant variance
Linear
O.0001
0.03
0.10
0.66/-1.32
74.39
12.16
9.27
Non-constant variance
Hill6
Linear
Polynomial (2-degree)d
Power"
NA
O.0001
O.0001
O.0001
0.52
0.52
0.52
0.05
NA
NA
0.45/-1.32
-0.07/0.12
-0.07/0.12
71.18
69.32
69.32
8.47
15.27
15.50
6.05
8.46
9.02
AIC = Akaike Information Criterion; BMD = maximum likelihood estimate of the dose associated with the selected
benchmark response; BMDL = 95% lower confidence limit on the BMD; NA = not applicable (insufficient degrees of
freedom to fit the model); SD = standard deviation
aValues >0.05 fail to meet conventional goodness-of-fit criteria.
Values <0.10 fail to meet conventional goodness-of-fit criteria.
°Scaled residuals at doses immediately below and immediately above the benchmark dose.
dCoefficients restricted to be positive.
ePower restricted to >1.
B-21
DRAFT - DO NOT CITE OR QUOTE
-------
Table B-8. Summary of benchmark dose modeling results for relative liver weight in
female rats
Model
Test for
significant
difference
/>-valuea
Variance
/>-valueb
Means
/>-valueb
Scaled
residuals of
interest0
AIC
BMD1SD
(mg/kg-day)
BMDL1SD
(mg/kg-day)
All dose groups included
Constant variance
Linear
O.0001
<0.0001
0.01
-0.66/-1.01
181.20
36.16
30.95
Non-constant variance
Lineard
O.0001
0.01
O.0001
<-10/<-10
6.00
0.003
NA
Highest dose group dropped
Constant variance
Lineard
O.0001
0.002
<0.0001
-0.52/-1.19
129.06
26.16
21.87
Non-constant variance
Lineard
O.0001
0.01
0.001
-0.12/-0.30
123.73
16.52
12.39
Two highest dose groups dropped
Constant variance
Hille
Lineard
Polynomial (2 -degree)
Polynomial (3-degree)A'f
Power6
O.0001
O.0001
<0.0001
<0.0001
<0.0001
0.11
0.11
0.11
0.11
0.11
NA
0.005
0.22
0.38
0.15
1.12/-0.72
1.31/-0.09
0.94/-0.70
0.69/-0.43
1.12/-0.72
74.32
78.98
71.76
70.98
72.32
25.33
13.20
23.57
21.90
25.31
17.12
10.81
15.68
14.78
17.12
AIC = Akaike Information Criterion; BMD = maximum likelihood estimate of the dose associated with the selected benchmark
response; BMDL = 95% lower confidence limit on the BMD; NA= not applicable (BMD/BMDL computation failed or
insufficient degrees of freedom to fit model); SD = standard deviation
aValues >0.05 fail to meet conventional goodness-of-fit criteria.
bValues <0.10 fail to meet conventional goodness-of-fit criteria.
°Scaled residuals at doses immediately below and immediately above the benchmark dose.
Coefficients restricted to be positive.
"Power restricted to >1.
Best-fitting model is displayed in boldface type. BMDLs for models providing adequate fit differed by < threefold, so the model
with the lowest AIC was selected.
B-22
DRAFT - DO NOT CITE OR QUOTE
-------
Polynomial Model with 0.95 Confidence Level
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
c
o
Q_
(/)
0
ce
c
ro
0
44
42
40
38
36
34
Polynomial
BMDL
BMD
10
20
30
40
dose
50
60
70
80
08:3403/292010
Polynomial Model. (Version: 2.13; Date: 04/08/2008)
Input Data File:
C:\USEPA\IRIS\TCE\NTP\rellivwt\female\ply_rellivwtF2HDD_Poly_3.(d)
Gnuplot Plotting File:
C:\USEPA\IRIS\TCE\NTP\rellivwt\female\ply_rellivwtF2HDD_Poly_3.pit
Mon Mar 29 09:34:20 2010
HMDS Model Run
The form of the response function is:
Y[dose] = beta 0 + beta l*dose + beta 2*doseA2
Dependent variable = mean
Independent variable = dose
rho is set to 0
The polynomial coefficients are restricted to be positive
A constant variance model is fit
Total number of dose groups = 4
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
B-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
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
Parameter Convergence has been set to: le-008
Default Initial Parameter Values
beta_0 =
beta 1 =
alpha
rho
35.07
0.115542
beta_2
beta 3
1.93677
0
0
= 2.848966-005
Specified
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -rho -beta_2
have been estimated at a boundary point, or have been specified by the user,
and do not appear in the correlation matrix )
alpha beta_0 beta_l beta_3
alpha 1 -6e-009 3.2e-oo9 -i.ve-009
beta_0 -6e-009 1 -0.76 0.56
beta_l 3.2e-009 -0.76 1 -0.92
beta_3 -1.76-009 0.56 -0.92 1
Parameter Estimates
Variable
alpha 1.77636
beta_0 35.1967
beta_l
beta_2
beta 3
Estimate Std. Err.
0.397207 0.997852
0.395218 34.4221
0.0567055 0.0185417
1.598986-026 NA
8.68894e-006 2.57808e-006
95.0% Wald Confidence Interval
Lower Conf. Limit Upper Conf. Limit
2 .55487
35.9713
0.0203645 0.0930465
3.636e-006
1.374196-005
NA - Indicates that this parameter has hit a bound
implied by some inequality constraint and thus
has no standard error.
Table of Data and Estimated Values of Interest
Dose N Obs Mean Est Mean Obs Std Dev Est Std Dev Scaled Res.
0
20
40
80
10
10
10
10
35 .1
36 .7
37 .8
44 .2
35 .2
36 .4
38
44 .2
1.77
1.14
1.61
0.85
1.33
1.33
1.33
1 .33
-0.301
0.687
-0 .43
0. 043
Model Descriptions for likelihoods calculated
Model Al: Yij
var{e(ij) }
Mu(i) + e (ij)
SigmaA2
B-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
Model A2 :
Yij = Mud) + e(ij)
Var{e(ij)} = Sigma
Model A3: Yij = Mu(i) + e(ij)
Var{e (ij ) } = SigmaA2
Model A3 uses any fixed variance parameters that
were specified by the user
Model
R: Yi = Mu + e (i)
Var{e (i) } = SigmaA2
Likelihoods of Interest
Model
Al
A2
A3
fitted
R
Log (likelihood)
-31 .113274
-28.050020
-31 .113274
-31.491356
-72.394938
# Param's
5
8
5
4
2
AIC
72 .226548
72 .100041
72 .226548
70. 982711
148.789876
Explanation of Tests
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
53
54
55
56
57
58
59
60
Test 1: Do responses and/or variances differ
(A2 vs. R)
Test 2: Are Variances Homogeneous? (Al vs A2)
Test 3: Are variances adequately modeled? (A2
Test 4: Does the Model for the Mean Fit? (A3
among Dose levels?
vs. A3)
vs. fitted)
(Note: When rho=0 the results of Test 3 and Test 2 will be the same.)
Tests of Interest
Test -2*log (Likelihood Ratio) Test df
Test 1 88.6898 6
Test 2 6.12651 3
Test 3 6.12651 3
Test 4 0.756163 1
The p-value for Test 1 is less than .05. There
p-value
<.0001
0.1056
0.1056
0.3845
appears to be a
difference between response and/or variances among the dose levels
It seems appropriate to model the data
The p-value for Test 2 is greater than .1. A homogeneous variance
model appears to be appropriate here
The p-value for Test 3 is greater than .1. The
to be appropriate here
The p-value for Test 4 is greater than .1. The
to adequately describe the data
Benchmark Dose Computation
Specified effect = 1
modeled variance appears
model chosen seems
B-25
DRAFT - DO NOT CITE OR QUOTE
-------
1 Risk Type = Estimated standard deviations from the control mean
2
3 Confidence level = 0.95
4
5 HMD = 21.8955
6
7
8 BMDL = 14.7785
9
10
11 Serum ALT activity in male and female rats (Tables B-9 and B-10)
12 All doses were retained in the BMD modeling of serum ALT in males and females. The
13 assumption of constant variance was not upheld for either dataset, but in each case, the power
14 model for variance built into the BMDS provided adequate fit to the variance data. With the
15 variance model applied, adequate fit to the means was provided by the Hill, power, and 2- and 5-
16 degree polynomial models for the males, and by the Hill model alone for the females. For the
17 males, estimated BMDLs from the adequately fitting models differed by less than threefold. In
18 accordance with U.S. EPA (2000), the model with the lowest AIC (i.e., 2-degree polynomial)
19 was selected as the basis for the BMDiso and BMDLiso estimates of 41 and 26 mg/kg-day. For
20 the females, BMDiso and BMDLiso estimates of 82 and 69 mg/kg-day were based on the Hill
21 model.
B-26 DRAFT - DO NOT CITE OR QUOTE
-------
Table B-9. Summary of benchmark dose modeling results for serum ALT activity
in male rats
Model
Test for significant
difference />-valuea
Variance
/>-valueb
Means
/>-valueb
Scaled
residuals of
interest0
AIC
BMD1SD
(mg/kg-
day)
BMDL1SD
(mg/kg-
day)
All dose groups included
Constant variance
Linear
O.0001
<0.0001
O.0001
-0.19/-1.55
486.88
43.91
37.37
Non-constant variance
Hille
Linear
Polynomial (2-degree) '
Polynomial (3 -degree)
Polynomial (4-degree)d
Polynomial (5-degree)d
Power6
O.0001
O.0001
<0.0001
O.0001
O.0001
<0.0001
<0.0001
0.72
0.72
0. 72
0.72
0.72
0.72
0.72
0.51
O.0001
0.84
<0.0001
<0.0001
0.47
0.73
0.10/0.77
>10
-0.21/1.00
>10
NA
-0.14/1.06
-0.11/0.76
370.02
6.00
366.08
10.00
606.63
370.17
367.96
42.19
0.00
40.98
0.00
NA
40.62
41.97
34.33
NA
26.35
NA
28.22
26.19
32.24
Akaike Information Criterion; BMD = maximum likelihood estimate of the dose associated with the selected benchmark
response; BMDL = 95% lower confidence limit on the BMD; NA= not applicable (BMD/BMDL computation failed); SD =
standard deviation
aValues >0.05 fail to meet conventional goodness-of-fit criteria.
Values <0.10 fail to meet conventional goodness-of-fit criteria.
°Scaled residuals at doses immediately below and immediately above the benchmark dose.
Coefficients restricted to be positive.
ePower restricted to >1.
fBest-fitting model is displayed in boldface type. BMDLs for models providing adequate fit differed by < threefold, so the
model with the lowest AIC was selected.
DRAFT - DO NOT CITE OR QUOTE
-------
Polynomial Model with 0.95 Confidence Level
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
c
o
Q_
(/)
0
ce
c
ro
0
350
300
250
200
150
100
50
Polynomial
BMDL
3MD
50
100
150
dose
200
250
300
09:5903/292010
Polynomial Model. (Version: 2.13; Date: 04/08/2008)
input Data File: C:\USEPA\lRlS\TCE\NTP\ALT\male\ply_ALTM_poly_2.(d)
Gnuplot Plotting File:
C:\USEPA\IRIS\TCE\NTP\ALT\male\ply_ALTM_poly_2.pit
Mon Mar 29 10:59:45 2010
HMDS Model Run
The form of the response function is:
Y[dose] = beta 0 + beta l*dose + beta 2*doseA2 + ...
Dependent variable = mean
Independent variable = dose
The polynomial coefficients are restricted to be positive
The variance is to be modeled as Var(i) = exp(lalpha + log(mean(i)) * rho)
Total number of dose groups = 6
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
B-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
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
Default Initial Parameter Values
lalpha = 6.52437
rho = 0
beta_0 = 48.8991
beta 1 = 0.00912505
beta_2 = 0.00233971
! ! ! Warning: optimum may not have been found. ! ! !
!!! You may want to try choosing different initial values. !!!
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter (s) -rho
have been estimated at a boundary point, or have been specified
and do not appear in the correlation matrix )
lalpha beta 0 beta 1 beta 2
lalpha 1 -0.0021 -0.015 0.027
beta_0 -0.0021 1 -0.71 0.49
beta_l -0.015 -0.71 1 -0.86
beta_2 0.027 0.49 -0.86 1
Parameter Estimates
95.0% Wald Confidence I
Variable Estimate Std. Err. Lower Conf. Limit Upper
lalpha -6.58334 0.182468 -6.94097
rho 2.62555 NA
beta_0 47.7312 1.57297 44.6483
beta 1 0.05625 0.0541054 -0.0497946
beta_2 0.00216953 0.000281829 0.00161716
NA - Indicates that this parameter has hit a bound
implied by some inequality constraint and thus
has no standard error.
Table of Data and Estimated Values of Interest
Dose N Obs Mean Est Mean Obs Std Dev Est Std Dev Scaled Res.
0 10 48 47.7 6.3 5.95 0.143
20 10 49 49.7 6.3 6.28 -0.365
40 10 53 53.5 6.3 6.9 -0.207
80 10 69 66.1 9.5 9.12 1
170 10 115 120 25.3 19.9 -0.792
320 10 292 288 56.9 62.9 0.206
Model Descriptions for likelihoods calculated
Model Al: Yij = Mu(i) + e(ij)
Var{e (ij ) } = SigmaA2
by the user
nterval
Conf. Limit
-6 .22571
50 .8142
0.162295
0 .0027219
B-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
Model A2 :
Yij = Mud) + e(ij)
Var{e(ij)} = Sigma
Model A3: Yij = Mu(i) + e(ij)
Var{e(ij)} = expdalpha + rho*ln (Mu (i) ) )
Model A3 uses any fixed variance parameters that
were specified by the user
Model R: Yi = Mu + e (i)
Var{e (i) } = SigmaA2
Likelihoods of Interest
Model
Al
A2
A3
fitted
R
Log (likelihood)
-222 .570247
-177 .293103
-178 .329731
-179.039110
-300.315008
# Param's
7
12
8
4
2
AIC
459. 140493
378.586206
372.659462
366.078220
604.630016
Explanation of Tests
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
53
54
55
56
57
58
59
60
Test 1: Do responses and/or variances differ among Dose levels?
(A2 vs. R)
Test 2: Are Variances Homogeneous? (Al vs A2)
Test 3: Are variances adequately modeled? (A2 vs. A3)
Test 4: Does the Model for the Mean Fit?
(Note: When rho=0 the results of Test 3
Tests of Interest
Test -2*log (Likelihood Ratio) Test
Test 1 246.044 10
Test 2 90.5543 5
Test 3 2.07326 4
Test 4 1.41876 4
The p-value for Test 1 is less than .05.
(A3 vs. fitted)
and Test 2 will be the same.)
df p-value
<.0001
<.0001
0.7223
0.8409
There appears to be a
difference between response and/or variances among the dose levels
It seems appropriate to model the data
The p-value for Test 2 is less than .1. A
model appears to be appropriate
The p-value for Test 3 is greater than .1.
to be appropriate here
The p-value for Test 4 is greater than .1.
to adequately describe the data
Benchmark Dose Computation
Specified effect = 1
Risk Type = Estimated standard
non-homogeneous variance
The modeled variance appears
The model chosen seems
deviations from the control me.
B-30
DRAFT - DO NOT CITE OR QUOTE
-------
1
2 Confidence level = 0.95
3
4 HMD = 40.9754
5
6
7 BMDL = 26.3459
B-31 DRAFT - DO NOT CITE OR QUOTE
-------
Table B-10. Summary of benchmark dose modeling results for serum ALT
activity in female rats
Model
Test for
significant
difference
/>-valuea
Variance
/>-valueb
Means
/>-valueb
Scaled
residuals of
interest0
AIC
BMD1SD
(mg/kg-
day)
BMDL1SD
(mg/kg-
day)
All dose groups included
Constant variance
Lineard
O.0001
<0.0001
O.0001
-0.12/2.54
512.92
45.04
38.30
Non-constant variance
HUf
Lineard
Polynomial (2 -degree/
Polynomial (3 -degree)
Polynomial (4-degree)d
Polynomial (5-degree)d
Power6
<0.0001
<0.0001
<0.0001
O.0001
<0.0001
<0.0001
<0.0001
0.23
0.23
0.23
0.23
0.23
0.23
0.23
0.16
<0.0001
<0.0001
O.0001
<0.0001
<0.0001
0.02
0.09/-0.29
0.79/3.84
-0.91/-0.16
-0.95/-0.20
-0.77/-0.40
-0.85/-0.14
-0.26/-1.58
351.50
444.14
413.32
415.39
392.73
432.77
355.84
82.49
142.23
65.95
71.30
71.75
79.16
64.07
68.61
12.12
19.55
15.90
22.50
13.16
55.45
AIC = Akaike Information Criterion; BMD = maximum likelihood estimate of the dose associated with the selected
benchmark response; BMDL = 95% lower confidence limit on the BMD; SD = standard deviation
aVallies >0.05 fail to meet conventional goodness-of-fit criteria.
bValues <0.10 fail to meet conventional goodness-of-fit criteria.
°Scaled residuals at doses immediately below and immediately above the benchmark dose.
, J J
Coefficients restricted to be positive.
ePower restricted to >1.
Best-fitting model is displayed in boldface type. In this case, Hill model was the only model that provided an adequate fit
to the data.
B-32
DRAFT - DO NOT CITE OR QUOTE
-------
Hill Model with 0.95 Confidence Level
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
c
o
Q_
(/)
0
ce
c
ro
0
400
350
300
250
200
150
100
50
300
10:0803/292010
Hill Model. (Version: 2.14; Date: 06/26/2008)
Input Data File: C:\USEPA\IRIS\TCE\NTP\ALT\female\hil_ALTF_Hill.(d)
Gnuplot Plotting File:
C:\USEPA\IRIS\TCE\NTP\ALT\female\hil_ALTF_Hill.pit
Mon Mar 29 11:08:43 2010
HMDS Model Run
The form of the response function is:
Y [dose] = intercept + v*doseAn/(kAn + dose^n)
Dependent variable = mean
Independent variable = dose
Power parameter restricted to be greater than 1
The variance is to be modeled as Var(i) = exp(lalpha + rho * In(mean(i)))
Total number of dose groups = 6
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
B-33
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
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
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
Default Initial Parameter Values
lalpha =
rho =
intercept =
v =
n =
k =
6 .46604
0
46
293
2 . 07344
416.806
Asymptotic Correlation Matrix of Parameter Estimates
lalpha
rho
intercept
lalpha
1
-0 .99
-0 .12
0.1
-0 .0074
0. 051
rho
-0 .99
1
0. 098
-0 .11
0 .0073
-0. 052
intercept
-0 .12
0. 098
1
-0 .41
0 .49
-0 .42
V
0 .1
-0.11
-0.41
1
-0.9
0 .98
n
-0. 0074
0. 0073
0.49
-0 .9
1
-0 .95
k
0 .051
-0 .052
-0.42
0. 98
-0. 95
1
Parameter Estimates
Variable
lalpha
rho
intercept
V
n
k
Estimate
-5.48513
2.36002
43 .8372
440. 049
3.71466
266.476
Std. Err.
1.18231
0 .272384
1. 06856
121.144
0 .661842
45 .4588
95.0% Wald Confidence Interval
Lower Conf. Limit Upper Conf. Limit
-7.80242 -3.16783
1.82615 2.89388
41.7428 45.9315
202.612 677.486
2.41747 5.01185
177.378 355.573
Table of Data and Estimated Values of Interest
Dose N Obs Mean Est Mean Obs Std Dev Est Std Dev Scaled Res.
0
20
40
80
170
320
10
10
10
10
10
10
46
42
44
49
112
339
43 .8
43 .9
44 .2
48 .8
114
336
6 .3
3.2
6.3
6.3
22 .1
56.9
5 .58
5 .58
5 .63
6 .33
17.1
61.6
1 .23
-1 .06
-0.124
0 .0904
-0 .29
0.159
Model Descriptions for likelihoods calculated
Model Al: Yij = Mu(i) + e(ij)
Var{e(ij) } = SigmaA2
Model A2: Yij = Mu(i) + e(ij)
Var{e(ij) } = Sigma(i)A2
Model A3: Yij = Mu(i) + e(ij)
Var{e(ij)} = exp(lalpha + rho*ln(Mu(i)))
Model A3 uses any fixed variance parameters that
B-34
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
we
Model
Test 1
Test 2
Test 3
Test 4
(Note:
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
were specified by the user
Yi = Mu + e(i)
Var{e(i)} = SigmaA2
Likelihoods of Interest
Log (likelihood)
-220.820465
-165 . 059425
-167.889045
-169.749216
-312 . 021870
# Param' s
7
12
8
6
2
AIC
455. 640931
354. 118851
351.778089
351.498431
628. 043741
Model
Al
A2
A3
fitted
R
Explanation of Tests
Do responses and/or variances differ among Dose levels?
(A2 vs. R)
Are Variances Homogeneous? (Al vs A2)
Are variances adequately modeled? (A2 vs. A3)
Does the Model for the Mean Fit? (A3 vs. fitted)
When rho=0 the results of Test 3 and Test 2 will be the same.)
Tests of Interest
-2*log(Likelihood Ratio) Test df
p-value
Test
Test 1
Test 2
Test 3
Test 4
The p-value for Test 1 is less than .05. There appears to be a
difference between response and/or variances among the dose levels
It seems appropriate to model the data
293.925
111.522
5. 65924
3 . 72034
10
5
4
2
<.0001
<.0001
0.2261
0.1556
The p-value for Test 2 is less than .1.
model appears to be appropriate
A non-homogeneous variance
The p-value for Test 3 is greater than .1. The modeled variance appears
to be appropriate here
The p-value for Test 4 is greater than .1.
to adequately describe the data
The model chosen seems
B-35
DRAFT - DO NOT CITE OR QUOTE
-------
1
2 Benchmark Dose Computation
3
4 Specified effect = 1
5
6 Risk Type = Estimated standard deviations from the control mean
7
8 Confidence level = 0.95
9
10 HMD = 82.493
11
12 BMDL = 68.6138
13
14
15 Serum SDH activity in male and female rats (Tables B-ll and B-12)
16 No model provided an adequate fit to the data for changes in serum SDH activity in male
17 rats. This was due to the difficulty in modeling the reported variances. As a result, these data
18 are considered unsuitable for BMD modeling. For females, only the power model provided an
19 adequate fit to the serum SDH activity data after the highest dose was dropped and the variance
20 was modeled using the non-constant variance model included in BMDS. This model served as
21 the basis for the BMDiso and BMDLiso estimates of 157 and 113 mg/kg-day for this endpoint.
B-3 6 DRAFT - DO NOT CITE OR QUOTE
-------
Table B-ll. Summary of benchmark dose modeling results for serum SDH
activity in male rats
Model
Test for
significant
difference
/>-valuea
Variance
/>-valueb
Means
/>-valueb
Scaled
residuals of
interest0
AIC
BMD1SD
(mg/kg-
day)
BMDL1SD
(mg/kg-
day)
All dose groups included
Constant variance
Linear
O.0001
O.0001
0.19
-0.75/-1.42
293.96
41.70
35.55
Non-constant variance
Lineard
O.0001
0.05
O.0001
-0.92/0.60
307.18
62.52
11.14
Highest dose group dropped
Constant variance
Lineard
O.0001
0.02
0.08
1.33/-1.16
212.18
34.45
28.37
Non-constant variance
Lineard
O.0001
0.03
0.05
1.09/-1.28
212.07
32.47
19.12
Two Highest dose groups dropped
Constant variance
Lineard
0.0004
0.04
0.26
-0.92/0.15
159.19
45.73
31.69
Non-constant variance
Linear
0.0004
0.03
0.17
-0.91/0.13
161.04
42.28
25.15
Three highest dose groups dropped
Constant variance
Linear
0.03
0.04
0.14
-0.60e
125.02
58.79
27.97
Non-constant variance
Lineard
0.03
0.05
0.64
1.20/-0.82
122.10
27.88
13.75
AIC = Akaike Information Criterion; BMD = maximum likelihood estimate of the dose associated with the selected
benchmark response; BMDL = 95% lower confidence limit on the BMD; SD = standard deviation
aValues >0.05 fail to meet conventional goodness-of-fit criteria.
Values <0.10 fail to meet conventional goodness-of-fit criteria.
°Scaled residuals at doses immediately below and immediately above the benchmark dose.
Coefficients restricted to be positive.
eResidual reported for highest dose tested.
B-37
DRAFT - DO NOT CITE OR QUOTE
-------
Table B-12. Summary of benchmark dose modeling results for serum SDH
activity in female rats
Model
Test for
significant
difference
/>-valuea
Variance
/>-valueb
Means
/>-valueb
Scaled
residuals of
interest0
AIC
BMD1SD
(mg/kg-
day)
BMDL1SD
(mg/kg-
day)
All dose groups included
Constant variance
Linear
O.0001
<0.0001
O.0001
0.18/-3.60
321.64
47.70
40.47
Non-constant variance
Lineard
O.0001
0.04
O.0001
NA
432.91
NA
24.11
Highest dose group dropped
Constant variance
Lineard
O.0001
0.0002
0.0001
-0.05/-3.48
244.99
63.45
48.93
Non-constant variance
Hille
Lineard
Polynomial (2 -degree)
Polynomial (3-degree)d
Polynomial (4-degree)d
Power*'*
<0.0001
<0.0001
<0.0001
<0.0001
O.0001
<0.0001
0.18
0.18
0.18
0.18
0.18
0.18
0.05
0.00
0.00
0.01
0.04
0.10
-1.34/0.00
-0.09/-2.36
-2.77/1.04
-2.19/0.42
-1.78/0.17
-1.34/0.00
217.37
229.76
224.39
219.90
217.52
215.37
153.80
67.45
87.97
106.18
118.22
156.52
NA
38.00
66.87
87.33
102.34
113.49
AIC = Akaike Information Criterion; BMD = maximum likelihood estimate of the dose associated with the selected
benchmark response; BMDL = 95% lower confidence limit on the BMD; NA = not applicable (BMD/BMDL
computation failed); SD = standard deviation
aValues >0.05 fail to meet conventional goodness-of-fit criteria.
Values <0.10 fail to meet conventional goodness-of-fit criteria.
°Scaled residuals at doses immediately below and immediately above the benchmark dose.
Coefficients restricted to be positive.
"Power restricted to >1.
fBest-fitting model is displayed in boldface type. Power model was the only model that provided an adequate fit to the
data.
B-38
DRAFT - DO NOT CITE OR QUOTE
-------
Power Model with 0.95 Confidence Level
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
c
o
Q_
(/)
0
ce
c
ro
0
50
45
40
35
30
25
20 L
Power
BMDL
BMD
0 20
40
60
80
dose
100
120
140
160
14:2003/292010
Power Model. (Version: 2.15; Date: 04/07/2008)
Input Data File:
C:\USEPA\IRIS\TCE\NTP\SDH\female\pOW_SDHFHDD_pOwer.(d)
Gnuplot Plotting File:
C:\USEPA\IRIS\TCE\NTP\SDH\female\pOW_SDHFHDD_pOwer.plt
Mon Mar 29 15:20:23 2010
HMDS Model Run
The form of the response function is:
Y[dose] = control + slope * dose^power
Dependent variable = mean
Independent variable = dose
The power is restricted to be greater than or equal to 1
The variance is to be modeled as Var(i) = exp(lalpha + log(mean(i)) * rho)
Total number of dose groups = 5
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
B-39
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
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
53
54
55
56
57
58
59
60
61
62
63
64
65
Parameter Convergence has been set to: le-008
Default Initial Parameter Values
lalpha = 3.46985
rho = 0
control = 25
Slope = 0.0617409
power = 1.1118
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter (s) -power
have been estimated at a boundary point, or have been specified
and do not appear in the correlation matrix )
lalpha rho control slope
lalpha 1 -1 -0.15 0.37
rho -1 1 0.14 -0.37
control -0.15 0.14 1 -0.22
Slope 0.37 -0.37 -0.22 1
Parameter Estimates
95.0% Wald Confidence I
Variable Estimate Std. Err. Lower Conf. Limit Upper
lalpha -7.0365 3.52075 -13.937
rho 3.00361 1.03813 0.968917
control 26.75 0.652491 25.4711
slope 1.29772e-039 2.07902e-040 8.90244e-040 1.
power 18 NA
NA - Indicates that this parameter has hit a bound
implied by some inequality constraint and thus
has no standard error.
Table of Data and Estimated Values of Interest
Dose N Obs Mean Est Mean Obs Std Dev Est Std Dev Scaled Res.
0 10 27 26.7 3.2 4.13 0.192
20 10 27 26.7 3.2 4.13 0.192
40 10 28 26.7 6.3 4.13 0.958
80 10 25 26.8 3.2 4.13 -1.34
170 10 45 45 9.5 9.01 3.88e-006
Model Descriptions for likelihoods calculated
Model Al: Yij = Mu(i) + e(ij)
Var{e (ij) } = SigmaA2
by the user
nterval
Conf. Limit
-0.135945
5 .0383
28.0289
7052e-039
B-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
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Model A2 :
Yij = Mud) + e(ij)
Var{e(ij)} = Sigma
Model A3: Yij = Mu(i) + e(ij)
Var{e(ij)} = expdalpha + rho*ln (Mu (i) ) )
Model A3 uses any fixed variance parameters that
were specified by the user
Model R: Yi = Mu + e (i)
Var{e (i) } = SigmaA2
Likelihoods of Interest
Model Log(likelihood)
Al -109.112298
A2 -98.178926
A3 -100.610596
fitted -103.685379
R -135.518801
Explanation of Tests
Test 1: Do responses and/or variances differ among Dose levels?
(A2 vs. R)
Test 2: Are Variances Homogeneous? (Al vs A2)
Test 3: Are variances adequately modeled? (A2 vs. A3)
Test 4: Does the Model for the Mean Fit? (A3 vs. fitted)
(Note: When rho=0 the results of Test 3 and Test 2 will be the same.)
# Param' s
6
10
7
4
2
AIC
230.224595
216.357851
215.221192
215.370759
275.037602
Test
Test 1
Test 2
Test 3
Test 4
Tests of Interest
-2*log (Likelihood Ratio) Test df
74.6798
21.8667
4.86334
6. 14957
p-value
<.0001
0.000213
0.1821
0.1046
The p-value for Test 1 is less than .05. There appears to be a
difference between response and/or variances among the dose levels
It seems appropriate to model the data
The p-value for Test 2 is less than .1.
model appears to be appropriate
A non-homogeneous variance
The p-value for Test 3 is greater than .1. The modeled variance appears
to be appropriate here
The p-value for Test 4 is greater than .1.
to adequately describe the data
The model chosen seems
Benchmark Dose Computation
Specified effect = 1
Risk Type = Estimated standard deviations from the control mean
Confidence level = 0.95
B-41
DRAFT - DO NOT CITE OR QUOTE
-------
1
2 HMD = 156.523
3
4
5 BMDL = 113.491
6
7 Serum bile acids in male and female rats (Tables B-13 and B-14)
8 All doses were retained in the modeling of serum bile acids in males and females. The
9 assumption of constant variance was not upheld for either dataset, but in each case, the power
10 model for variance included in BMDS provided adequate fit to the variance data. With the
11 variance model applied, adequate fit to the mean data was provided by several models for each
12 sex, and for both datasets, BMDL estimates across models with adequate fit differed by less than
13 threefold. In accordance with U.S. EPA (2000), the models with the lowest AIC (power model
14 for males and 5-degree polynomial model for females) were selected as the basis for the BMDiso
15 and BMDLiso estimates for these endpoints (respectively, 72 and 57 mg/kg-day for males and
16 188 and 170 mg/kg-day for females).
17
B-42 DRAFT - DO NOT CITE OR QUOTE
-------
Table B-13. Summary of benchmark dose modeling results for serum bile acid
levels in male rats
Model
Test for
significant
difference
/>-valuea
Variance
/>-valueb
Means
/>-valueb
Scaled
residuals of
interest0
AIC
BMD1SD
(mg/kg-
day)
BMDL1SD
(mg/kg-
day)
All dose groups included
Constant variance
Linear
O.0001
<0.0001
0.002
-0.10/-1.38
578.68
76.00
62.75
Non-constant variance
Hille
Linear
Polynomial (2-degree)d
Polynomial (3-degree)d
Polynomial (4-degree)
Polynomial (5-degree)d
Power'*
O.0001
O.0001
O.0001
<0.0001
<0.0001
<0.0001
<0.0001
0.77
0.77
0.77
0.77
0.77
0.77
0.77
0.69
O.0001
0.21
0.32
0.32
O.0001
0.46
0.17/-0.74
0.48/2.69
-0.88/-1.16
-0.65/-0.56
-0.65/-0.56
-1.08/0.17
-0.56/-0.43
427.84
454.67
428.95
428.58
428.58
449.32
427.70
82.84
115.63
58.37
69.21
69.21
76.72
72.45
66.69
36.05
50.80
54.31
54.31
25.65
57.77
AIC = Akaike Information Criterion; BMD = maximum likelihood estimate of the dose associated with the selected
benchmark response; BMDL = 95% lower confidence limit on the BMD; SD = standard deviation
aValues >0.05 fail to meet conventional goodness-of-fit criteria.
Values <0.10 fail to meet conventional goodness-of-fit criteria.
°Scaled residuals at doses immediately below and immediately above the benchmark dose.
Coefficients restricted to be positive.
"Power restricted to >1.
fBest-fitting model is displayed in boldface type. BMDLs for models providing adequate fit differed by < threefold,
so the model with the lowest AIC was selected.
B-43
DRAFT - DO NOT CITE OR QUOTE
-------
Power Model with 0.95 Confidence Level
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
c
o
Q_
(/)
0
ce
c
ro
0
400
300
200
100
300
14:3903/292010
Power Model. (Version: 2.15; Date: 04/07/2008)
input Data File: C:\USEPA\lRlS\TCE\NTP\bile\male\pow_BileM_power.(d)
Gnuplot Plotting File:
C:\USEPA\lRlS\TCE\NTP\bile\male\pow_BileM_power.plt
Mon Mar 29 15:39:39 2010
HMDS Model Run
The form of the response function is:
Y [dose] = control + slope * dose^power
Dependent variable = mean
Independent variable = dose
The power is restricted to be greater than or equal to 1
The variance is to be modeled as Var(i) = exp(lalpha + log(mean(i)) * rho)
Total number of dose groups = 6
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
B-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
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
Default Initial Parameter Values
lalpha =
rho =
control =
slope =
power =
8.35885
0
27.2
0.000160062
2.50584
Asymptotic Correlation Matrix of Parameter Estimates
lalpha
lalpha
rho
control
slope
power
-0.
-0.
-0.
0.
1
.98
.31
.17
.22
rho
-0.
0.
0.
-0.
.98
1
.25
.18
.23
control
-0.31
0.25
1
-0.3
0.28
Slope
-0.17
0.18
-0.3
1
-1
power
0 .22
-0.23
0.28
-1
1
Variable
lalpha
rho
control
slope
power
Estimate
-3.601
2.39924
26.8064
0.000289806
2.40282
Parameter Estimates
95.0% Wald Confidence Interval
Std. Err. Lower Conf. Limit Upper Conf. Limit
1.08576 -5.72905 -1.47295
0.272426 1.86529 2.93318
1.58205 23.7056 29.9071
0.000360688 -0.00041713 0.000996743
0.233505 1.94515 2.86048
Table of Data and Estimated Values of Interest
Dose N Obs Mean Est Mean Obs Std Dev Est Std Dev Scaled Res.
0
20
40
80
170
320
10
10
10
10
10
10
29.2
27 .5
27 .2
35 .9
92
332
26 .8
27 .2
28 .9
37 .6
93 .1
330
9.2
8 .5
8 .5
12 .3
52 .5
150
8.54
8.69
9.33
12 .8
38
173
0 .886
0 .111
-0 .561
-0 .429
-0. 0914
0. 0463
Model Descriptions for likelihoods calculated
Model Al: Yij
var{e(ij) }
Model A2: Yij
var{e(ij)}
Model A3: Yij
var{e(ij)}
Mu(i) + e (ij)
SigmaA2
Mu(i) + e (ij)
Sigma(i)A2
Mu(i) + e (ij)
exp(lalpha + rho*ln(Mu(i)
B-45
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
MO
we
Model
Test 1
Test 2
Test 3
Test 4
(Note:
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
Model A3 uses any fixed variance parameters that
were specified by the user
R: Yi = Mu + e(i)
Var{e(i)} = SigmaA2
Likelihoods of Interest
Log (likelihood)
-277.604668
-206.636351
-207.553828
-208.851786
-320.497188
# Param' s
7
12
8
5
2
AIC
569.209336
437.272702
431. 107657
427. 703572
644. 994376
Model
Al
A2
A3
fitted
R
Explanation of Tests
Do responses and/or variances differ among Dose levels?
(A2 vs. R)
Are Variances Homogeneous? (Al vs A2)
Are variances adequately modeled? (A2 vs. A3)
Does the Model for the Mean Fit? (A3 vs. fitted)
When rho=0 the results of Test 3 and Test 2 will be the same.)
Test
Test 1
Test 2
Test 3
Test 4
Tests of Interest
-2*log (Likelihood Ratio) Test df
227.722
141. 937
1.83495
2 . 59591
10
5
4
3
p-value
<.0001
<.0001
0.7661
0.4582
The p-value for Test 1 is less than .05. There appears to be a
difference between response and/or variances among the dose levels
It seems appropriate to model the data
The p-value for Test 2 is less than .1.
model appears to be appropriate
A non-homogeneous variance
The p-value for Test 3 is greater than .1. The modeled variance appears
to be appropriate here
The p-value for Test 4 is greater than .1.
to adequately describe the data
The model chosen seems
B-46
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Benchmark Dose Computation
Specified effect = 1
Risk Type = Estimated standard deviations from the control mean
Confidence level = 0.95
HMD = 72.4471
BMDL = 57.1682
Table B-14. Summary of benchmark dose modeling results for serum bile acid
levels in female rats
Model
Test for
significant
difference p-
valuea
Variance
/>-valueb
Means
/>-valueb
Scaled
residuals of
interest0
AIC
BMD1SD
(mg/kg-
day)
BMDL1SD
(mg/kg-
day)
All dose groups included
Constant variance
Lineard
O.0001
O.0001
O.0001
-1.13/-3.83
596.57
101.36
81.28
Non-constant variance
Hill6
Lineard
Polynomial (2 -degree/
Polynomial (3 -degree)
Polynomial (4-degree)d
Polynomial (5-degree) 'g
Powere
O.0001
O.0001
O.0001
O.0001
O.0001
<0.0001
O.0001
0.47
0.47
0.47
0.47
0.47
0.47
0.47
0.38
O.0001
<0.0001
0.003
0.08
0.33
0.38
-0.51/0.02
3.70f
3.09f
-0.71/-2.18
-0.42/-1.95
-1.34/0.34
-0.50/0.02
466.68
505.52
485.36
477.39
469.90
466.14
466.68
186.94
343.48
344.76
149.70
168.35
187.71
216.74
177.64
139.12
145.95
129.07
152.78
169.55
177.00
AIC = Akaike Information Criterion; BMD = maximum likelihood estimate of the dose associated with the selected
benchmark response; BMDL = 95% lower confidence limit on the BMD; SD = standard deviation
aValues >0.05 fail to meet conventional goodness-of-fit criteria.
Values <0.10 fail to meet conventional goodness-of-fit criteria.
°Scaled residuals at doses immediately below and immediately above the benchmark dose.
Coefficients restricted to be positive.
"Power restricted to >1.
fResidual at highest dose tested.
8Best-fitting model is displayed in boldface type. BMDLs for models providing adequate fit differed by < threefold,
so the model with the lowest AIC was selected.
16
B-47
DRAFT - DO NOT CITE OR QUOTE
-------
Polynomial Model with 0.95 Confidence Level
1
2
^
^J
4
5
450
400
350
300
8
c
EL 250
0
ce
i 200
0
150
100
50
0
'• rl_i,.___l:li
Polynomial
-
-
-
-
.
-
•
- ' T T -r- ^-
1 I
: BMDL
0 50 100 150
dose
14:4703/292010
1 ' '
/
/
^/
-
-
-
-
-
-
-
BMD :
200 250 300
Polynomial Model. (Version: 2.13; Date: 04/08/2008)
Input Data File:
6 C:\USEPA\lRlS\TCE\NTP\bile\female\ply BileF Poly 5.(d)
1
8
9
10
_L *_*
11
12
13
_L —*
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Gnuplot Plotting File:
C:\USEPA\lRlS\TCE\NTP\bile\female\ply BileF Poly 5. pit
Mon Mar 29 15:47:49 2010
BMDS Model Run
The form of the response function is:
Y[dose] = beta 0 + beta l*dose + beta 2*doseA2 + ...
Dependent variable = mean
Independent variable = dose
The polynomial coefficients are restricted to be positive
The variance is to be modeled as Var(i) = exp(lalpha + log (mean (i))
Total number of dose groups = 6
Total number of records with missing values
Maximum number of iterations = 250
= 0
* rho)
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
B-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
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
Default Initial Parameter Values
lalpha = 8.43454
rho = 0
beta_0 = 37
beta_l = 0
beta_2 = 0
beta_3 = 0
beta_4 = 0
beta 5 = 0
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -beta_l -beta_2 -beta_3 -beta_4
have been estimated at a boundary point, or have been specified by the user,
and do not appear in the correlation matrix )
lalpha
lalpha 1
rho -0.98
beta_0 -0.049
beta 5 0.16
rho beta_0
-0.98 -0.049
1 0.049
0.049 1
-0.16 -0.15
beta_5
0.
-0.
-0.
.16
.16
.15
1
Parameter Estimates
Variable
lalpha
rho
beta_0
beta 1
beta_2
beta 3
beta_4
beta 5
Estimate
-1.58198
2. 03725
38 .2101
1.251286-026
0
0
0
7. 95519e-011
Std. Err.
1. 00675
0 .245366
2.76802
NA
NA
NA
NA
1.432946-011
95.0% Wald Confidence Interval
Lower Conf. Limit Upper Conf. Limit
-3.55517 0.391218
1.55634 2.51816
32.7849 43.6353
5.14667e-011
1.076376-010
NA - Indicates that this parameter has hit a bound
implied by some inequality constraint and thus
has no standard error.
B-49
DRAFT - DO NOT CITE OR QUOTE
-------
1
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 of Data and Estimated Values of Interest
Dose N Obs Mean Est Mean Obs Std Dev Est Std Dev
0 10
20 10
40 10
80 10
170 10
320 10
37
46 .6
39.1
36 .3
39.3
322
Model Descriptions for
Model Al:
Model A2 :
Model A3 :
Yij =
var{e(ij)} =
Yij =
var{e(ij)} =
Yij =
var{e(ij)} =
38.2 22.5 18.5
38.2 20.6 18.5
38.2 17.7 18.5
38.5 12.3 18.7
49.5 25 24.1
305 160 154
likelihoods calculated
Mu (i ) + e (i j )
SigmaA2
Mu (i ) + e (i j )
Sigma (i) A2
Mu (i ) + e (i j )
expdalpha + rho*ln (Mu (i) ) )
Scaled Res
-0 .206
1 .43
0 .15
-0 .368
-1 .34
0.336
26
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
53
54
55
56
57
58
59
60
61
62
Model R
Test 1:
Test 2:
Test 3:
Test 4:
(Note: i
Test
Test 1
Test 2
Test 3
Test 4
The p-vali
difference
Model A3 uses any fixed variance parameters that
were specified by the user
Yi = Mu + e(i)
Var{e(i)} = SigmaA2
Likelihoods of Interest
Log (likelihood)
-279 .875470
-224.999384
-226.787639
-229.071113
-318.845182
# Param' s
7
12
8
4
2
AIC
573 . 750939
473 . 998768
469. 575277
466. 142225
641.690364
Model
Al
A2
A3
fitted
R
Explanation of Tests
Do responses and/or variances differ among Dose levels?
(A2 vs. R)
Are Variances Homogeneous? (Al vs A2)
Are variances adequately modeled? (A2 vs. A3)
Does the Model for the Mean Fit? (A3 vs. fitted)
When rho=0 the results of Test 3 and Test 2 will be the same.)
Tests of Interest
-2*log(Likelihood Ratio) Test df
p-value
187.692
109.752
3.57651
4.56695
10
5
4
4
<.0001
<.0001
0.4663
0.3347
The p-value for Test 1 is less than .05. There appears to be a
difference between response and/or variances among the dose levels
B-50
DRAFT - DO NOT CITE OR QUOTE
-------
1 It seems appropriate to model the data
2
3 The p-value for Test 2 is less than . 1. A non-homogeneous variance
4 model appears to be appropriate
5
6 The p-value for Test 3 is greater than .1. The modeled variance appears
7 to be appropriate here
8
9 The p-value for Test 4 is greater than .1. The model chosen seems
10 to adequately describe the data
11
12
13 Benchmark Dose Computation
14
15 Specified effect = 1
16
17 Risk Type = Estimated standard deviations from the control mean
18
19 Confidence level = 0.95
20
21 HMD = 187.713
22
23
24 BMDL = 169.553
25
26
27 Fetal body weights in Sprague-Dawley rats (Tables B-15 and B-16)
28 Fetal body weight data from Gulati et al. (1991) in Sprague-Dawley rats administered
29 1,1,2,2-tetrachloroethane in the diet on GD 4 - 20 are shown in Table B-15. BMD modeling
30 results based on these data are shown in Table B-l6. Adequate model fit was achieved for the
31 fetal body weight data only after the highest two dose groups were dropped. This was due to
32 difficulty in modeling the reported variances. After dropping the two highest dose groups, the
33 remaining dose groups satisfied the assumption of constant variance. Assuming constant
34 variance, the linear model provided adequate fit to the mean fetal body weight data. The higher
35 order models either did not fit (p < 0.1: higher order polynomial, power) or failed due to too
36 many parameters for the available data points (Hill). The linear model is the basis for the
37 BMDiso and BMDLiso estimates of 83 and 60 mg/kg-day, respectively, for this endpoint shown
38 in Table B-l6.
B-51 DRAFT - DO NOT CITE OR QUOTE
-------
Table B-15. Fetal body weight in Sprague-Dawley rats administered
1,1^2,2-tetrachloroethane in the diet on gestation days 4-20
Dose (mg/kg-day)
0
34
98
180
278
330
Number of animals
9
8
8
9
9
5
Mean (g)
2.28
2.17
2.19
1.99
2.04
1.81
Standard error
0.04
0.04
0.03
0.05
0.14
0.12
Source: Gulati et al. (1991).
B-52
DRAFT - DO NOT CITE OR QUOTE
-------
Table B-16. Summary of benchmark dose modeling results for fetal body weight
following exposure of pregnant Sprague-Dawley rats on gestational days 4—20
Model
Test for
significant
difference
/>-valuea
Variance
/>-valueb
Means
/>-valueb
Scaled residuals of
interest0
AIC
BMD1SD
(mg/kg-
day)
BMDL1SD
(mg/kg-
day)
All dose groups included
Constant variance
Lineard
O.0001
O.0001
0.40
-0.92/1.23
-91.54
201.09
139.17
Non constant variance
Lineard
O.0001
0.07
0.20
-1.25/0.88
-112.47
84.64
56.25
Highest dose group dropped
Constant variance
Lineard
O.0001
O.0001
0.40
-1.24/0.70
-83.65
238.24
147.87
Non constant variance
Linear
O.0001
0.05
0.18
-1.27/0.83
-105.40
84.31
53.36
Two highest dose groups dropped
Constant variance
Hille
T • d,f
Linear
Polynomial (2-degree)d
Polynomial (3 -degree)
Powere
0.0002
0.0002
0.0002
0.0002
0.0002
0.35
0.35
0.35
0.35
0.35
NA
0.12
0.06
0.08
0.06
0.38/-0.06
-1.19/1.46
0.87/-0.20
0.65/-0.09
0.38/-0.06
-101.33
-104.84
-103.53
-103.98
-103.33
129.74
83.10
110.21
118.06
129.71
61.35
59.73
62.16
64.06
61.40
AIC = Akaike Information Criterion; BMD = maximum likelihood estimate of the dose associated with the selected benchmark
response; BMDL = 95% lower confidence limit on the BMD; SD = standard deviation
aValues >0.05 fail to meet conventional goodness-of-fit criteria.
bValues <0.10 fail to meet conventional goodness-of-fit criteria.
°Scaled residuals at doses immediately below and immediately above the benchmark dose.
Coefficients restricted to be negative.
"Power restricted to >1.
Best-fitting model is displayed in boldface type. The linear model is the only model providing an adequate fit to the data.
2
B-53
DRAFT - DO NOT CITE OR QUOTE
-------
Linear Model with 0.95 Confidence Level
c
o
Q_
(/)
0
ce
c
ro
0
2.4
2.3
2.2
2.1
1.9
Linear
BMDL
BMD
20 40 60 80 100 120 140 160
dose
180
1
2
^
^J
4
5
6
1
8
9
10
_L *_*
11
12
13
_L —*
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
15:0203/292010
Polynomial Model. (Version: 2.13; Date: 04/08/2008)
Input Data File:
C:\USEPA\lRlS\TCE\gulati\fetalbdwt\lin f etalbdwt2HDD linear.
Gnuplot Plotting File:
C:\USEPA\lRlS\TCE\gulati\fetalbdwt\lin f etalbdwt2HDD linear.
Mon Mar 29 16 : 02 :5
HMDS Model Run
The form of the response function is:
Y[dose] = beta 0 + beta l*dose + beta 2*doseA2 + ...
Dependent variable = mean
Independent variable = dose
rho is set to 0
The polynomial coefficients are restricted to be negative
A constant variance model is fit
Total number of dose groups = 4
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
(d)
pit
7 2010
B-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
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
Parameter Convergence has been set to: le-008
Default Initial Parameter Values
alpha = 0.0141567
rho = 0 Specified
beta_0 = 2.26747
beta 1 = -0.0014099
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -rho
have been estimated at a boundary point, or have been specified by the user,
and do not appear in the correlation matrix )
alpha
beta_0
beta 1
alpha
1
-l.Se-OlO
2e-010
beta_0
-l.Se-OlO
1
-0. 75
beta_l
2e-010
-0 .75
1
Variable
alpha
beta_0
beta 1
Parameter Estimates
95.0% Wald Confidence Interval
Estimate Std. Err. Lower Conf. Limit Upper Conf. Limit
0.0141234 0.00342543 0.00740968 0.0208371
2.26874 0.0306445 2.20868 2.3288
-0.00143017 0.000290756 -0.00200004 -0.000860296
2.28
2.17
2.19
1. 99
2.27
2.22
2.13
2. 01
0.12
0.11
0. 08
0.15
0 .119
0 .119
0 .119
0 .119
0.284
-1 .19
1 .46
-0.538
Table of Data and Estimated Values of Interest
Dose N Obs Mean Est Mean Obs Std Dev Est Std Dev Scaled Res.
0
34
98
180
Model Descriptions for likelihoods calculated
Model Al: Yij = Mu(i) + e(ij)
Var{e(ij) } = SigmaA2
Model A2: Yij = Mu(i) + e(ij)
Var{e(ij) } = Sigma(i)A2
Model A3: Yij = Mu(i) + e(ij)
Var{e(ij)} = SigmaA2
Model A3 uses any fixed variance parameters that
were specified by the user
Model R:
Yi = Mu
B-55
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Test 1
Test 2
Test 3
Test 4
(Note:
Test
Var{e(i)} = SigmaA2
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
Likelihoods of Interest
Log (likelihood)
57 .506457
59 .148779
57 .506457
55 .418685
46.282389
# Param' s
5
8
5
3
2
AIC
-105. 012914
-102 .297557
-105. 012914
-104.837369
-88. 564779
Model
Al
A2
A3
fitted
R
Explanation of Tests
Do responses and/or variances differ among Dose levels?
(A2 vs. R)
Are Variances Homogeneous? (Al vs A2)
Are variances adequately modeled? (A2 vs. A3)
Does the Model for the Mean Fit? (A3 vs. fitted)
When rho=0 the results of Test 3 and Test 2 will be the same.)
Tests of Interest
-2*log (Likelihood Ratio) Test df
p-value
25 .7328
3.28464
3.28464
4. 17554
6
3
3
2
0.0002497
0.3498
0.3498
0.124
Test 1
Test 2
Test 3
Test 4
The p-value for Test 1 is less than .05. There appears to be a
difference between response and/or variances among the dose levels
It seems appropriate to model the data
The p-value for Test 2 is greater than .1.
model appears to be appropriate here
The p-value for Test 3 is greater than .1.
to be appropriate here
The p-value for Test 4 is greater than .1.
to adequately describe the data
A homogeneous variance
The modeled variance appears
The model chosen seems
B-56
DRAFT - DO NOT CITE OR QUOTE
-------
1
2 Benchmark Dose Computation
3
4 Specified effect = 1
5
6 Risk Type = Estimated standard deviations from the control mean
7
8 Confidence level = 0.95
9
10 HMD = 83.0965
11
12
13 BMDL = 59.7345
14
15
B-57 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
APPENDIX C. BENCHMARK DOSE MODELING RESULTS FOR THE DERIVATION
OF THE ORAL SLOPE FACTOR
Hepatocellular carcinomas in male and female B6C3F] mice (Tables C-l and C-2)
The incidence data for hepatocellular carcinomas in male and female B6C3Fi mice
exposed via gavage to 1,1,2,2-tetrachloroethane 5 days/week for 78 weeks are shown in Table C-
1 (NCI, 1978).
Table C-l. Incidence of hepatocellular carcinomas in B6C3Fi mice
administered 1,1^2,2-tetrachloroethane by gavage for 78 weeks
Endpoint
Hepatocellular carcinomas
Sex
M
F
Dose (mg/kg-day)a
0"
3/36
1/40
8.22
13/50
30/48
16.5
44/49
43/47
10
11
12
13
14
15
16
17
18
aHED as calculated in Section 5.4.3 and shown in Table 5-5.
bPooled vehicle controls
Source: NCI (1978).
The BMD modeling results from the data in Table C-l are summarized in Tables C-2 (for
males) and C-3 (for females) followed by the standard BMDS output for the selected models
from version 2.1.1 of the software. The multistage cancer model did not provide an adequate fit
to the incidence data for hepatocellular carcinomas in male mice; these data are considered
unsuitable for BMD modeling. The one-stage multistage model provided the best fit to the
incidence data for hepatocellular carcinomas in females, and this model was used as the basis for
the BMDio and BMDLio estimates (0.81 and 0.65 mg/kg-day, respectively, as HEDs) for this
endpoint.
C-l
DRAFT - DO NOT CITE OR QUOTE
-------
Table C-2. Summary of benchmark dose modeling results for the incidence
of hepatocellular carcinomas in male mice
Model
Multistage (1 -degree
polynomial)0
Multistage (2 -degree
polynomial)0
DF
1
1
x2
18.30
5.24
% Goodness
of fit
/>-valuea
O.001
0.02
Scaled
residuals of
interest1"
0.51/-3.27
0.53/-1.83
AIC
134.58
119.87
BMD10[HED]
(mg/kg-day)
1.42
4.10
BMDL10[HED]
(mg/kg-day)
1.11
3.08
2
3
AIC = Akaike Information Criterion; BMD = maximum likelihood estimate of the dose associated with the selected
benchmark response; BMDL = 95% lower confidence limit on the BMD; DF = degrees of freedom
"Values < 0.1 fail to meet conventional goodness-of-fit criteria.
Scaled residuals at doses immediately below and immediately above the benchmark dose.
°Betas restricted to > 0.
Table C-3. Summary of benchmark dose modeling results for the incidence
of hepatocellular carcinomas in female mice
Model
Multistage (1-degree
polynomial)*'*
Multistage (2 -degree
polynomial)0
DF
1
0
x2
0.74
0.00
X2 Goodness
of fit
/>-valuea
0.39
NA
Scaled
residual of
interest
0.04/-0.61
0.00/0.00
AIC
104.99
106.22
BMD10[HED]
(mg/kg-day)
0.81
1.18
BMDL10[HED]
(mg/kg-day)
0.65
0.67
AIC = Akaike Information Criterion; BMD = maximum likelihood estimate of the dose associated with the selected
benchmark response; BMDL = 95% lower confidence limit on the BMD; DF = degrees of freedom; NA= not
applicable (/>value was not generated due to insufficient DF)
"Values < 0.1 fail to meet conventional goodness-of-fit criteria.
bScaled residuals at doses immediately below and immediately above the benchmark dose.
°Betas restricted to > 0.
dSelected model is displayed in boldface type.
C-2
DRAFT - DO NOT CITE OR QUOTE
-------
Multistage Cancer Model with 0.95 Confidence Level
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
C
o
•
0.8
0.6
0.4
0.2
Multistage Cancer
Linear extrapolation
BMDL BMD
8
dose
10
12
14
16
15:11 03/292010
Multistage Cancer Model. (Version: 1.7; Date: 05/16/2008)
Input Data File:
C:\USEPA\lRlS\TCE\NCl\hepcarc\female\msc_hepcarcF_MS_l.(d)
Gnuplot Plotting File:
C:\USEPA\lRlS\TCE\NCl\hepcarc\female\msc_hepcarcF_MS_l.pit
Mon Mar 29 16:11:43 2010
HMDS Model Run
The form of the probability function is:
P [response] = background + (1-background)*[1-EXP(
-betal*doseAl)]
The parameter betas are restricted to be positive
Dependent variable = incidence
Independent variable = dose
Total number of observations = 3
Total number of records with missing values
Total number of parameters in model = 2
Total number of specified parameters = 0
Degree of polynomial = 1
= 0
C-3
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
9
10
11
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
T-O
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial Parameter Values
Background = 0
Beta(l) = 0.147828
Asymptotic Correlation Matrix of Parameter Estimates
Background Beta ( 1 )
Background 1 -0.54
Beta(l) -0.54 1
Parameter Estimates
95.0% Wald Confidence
Variable Estimate Std. Err. Lower Conf. Limit Upper
Background 0.0240983 * *
Beta(l) 0.130589 * *
* - Indicates that this value is not calculated.
Analysis of Deviance Table
Model Log (likelihood) # Param's Deviance Test d.f. P- value
Full model -50.1115 3
Fitted model -50.4931 2 0.763231 1 0.3823
Reduced model -92.948 1 85.673 2 <.0001
AIC: 104.986
Goodness of Fit
Dose Est. Prob. Expected Observed Size
0.0000 0.0241 0.964 1.000 40
8.2200 0.6664 31.988 30.000 48
16.5000 0.8869 41.682 43.000 47
ChiA2 = 0.74 d.f. = 1 P-value = 0.3897
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Extra risk
Confidence level = 0.95
Interval
Conf. Limit
*
*
Scaled
Residual
0.037
-0.608
0.607
C-4
DRAFT - DO NOT CITE OR QUOTE
-------
1
2 HMD = 0.806812
3
4 BMDL = 0.648049
5
6 BMDU = 1.01577
7
8 Taken together, (0.648049, 1.01577) is a 90 % two-sided confidence
9 interval for the HMD
10
11 Multistage Cancer Slope Factor = 0.154309
12
13
14
15
C-5 DRAFT - DO NOT CITE OR QUOTE
------- |