DRAFT - DO NOT CITE OR QUOTE                         EPA/635/R-09/001
                                                    www.epa.gov/iris
oEPA
           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)

                            August 2009
                              NOTICE


This document is an External 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.
August, 2009                           ii           DRAFT - DO NOT CITE OR QUOTE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	xii
AUTHORS, CONTRIBUTORS, AND REVIEWERS	xiii

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	12

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	37
   4.4. OTHER DURATION- OR ENDPOINT-SPECIFIC STUDIES	38
       4.4.1.  Acute Studies (Oral and Inhalation)	38
          4.4.1.1.  Oral Studies	38
          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	47
       4.4.3.  Acute Injection Studies	49
August, 2009                        iii          DRAFT - DO NOT CITE OR QUOTE

-------
       4.4.4. Immunotoxicological Studies	49
   4.5. MECHANISTIC DATA AND OTHER STUDIES IN SUPPORT OF THE MODE
       OF ACTION	50
       4.5.1. Genotoxicity	50
       4.5.2. Short-Term Tests of Carcinogenicity	53
   4.6. SYNTHESIS OF MAJOR NONCANCER EFFECTS	55
       4.6.1. Oral	55
           4.6.1.1. Human Data	55
           4.6.1.2. Animal Data	55
       4.6.2. Inhalation	61
           4.6.2.1. Human Data	61
           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.)	83
           5.1.2.3. RfD Derivation—Including Application of UFs	83
       5.1.3. RfD Comparison Information	85
       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.)	92
       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
August, 2009                          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-l

APPENDIX C. BENCHMARK DOSE MODELING RESULTS FOR THE DERIVATION
            OF THE ORAL SLOPE FACTOR	C-l
August, 2009                      v         DRAFT - DO NOT CITE OR QUOTE

-------
                                   LIST OF TABLES
2-1.  Chemical and physical properties of 1,1,2,2-tetrachloroethane	3

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	31

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


August, 2009                            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	50

4-16. Pulmonary adenomas from 1,1,2,2-tetrachloroethane exposure in mice	54

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 maj or human studies of inhalation exposure to
      1,1,2,2-tetrachloroethane	63

4-20. Summary of noncancer results of maj or 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	81

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-2

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-3

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-10

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-ll
August, 2009                            vii           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-17

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-18

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-26

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-27

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-35

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-36

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-44

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-45

B-13. 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-50

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-l

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-3
August, 2009                           viii           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
August, 2009                            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
LC50         median lethal concentration
LD50         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
August, 2009
DRAFT - DO NOT CITE OR QUOTE

-------
U.S. EPA    U.S. Environmental Protection Agency
WBC       white blood cell
August, 2009                         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).
August, 2009                            xii           DRAFT - DO NOT CITE OR QUOTE

-------
                  AUTHORS, CONTRIBUTORS, AND REVIEWERS
CHEMICAL MANAGER

Martin W. Gehlhaus, M.H.S.
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Washington, DC

AUTHORS

AmbujaBale, 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
August, 2009                          xiii          DRAFT - DO NOT CITE OR QUOTE

-------
Marc Odin, M.S.
Environmental Science Center
Syracuse Research Corporation
Syracuse, NY

REVIEWERS
       This document has been reviewed by EPA scientists, interagency reviewers from other
federal agencies, 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

SueRieth, M.S.
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
August, 2009                          xiv          DRAFT - DO NOT CITE OR QUOTE

-------
 1                                     1. INTRODUCTION
 2
 O
 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
1>1   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
      August, 2009                            1            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).
20
     August, 2009                          2           DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
                  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—C-
                                       Cl
-C-
 Cl
-H
       Figure 2-1. Structure of l,l?2,2-tetrachloroethane.
       Table 2-1. Chemical and physical properties of 1,1^2,2-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
Riddick et al., 1986
Lide, 1993
Riddick et al., 1986
Lide, 1993
Merck Index, 1989
Riddick et al., 1986
Lide, 1993
HSDB, 2009; Amoore and
Hautala, 1983
Amoore and Hautala, 1983
HSDB, 2009
Riddick et al., 1986
Merck Index, 1989
HSDB, 2009; Merck Index,
1989; Hawley, 1981
August, 2009
       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 10'4 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
Riddick et al., 1986
HSDB, 2009; Flick, 1985
Hansch and Leo, 1985
Chiouetal., 1979
ASTER, 1995
Mackay and Shiu, 1981
HSDB, 2009
ASTER, 1995
HSDB, 2009; Hawley, 1981
Calculated
Calculated
 2          In the past, the major use for 1,1,2,2-tetrachloroethane was in the production of
 3   trichloroethylene, tetrachloroethylene, and 1,2-dichloroethylene (Archer, 1979).  With the
 4   development of new processes for manufacturing chlorinated ethylenes and the availability of
 5   less toxic solvents, the production of 1,1,2,2-tetrachloroethane as a commercial end-product in
 6   the United States and Canada has steadily declined since the late 1960s, and production ceased
 7   by the early-1990s (HSDB, 2009; Environment Canada and Health Canada, 1993).
 8   1,1,2,2-Tetrachloroethane may still appear as a chemical intermediate in the production of a
 9   variety of other common chemicals. It was also used as a solvent, in cleaning and degreasing
10   metals, in paint removers, varnishes, and lacquers, in photographic films, and as an extractant for
11   oils and fats (Hawley, 1981).  Although at one time it was used as an insecticide, fumigant, and
12   weed killer (Hawley, 1981), it presently is not registered for any of these purposes. It was once
13   used as an ingredient in an insect repellent, but registration was canceled in the late 1970s.
14
     August, 2009
DRAFT - DO NOT CITE OR QUOTE

-------
 1                                    3. TOXICOKINETICS
 2
 O
 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    microsome 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
29    5 days/week for 4 weeks, followed by a single radiolabeled dose of the compound, and evaluated
30    the disposition of the radiolabeled 1,1,2,2-tetrachloroethane over the next 48 hours. While
31    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
      August, 2009                            5            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
 5    containing [38Cl]-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 jimol 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  considerable
16    absorption occurred in both species, mice absorbed proportionally more 1,1,2,2-tetrachloro-
17    ethane on a per-body-weight basis. Ikeda and Ohtsuji (1972) detected metabolites, measured as
18    total trichlorocompounds, trichloroacetic acid, and trichloroethanol, in the urine of rats exposed
19    to 200 ppm (1,370 mg/m3) 1,1,2,2-tetrachloroethane,  indicating that absorption had occurred;
20    however, they did not provide a  quantitative estimate of absorption rate or fraction.  Similarly,
21    Gargas and Anderson (1989) followed the elimination of 1,1,2,2-tetrachloroethane as exhaled
22    breath from the blood after a 6-hour exposure to 350 ppm (2,400 mg/m3), but did not provide
23    quantitative 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 high and selective uptake of nonvolatile radioactivity in
37    the 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

      August, 2009                            6           DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
forestomach. High levels of activity were also found in the liver, bile, inner zone of the adrenal
cortex, and interstitium of the testis, 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.
                   Cl
                            Cl
10
11
12
13
14
15
16
17
18
19
20
21
22
          free radical
                                            Cl
                                                     Cl
ct
reductive /
dechlorination Cl
\,
Cl Cl _ 1,1,2,2-tetrachloroeth
C
Trichloroethylene
C
trichloroacetaldehyde
s
Cl OH
Cl
trichloroethanol
; 	 ' I^Jmatic (oxidation
/ 1 riehvrirnchlnrinatinn 1
' Cl Cl
°l 0 C/
// tetrachloroet
: ci /
P450/
/ \ / :
Cl O
// o
n' \ ^\
Cl OH
oxalic a
Trichloroacetic acid
Cl
nylene
O
J/
cid
                                                            hydrolytic  °.,
                                                                                dichloroacetic acid
                                                                                   NH2
HO Glyoxylic
acid 	
Io
"), + C OH
\
— + ^ 	 '
glycine
h. r.r
                                                                       formic acid
       Source: Adapted from ATSDR (1996).

       Figure 3-1.  Suggested metabolic pathways of l,l?2,2-tetrachloroethane.

       In vivo and in vitro studies indicate that the metabolism of 1,1,2,2-tetrachloroethane
proceeds via multiple pathways in rodents (Mitoma et al., 1985; Casciola and Ivanetich, 1984;
Halpert, 1982; Koizumi et al., 1982; Halpert and Neal, 1981; Ikeda and Ohtsuji, 1972; Yllner,
1971). The predominant pathway appears to involve production of dichloroacetic acid, formed
as an initial metabolite via stagewise hydrolytic cleavage of 1,1,2,2-tetrachloroethane, yielding
dichloroacetyl chloride and dichloroacetaldehyde as intermediates, or by cytochrome P450-based
oxidation of 1,1,2,2-tetrachloroethane (Casciola and Ivanetich, 1984; Halpert and Neal, 1981;
      August, 2009
                                                     DRAFT - DO NOT CITE OR QUOTE

-------
 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% 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).  Mean
19    dichloroacetic acid, trichloroacetic acid, trichloroethanol, oxalic acid, glyoxylic acid, and urea
20    accounted for 27, 4, 10, 7, 0.9, and 2% of the urinary radioactivity excreted by the mice in
21    24 hours, respectively (Yllner et al.,  1971). Yllner et al. (1971) also demonstrated that 20-23%
22    of the [14C]-tetrachloroethane was converted to  glycine following the simultaneous 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 cause inactivation of cytochrome
38    P450. 1,1,2,2-Tetrachloroethane  effectively inactivated the major phenobarbital-inducible P450

      August, 2009                            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    reduced total cytochrome P450 activity 44 and 37% in males and females, respectively, at
 8    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 sexes, 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 sexes.
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 testis (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

      August, 2009                            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]-1,1,2,2-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 CO2 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 CO2 and 2% as unchanged compound), 26% in urine, and 6% in feces.
      August, 2009                            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 et al.,  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    Milman et al. (1984) reported in an abstract that 45% of the activity from a single radiolabeled
22    oral dose of 1,1,2,2-tetrachloroethane was recovered in the carcass, although the evaluation time
23    was not reported.  Mitoma et al. (1985) reported a 30.75% retention in the carcass of rats and a
24    27.44% retention in the carcass of mice 48 hours after exposure to a single labeled dose of
25    1,1,2,2-tetrachloroethane. Dow Chemical Company (1988) reported 30% retention in the carcass
26    in rats exposed to 10 ppm by inhalation, 25% in mice exposed to 10 ppm by inhalation, 23% in
27    rats  exposed to 150 mg/kg by gavage, and 17.3% in mice exposed to 150 mg/kg by gavage.
28    Colacci et al. (1987) reported covalent binding of radiolabeled 1,1,2,2-tetrachloroethane to DNA,
29    RNA, and protein in the liver, kidney, lung, and stomach of rats and mice exposed to a single
30    intravenous dose and analyzed 22 hours postexposure. In vitro binding to calf thymus DNA was
31    found to be greatest when the microsomal fraction was present, and was inhibited by SKF-525 A,
32    indicating that metabolic  activation was likely required for DNA binding (Colacci et al., 1987).
33    However, Collaci et al. (1987) did not distinguish between covalent binding and whether the
34    presence of radiolabel in the DNA, RNA, and protein was the result of incorporated radiolabeled
35    carbon into the biomolecules through normal biochemical processes.
36
      August, 2009                            11          DRAFT - DO NOT CITE OR QUOTE

-------
 1    3.5. PHYSIOLOGICALLY BASED TOXICOKINETIC MODELS
 2          No physiologically based toxicokinetic (PBTK) models for 1,1,2,2-tetrachloroethane
 3    were located for humans. Muelenberg et al. (2003) used saline:air, rat brain:air, and olive oil:air
 4    partition coefficients to model 28 chemicals from three distinct chemical classes, including
 5    alkylbenzenes, chlorinated hydrocarbons, and ketones.  The saline:air, rat brain:air, and olive
 6    oil:air partition coefficients derived for 1,1,2,2-tetrachloroethane were 35.6 ± 6.05, 344 ± 21.0,
 7    and 10,125 ± 547, respectively.  The brain partition coefficients for the 28 chemicals were
 8    predicted with accuracy within a factor of 2.5 for 95% of the chemicals.  While the study
 9    demonstrates the ability to predict rat brain partition coefficients using a bilinear equation, the
10    utility of the information for this assessment is limited.  Similarly, several physiologically based
11    pharmacokinetic (PBPK) investigations of 1,1,2,2-tetrachloroethane exposure in fish (McKim et
12    al., 1999; Nichols et al., 1993) provide little utility  for this assessment. In sum, adequate
13    information for PBTK modeling of 1,1,2,2-tetrachloroethane remains a research need.
14          Chiu and White (2006) presented an analysis of steady-state solutions to a PBPK model
15    for a generic volatile organic chemical (VOC) metabolized in the liver.  The only parameters
16    used to determine the system state for a given oral  dose rate or inhalation exposure concentration
17    were the blood-air partition coefficient, metabolic constants, and the rates of blood flow to the
18    liver and of alveolar ventilation. At exposures where metabolism is close to linear (i.e.,
19    unsaturated), it was demonstrated that only the effective first order metabolic rate constant was
20    needed.  Additionally, it was found that the relationship between cumulative exposure and
21    average internal dose (e.g., areas under the curve [AUCs]) remains the same for time-varying
22    exposures. The study authors concluded that steady-state solutions can reproduce or closely
23    approximate the solutions using a full PBPK model. Section 5.2.2 addresses the applicability of
24    using this model  to conduct a route-to-route extrapolation in this assessment.
25
      August, 2009                            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/m3) was tolerated without effect for 10 minutes, while 146 ppm (1,003 mg/m3) for
35    30 minutes or 336 ppm  (2,308 mg/m3) for 10 minutes caused irritation of the mucous membranes,
36    pressure in the head, vertigo, and fatigue. No other relevant information was reported.
37          Minot and Smith (1921) reported that symptoms of industrial 1,1,2,2-tetrachloroethane
38    poisoning (concentrations not specified) included fatigue, perspiration, drowsiness, loss of

      August, 2009                           13           DRAFT - DO NOT CITE  OR QUOTE

-------
 1    appetite, nausea, vomiting, constipation, headache, and jaundice. Hematological changes
 2    included increased large mononuclear cells, elevated white blood cell (WBC) count, a slight but
 3    progressive anemia, and a slight increase in platelet number. Similar symptoms were reported by
 4    Parmenter (1921) and Wilcox et al. (1915).  Horiguchi et al. (1964) reported that in 127 coating
 5    workers employed in artificial pearl factories and exposed to 75-225 ppm (500-1,500 mg/m3)
 6    1,1,2,2-tetrachloroethane (along with other solvents), observed effects included decreased
 7    specific gravity of the whole blood, decreased RBC count, relative lymphocytosis, neurological
 8    findings (not specified), and a positive urobilinogen test.
 9           Lobo-Mendonca (1963) observed a number of adverse health effects in a mixed-sex
10    group of 380 workers at 23 Indian bangle manufacturing facilities (80% of workers employed at
11    these facilities were examined). In addition to the inhalation exposure, approximately 50% of
12    the examined workers had a substantial amount of dermal exposure to 1,1,2,2-tetrachloroethane.
13    Some of the workers were exposed to a mixture of equal parts acetone and 1,1,2,2-tetrachloro-
14    ethane. Air samples were collected at several work areas in seven facilities. Levels of
15    1,1,2,2-tetrachloroethane in the air ranged from 9.1 to 98  ppm (62.5-672 mg/m3). High
16    incidences of a number of effects were reported, including anemia (33.7%), loss of appetite
17    (22.6%), abdominal pain (23.7%), headaches (26.6%), vertigo (30.5%), and tremors (35%). The
18    significance of these effects cannot be determined because a control group of unexposed workers
19    was not examined and coexposure to  acetone was possible. The study authors noted that the
20    incidence of tremors appeared to be directly related to 1,1,2,2-tetrachloroethane exposure
21    concentrations, as the percentage of workers handling tetrachloroethane and displaying tremors
22    increased as the air concentration of 1,1,2,2-tetrachloroethane increased.
23           Over a 3-year period, Jeney et al. (1957) examined 34-75 workers employed at a
24    penicillin production facility.  1,1,2,2-Tetrachloroethane was used as an emulsifier, and wide
25    fluctuations in atmospheric levels occurred throughout the day.  The investigators noted that the
26    workers were only in the areas with high 1,1,2,2-tetrachloroethane concentrations for short
27    periods of time, and gauze masks with organic solvent filters were worn in these areas. During
28    the first year of the study, 1,1,2,2-tetrachloroethane  levels ranged from 0.016 to 1.7 mg/L (16-
29    1,700 mg/m3; 2-248 ppm). In the second year of the study, ventilation in the work room was
30    improved and 1,1,2,2-tetrachloroethane levels ranged from 0.01 to 0.85 mg/L (10-850 mg/m3;
31    1-124 ppm). In the third year of the study, the workers were transferred to a newly built facility
32    and 1,1,2,2-tetrachloroethane levels in the new facility ranged from 0.01 to 0.25 mg/L (10-
33    250 mg/m3; 1-36 ppm). At 2-month intervals, the workers received general physical
34    examinations, and blood was drawn for measurement of hematological parameters, serum
35    bilirubin levels, and liver function tests; urinary hippuric acid levels were measured every
36    6 months.  It appears that workers with positive signs of liver damage, including palpability of
37    the liver, rise in bilirubin levels, positive liver function tests, and urobilinogenuria, were
38    transferred to other areas of the facility and were not examined further.

      August, 2009                           14           DRAFT - DO NOT CITE OR QUOTE

-------
 1          In the first year of the study, 31% of the examined workers had "general or gastro-
 2    intestinal symptoms." Loss of appetite, bad taste in the mouth, epigastric pain, and a "dull
 3    straining pressure feeling in the area of the liver" were reported by 66% of the workers
 4    experiencing gastrointestinal symptoms.  Other symptoms included headaches, general weakness,
 5    and fatigue in 29%, severe weight loss in 4%, and "tormenting itching" in 1%. Enlargement of
 6    the liver was observed in 38% of the screened workers. Urobilinogenuria was detected in 50%
 7    of the workers, most often following more than 6 months of employment, and 31% of the
 8    workers with urobilinogenuria also had palpable livers.
 9          In the second year of the study, there was a decline in the number of symptomatic
10    workers (13% of examined workers) and in workers with positive urobilinogenuria findings
11    (24%).  Liver enlargement was observed in 20% of the examined workers.  In the third year, the
12    number of workers reporting symptoms decreased to 2%, and positive urobilinogen findings
13    were found in 12%.  The investigators noted that the increased urobilinogen levels during the
14    third year of observation may have been secondary to excessive alcohol consumption or dietary
15    excess.  Enlarged livers were found in 5% of the examined workers.
16          During the course of the study, no alterations in erythrocyte or hemoglobin (Hb) levels
17    were found. Leukopenia (defined as leukocyte levels of <5,800 cells/mL) was found in 20% of
18    the workers, but no relationship between the number of cases and duration of 1,1,2,2-tetrachloro-
19    ethane exposure was found. A positive relationship between duration of exposure and frequency
20    of abnormal liver function test results  was observed, as statistically significant correlations were
21    found on the thymol and Takata-Ucko liver function tests, but not the gold sol reaction test. The
22    thymol liver function test measures the direct precipitation of both lipids and abnormal lipid
23    protein complexes appearing in liver disease by the addition of a thymol  solution (Kunkel and
24    Hoagland, 1947). The Takata-Ucko (or Takata-Ara) test detects  an increase in the amounts of
25    the globulin components of the serum, signifying liver disease (Kunkel and Hoagland, 1947).
26    Abnormal hippuric acid  levels were only detected in 1% of the examined workers during the first
27    2 years, and no abnormalities were observed during the third year. Increased serum bilirubin
28    levels (>1 mg/dL) were observed in 20, 18.7, and 7.6% of the workers during the first, second,
29    and third years, respectively. The prevalence of hepatitis was assessed using sickness benefit
30    files. In the 1-year period prior to the study, 21 cases of hepatitis were found (total number of
31    workers not reported). Three cases of hepatitis were found in the first year of the  study, eight
32    cases in the second year, and four cases in the third year. The lack of a control group and poor
33    reporting of study design and results precludes using this study for quantitative dose-response
34    analysis.
35          Norman et al. (1981) examined the mortality of the employees of 39 chemical processing
36    plants used by the Army during World War II.  Ten plants used 1,1,2,2-tetrachloroethane to help
37    treat clothing, while the  others plants used water in the same process. Estimates of exposure
38    levels were not reported, and coexposure to dry-cleaning chemicals was expected.  At the time of

      August, 2009                            15           DRAFT - DO  NOT CITE OR QUOTE

-------
 1    evaluation, 2,414 deaths were reported in the study cohort. No differences in standard mortality
 2    ratios were seen between the tetrachloroethane and water groups for total mortality,
 3    cardiovascular disease, cirrhosis of the liver, or cancer of the digestive and respiratory systems.
 4    The mortality ratio for lymphatic cancers in the tetrachloroethane group was increased relative to
 5    controls or the water group, although the number of deaths was small (4 cases, with an expected
 6    number of 0.85). No other differences were seen between the groups.
 7
 8    4.2. SUBCHRONIC AND CHRONIC STUDIES AND CANCER BIOASSAYS IN
 9    ANIMALS—ORAL AND INHALATION
10    4.2.1.  Oral Exposure
11    4.2.1.1. Subchronic Studies
12          NTP (2004) fed groups of male and female F344 rats (10/sex/group) diets containing 0,
13    268, 589, 1,180, 2,300, or 4,600 ppm of microencapsulated 1,1,2,2-tetrachloroethane for
14    14 weeks. NTP (2004) reported that the microcapsules containing 1,1,2,2-tetrachloroethane
15    were specified to be no greater than 420 jim in diameter, and were not expected to have  any
16    significant effect on the study. The reported average daily doses were 0, 20, 40, 80, 170, or
17    320 mg/kg-day, and vehicle control (feed with empty microcapsules) and untreated control
18    groups were used for both  sexes. Endpoints evaluated throughout the study included clinical
19    signs, body weight, and feed consumption. Hematology and clinical chemistry were assessed on
20    days 5 and 21 and at the end of the study; urinalyses were not performed. Necropsies were
21    performed on all animals, and selected organs (liver, heart, right kidney, lung, right testis, and
22    thymus) were weighed.  Comprehensive histological examinations were performed on untreated
23    control, vehicle control, and high dose groups. Tissues examined in the lower dose groups were
24    limited to bone with marrow, clitoral gland, liver,  ovary, prostate  gland, spleen, testis with
25    epididymis and seminal vesicle, and uterus.  A functional observational battery (FOB) was
26    performed on rats in the control groups and the 20, 40, and 80 mg/kg-day groups during weeks 4
27    and 13. Sperm motility, vaginal cytology, estrous cycle length, and percentage of time spent in
28    the various estrus stages were evaluated in control groups and the 40, 80, and 170 mg/kg-day
29    groups.
30          All animals  survived to the end of the study, but clinical signs of thinness and pallor were
31    observed in all animals in the 170 and 320 mg/kg-day groups (NTP, 2004).  Final body weights
32    (Table 4-1) were statistically significantly lower than vehicle controls in males at 80,  170, and
33    320 mg/kg-day (7, 29, and 65% lower, respectively) and females at 80, 170, and 320 mg/kg-day
34    (9, 29, and 56% lower, respectively), with both sexes at 320 mg/kg-day losing weight over the
35    course of the study. However, feed consumption by the rats also decreased with increasing dose
36    level (NTP, 2004).
37
      August, 2009                            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%
o
-J
-9
-29
-56
2
O
4
5
6
7
8
9
      "Mean ± 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 3 20 mg/kg-day.
            Table 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
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).
10
     August, 2009
                                            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
      aMean ± standard error.
      V<0.05.
      Source: NTP (2004).
 1
 2          Results of the FOB showed no exposure-related findings of neurotoxicity. The
 3    hematology evaluations indicated that 1,1,2,2-tetrachloroethane affected the circulating erythroid
 4    mass in both sexes (Table 4-3). There was evidence of a transient erythrocytosis, as shown by
 5    increases in hematocrit values, Hb concentration, and erythrocyte counts on days 5 and 21 at
 6    >170 mg/kg-day.  The erythrocytosis was not considered clinically significant and disappeared
 7    by week 14, at which time minimal to mild, dose-related anemia was evident, as shown by
 8    decreases in hematocrit and Hb at >40 mg/kg-day. For example, although males exposed to
 9    40 mg/kg-day  showed a statistically significant decrease in Fib at week 14, the magnitude of the
10    change was small (3.8%).  The anemia was characterized as microcytic based on evidence
11    suggesting that the circulating erythrocytes were smaller than expected, including decreases in
12    mean cell volumes, mean cell Fib values, and mean cell Fib concentration in both sexes at
13    >80 mg/kg-day at various time points. At week 14, there were no  changes in reticulocyte counts,
14    suggesting that there was no erythropoietic response to the anemia, which was in turn supported
15    by the bone marrow atrophy observed microscopically.  As discussed by NTP (2004), the
16    erythrocytosis suggested a physiological response consistent with hemoconcentration due to
17    dehydration, as well as compromised nutritional status due to the reduced weight gain and food
18    consumption, both of which may have contributed to the development of the anemia.
19
     August, 2009
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 (IL)
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 (IL)
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 ± Q.4b
16.0±0.2b
662.5 ± 19.4b
1
2
3
4
5
6
"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 sexes, 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) in males at
>80 mg/kg-day (41, 134, and 496%, and 15, 74, and 174%, respectively) and females at
     August, 2009
                                        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) in both sexes at > 170 mg/kg-day (36 and 66% in males and 58 and 117% in
 3    females), increases in bile acids in males at >170 mg/kg-day (233 and 1,110%) and females at
 4    320 mg/kg-day (590%), and decreases in serum cholesterol in females at >80 mg/kg-day (23, 39,
 5    and 48%, respectively) and males at 320 mg/kg-day (12%).  There were no exposure-related
 6    changes in rat serum 5'-nucleotidase at week 14, although increases occurred on day 5 in females
 7    at >20 mg/kg-day and on day 21 in males and females at 80, 170, and/or 320 mg/kg-day.
 8          A summary of histopathological alterations following 1,1,2,2-tetrachloroethane exposure
 9    is presented in Table 4-4. Hepatic cytoplasmic vacuolization was noted in males exposed to
10    >20 mg/kg-day and in females exposed to >40 mg/kg-day.  Although incidence of this alteration
11    was high in affected groups, severity was only minimal-to-mild and only increased with dose
12    from 20 to 40 mg/kg-day in males and 40 to 80 mg/kg-day in females.  Females exposed to
13    >80 mg/kg-day showed an increase in the incidence of hepatocyte hypertrophy with an increase
14    in severity and incidence with increasing exposure level, and males showed similar results at
15    exposures >170 mg/kg-day. A statistically significant increase in the incidence of hepatocellular
16    necrosis was observed in male and female rats at 170 and 320 mg/kg-day, accompanied by an
17    increased severity with an increase in dose. At > 170 mg/kg-day, additional effects in the liver in
18    both sexes were hepatocyte pigmentation and mitotic alteration and mixed cell foci, with bile
19    duct hyperplasia observed in females only.  Pigmentation of the spleen was statistically
20    significantly increased in male rats exposed to >80 mg/kg-day and in female rats exposed to
21    >170 mg/kg-day. Other histological effects included statistically significantly increased
22    incidences of atrophy (red pulp and lymphoid follicle) in the spleen of males at 170 and 320
23    mg/kg-day and the spleen of females at 320 mg/kg-day.  A statistically significant increase in
24    atrophy of bone (metaphysis) and bone marrow, prostate gland, preputial gland, seminal vesicles,
25    testes (germinal epithelium), uterus, and clitoral gland, as well as an increase in ovarian
26    interstitial  cell cytoplasmic alterations, was observed in females at >170 mg/kg-day and in males
27    at 320 mg/kg-day.
28
      August, 2009                            20          DRAFT - DO NOT CITE OR QUOTE

-------
             Table 4-4.  Incidences of selected histopathological lesions in rats exposed to
             dietary l,l?2,2-tetrachlor ethane 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. Hepatocellular
10    hypertrophy and spleen pigmentation were observed at 80 mg/kg-day, although these changes
11    were generally of minimal severity.  Increases in serum ALT, SDH, and cholesterol, and
      August, 2009
21
DRAFT - DO NOT CITE OR QUOTE

-------
 1    decreases in body weight gain (<10%), were observed at >80 mg/kg-day.  Increases in serum
 2    ALP and bile acids, hepatocellular necrosis, bile duct hyperplasia, hepatocellular mitotic
 3    alterations, foci of cellular alterations, and liver pigmentation occurred at  170 and/or 320 mg/kg-
 4    day. A no-observed-adverse-effect level (NOAEL) of 20 mg/kg-day and a lowest-observed-
 5    adverse-effect level (LOAEL) of 40 mg/kg-day was identified by EPA for increased relative
 6    liver weight in male and female rats. NTP (2004) identified a NOAEL of 20 mg/kg-day in rats
 7    based on survival and body weight changes and increased lesion incidences. There were no
 8    clinical signs of neurotoxicity at doses as high as 320 mg/kg-day or exposure-related findings in
 9    the FOB at doses as high as 80 mg/kg-day (highest tested dose in the FOB), indicating that the
10    nervous system may be less sensitive than the liver for subchronic dietary exposure.
11          NTP (2004) also exposed groups of male and female B6C3Fi mice (10/sex/group) to
12    diets containing 0, 589, 1,120, 2,300, 4,550, or 9,100 ppm of microencapsulated 1,1,2,2-tetra-
13    chloroethane for 14 weeks, with vehicle and untreated control groups for each sex. The reported
14    average daily doses were 0, 100, 200, 370, 700, or 1,360 mg/kg-day for males and 0, 80, 160,
15    300, 600, or 1,400 mg/kg-day for females. Endpoints evaluated throughout the study included
16    clinical signs, body weight, and feed consumption.  Clinical chemistry was assessed at the end of
17    the study, but hematological evaluations and urinalyses were not performed. Necropsies were
18    conducted on all animals and selected organs (liver, heart, right kidney, lung, right testis, and
19    thymus) were weighed.  Comprehensive histological examinations were performed on untreated
20    control, vehicle control,  and high dose groups. Tissues examined in the lower dose groups were
21    limited to the liver, spleen, and thymus in both sexes; preputial gland in males; and lungs in
22    females. An FOB (21 parameters) was performed on mice in both  control and 160/200, 300/370,
23    and 600/700 mg/kg-day  (1,120, 2,300, and 4,550 ppm, respectively) dose groups during weeks 4
24    and 13. Sperm motility, vaginal cytology, estrous cycle length, and percentage of time spent in
25    the various estrus stages were evaluated in both control and  160/200, 600/700, and 1,360/
26    1,400 mg/kg-day (1,120, 2,300, and 4,550 ppm, respectively) dose groups.
27          All mice survived to the end of the study (NTP, 2004). Thinness was observed  clinically
28    in male mice (3/10,  9/10, 10/10) at 370, 700, and 1,400 mg/kg-day, respectively, and in female
29    mice (1/10, 2/10, 10/10) at 300, 600, and 1,360 mg/kg-day, respectively. Final body weights
30    were statistically significantly lower than vehicle controls in male mice at 370, 700, and
31    1,360 mg/kg-day (12, 16, and 23%, respectively) and female mice at 600 and 1,400 mg/kg-day
32    (11 and 12%, respectively) (Table 4-5).  Feed consumption was slightly less than controls in
33    males at >700 mg/kg-day, but similar to controls in females.
34
      August, 2009                            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 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
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 sexes 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.
     August, 2009
                                       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 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
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
aMean ± 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).
August, 2009
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
       Clinical chemistry findings in the mice are summarized in Tables 4-7 and 4-8 and
included statistically significant decreases in total serum protein in males at >200 mg/kg-day,
total serum protein in females at >300 mg/kg-day, and serum albumin in females at 1,400 mg/kg-
day (NTP, 2004). Decreased serum albumin could not fully account for the decreased total
protein concentrations, 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 SDH levels 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 was observed in
females at >160 mg/kg-day (22, 38, 41, and 16%, respectively), and a statistically significant
increase in ALT 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 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 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 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
5.4±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.
      bStatistically significantly different from control value.
      Source:  NTP (2004).
20
      August, 2009
                                        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
      aMean ± standard error.
      bStatistically significantly different from control value.
      Source:  NTP (2004).
 1
 2           The histopathological results in the B6C3Fi mice are summarized in Table 4-9. A
 3    statistically significant increased incidence of minimal to moderate hepatocyte hypertrophy was
 4    observed at >160 mg/kg-day in females and >200 mg/kg-day in males.  The incidence of
 5    hepatocellular necrosis was statistically significantly increased in male mice at >370 mg/kg-day
 6    and in female mice at >300 mg/kg-day. A statistically significant increased incidence of
 7    pigmentation and bile duct hyperplasia occurred at >300 mg/kg-day in females and >370 mg/kg-
 8    day in males. Additionally, the histological findings included an increased incidence of preputial
 9    gland atrophy in males in the 100, 700, and  1,360 mg/kg-day dose groups (Table 4-9), but this
10    effect did not appear dose-related. Based on the serum chemistry changes at > 160 mg/kg-day
11    and clear evidence of histopathology at higher doses, a NOAEL of 80 mg/kg-day  and a LOAEL
12    of 160 mg/kg-day were identified based on liver toxicity.
13
     August, 2009
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)
      "Values 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    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 sexes 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%,
     August, 2009
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 during the first 5 weeks of the study. The male and female rats also demonstrated
 4    an increased incidence of endemic chronic murine pneumonia. Incidences of chronic murine
 5    pneumonia in the vehicle control, low-, and high-dose groups were 40, 68, and 76% in females
 6    and 55, 50, and 65% in males. Clinical observations included squinted or reddened eyes in all
 7    control and treated groups of both sexes, but these effects occurred with greater frequency in the
 8    exposed rats. There was a low or moderate incidence of labored breathing, wheezing, and/or
 9    nasal discharge in all control and treated groups during the first year of the study, and near the
10    end of the study these signs were observed more frequently in the exposed animals.
11          Dose-related decreases in body weight gain were observed. However,  as the study
12    approached termination (weeks 100-110), the differences in body weight across the dose groups
13    decreased.
14          Histopathological effects included a dose-related increased incidence of hepatic fatty
15    metamorphosis in high-dose males (2/20, 0/20, 2/50, and 9/49 in the untreated control, vehicle
16    control, low-dose, and high-dose groups,  respectively). In addition, inflammation, focal cellular
17    changes, and angiectasis were observed in male and female rats but were not statistically
18    significant or biologically relevant. NCI (1978) stated that the inflammatory, degenerative, and
19    proliferative lesions observed in the control and dosed animals were similar in incidence and
20    type to those occurring in naturally aged rats.
21          A statistically significant increase in tumor incidence was not observed in the rats;
22    however, two hepatocellular carcinomas,  which are rare tumors in male Osborne-Mendel rats
23    (NCI, 1978), as well as one neoplastic nodule, were observed in the high-dose males
24    (Table 4-10). A hepatocellular carcinoma was also observed in an untreated female control.
25    Although interpretation of this study is confounded by the chronic murine pneumonia, it is
26    unlikely to have contributed to the fatty metamorphosis observed in the liver of male rats.
27
      August, 2009                           28           DRAFT - DO NOT CITE OR QUOTE

-------
            Table 4-10. Incidence of neoplasms in male Osborne-Mendel rats exposed to
            l,l?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
1/70
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
     August, 2009
29
DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
vehicle control group was statistically significantly increased over the incidence in historical
controls (NCI, 1978).
             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
 6
 7
 8
 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 sexes. 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.
       A very  slight decrease in body weight gain occurred in the high-dose male mice.  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
      August, 2009
                                       30
DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
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 that were statistically significantly increased were limited to hydronephrosis (16/46) and
chronic inflammation in the kidneys (5/46) in high-dose females, chronic inflammation in the
low- (13/39) and high-dose (10/47) males, and acute toxic tubular nephrosis in high-dose male
mice that died during weeks 69 and 70.
       Statistically significant (p < 0.001) increases in the incidences of hepatocellular
carcinomas occurred in both sexes and at both dose levels (Table 4-12).  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 sexes 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.
             Table 4-12. 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
21
22
23
24
25
"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 4-13).  Lymphomas were observed in
      August, 2009
                                       31
DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
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 4-13. Incidence of additional neoplasms in male and female B6C3Fi
            mice exposed to l,l?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
 4
 5
 6
 7
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.
 9    4.2.2.  Inhalation Exposure
10    4.2.2.1. Subchronic Studies
11          Truffert et al. (1977) exposed groups of female Sprague-Dawley rats (55/dose) to
12    1,1,2,2-tetrachloroethane vapor at reported calculated atmospheric concentrations of 0 or
13    560 mL/m3 5 days/week for 15 weeks (78 exposures).  The daily exposure duration was 6 hours
14    for the first 8 exposures and 5 hours for the remaining 70 exposures. There is uncertainty
15    regarding the actual concentration employed due to the unusual unit of exposure (i.e., mL/m3). It
16    is assumed that mL/m3 is a volume/volume vapor concentration, so the reported concentration is
17    equivalent to 560 ppm (3,909 mg/m3).  Interim sacrifices were conducted after 2, 4, 9, 19, 39,
18    and 63 exposures, although the number of animals killed at each time period was not reported.
19          This study is limited by poor reporting quality and minimal  quantitative data.
20    Pronounced prostration was observed "after the first exposures to 1,1,2,2-tetrachloroethane,
21    followed by recovery". Body weight gain was decreased at the end of the study, but the
22    magnitude of the change was not reported.  Increases in relative liver weights were observed
23    beginning 15 days after exposure initiation, but were not quantified. Hematological alterations
24    consisting of a slight decrease in hematocrit "confirmed by the joint RBC and WBC counts"
25    were observed at the end of the study, but were not quantified. A marked increase (313%) in
26    thymidine uptake in hepatic DNA was observed after four exposures, but by the ninth exposure
27    the thymidine uptake had decreased to levels similar to controls. Histological alterations were
      August, 2009
                                       32
DRAFT - DO NOT CITE OR QUOTE

-------
 1    observed in the liver after nine exposures and included granular appearance, cytoplasmic
 2    vacuolization, and evidence of hyperplasia (increase in the number of binucleated cells and the
 3    appearance of mitosis), but the alterations regressed after 19 exposures and were no longer
 4    observed after 39 exposures. Incidences and severity of the liver lesions were not reported.
 5    Considering the lack of incidence and severity data and other inadequately reported results, lack
 6    of information on dose-response due to the use of a single exposure level, and uncertainty
 7    regarding the exposure concentration, a NOAEL or LOAEL cannot be identified from this study.
 8          Horiuchi et al. (1962) exposed one adult male monkey (Macaco, cynomolga Linne) to
 9    1,1,2,2-tetrachloroethane for 2 hours/day, 6 days/week for a total of 190 exposures in 9 months.
10    The exposure level was 2,000-4,000 ppm (13,700-27,500 mg/m3) for the first 20 exposures,
11    1,000-2,000 ppm (6,870-13,700 mg/m3) for the next 140 exposures, and 3,000-4,000 ppm
12    (20,600-27,500 mg/m3) for the last 30 exposures. The TWA concentration was 1,974 ppm
13    (13,560 mg/m3). The  authors noted that the monkey was weak after approximately seven
14    exposures and had diarrhea and anorexia between the 12th and 15th exposures. Beginning at the
15    15th exposure, the monkey was "almost completely unconscious falling upon his side" for 20-
16    60 minutes after each  exposure. The authors noted a gradual increase in body weight during
17    months 3-5 followed by a gradual decrease until the study  was terminated.  Hematological
18    parameters demonstrated sporadic changes in hematocrit and RBC and WBC counts, but the
19    significance of these findings  cannot be determined because there were no clear trends, only one
20    monkey was tested, and there was no control group. Histological alterations consisted of fatty
21    degeneration in the liver and splenic congestion,  and no effects were observed in the heart, lung,
22    kidney, pancreas, or testis. This study cannot be used to identify a NOAEL or LOAEL for
23    subchronic exposure due to the use of a single animal without a control.
24          A 6-month inhalation study in rats was performed by the Mellon Institute of Industrial
25    Research (1947). Groups of 12 male and 12 female albino rats were exposed to 0 or 167 ppm
26    (1,150 mg/m3) of 1,1,2,2-tetrachloroethane for 7  hours/day on alternate days for the 6-month
27    study period. A statistically significant increase (15%) in kidney weight was observed in the
28    1,1,2,2-tetrachloroethane-exposed rats. The rats  also appeared to develop lung lesions following
29    exposure to tetrachloroethane; however, the study authors stated that the pathology reported for
30    tetrachloroethane must be discounted due to approximately 50% of the control  animals
31    demonstrating major pathology of the kidney, liver, or lung. Meaningful interpretation of these
32    results is precluded by the observed endemic lung infection, which resulted in significant early
33    mortality in all of the rats (57  and 69% mortality in the control and tetrachloroethane-exposed
34    groups, respectively).  This study also included one mongrel dog that followed the same study
35    design and evaluation as the rats. Serum phosphatase levels,  mean of 33 units/100 mL, and
36    blood urea nitrogen levels, mean of 20.66%, were increased in the treated dog compared to
37    control values of 5.72/100 mL and  14.94%, respectively. The dog survived the 6-month
38    exposure with effects that included cloudy swelling of the liver and of the convoluted tubules of

      August, 2009                           33          DRAFT - DO NOT CITE OR QUOTE

-------
 1    the kidney, and light congestion of the lungs. Identification of a LOAEL or NOAEL is
 2    precluded by poor study reporting, high mortality in the rats, and the use of a single treated
 3    animal in the dog study.
 4          Kulinskaya and Verlinskaya (1972) examined effects of 1,1,2,2-tetrachloroethane on the
 5    blood acetylcholine system in Chinchilla rabbits exposed to 0 or 10 mg/m3 (0 or 1.5 ppm)
 6    3 hours/day, 6 days/week for 7-8.5 months. The animals were immunized twice, at 1.5-2 and
 7    4 months, subcutaneously with a 1.2 and 1.5 billion microbe dose of typhoid vaccine in an
 8    attempt to reveal changes in the immunological reactivity following 1,1,2,2-tetrachloroethane
 9    exposures. The exposed group contained six animals, and the size of the control group was not
10    specified. In comparison with both initial and control levels, serum acetylcholine levels were
11    decreased after 1.5 months, significantly increased after 4.5 months, and significantly decreased
12    at the end of the study. The concentration of acetylcholine in the blood was increased following
13    the first immunization. No changes in serum acetylcholinesterase activity were reported,
14    although serum butyrylcholinesterase activity was reduced after 5-6 months of exposure. This is
15    a poorly reported study that did not examine any other relevant endpoints.  A NOAEL or
16    LOAEL could not be identified because the changes in acetylcholine were inconsistent across
17    time and incompletely quantified, and the biological significance of the change is unclear.
18
19    4.2.2.2. Chronic Studies
20          In a chronic inhalation study by Schmidt et al. (1972), groups of 105 male rats were
21    exposed to 0 or 0.0133 mg/L (13.3 mg/m3) 1,1,2,2-tetrachloroethane for 4 hours daily for up to
22    265 days. Subgroups of seven treated and seven control rats were killed after 110 or 265 days of
23    exposure and 60 days after exposure termination, with the remaining animals observed until
24    natural death.  There were no significant alterations in survival. Weight gain in exposed rats was
25    2.1,11.6, and 12.2% less than controls on study days 110, 260, and 324, although the only
26    statistically significant decreases in body weight gain occurred between days 90  and 170. Other
27    statistically significant changes included increased leukocyte (89%) and Pi-globulin (12%) levels
28    compared to controls after 110 days, and an increased percentage of segmented nucleated
29    neutrophils (36%), decreased percentage of lymphocytes (17%), and increased percentage of
30    liver total fat content (34%) after 265 days. There was a statistically significant decrease in
31    y-globulin levels (32%) at 60 days postexposure and a decrease in adrenal ascorbic acid content
32    (a measure of pituitary adrenocorticotropic hormone [ACTH] activity) at all three time periods
33    (64, 21, and 13%,  respectively).  This study is insufficient for identification of a NOAEL or
34    LOAEL for systemic toxicity because the experimental design and results were poorly reported,
35    and histological examinations were not conducted.
36
      August, 2009                           34           DRAFT - DO NOT CITE OR QUOTE

-------
 1    4.3. REPRODUCTIVE/DEVELOPMENTAL STUDIES—ORAL AND INHALATION
 2    4.3.1.  Oral Exposure
 3          Gulati et al. (1991a) exposed timed-pregnant CD Sprague-Dawley rats (8-9 animals/
 4    group) to diets containing 0, 0.045, 0.135, 0.27, 0.405, or 0.54% microencapsulated
 5    1,1,2,2-tetrachloroethane from gestation days (GDs) 4 through 20.  Based on body weight and
 6    food consumption data, the reported estimated doses of 1,1,2,2-tetrachloroethane were 0, 34, 98,
 7    180, 278, or 330 mg/kg-day. Dams were sacrificed and litters were evaluated on GD 20.
 8    Evaluations included maternal body weight, feed consumption and clinical signs, uterine weight,
 9    and numbers of implantations, early and late resorptions, live fetuses, and dead fetuses.
10    Necropsies were performed on the maternal animals, but fetuses were not examined for
11    malformations.
12          All dams survived to study termination on GD 20. Maternal body weight was
13    statistically significantly decreased 9, 11, 14, and 24% at 98, 108, 278, and 330 mg/kg-day,
14    respectively, compared to controls, and demonstrated a dose-dependent and time-dependent
15    decrease in all dose groups. However, an increase in maternal body weight on day 20, compared
16    to body weight on day 4, was  apparent for all dose groups.  Daily food consumption was
17    significantly decreased in all dose groups, and this may have contributed to the decreased body
18    weights observed in the study. Four out of nine rats in the 278 mg/kg-day dose group had
19    slightly rough fur beginning on GD 10, while rough fur was present in all animals in the
20    330 mg/kg-day dose group. No  statistically significant changes were observed in the numbers of
21    live fetuses/litter, dead fetuses/litter, resorptions/litter, or implants/litter. One dam in the
22    98 mg/kg-day group and four  of nine dams in the 330 mg/kg-day group completely resorbed
23    their litters.  At scheduled sacrifice, average fetal weights were statistically significantly
24    decreased 3.9, 12.7, 10.5, and 20.6% in the 98, 108, 278, and 330 mg/kg-day dose groups,
25    respectively. Gravid uterine weight was statistically significantly reduced only in the
26    330 mg/kg-day animals. Small but statistically significant decreases were seen in maternal body
27    weight and average fetal weight at >98 mg/kg-day. Using statistical significance and a 10%
28    change as the criterion for an adverse change in maternal body weight, a NOAEL of 34 mg/kg-
29    day and LOAEL of 98 mg/kg-day were selected for changes in maternal body weight. A
30    NOAEL of 34 mg/kg-day and LOAEL of 98 mg/kg-day were selected for developmental toxicity
31    based  on the lowest dose that caused a statistically significant decrease in fetal body weight.
32          Gulati et al. (1991b) exposed timed-pregnant Swiss CD-I mice (n = 5-11) to diets
33    containing 0, 0.5, 1, 1.5, 2, or  3% microencapsulated 1,1,2,2-tetrachloroethane from GDs 4
34    through 17.  Based on body weight and food consumption data, the reported estimated doses of
35    1,1,2,2-tetrachloroethane were 0, 987, 2,120, 2,216, or 4,575 mg/kg-day; an average dose could
36    not be calculated for the 3% group due to early mortality. Dams were sacrificed and litters were
37    evaluated on GD 17. Evaluations included maternal body weight, feed consumption and clinical
38    signs,  uterine weight,  and numbers of implantations, early and late resorptions, live fetuses, and

      August, 2009                           3 5          DRAFT - DO NOT CITE OR QUOTE

-------
 1    dead fetuses. Necropsies were performed on the maternal animals, but fetuses were not
 2    examined for malformations.
 3          All animals (9/9) in the 3% group died prior to the end of the study. Mortality was 0/11,
 4    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
 5    the mortality in the higher dose groups affected the statistical power of the study for those groups.
 6    Maternal body weights were statistically significantly decreased compared to controls at
 7    >2,120 mg/kg-day beginning on study day 9, although the day 17 data were not statistically
 8    significantly different from controls for any treatment group. Average daily feed consumption
 9    was statistically significantly decreased in all treated groups except in the 987 mg/kg-day
10    animals. Gross hepatic effects were reported in dams from all groups except the 987 mg/kg-day
11    group and included pale or grey and/or enlarged livers and a prominent lobulated pattern.
12    Complete litter resorption occurred in 1/11, 0/9, 2/8, 1/1, and 1/2 dams in the 0, 987, 2,120,
13    2,216, and 4,575 mg/kg-day groups, respectively.  No changes in developmental endpoints were
14    noted in the 987 or 2,120 mg/kg-day groups. The 2,120 and 4,575 mg/kg-day groups had too
15    few litters, due to maternal toxicity, to permit statistical analysis of the findings. The high
16    mortality in the exposed mice precluded the identification of a NOAEL or LOAEL for this study.
17          NTP (2004) conducted a 14-week study in which groups of 10 male and 10 female
18    F344 rats were fed  diets containing microencapsulated 1,1,2,2-tetrachloroethane at reported
19    average  daily doses of 0, 20, 40, 80, 170,  or 320 mg/kg-day. The main part of this study is
20    summarized in Section 4.2.1.1.  Reproductive function (fertility) was not evaluated. Endpoints
21    relevant to reproductive toxicity included histology (testis with epididymis and seminal vesicle,
22    preputial gland, prostate gland, clitoral gland, ovary, and uterus) and weights (left cauda
23    epididymis, left epididymis, and left testis) of selected reproductive tissues in all control  and
24    treated groups. Sperm evaluations and vaginal cytology evaluations were performed in animals
25    in the 0, 40, 80, and 170 mg/kg-day dose groups. The sperm evaluations consisted of spermatid
26    heads per testis and per gram testis, spermatid counts, and epididymal spermatozoal motility and
27    concentration.  The vaginal cytology evaluations consisted of measures of estrous cycle length.
28          Sperm motility was 17.1, 14.9, and 24.0% lower than in vehicle controls at 40, 80, and
29    170 mg/kg-day, respectively. Other statistically significant effects in the males included
30    reductions in absolute epididymis weight at >80 mg/kg-day and absolute left cauda epididymis
31    weight at 170 mg/kg-day,  and statistically significant increases in the incidences (90-100%) of
32    minimal to moderate atrophy of the preputial and prostate gland, seminal vesicle, and testicular
33    germinal epithelium at 320 mg/kg-day. Effects in the females included statistically significant
34    increases in incidences of minimal to mild uterine atrophy (70-90%) at > 170 mg/kg-day and
35    clitoral gland atrophy (70%) and ovarian interstitial cell cytoplasmic alterations (100%) at
36    320 mg/kg-day. The vaginal cytology evaluations indicated that the females in the 170 mg/kg-
37    day group spent more time in diestrus and less time in proestrus, estrus, and metestrus than did
      August, 2009                           36          DRAFT - DO NOT CITE OR QUOTE

-------
 1    the vehicle controls. Body weight loss and reduced body weight gain at the lower dose levels
 2    may have contributed to the atrophy and other effects observed in both sexes (NTP, 2004).
 3          NTP (2004) also tested groups of 10 male and 10 female B6C3Fi mice that were
 4    similarly exposed to 1,1,2,2-tetrachloroethane for 14 weeks at reported average daily dietary
 5    doses of 0, 100, 200, 370, 700, or 1,360 mg/kg-day (males) or 0, 80, 160, 300, 600, or
 6    1,400 mg/kg-day (females).  The main part of this study is summarized in Section 4.2.1.1.
 7    Reproductive function (fertility) was not evaluated, and toxicity endpoints in reproductive organs
 8    are the same as those evaluated in the rat part of the study summarized above.  The sperm and
 9    vaginal cytology evaluations were performed in the 0, 1,120, 4,550, or 9,100 mg/kg-day dose
10    groups.
11          Effects observed in the male mice included statistically significant increases in the
12    incidence of preputial gland atrophy at 100, 700, and 1,360 mg/kg-day (incidences in the control
13    to high dose groups were 0/10, 4/10, 2/10, 0/10, 4/10, and 5/10, respectively), decreased absolute
14    testis weight at >700 mg/kg-day and absolute epididymis and cauda epididymis weights at
15    1,360 mg/kg-day, and decreased epididymal spermatozoal motility at 1,360 mg/kg-day (3.1%
16    less than vehicle controls). In female mice, the length of the estrous cycle was significantly
17    increased at 9,100 pm (1,400 mg/kg-day) (8.7% longer than vehicle controls).  The pronounced
18    decreases in body weight gain or body weight loss were similar to those observed in rats.
19
20    4.3.2.  Inhalation Exposure
21          Male rats were exposed to 0 or 15 mg/m3 (2.2 ppm) 1,1,2,2-tetrachloroethane 4 hours/day
22    for up to 8 days in a 10-day period  (Gohlke and Schmidt, 1972; Schmidt et al., 1972).
23    Reproductive function was not tested, but evaluations included histological examinations of the
24    testes in groups of seven control and seven treated males following the second, fourth, and eighth
25    exposures, as detailed in Schmidt et al. (1972) in Section  4.2.2.2. This study is limited by
26    imprecise and incomplete reporting of results. It was noted that testicular histopathology,
27    described as atrophy of the seminal tubules with strongly restricted or absent spermatogenesis,
28    was observed in five exposed animals following the fourth exposure; data for the other time
29    periods and the control group were not reported.
30          The Schmidt et al. (1972) chronic inhalation study, summarized in Section 4.2.2.2,
31    included a limited reproductive function/developmental toxicity assessment. Male rats were
32    exposed to 0 or 13.3 mg/m3 (1.9 ppm) 1,1,2,2-tetrachloroethane 4 hours/day for 265 days, as well
33    as during the mating period.  One week before the end of the exposure period, seven control and
34    seven exposed males were each mated with five unexposed virgin females. Dams were
35    permitted to deliver and the offspring were observed for 84 days and were examined
36    macroscopically for malformations. The percentage  of mated females having offspring, littering
37    interval, time to 50% littered, total  number of pups, pups/litter, average birth weight,  postnatal
38    survival on days 1, 2, 7,  14, 21, and 84, sex ratio, and average body weight on postnatal day 84

      August, 2009                            37           DRAFT - DO NOT CITE OR QUOTE

-------
 1    were also measured. No macroscopic malformations or significant group differences in the other
 2    indices were found, indicating that 13.3 mg/m3 was a NOAEL for male reproductive toxicity.
 3          No effects attributable to 1,1,2,2-tetrachloroethane were reported in rats exposed to 5 or
 4    50 ppm (34.3 or 343 mg/m3, respectively) 7 hours/day for 5 days in a dominant lethal test
 5    (McGregor, 1980).  A viral infection may have resulted in increased numbers of early deaths in
 6    all groups, including the control group, possibly affecting study sensitivity. The frequency of
 7    sperm with hook abnormalities was statistically significantly increased in the 343 mg/m3 group,
 8    but not at 34.3 mg/m3.
 9
10    4.4. OTHER DURATION- OR ENDPOINT-SPECIFIC STUDIES
11    4.4.1. Acute Studies (Oral and Inhalation)
12    4.4.1.1.  Oral Studies
13          Oral (single-dose gavage) median lethal dose (LD50) values of 250-800 mg/kg have been
14    reported in rats (NTP, 2004; Schmidt et al., 1980b; Gohlke et al., 1977; Smyth et al., 1969).
15    Cottalasso et al. (1998) described a series of experiments  evaluating the  effect of a single gavage
16    dose of 1,1,2,2-tetrachloroethane on the liver of exposed rats. In the first experiment, male
17    Sprague-Dawley rats (5/group) were given a single gavage dose of 0, 143.5, 287, 574, or
18    1,148 mg/kg in mineral oil and five animals from each group were sacrificed 5, 15, 30, or
19    60 minutes later. Sixty minutes after treatment, statistically significant (p < 0.10), dose-related
20    increases in serum levels of AST (66, 129, and 201%, respectively) and ALT (54, 88, and 146%,
21    respectively) were observed at >287 mg/kg. The increase in rat serum activities of AST and
22    ALT were also increased in a time-dependent manner.  AST increased 13-130% from 5 to
23    60 minutes in rats at 574 mg/kg-day and ALT increased 8-88% from 5 to 60 minutes. A
24    statistically significant decrease in microsomal G6Pase activity (19, 36, and 47%, respectively)
25    was observed at >287 mg/kg. A statistically significant decrease in levels of dolichol, a
26    polyisoprenoid compound believed to be important in protein glycosylation reactions, in the liver
27    (41 and 56%, respectively) and a statistically significant increase in triglyceride levels in liver
28    homogenate (60 and 83%, respectively) were observed at >574 mg/kg. A statistically significant
29    increase in the trigylceride levels in liver microsomes (46, 65, and 97%,  respectively) was
30    observed at >287 mg/kg. See Table 4-14 for a summary of these acute liver toxicity results. A
31    time-dependent effect was observed in the decrease in G6Pase, in the increase in triglyceride
32    levels, and in the decrease in levels of dolichol in the liver at 574 mg/kg-day from 5 to
33    60 minutes.
34
      August, 2009                           3 8           DRAFT - DO NOT CITE OR QUOTE

-------
             Table 4-14.  Effects of acute (60 minutes) l,l?2,2-tetrachloroethane
             treatment on rat liver
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
      "Significantly different from control.
      Source: Cottalasso et al. (1998).
 1
 2           Schmidt et al. (1980b) administered 0 or 100 mg/kg doses of 1,1,2,2-tetrachloroethane in
 3    corn oil by gavage to groups of 10 male Wistar rats, followed immediately by increased
 4    environmental temperatures, and evaluated hepatic effects 20-22 hours post administration.
 5    Statistically significant increases in serum leucine aminopeptidase, hepatic ascorbic acid, and
 6    hepatic triglyceride levels (10.5, 22.3, and 125% greater than control levels, respectively) were
 7    observed, but changes in body weight, liver weight, hepatic N-demethylation of aminopyrine,
 8    and serum ALT were not observed.  The report includes a general statement that all chemicals
 9    tested in this study led to necrosis and fatty degeneration, which suggests that 100 mg/kg was a
10    hepatotoxic dose of 1,1,2,2-tetrachloroethane. However, the significance of the histology results
11    cannot be assessed due to a lack of incidence and severity measures. No other 1,1,2,2-tetra-
12    chloroethane-related histological data were reported in this study.
13           Wolff (1978) exposed 8- to 10-week-old, female Wistar rats in groups of 8-10 animals,
14    to a single gavage dose of 0, 25, or 50 mg/kg of 1,1,2,2-tetrachloroethane 30 minutes prior to
15    testing for passive avoidance (shock level of 0.4 milliamperes [mA]). Passive avoidance was
16    measured by allowing the test rats to explore the test apparatus, which consisted of a larger, lit
17    box and a smaller, dark box. After 180 seconds, the darkened box received an electrical shock
18    through the grid floor. During the 180 seconds, the rats remained in the darkened box
19    approximately 80% of the time. The test was repeated 24 hours later. No differences in
20    avoidance were observed between the control and 25 mg/kg groups, but decreased passive
21    avoidance behavior was reported following exposure to 50 mg/kg. In the second test series, the
22    shock level  was increased to 0.8 mA and the 1,1,2,2-tetrachloroethane dose was increased to
23    50 mg/kg.  The 1,1,2,2-tetrachloroethane doses were then increased to 80 mg/kg and then to
24    100 mg/kg.  Increasing the shock level to 0.8 mA resulted in no significant differences in
25    avoidance between the controls and the 50 mg/kg-day dose group (n = 10).  Passive avoidance
26    was altered  at 80 mg/kg (n = 10),  and at 100 mg/kg, the animals (n = 10) were ataxic  and did not
27    learn to avoid the shock. The authors stated that the treatment with 1,1,2,2-tetrachloroethane
      August, 2009
39
DRAFT - DO NOT CITE OR QUOTE

-------
 1    may have affected the threshold of perception of the shock, rather than memory (Wolff, 1978).
 2    This conclusion would be consistent with the high-dose anesthetic effects characteristic of
 3    volatile organic compounds in general.
 4
 5    4.4.1.2. Inhalation Studies
 6          Schmidt et al. (1980a) established a 24-hour median lethal concentration (LCso) of
 7    8,600 mg/m3 (1,256 ppm) for 1,1,2,2-tetrachloroethane in rats for a single 4-hour exposure.
 8    Carpenter et al. (1949) found that a 4-hour exposure to 1,000 ppm 1,1,2,2-tetrachloroethane
 9    (6,870 mg/m3) was lethal in Sherman rats, with mortality in "2/6, 3/6, or 4/6" animals.
10          Price et al. (1978) exposed rats and guinea pigs to 576, 5,050, and 6,310 ppm
11    1,1,2,2-tetrachloroethane for 30 minutes.  Rats exposed to 576 ppm (3,950 mg/m3) for
12    30 minutes showed a slight reduction in activity and alertness, while increasing the concentration
13    to 5,050 or 6,310 ppm (34,700 or 43,350 mg/m3) caused lacrimation, ataxia, narcosis, labored
14    respiration, and 30-50% mortality (Price et al., 1978).  Eye closure, squinting, lacrimation, and
15    decreased activity were observed in guinea pigs exposed to 576 ppm for 30 minutes; exposure to
16    5,050 ppm resulted in tremors, narcosis, and labored breathing, and exposure to 6,310 ppm
17    caused 30% mortality (Price et al., 1978). Organ weight measurements and gross pathology and
18    histology evaluations performed 14 days following the 30-minute exposures did not result in
19    chemical-related effects in the lungs, liver, kidneys, heart, brain, adrenals, testes, epididymides,
20    ovaries, or uterus in either species.
21          Pantelitsch (1933) exposed groups of three mice to 1,1,2,2-tetrachloroethane concent-
22    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,
23    5,220, or 6,120 ppm, respectively) for approximately 1.5-2 hours and examined changes in
24    clinical status of the animals.  All concentrations resulted in disturbed equilibrium, prostration,
25    and loss of reflexes, with deaths occurring at >8,000-10,000 mg/m3; increasing the concentration
26    resulted in a more rapid onset of symptoms.
27          Horvath and Frantik (1973) determined that effective concentrations of 1,1,2,2-tetra-
28    chloroethane following a single 6-hour exposure in rats were  360 ppm (2,470 mg/m3) for a 50%
29    decrease in spontaneous motor activity and 200  ppm (1,370 mg/m3) for a 50% increase in
30    pentobarbital sleep time. No additional relevant information was reported.
31          Schmidt et al. (1980a) exposed groups  of 10 male Wistar rats to 0, 410, 700, 1,030, 2,100,
32    or 4,200 mg/m3 (0, 60, 102, 150, 307,  or 613 ppm, respectively) 1,1,2,2-tetrachloroethane (mean
33    concentrations) for 4 hours and evaluated the animals immediately (within 15-100 minutes), at
34    24 hours, or at 120 hours following exposure.  The purpose of this study was to determine a
35    threshold concentration for effects on the liver following inhalation exposure. Evaluation of this
36    study is complicated by imprecise and incomplete reporting of results, exposure levels, and
37    observation durations.  For example, results for endpoints other than liver histology, ascorbic
38    acid content, and histochemistry were not reported for the lowest concentration (410 mg/m3), and

      August, 2009                           40          DRAFT - DO NOT CITE OR QUOTE

-------
 1    liver ascorbic acid content and serum and liver triglyceride levels were the only results reported
 2    quantitatively.  Histological effects included diffuse fine droplet fatty degeneration in the liver at
 3    410 and 700 mg/m3 (24 hours postexposure), nonspecific inflammation and Councilman bodies
 4    (eosinophilic globules derived from necrosis of single hepatocytes) in the liver at 4,200 mg/m3
 5    (24 hours postexposure), and interstitial nephritis in the kidneys at 700 mg/m3 (120 hours
 6    postexposure).  Additional information on these findings, including incidences and results for
 7    other exposure concentrations, was not reported.
 8          Hepatic ascorbic acid levels were statistically significantly increased in groups exposed
 9    to >700 mg/m3 immediately after exposure (2, 64, 29, 167, and 182% higher than controls at 410,
10    700, 1,030, 2,100, and 4,200 mg/m3, respectively), but returned to control levels within 24 hours.
11    Serum triglyceride concentrations were statistically significantly decreased at >700 mg/m3 after
12    24 hours (35, 23, 29, and 56% at 700, 1,030, 2,100, and 4,200 mg/m3, respectively) and at
13    2,100 and 4,200 mg/m3  (39 and 42%, respectively) after 120 hours. Hepatic triglyceride levels
14    were significantly increased at 2,100 and 4,200 mg/m3 (92 and 76%, respectively) at 24 hours
15    postexposure. Hexobarbital sleep time was increased at 2,100 and 4,200 mg/m3 (not quantified).
16    Assessing the biological significance and adversity of the effects in this study is complicated by
17    factors that include the lack of liver lesion incidence data, the paucity of other quantitative data,
18    and other reporting insufficiencies. The authors concluded that the threshold for effects on the
19    liver was between 410 and 700 mg/m3 because the fine droplet fatty degeneration was not
20    considered to be biologically significant in the absence of accompanying serum and liver
21    biochemical changes.
22          Hepatic effects were also reported by Tomokuni (1969), who administered a single
23    3-hour exposure of 600  ppm (4,120 mg/m3) 1,1,2,2-tetrachloroethane to female Cb mice.  Total
24    hepatic lipids and triglycerides were statistically significantly increased following exposure and
25    continued to increase for 8 hours postexposure. Hepatic triglyceride levels increased more than
26    total lipid levels for 8 hours postexposure. Total hepatic adenosine triphosphate (ATP) levels
27    were decreased immediately following exposure and continued to decrease over the next 8 hours.
28    A later study by the same investigator (Tomokuni, 1970) evaluated female Cb mice (5-8/group)
29    exposed to 800 ppm (5,490 mg/m3) 1,1,2,2-tetrachloroethane for 3 hours and then followed the
30    time-course of the changes  in hepatic lipids and phospholipids over the next 90 hours. Increased
31    tricglyceride and decreased phospholipid levels were seen for the first 30-45 hours postexposure,
32    but the effects generally resolved by 90 hours postexposure,  demonstrating that hepatic effects
33    resolved after exposure  was terminated.
34          Horiuchi et al. (1962) exposed 10 male mice for a single 3-hour period to an atmosphere
35    containing 5,900 ppm (-40,500 mg/m3) or 6,600 ppm (-45,300 mg/m3) 1,1,2,2-tetrachloroethane
36    and then observed the animals for 1 week following exposure.  Tissues were obtained for
37    histologic evaluation from animals at sacrifice or when discovered dead. Three mice exposed to
38    5,900 ppm and four mice exposed to 6,600 ppm died prior to the end of the study.

      August, 2009                           41           DRAFT - DO NOT CITE OR QUOTE

-------
 1          Deguchi (1972) administered a single 6-hour exposure of 0, 10, 100, or 1,000 ppm (0, 69,
 2    690, or 6,900 mg/m3, respectively) of 1,1,2,2-tetrachloroethane to male rats and evaluated serum
 3    AST and ALT levels up to 72 hours postexposure.  This study was reported in Japanese and
 4    included an English translation of the abstract.  Based on information in the English abstract and
 5    data graphs in this Japanese study, there was a slight increase in serum AST at all exposure
 6    concentrations 72 hours postexposure.
 7
 8    4.4.2.  Short-term Studies (Oral and Inhalation)
 9    4.4.2.1. Oral Studies
10          Dow Chemical Company (1988) exposed groups of male Osborne-Mendel rats (n = 5) to
11    daily gavage doses of 0, 25, 75, 150, or 300 mg/kg-day 1,1,2,2-tetrachloroethane every 24 hours
12    for 4 days, followed by an injection of [3H]-thymidine, for DNA incorporation studies,  24 hours
13    following the last 1,1,2,2-tetrachloroethane dose. The fourth dose was not administered to the
14    300 mg/kg-day group due to signs of central nervous system (CNS) depression and debilitation,
15    and one animal in this group died before [3H]-thymidine injection. Terminal body weights of the
16    300 mg/kg-day animals were statistically significantly decreased 17% compared to controls.
17    Absolute liver weights at the highest dose were decreased and relative liver weights were
18    statistically significantly increased 14% in the 150 mg/kg-day dose group.
19          Histological examinations of the livers showed increased numbers of hepatocytes in
20    mitosis in the 75,  150, and 300 mg/kg-day groups, although this response was variable in high-
21    dose rats due, possibly, to the increased toxicity observed in this group (Dow Chemical
22    Company, 1988).  Increased numbers of reticuloendothelial cells were seen at 300 mg/kg-day.
23    Increased glycogen was found in hepatocytes of 75 and  150 mg/kg-day animals, although this
24    could be an outcome of altered feeding patterns resulting from sedative effects of dosing (Dow
25    Chemical Company, 1988).
26          Hepatic DNA synthesis ([3H]-thymidine incorporation) was increased 2.8-, 4.8-, and
27    2.5-fold at 75, 150, and 300 mg/kg-day, respectively; the decline at 300 mg/kg-day may have
28    been due to the poor clinical status of the rats in this group (Dow Chemical Company, 1988).
29    Total hepatic DNA content was not increased.  Other endpoints were not evaluated.  The  300
30    mg/kg-day dose is a frank effect level (PEL) based on the CNS depression and mortality. The 75
31    mg/kg dose may represent a NOAEL for increased relative liver weight in rats. However, the
32    increase in DNA synthesis and mitosis are not necessarily indicative of hepatotoxicity,  and the
33    histological examinations showed no accompanying degeneration or other adverse liver lesions.
34          Dow Chemical Company (1988) similarly exposed groups of male B6C3Fi mice (n = 5)
35    to daily gavage doses of 0, 25, 75, 150, or 300 mg/kg-day 1,1,2,2-tetrachloroethane for 4  days,
36    followed by [3H]-thymidine injection for the DNA incorporation studies. All animals survived
37    treatment, and changes in body weight were  not observed at any dose level. Absolute and
38    relative liver weights were increased 13 and  11%, respectively, at 150 mg/kg-day and 19  and

      August, 2009                           42            DRAFT - DO NOT CITE OR QUOTE

-------
 1    72%, respectively, at 300 mg/kg-day, although only the increase in relative liver weight at 300
 2    mg/kg-day was statistically significantly (p = 0.05).
 3          Histopathologic examination of the liver revealed centrilobular swelling, with a
 4    corresponding decrease in hepatocyte size in the periportal region due to decreased glycogen
 5    content, in mice at >75 mg/kg-day. Increased hepatocyte mitosis was also observed in mice at
 6    300 mg/kg-day. Hepatic DNA synthesis was increased 1.7-fold at 150 mg/kg-day and 4.4-fold at
 7    300 mg/kg-day, although total hepatic DNA content was not increased. Other endpoints were
 8    not evaluated.
 9          TSI Mason Laboratories (1993a, unpublished) administered 1,1,2,2-tetrachloroethane  in
10    corn oil to groups of male and female (n = 5) F344/N rats at 0, 135, 270, or 540 mg/kg for
11    12 days over a  16-day period.  Rats were weighed prior to dosing, after 7 days, and prior to
12    euthanasia, and all surviving rats were euthanized and subject to necropsy. Study endpoints
13    included clinical observations, body weight, necropsy, selected organ weights (liver, kidney,
14    thymus, lung, heart, and testes), and histology of gross lesions. All of the high-dose rats died by
15    day 5 of the study. Male rats exposed to 270 mg/kg displayed an increase in body weight from
16    day 1 through day 17 of 37%, compared to an increase of 64% in controls. Female rats exposed
17    to 270 mg/kg displayed a decrease in body weight from day 1 through day 17 of 3%, compared
18    with an increase of 30% in controls. The automatic watering system  for the low- and high-dose
19    males failed prior to the administration of 1,1,2,2-tetrachloroethane, and the low and high doses
20    of the study were repeated in a subsequent study by TSI Mason Laboratories (1993b,
21    unpublished).
22          Clinical signs were absent in the 135 mg/kg animals, but animals exposed to 270 or
23    540 mg/kg were lethargic following treatment. Absolute liver weights were statistically
24    significantly increased (19%) in the 135 mg/kg-day female rats, while relative liver weights were
25    statistically significantly increased at both 135 and 270 mg/kg-day (16 and 34%, respectively).
26    No changes in absolute or relative liver weights were seen in exposed male rats.  Absolute right
27    kidney weight was significantly increased 9 and 37% in females at 135 and 270 mg/kg-day,
28    respectively. Absolute thymus weight was statistically significantly decreased in the mid-dose
29    group of male rats (33% at 270 mg/kg-day) while absolute (45%)  and relative (32%) thymus
30    weights were statistically significantly decreased in only the mid-dose females. Relative right
31    testis weight was statistically significantly increased (10% at 270 mg/kg-day) in male rats.
32    Absolute, but not relative, lung weights were statistically significantly decreased in 270  mg/kg-
33    day females (17%), while relative heart weights were statistically  significantly increased (14%)
34    in females.
35          Gross and microscopic lesions were observed in the liver (i.e., hepatodiaphragmatic
36    nodules) of one control, one mid-dose, and one high-dose rat, but these were common
37    spontaneous lesions.
      August, 2009                           43           DRAFT - DO NOT CITE OR QUOTE

-------
 1          In another study, TSI Mason Laboratories (1993b, unpublished) exposed groups of male
 2    F344/N rats (n = 5) to 0, 135, 270, or 540 mg/kg-day 1,1,2,2-tetrachloroethane by gavage in corn
 3    oil on 12 days in a 16-day period. Study endpoints included clinical observations, body weight,
 4    necropsy, selected organ weights (liver, kidney, thymus, lung, heart, and testes), and histology of
 5    gross lesions.  All animals exposed to 540 mg/kg-day died by day 3 of the study. Rats in the
 6    270 and 540 mg/kg-day groups were extremely lethargic following administration of the test
 7    article, with recovery observed only in the 270 mg/kg-day rats.
 8          The weight gain observed in the low- and mid-dose rats was 55.2 and 28%, respectively.
 9    At 135 mg/kg, statistically significant increases of 17 and 13% in absolute and relative liver
10    weights, respectively, were observed compared to controls. In the mid-dose group, statistically
11    significant decreases in absolute testes weight (7%), absolute kidney weight (9%), absolute and
12    relative heart weight (10 and 6%, respectively), and absolute and relative thymus weight (33 and
13    21%, respectively) were observed.  Statistically significant increases in relative thymus (10%),
14    liver (16%), and kidney weights (7%) were observed at 270 mg/kg compared to controls.
15          Gross and microscopic lesions were observed in the liver of one 270 mg/kg-day male and
16    in the glandular stomach of one 540 mg/kg-day male, but these were diagnosed as spontaneous
17    lesions commonly observed in F344/N rats.  The lesion observed in the liver was a dark nodule
18    on the median lobe and corresponded histomorphologically to a hepatodiaphragmatic nodule,
19    and the lesion observed in the glandular stomach was a pale foci.
20          TSI Mason Laboratories (1993c, unpublished) exposed groups of five male and five
21    female B6C3Fi mice to 0, 337.5, 675, or 1,350 mg/kg-day 1,1,2,2-tetrachloroethane by gavage in
22    corn oil on 12 days during a 16-day period.  Study endpoints included clinical observations, body
23    weight, necropsy, selected organ weights (liver, kidney, thymus, lung, heart, and testes),  and
24    histology of gross lesions. All mice of both sexes in the 1,350 mg/kg-day groups were found
25    dead or euthanized by day 3 of the study. Additionally, one 675 mg/kg-day female died  and one
26    337.5 mg/kg-day female was euthanized prior to the end of the study.
27          No significant changes in body weight were reported in treated groups.  Animals  in the
28    675 and 1,350 mg/kg-day groups appeared lethargic within 15 minutes of dosing, and the
29    1,350 mg/kg-day mice failed to recover after the third treatment. Lethargy also occurred in the
30    337.5 mg/kg-day female that was sacrificed, but not in other animals in that exposure group.  In
31    male mice, relative liver weight was statistically significantly increased 9% at 337.5 mg/kg, and
32    absolute and relative liver weights were statistically significantly increased 28 and 37%,
33    respectively, at 675 mg/kg-day. In female mice, absolute and relative liver weights were
34    statistically significantly increased by 50 and 42%, respectively, at 675 mg/kg.
35          Gross hepatic changes, described as pale livers, were noted in one male  and three females
36    at 337.5 mg/kg-day and in four males and three females at 675 mg/kg-day.  Histological
37    examination of the gross lesions showed that they correlated with centrilobular  hepatocellular
38    degeneration characterized by hepatocellular swelling, cytoplasmic rarefaction, and

      August, 2009                            44           DRAFT - DO NOT CITE OR QUOTE

-------
 1    hepatocellular necrosis in the 675 and 1,350 mg/kg-day males and the 337.5, 675, and
 2    1,350 mg/kg-day females. Hepatocellular necrosis was the most common lesion observed at
 3    675 mg/kg-day.
 4          In a study examining the potential renal toxicity of orally administered halogenated
 5    ethanes, groups of five male F344/N rats received 0, 0.62, or 1.24 mmol/kg-day 1,1,2,2-tetra-
 6    chloroethane by gavage in corn oil (0, 104, or 208 mg/kg-day, respectively) for 21 consecutive
 7    days (NTP, 1996).  All rats in the high-dose group died or were killed moribund on days 13-14
 8    and were not evaluated further. Evaluations of the 0 and 104 mg/kg-day animals included
 9    weekly body weights, end-of-study urinalysis (volume, specific gravity, creatinine, glucose, total
10    protein, AST, y-glutamyl transpeptidase, and N-acetyl-p-D-glucosaminidase), gross necropsy,
11    selected organ weights (right kidney, liver, and right testis), selected histopathology (right kidney,
12    left liver lobe, and gross lesions),  and kidney cell proliferation analysis (proliferating cell nuclear
13    antigen [PCNA] labeling index for proximal and distal tubule epithelial cells in S phase).
14    Clinical signs in the high-dose animals included thinness and lethargy (5/5 rats), diarrhea,
15    abnormal breathing, and ruffled fur (3/5 rats). In the low-dose group,  no effects on survival,
16    body weight gain, urinalysis parameters, absolute or relative kidney weights, renal or testicular
17    histopathology, or kidney cell PCNA labeling index were observed.
18          Hepatic effects in the low-dose group included increased absolute and relative liver
19    weights (24 and 29% greater than controls, respectively) and cytoplasmic vacuolization of
20    hepatocytes. The vacuolation occurred in hepatocytes of all low-dose rats and consisted of
21    multifocal areas with clear droplets within the cytoplasm. Changes in the kidneys of the male
22    rats were not observed.
23          In a range-finding study, the NTP (NTP, 2004; TSI Mason Laboratories, 1993d) exposed
24    male and female F344/N rats (5/sex/group) to 0, 3,325, 6,650, 13,300, 26,600, or 53,200 ppm
25    1,1,2,2-tetrachloroethane in the diet (microcapsules) for 15 days.  Unexposed and vehicle control
26    groups were also evaluated, with the latter being given feed with empty microcapsules.  Study
27    endpoints included clinical observations, body weight, food consumption, necropsy, selected
28    organ weights (liver, kidney, thymus, lung, heart, and testes), and histology of gross lesions;
29    histology was not evaluated in animals without gross lesions. The study authors reported that
30    average daily doses for the three lowest concentrations were 300, 400, or 500 mg/kg-day for both
31    sexes. All rats exposed to 26,600 or 53,200 ppm were killed moribund on day 11.  The average
32    daily doses for these groups were not reported.
33          Female rats exposed to 400 mg/kg-day and both sexes exposed to 500 mg/kg-day were
34    thin and displayed ruffled fur. Body weight at study termination was statistically significantly
35    lower than controls in both sexes of all treated groups. Male rats exposed to 300 mg/kg-day
36    showed decreased weight gain compared to controls and those exposed to higher doses lost
37    weight, with final body weights in male rats 28, 46, and 53% less than vehicle controls at 300,
38    400, and 500 mg/kg-day, respectively. Females lost weight at doses of >300 mg/kg-day, with

      August, 2009                           45           DRAFT - DO NOT CITE OR QUOTE

-------
 1    final body weights in female rats 25, 38, and 47% less than vehicle controls at 300, 400, and
 2    500 mg/kg-day, respectively. Decreased feed consumption likely contributed to the decreased
 3    weight gains because consumption was reduced in a dose-related manner in both sexes of all
 4    treated groups (NTP, 1996).
 5          Absolute thymus weights were decreased 24, 69, and 84% in male rats and 37, 61, and
 6    81% in female rats at doses of >300 mg/kg-day and relative thymus weights were decreased
 7    42 and 65% in male rats and 38 and 65% in female rats at >400 mg/kg-day (NTP, 2004; TSI
 8    Mason Laboratories, 1993d). In male rats, absolute liver weights were decreased 22, 49, and
 9    60% compared to controls at 300, 400, and 500 mg/kg-day, respectively.  Relative liver weight
10    was increased 7% compared to controls at 300 mg/kg-day and decreased  14% compared to
11    controls at 500 mg/kg-day. In female rats, absolute liver weight was decreased 25  and 34%
12    compared to controls at 400 and 500 mg/kg-day, respectively, and relative liver weight was
13    increased 34 and 23% compared to controls at 300 and 500 mg/kg-day, respectively. Relative
14    kidney weights were increased 14, 26, and 18% in male rats at 300, 400, and 500 mg/kg-day,
15    respectively, and 17 and 36% in female rats at 400 and 500 mg/kg-day, respectively. Absolute
16    kidney weights were decreased 17, 32, and 45% in males and 16, 27, and 27% in females at 300,
17    400, and 500 mg/kg-day, respectively. Other organ weight decreases were considered a
18    reflection of the decreased body weights.
19          Focal areas of alopecia occurred on the  skin of four female rats in the 500 mg/kg-day
20    group, and these lesions correlated with minimal to moderate acanthosis,  which is an abnormal
21    benign increase in the thickness of the stratum spinosum, a layer of cells that is capable of
22    undergoing mitotic cell division, of the epidermis.  In the liver, mild or moderate centrilobular
23    degeneration was observed microscopically in the exposed male and female rats.
24          Groups of five male and five female B6C3Fi mice were exposed to 0, 3,325, 6,650,
25    13,300, 26,600, or 53,200 ppm of encapsulated 1,1,2,2-tetrachloroethane  in the diet for 15 days
26    (NTP, 2004; TSI Mason Laboratories, 1993d).  Organ weights, gross necropsy, and histology of
27    gross lesions were evaluated in  surviving mice  at the termination of the study. Average daily
28    doses were not determined by the study authors because feed consumption could not be
29    measured accurately due to excessive scattering of feed.  All male and female mice exposed to
30    53,200 ppm, all males exposed to 26,600 ppm,  and two males exposed to 13,300 ppm were
31    sacrificed in extremis before the end of the study. Final body weights were decreased 16, 24,
32    and 22%, in comparison to vehicle controls, in males at 3,325, 6,650, and 13,300 ppm,
33    respectively. In females, final body weights were decreased  9, 20, 31, and 34% at  3,325, 6,650,
34    13,300, and 26,600 ppm, respectively.
35          Clinical findings included hyperactivity in males and females exposed to 3,325, 6,650, or
36    13,300 ppm and in females in the 26,600 ppm group. Males in the 26,600 and 53,200 ppm
37    groups were lethargic. Males exposed to >6,650 ppm and females exposed to 26,600 and
38    53,200 ppm were thin and had ruffled fur.  A statistically significant decrease in absolute (31, 47,

      August, 2009                           46          DRAFT - DO NOT CITE OR QUOTE

-------
 1    82, and 81%, respectively) and relative (22, 33, 74, and 72%, respectively) thymus weights
 2    compared to controls was observed in all exposed female mice. Relative liver weights were
 3    statistically significantly increased 22, 31, and 34% in male mice at 3,325, 6,650, and
 4    13,300 ppm, respectively.  Absolute liver weights were statistically significantly decreased 11, 9,
 5    and 5% in female mice at 6,650, 13,300, and 26,600 ppm, respectively, and relative liver weight
 6    increased 30 and 44% at 13,300 and 26,600 ppm, respectively.  Other organ weight changes
 7    were associated with changes in body weight.  Pale or mottled livers were noted in all exposed
 8    groups of male and female mice and correlated microscopically with hepatocellular degeneration,
 9    which was characterized by hepatocellular swelling, cytoplasmic rarefaction, single paranuclear
10    vacuoles, hepatocellular necrosis, and infrequent mononuclear infiltrates. The severity of the
11    hepatic changes increased with increasing exposure concentration.
12          The histological examinations in the surviving mice showed hepatocellular degeneration
13    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,
14    13,300, 26,600, and 53,200 ppm, respectively (TSI Mason Laboratories, 1993d).  For both sexes,
15    the lesions tended to be minimal to mild at 3,325 and 6,650 ppm, with more moderate to marked
16    severity observed at the higher doses.
17          The National Cancer Institute (NCI, 1978) conducted a range-finding study in rats and
18    mice in order to estimate the maximum tolerated dose for administration in the chronic bioassay.
19    In this study, Osborne-Mendel  rats (5/sex/group) received gavage doses of 0 (vehicle control
20    group), 56,  100, 178, 316, or 562 mg/kg 1,1,2,2-tetrachloroethane in corn oil 5 days/week for
21    6 weeks, followed by a 2-week observation period. B6C3Fi mice (5/sex/group) were similarly
22    exposed to 0, 32, 56, 100, 178, or 316 mg/kg 1,1,2,2-tetrachloroethane. It appears that mortality
23    and body weight gain were the only endpoints used to assess toxicity and determine the high-
24    dose levels for the NCI (1978)  chronic bioassays in rats and mice.  In the rats, one male exposed
25    to 100 mg/kg and all five females exposed to 316 mg/kg died (mortality rates in the 562 mg/kg
26    groups were not reported). Body weight gain was reduced 3, 9, and 38% in male rats and 9, 24,
27    and 41% in female rats at 56, 100, and 178 mg/kg-day, respectively. No deaths or significant
28    alterations in body weight gain were observed in the mice. In male rats, 100 and 178 mg/kg-day,
29    were selected as the NOAEL and LOAEL, respectively, for the observed decrease in body
30    weight, while in female rats the NOAEL and LOAEL were 56 and  100 mg/kg-day, respectively,
31    for the same endpoint.  The highest dose in mice, 316 mg/kg-day, was selected as the NOAEL
32    for body weight changes and mortality.
33
34    4.4.2.2. Short-term Inhalation Studies
35          Rats (n = 84) were exposed to 0  or 15  mg/m3 (2.2 ppm) 1,1,2,2-tetrachloroethane
36    4 hours/day for up to 8 days in a 10-day period (Gohlke and Schmidt, 1972; Schmidt et al., 1972).
37    Following the first, third, and seventh exposures, seven control and exposed rats were given an
      August, 2009                            47          DRAFT - DO NOT CITE OR QUOTE

-------
 1    unknown amount of ethanol.  Evaluations were performed on seven males from the control and
 2    treated groups, with and without ethanol, following the second, fourth, and eighth exposures.
 3          Statistically significant changes included increased serum total protein and decreased
 4    serum ai- and a2-globulin fractions compared to controls after the eighth exposure (day 10),
 5    although the difference was not quantified (Schmidt et al., 1972). Histological effects included a
 6    fine to medium droplet fatty degeneration of the liver that involved increasing numbers of
 7    animals with increasing duration of exposure, although the incidences and severity were not
 8    reported (Gohlke and Schmidt, 1972). The results of the serum and histochemical evaluations
 9    were illegible  in the best copy of the translated reference available.  Testicular atrophy in the
10    seminal tubules was observed in five treated animals following the fourth exposure (Gohlke and
11    Schmidt,  1972). This study is limited by imprecise and incomplete reporting of results.
12    Assessment of the adversity of liver and other effects in this study is complicated by the
13    reporting insufficiencies, particularly the paucity of incidence and other quantitative data, as well
14    as effects that were not consistently observed in the three time periods and a lack of information
15    on dose-response due to the use of a single exposure level.
16          Horiuchi et al. (1962)  exposed nine male mice to an average concentration of
17    approximately 7,000 ppm (48,000 mg/m3) 1,1,2,2-tetrachloroethane for 2 hours once/week for a
18    total of five exposures over 29 days.  All animals died during the study with none of the deaths
19    occurring during exposure, and most (5/9) of the mice died within 5 days of the first exposure.
20    The only other reported findings in the exposed animals were moderate congestion and fatty
21    degeneration of the liver and congestion of "other main tissues."
22          Horiuchi et al. (1962)  exposed six male rats to an average concentration of 9,000 ppm
23    (62,000 mg/m3) 1,1,2,2-tetrachloroethane 2 hours/day, 2-3 times a week for 11 exposures in
24    29 days. All rats died during  the study. No changes in body weight were reported. Exposed
25    animals generally showed hypermotility within the first few minutes of exposure,  followed by
26    atactic gait within approximately 20 minutes and eventual near-complete loss of consciousness
27    1-1.5 hours after the onset of exposure. Hematology was assessed in three rats that survived
28    beyond 2 weeks, and two of these  animals showed a decrease in RBC count  and Hb content.
29    Exposed animals generally showed moderate congestion and fatty degeneration of the liver and
30    congestion of "other main tissues."
31          As discussed in Section 4.2.2.1, one monkey was exposed to varying concentrations
32    (2,000-4,000 ppm for the first 20 exposures,  1,000-2,000 ppm for the 20th-160th exposure, and
33    3,000-4,000 ppm for the  remaining exposures)  of 1,1,2,2-tetrachloroethane for 2 hours/day,
34    6 days/week for 9 months (Horiuchi et al., 1962). Effects of short-term exposure included
35    weakness after seven exposures, diarrhea and anorexia between the 12th and 15th exposures, and
36    beginning at the 15th exposure, near-complete unconsciousness for 20-60 minutes after each
37    exposure.
38

      August, 2009                            48           DRAFT - DO NOT CITE OR QUOTE

-------
 1    4.4.3. Acute Injection Studies
 2          Paolini et al. (1992) exposed groups of male and female Swiss Albino mice to a single i.p.
 3    dose of 0, 300, or 600 mg/kg 1,1,2,2-tetrachloroethane and sacrificed the animals 24 hours after
 4    dosing to assess hepatotoxicity. An LDso of 1,476 mg/kg for 1,1,2,2-tetrachloroethane was
 5    calculated using six animals/dose and eight dose groups.  At 600 mg/kg, absolute and relative
 6    liver weights were statistically significantly decreased 16 and 37%, respectively, in female mice.
 7    No changes in total microsomal protein were noted.  Statistically significant decreases (37-74%)
 8    in hepatic cytochrome P450 enzymes of numerous classes were reported at both dose levels in
 9    male and female mice (see Section 3.3).  Other hepatic enzymes with statistically significantly
10    decreased activity included NADPH-cytochrome c-reductase, 5-aminolevulinic acid-synthetase,
11    ethoxyresorufm-O-deethylase, pentoxyresorufm O-depentylase, GST (600 mg/kg only), and
12    epoxide hydrolase.  Total hepatic heme was reduced at both doses, and heme oxygenase activity
13    was increased in a dose-related manner, but was statistically significant only in high-dose males
14    and females.
15          Wolff (1978) exposed groups of female Wistar rats to a single i.p.  dose of 0, 20, or
16    50 mg/kg 30 minutes prior to testing for passive avoidance of a 0.4 mA electric shock. No
17    differences between the control and 25 mg/kg groups were reported, but doses of 50 mg/kg
18    resulted in decreased passive avoidance behavior.  Similarly, no differences were seen in the
19    open-field test at any dose level.  In male ICR-mice, a single i.p. dose of 20 mg/kg resulted in a
20    significant reduction in spontaneous locomotor activity, and 50-60 mg/kg resulted in a 50%
21    reduction (Wolff, 1978).
22          In an abstract, Andrews et al. (2002) described the exposure of a rat whole embryo
23    culture system to 1,1,2,2-tetrachloroethane.  Gestational day 9 embryos were exposed to
24    concentrations between 0.5 and 2.9 mM  1,1,2,2-tetrachloroethane for 48 hours and then
25    evaluated for morphological changes. At concentrations >1.4 mM, 1,1,2,2-tetrachloroethane
26    resulted in rotational defects and anomalies of the heart and eye.  Embryo lethality was observed
27    at >2.4 mM.
28
29    4.4.4. Immunotoxicological Studies
30          Shmuter (1977) exposed groups of 12 Chinchilla rabbits to 0, 2, 10, or 100 mg/m3 (0, 0.3,
31    1.5, or 14.6  ppm, respectively) 1,1,2,2-tetrachloroethane 3 hours/day, 6 days/week for 8-
32    10 months.  Animals were vaccinated with 1 mL of a 1.5  x 109 suspension of heated typhoid
33    vaccine 1.5, 4.5-5, and 7.5-8 months after the initiation of 1,1,2,2-tetrachloroethane exposure.
34    Significant increases and decreases in total antibody levels were observed in the 2 and
35    100 mg/m3 groups, respectively.  No significant changes in 7S-typhoid antibody levels were
36    observed. Significant alterations in the levels of "normal" hemolysins to the Forsman's antigen
37    of sheep erythrocytes were observed in the 10 and 100 mg/m3 groups, as levels were increased in
38    the 10 mg/m3  group after 1.5, 2, and 2.5 months of exposure, decreased after 4 months, and

      August, 2009                            49           DRAFT - DO NOT CITE OR QUOTE

-------
 1   absent at 5 months of exposure. Levels of these hemolysins were decreased in the 100 mg/m
 2   group during the first 6 months of exposure.  Increases in the electrophoretic mobility of specific
 3   antibodies following 1,1,2,2-tetrachloroethane were also reported.  Exposure to 100 mg/m3
 4   1,1,2,2-tetrachloroethane resulted in a decrease in the relative content of antibodies in the
 5   y-globulin fraction and an increase in the T and [3 fractions.  This is a poorly reported study that
 6   provides inadequate quantitative data. The inconsistent dose-response patterns preclude
 7   assessing biological significance and identification of a NOAEL or LOAEL.
 9
10
11
12
13
14
15
4.5.  MECHANISTIC DATA AND OTHER STUDIES IN SUPPORT OF THE MODE OF
ACTION
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 4-15.
             Table 4-15.  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
Reverse
mutation
Forward
mutation
DNA damage
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)
NA
10 uL/plate
10 uL/plate
NA
10-3,333 uL/plate
NA
5-1,000 uL/plate
NA
0.06-2,979 nmol/
plate
10 uL/plate
15-236 mM
-
NP
-
-
-
-
-
-
-
NP
+
-
+
-
-
-
-
-
-
-
+
—
Nestmann et al., 1980
Rosenkranz, 1977;
Bremetal., 1974
Mitoma et al., 1984
Onoetal., 1996
NTP, 2004
Milman et al., 1988
Haworth et al., 1983
Warner etal., 1988
Roldan-Arjona et al.,
1991
Rosenkranz, 1977;
Bremetal., 1974
DeMarini and Brooks,
1992
     August, 2009
                                        50
DRAFT - DO NOT CITE OR QUOTE

-------
       Table 4-15. Results of in vitro and in vivo genotoxicity studies of
       l,l?2,2-tetrachloroethane
Saccharomyces Gene
cerevisisae conversion
Gene
reversion
Gene
recombina-
tion
Aspergillus Mitotic
nidulans crossover
(b) Mammalian cell assays
Mouse Lymphoma Gene
mutation
Hepatocytes DNA repair
(primary)

Test system
D7

D7

D7
PI
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
+
-
+
-
+
+
Callenetal., 1980
Nestmann and Lee, 1983
Callenetal., 1980
Nestmann and Lee, 1983
Callenetal., 1980
Crebellietal., 1988

L5178Y
Osborne
Mendel rats
B6C3FJ mice
25-500 nL/mL
NA
NA
-
NP
NP
-
-
-
NTP, 2004
Milmanetal., 1988;
Williams, 1983
In vitro chromosomal damage assays
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
50ppm
50ppm
589-9, 100 ppm
200 mg/kg
50-1,000 (mg/kg)
50-1,000 mg/kg
150 mg/kg
-
+
+
+
-
-
+
McGregor, 1980
NTP, 2004
Miyagawa et al., 1995
Mirsalis et al., 1989
Dow Chemical Co., 1988
August, 2009
51
DRAFT - DO NOT CITE OR QUOTE

-------
             Table 4-15. Results of in vitro and in vivo genotoxicity studies of
             l,l?2,2-tetrachloroethane
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
      + = positive; - = negative/no change; CHO = Chinese hamster ovary; NA = not available; NP = assay not
      performed; UDS = unscheduled DNA synthesis
 1
 2           1,1,2,2-Tetrachloroethane has been shown to be predominantly inactive in reverse
 3    mutation assays in Salmonella typhimurium (strains TA97, TA98, TA100, TA1530, TA1535,
 4    TA1537, and TA1538), either with or without the addition of S9 metabolic activating mixture,
 5    even at concentrations that lead to cytotoxicity (NTP, 2004; Ono et al., 1996; Milman et al.,
 6    1988; Warner et al., 1988; Mitoma et al., 1984; Haworth et al., 1983; Nestmann et al., 1980).
 7    Two studies reported reverse mutation activity in S. typhimurium (Rosenkranz,  1977; Brem  et al.,
 8    1974).  Results of studies employing methods to prevent volatilization were not notably different
 9    from those that did not use those methods.  1,1,2,2-Tetrachloroethane also did not induce
10    forward mutations (L-arabinose resistance) in S. typhimurium strain BA13 (Roldan-Arjona et al.,
11    1991).  Assays with Escherichia coli indicated that 1,1,2,2-tetrachloroethane induced DNA
12    damage, as shown by growth inhibition in DNA polymerase deficient E. coli (Rosenkranz, 1977;
13    Brem et al., 1974) and induction of prophage lambda (DeMarini and Brooks, 1992).  In
14    Saccharomyces cerevisiae, 1,1,2,2-tetrachloroethane induced gene conversion, reversion, and
15    recombination in  one study (Callen et al., 1980), whereas another study found no conversion or
16    reversion (Nestmann and Lee, 1983).  In Aspergillus nidulans, 1,1,2,2-tetrachloroethane induced
17    aneuploidy, but no crossing over (Crebelli et al., 1988).
18           1,1,2,2-Tetrachloroethane did not induce trifluorothymidine resistance in L5178Y mouse
19    lymphoma cells, with or without S9, at concentrations up to those producing lethality (NTP,
20    2004).  Primary hepatocytes  from rats and mice exposed in vitro to 1,1,2,2-tetrachloroethane did
21    not show altered DNA  repair at concentrations that were not cytotoxic (Milman et al., 1988;
22    Williams, 1983).  McGregor (1980) reported no increase in unscheduled DNA synthesis (UDS)
23    in human embryonic intestinal fibroblasts exposed to 1,1,2,2-tetrachloroethane.  Treatment of
24    Chinese hamster ovary (CHO) cells with up to 653 ug/mL  (which was cytotoxic) did not result in
25    increased induction of chromosomal aberrations (NTP, 2004; Galloway et al., 1987) but did
26    produce an increased frequency  of sister chromatid exchanges (SCEs) at concentrations of
27    >55.8 ug/mL (NTP, 2004; Galloway et al., 1987). SCEs were also induced in BALB/C-3T3  cells
      August, 2009
52
DRAFT - DO NOT CITE OR QUOTE

-------
 1    treated in vitro with high concentrations (>500 ug/mL) of 1,1,2,2-tetrachloroethane, either with
 2    or without S9 activating mixture (Colacci et al., 1992).
 3          In BALB/C-3T3 cells, 1,1,2,2-tetrachloroethane exposure of up to 250 ug/mL in the
 4    absence of exogenous metabolic activation did not result in increased numbers of transformed
 5    cells (Colacci et al., 1992; Milman et al., 1988; Tu et al., 1985; Arthur Little, Inc., 1983);
 6    survival was generally >70%. Higher doses (>500 ug/mL) were capable of transforming the
 7    cells, but also showed higher levels of cytotoxicity (Colacci et al.,  1990). However, even
 8    relatively low levels (31.25 ug/mL) of 1,1,2,2-tetrachloroethane used as an initiating agent,
 9    followed by promotion with 12-O-tetradecanoylphorbol-13-acetate, resulted in increased
10    numbers of transformed cells (Colacci et al., 1992). 1,1,2,2-Tetrachloroethane did not act as a
11    promoter in BALB/c-313 cells in vitro without metabolic activation (Colacci et al., 1996).
12          1,1,2,2-Tetrachloroethane tested negative for sex-linked recessive lethal mutations and
13    mitotic recombination inD. melanogaster (NTP, 2004; Vogel and Nivard, 1993; Woodruff et al.,
14    1985; McGregor, 1980).  Replicative DNA synthesis was increased in hepatocytes isolated from
15    male B6C3Fi mice exposed to a single gavage dose of 200 mg/kg  (24 and 48 hours
16    postexposure) or 400 mg/kg (24, 39, and 48 hours postexposure) relative to hepatocytes from
17    unexposed mice (Miyagawa et al., 1995).  Hepatocytes isolated from mice following a single
18    gavage dose of up to 1,000 mg/kg did not show an increase in UDS or S-phase DNA synthesis
19    (Mirsalis et al., 1989). Hepatocytes isolated from B6C3Fi mice 6  hours after a single gavage
20    dose of 150 mg/kg in corn oil demonstrated irreversible alkylation of hepatic DNA (Dow
21    Chemical Co., 1988).  Inhalation exposure to 5 or 50 ppm (34.3 or 343 mg/m3) for 7 hours/day,
22    5 days/week did not result in increased frequency of chromosomal aberrations in bone marrow
23    cells isolated from male rats (McGregor, 1980); female rats exposed to 50 ppm (343 mg/m3), but
24    not to 5 ppm (34.3 mg/m3), showed an increase in bone marrow cell aberrations other than gaps
25    (McGregor, 1980).
26          In summary, genotoxicity studies provide limited evidence of a mutagenic mode of action.
27    1,1,2,2-Tetrachloroethane has some genotoxic activity, but in vitro genotoxicity tests generally
28    reported negative results.  Similarly, in vivo studies had mostly negative results with the
29    exception of chromosomal aberrations in female rat bone marrow cells and micronucleus
30    formation in mouse bone marrow peripheral erythrocytes.  The results of rat liver preneoplastic
31    foci and mouse BALB/C-3T3 cell neoplastic transformation assays suggest that  1,1,2,2-tetra-
32    chloroethane may have initiating and promoting activity.  Overall, results of genotoxicity studies
33    for 1,1,2,2-tetrachloroethane are mixed and insufficient for establishing  a mutagenic mode of
34    action.
35
36    4.5.2.  Short-Term Tests of Carcinogenicity
37          Treatment of partially hepatectomized male Osborne-Mendel rats with a single
38    100 mg/kg gavage dose of 1,1,2,2-tetrachloroethane, followed by 7 weeks of promotion with

      August, 2009                           53           DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
phenobarbital in the diet, did not result in increased numbers of preneoplastic (GGT-positive)
foci in the liver (Milman et al., 1988; Story et al., 1986).  Exposure of partially hepatectomized
male Osborne-Mendel rats to a single i.p. dose of diethylnitrosamine (DEN) as an initiating agent
followed by promotion with 100 mg/kg-day of 1,1,2,2-tetrachloroethane by gavage 5 days/week
for 7 weeks caused a significantly increased number of GGT-positive foci in the liver (Milman et
al., 1988; Story et al., 1986).  1,1,2,2-Tetrachloroethane also significantly increased the number
of GGT-positive foci in rats administered the promotion protocol in the absence  of the DEN
initiator.  The study authors concluded that 1,1,2,2-tetrachloroethane induces hepatocarcino-
genesis primarily through a promoting mechanism (Story et al., 1986).
      Using a mouse strain that had been shown to be susceptible to pulmonary adenomas
when exposed to organic chemicals, Theiss et al. (1977) administered i.p. injections of 80, 200,
or 400 mg/kg 1,1,2,2-tetrachloroethane in Tricaprylin 5-18 times to groups of 20 male A/St mice
for 8 weeks. There was a dose-related increase in number of lung tumors/mouse (Table 4-16),
and the dose-response was nearly statistically significant (p < 0.05) (Theiss et al., 1977).

       Table 4-16. 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
£>-value
0
24
0
15/20
0.27 ±0.15

80
5
400
10/20
0.30 ±0.21
0.897
200
18
3,600
15/20
0.50 ±0.14
0.271
400
16
6,400
5/20
1.00 ±0.45
0.059
16
17
18
19
20
21
22
23
Source: Thiessetal. (1977).

       Maronpot et al. (1986) tested 65 chemicals at three doses in 6- to 8-week-old male and
female strain A/St or A/J mice housed 10/cage. Doses were set based on the highest dose
exhibiting a lack of overt toxicity from a preliminary dose-setting study, with the mid and low
dose as half the higher dose.  Mice were injected i.p. 3 times/week for 8 weeks. Lungs were
examined histologically. The data for 1,1,2,2-tetrachloroethane-exposed male and female strain
A/St are presented in Table 4-17.
      August, 2009
                                       54
DRAFT - DO NOT CITE OR QUOTE

-------
            Table 4-17. 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 mouseb
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
      "Numerator is number of mice alive at study termination; denominator is number of mice started on study.
      bBased on all surviving mice at study termination.
      Source: Maronpot et al. (1986).
 1
 2    4.6. SYNTHESIS OF MAJOR NONCANCER EFFECTS
 3    4.6.1. Oral
 4    4.6.1.1. Human Data
 5          Information on the acute oral toxicity of 1,1,2,2-tetrachloroethane in humans is available
 6    from several case reports.  Based on amounts of 1,1,2,2-tetrachloroethane recovered from the
 7    gastrointestinal tract of deceased subjects following intentional ingestion (Mant, 1953; Sherman,
 8    1953; Lilliman, 1949; Forbes, 1943; Elliot, 1933; Hepple,  1927), estimated lethal doses ranged
 9    from 273 to 9,700 mg/kg.  Patients who accidentally consumed a known volume of 1,1,2,2-tetra-
10    chloroethane, corresponding to single doses ranging from 68 to 117 mg/kg, as medicinal
11    treatment for hookworm experienced loss of consciousness and other clinical signs of narcosis
12    (Ward, 1955; Sherman, 1953).  Chronic oral effects of 1,1,2,2-tetrachloroethane in humans have
13    not been reported in the literature.
14
15    4.6.1.2. Animal Data
16          Few studies have evaluated acute oral toxicity  in animals, and the endpoints assessed
17    consist of data on lethality and neurological and liver effects (Table 4-18).  Oral LD50 values
18    ranged from 250 to 800 mg/kg in rats (NTP, 2004; Schmidt et al., 1980a; Gohlke et al., 1977;
19    Smyth et al., 1969). Neurological effects of acute, oral 1,1,2,2-tetrachloroethane administration
     August, 2009
55
DRAFT - DO NOT CITE OR QUOTE

-------
1   revealed ataxic effects and decreased passive avoidance behavior (Wolff, 1978). Hepatic
2   changes were noted in two separate acute oral toxicity studies. Male Sprague-Dawley rats
3   administered between 287 and 1,148 mg/kg 1,1,2,2-tetrachloroethane had dose-dependent
4   increases in the hepatic enzymes AST and ALT as well as a decrease in microsomal G6Pase
5   activity (Cottalasso et al., 1998).  Male Wistar rats were administered 100 mg/kg 1,1,2,2-tetra-
6   chloroethane and had increases in hepatic ascorbic acid and serum leucine aminopeptidase, but
7   no changes in serum ALT (Schmidt et al.,  1980a, b).  Both studies noted increases in triglyceride
8   levels in the liver.
    August, 2009                            56          DRAFT - DO NOT CITE OR QUOTE

-------
   Table 4-18.  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 and
ALT, increased liver
triglycerides; decreased
liver dolichol.
Increased hepatic ascorbic
acid and serum leucine
aminopeptidase
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 etal.,
1998
Schmidt etal.,
1980 a, 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;
August, 2009
57
DRAFT - DO NOT CITE OR QUOTE

-------
   Table 4-18.  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,
SDH, and cholesterol,
reduced epididymis weight.
Increased liver weight,
increased ALT, ALP, SDH,
and bile acids.
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
August, 2009
58
DRAFT - DO NOT CITE OR QUOTE

-------
       Table 4-18.  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)
CDs 4-20
CDs 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.
   August, 2009
59
DRAFT - DO NOT CITE OR QUOTE

-------
 1          Short-term oral exposure (Table 4-18) to 1,1,2,2-tetrachloroethane caused clinical signs
 2    of neurotoxicity and mortality at doses as low as 208-300 mg/kg-day by gavage in rats (NTP,
 3    1996; TSI Mason Laboratories, 1993a, b, unpublished; Dow Chemical Company, 1988). Body
 4    weight gain was decreased at similar dose levels in rats exposed by gavage or diet (NTP, 2004;
 5    TSI Mason Laboratories, 1993a, b, unpublished; Dow Chemical Company, 1988; NCI, 1978).
 6    Hepatic effects consisted of increased DNA synthesis and centrilobular swelling  in mice exposed
 7    to 75 mg/kg-day in the diet (Dow Chemical Company, 1988) and hepatocellular cytoplasmic
 8    vacuolation in rats exposed to 104 mg/kg-day (NTP, 1996).  At higher doses (337.5 mg/kg-day),
 9    hepatocellular degeneration was observed in mice (TSI Mason Laboratories, 1993c, unpublished).
10          Subchronic and chronic oral administration studies (Table 4-18) with 1,1,2,2-tetrachloro-
11    ethane in animals indicated that the liver is the most sensitive organ for toxicity.  Oral toxicity
12    studies in F344 and Osborne-Mendel rats and B6C3Fi mice were evaluated (NTP, 2004, NCI,
13    1978). The 14-week subchronic study by the National Toxicology Program (NTP, 2004) in both
14    F344 rats and B6C3Fi mice was the most comprehensive evaluation of 1,1,2,2-tetrachloroethane-
15    mediated toxicity through an orally administered route. NCI (1978) conducted a chronic study
16    on Osborne Mendel rats  and B6C3Fi mice in which dosing regimens were modified during the
17    course of the study.
18          In F344 rats, an increased incidence of hepatocellular cytoplasmic vacuolization was
19    observed at 20 mg/kg-day in males and 40 mg/kg-day in females, increased relative liver weights
20    were observed at 40 mg/kg-day, and hepatocellular hypertrophy was observed at 80 mg/kg-day
21    in the subchronic NTP (2004) study. Additional hepatic effects included increases in serum ALT
22    and SDH at 80 mg/kg-day, decreases in serum cholesterol at 80 mg/kg-day, and increases in
23    serum ALP and bile acids, hepatocellular necrosis, bile duct  hyperplasia, hepatocellular mitotic
24    alterations, foci of cellular alterations, and hepatocyte pigmentation at  170 and 320 mg/kg-day.
25    A NOAEL of 20 mg/kg-day and a LOAEL of 40 mg/kg-day  was selected based on the increase
26    in relative liver weight.  In the Osborne-Mendel rats, significant increases in hepatic fatty
27    metamorphosis were observed in male rats following a chronic exposure to 108 mg/kg-day
28    (TWA, based on changes in dosing regimen) (NCI, 1978). Mortality was significantly decreased
29    in female rats dosed at a TWA dose of 43 and 76 mg/kg-day. A NOAEL of 62 mg/kg-day and a
30    LOAEL of 108 mg/kg-day were identified in male rats based on an increased incidence of
31    hepatic fatty metamorphosis (NCI, 1978).
32          Mice  appear to be less sensitive than rats to noncancer effects mediated by orally
33    administered 1,1,2,2-tetrachloroethane. Relative liver weight was statistically significantly
34    increased in female and male B6C3Fi mice at 80 and 200 mg/kg-day, respectively. Effects in  the
35    mice also included minimal hepatocellular hypertrophy, increased serum SDH, ALT, and bile
36    acids, and decreased  serum cholesterol at 160-200 mg/kg-day, and  increased serum ALP and
37    5'-nucleotidase, necrosis, pigmentation, and bile duct hyperplasia at 300-370 mg/kg-day. Based
38    on the increase in relative liver weight observed in the NTP (2004)  study, a NOAEL of

      August, 2009                          60          DRAFT - DO NOT CITE OR QUOTE

-------
 1    100 mg/kg-day and a LOAEL of 200 mg/kg-day in male mice and a LOAEL of 80 mg/kg-day in
 2    female mice was identified .  Male and female B6C3Fi mice were also evaluated for chronic oral
 3    toxicity. A FEL for males and females of 284 mg/kg-day (TWA dose) was identified for kidney
 4    toxicity as measured by increases in hydronephrosis, kidney inflammation, and acute tubular
 5    nephrosis (NCI, 1978).
 6          Comprehensive neurobehavioral testing showed no evidence of neurotoxicity in either
 7    species at doses equal to or higher than the LOAELs based on liver effects (NTP, 2004),
 8    indicating that the liver is more sensitive than the nervous system to subchronic dietary exposure
 9    to 1,1,2,2-tetrachloroethane.
10          Developmental parameters were significantly affected by oral administration of
11    1,1,2,2-tetrachloroethane in rats and mice. Significant decreases in rat maternal and fetal body
12    weights were noted at doses of >98 mg/kg-day (Gulati et al., 1991a). Using statistical
13    significance and a 10% change as the criteria for establishing an adverse effect in maternal body
14    weight, a NOAEL of 34 mg/kg-day and LOAEL of 98 mg/kg-day were selected.  A NOAEL of
15    34  mg/kg-day and LOAEL of 98 mg/kg-day were selected for developmental toxicity based on
16    the lowest dose that caused a statistically significant decrease in fetal body weight.  In mice, the
17    FEL based on maternal toxicity and resorption of litters is 2,120 mg/kg-day (Gulati et al., 1991b).
18    The high mortality in the exposed mice precluded the identification of a NOAEL or LOAEL
19    from this study.
20          Toxicity to reproductive tissues following 1,1,2,2-tetrachloroethane exposure to adult rats
21    and mice was observed at dose levels as low as  40 mg/kg-day (NTP, 2004). In male rats, sperm
22    motility was decreased at >40 mg/kg-day. Higher doses resulted in decreased epididymal
23    absolute weight and increased  atrophy of the preputial and prostate gland, seminal vesicle, and
24    testicular germinal epithelium. In female rats, minimal to mild uterine atrophy was increased at
25    >170 mg/kg-day and clitoral gland atrophy and  ovarian interstitial cell cytoplasmic alterations
26    were increased at 320 mg/kg-day.  Female F344 rats in the 170 mg/kg-day group spent more
27    time in diestrus than did the vehicle controls.
28          Male B6C3Fi mice had increased incidences of preputial gland atrophy at >100 mg/kg-
29    day. Less sensitive effects included decreases in absolute testis weight (>700 mg/kg-day) and
30    absolute epididymis and cauda epididymis weights (1,360 mg/kg-day) and a decrease in
31    epididymal spermatozoal motility (1,360 mg/kg-day).  The only noted reproductive toxicity
32    parameter in female mice affected was a significant increase in the length of the estrous cycle at
33    a dose of 1,400 mg/kg-day (NTP, 2004).
34
35    4.6.2.  Inhalation
36    4.6.2.1. Human Data
37          Limited information is  available on the acute inhalation toxicity of 1,1,2,2-tetrachloro-
38    ethane in humans (Table 4-19). The results of an early, poorly reported experimental study with

      August, 2009                           61          DRAFT - DO NOT CITE OR QUOTE

-------
1   two volunteers suggest that 3 ppm (6.9 mg/m3) was the odor detection threshold. Irritation of the
2   mucous membranes, pressure in the head, vertigo, and fatigue were observed at 146 ppm (1,003
3   mg/m3) for 30 minutes or 336 ppm (2,308 mg/m3) for 10 minutes. Common reported symptoms
4   of high-level acute inhalation exposure to 1,1,2,2-tetrachloroethane in humans include
5   drowsiness, nausea, headache,  and weakness, and at extremely high concentrations, jaundice,
6   unconsciousness, and respiratory failure (Coyer, 1944; Hamilton, 1917).
    August, 2009                           62          DRAFT - DO NOT CITE OR QUOTE

-------
        Table 4-19. Summary of noncancer results of major human studies of inhalation exposure to l,l?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
39 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 etal., 1957
 ND = not determined; NS = not stated
1
    August, 2009
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 4-20) 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/m3) 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).
      August, 2009                           64           DRAFT - DO NOT CITE OR QUOTE

-------
   Table 4-20.  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
LCso
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
slight increase in serum AST at all exposure
concentrations 72 hours postexposure
Schmidt etal., 1980a
Schmidt et al., 1980a
Carpenter et al., 1949
Price etal., 1978
Price etal., 1978
Horvath and Frantik, 1973
Tomokuni, 1969
Tomokuni, 1970
Pantelitsch, 1933
Horiuchietal., 1962
Deguchi, 1970
August, 2009
65
DRAFT - DO NOT CITE OR QUOTE

-------
   Table 4-20.  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
Oorl5
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
0 or 1,150
0 or 1,150
OorlO
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 et al., 1962
Mellon Institute of
Industrial Research, 1947
Mellon Institute of
Industrial Research, 1947
Kulinskaya and
Verlinskaya, 1972
August, 2009
66
DRAFT - DO NOT CITE OR QUOTE

-------
       Table 4-20. 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
   August, 2009
67
DRAFT - DO NOT CITE OR QUOTE

-------
 1          Acute and short-term inhalation exposure (Table 4-20) to high concentrations (>7,000
 2    ppm) of 1,1,2,2-tetrachloroethane caused mortality and neurological and liver effects in animals.
 3    Mortality occurred in mice exposed to 7,000 ppm (48,000 mg/m3) for 2 hours once/week for 4
 4    exposures in 29 days and in rats exposed to 9,000 ppm (62,000 mg/m3) for 2 hours/day, 2-3
 5    times/week for 11 exposures in 29 days. Congestion and fatty degeneration in the liver (mice
 6    and rats), as well as a biphasic change in neurological motor activity (hyperactivity followed by
 7    ataxia, rats only), were also reported (Horiuchi et al., 1962). At the lowest inhalation exposure
 8    of 2.2 ppm (15 mg/m3) for 4 hours/day (8-10 days), rats had fine droplet fatty degeneration in
 9    the liver and changes in levels of serum proteins, but no neurological changes were reported
10    (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/m3) 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 kidney, 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, were
31    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 kidney 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.

      August, 2009                           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 aNOAEL 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 to tissues, further implicating metabolic activation
31    as a possible step in the mode of action. However, there is uncertainty  as to whether the
32    presence of radiolabel in proteins, DNA, and RNA may be radiolabeled carbon that has been
33    incorporated into biomolecules through normal biochemical processes.  Studies describing  the
34    mechanism of 1,1,2,2-tetrachloroethane-induced noncancer toxicological effects are not
35    available.
36
      August, 2009                           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 sexes was
 7    observed at doses of 142 and 284 mg/kg-day. A decrease in the time to tumor in both sexes of
 8    mice was also observed. In this same bioassay, male Osborne-Mendel rats exhibited an
 9    increased incidence of hepatocellular carcinomas, a rare tumor in this strain (NCI, 1978), at the
10    high dose only, although this increased incidence was not statistically significant.  An untreated
11    female control rat also developed a hepatocellular carcinoma. Limitations in the study included
12    increased mortality in male and female mice and the variable doses given to the mice over the
13    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 sexes 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 cause hepatocellular carcinomas and
30    hepatocellular adenomas in male and female B6C3Fi mice, respectively, but did not demonstrate
31    carcinogenicity in Osborne-Mendel or Sprague-Dawley rats due to inadequate study designs
32    (NTP, 1990; NCI, 1976).  Tetrachloroethylene, another metabolite of 1,1,2,2-tetrachloroethane,
33    was characterized by NCI (1977) as a liver carcinogen in B6C3Fi mice, but an evaluation of
34    carcinogenicity in Osborne-Mendel rats was inadequate due to early mortality. In a study by
35    NTP (1986), tetrachloroethylene demonstrated evidence of carcinogenicity in F344 rats, as
36    shown by increased incidences of mononuclear cell leukemia, and in B6C3Fi mice, as shown by
37    increased incidences of hepatocellular adenomas and carcinomas in males and carcinomas in
38    females.

      August, 2009                            70           DRAFT  - DO NOT CITE OR QUOTE

-------
 1          Additional support for this cancer descriptor comes from studies on the tumor initiating
 2    and promoting activity in mammalian cells (Colacci et al., 1996, 1992).
 3          No animal cancer bioassay data following inhalation exposure to 1,1,2,2-tetrachloro-
 4    ethane are available. However, U.S. EPA's Guidelines for Carcinogen Risk Assessment (2005a)
 5    indicates that, for tumors occurring at a site other than the initial point  of contact, the cancer
 6    descriptor generally applies to  all routes of exposure that have not been adequately studied unless
 7    there is convincing information to indicate otherwise.  No additional information is available for
 8    1,1,2,2-tetrachloroethane.  Thus, 1,1,2,2-tetrachloroethane is "likely to be carcinogenic to
 9    humans" by any route of exposure.
10          The weight-of-evidence for the carcinogenicity of 1,1,2,2-tetrachloroethane could be
11    strengthened by additional cancer bioassays demonstrating tumor development.  Currently, the
12    NCI (1978) bioassay is the only study available demonstrating 1,1,2,2-tetrachloroethane
13    tumorgenicity. The NCI (1978) study was a 78-week study, compared to a 104-week bioassay,
14    and the limitations of the study included increased mortality in male and female mice,  the
15    variable doses given to the mice over the course of the 78-week exposure period, and the acute
16    toxic tubular nephrosis, characterized as the cause of death, in the high-dose male mice that died
17    prior to study termination (although hepatocellular carcinomas were observed in most  of these
18    mice).
19
20    4.7.2. Synthesis of Human, Animal, and Other Supporting Evidence
21          Only  one study in humans evaluated the possible carcinogenic effects of  1,1,2,2-tetra-
22    chloroethane. Norman et al. (1981) evaluated groups of clothing-treatment workers employed
23    during World War II in which  some workers used 1,1,2,2-tetrachloroethane and  some used water.
24    Inhalation exposure concentrations and durations were not reported and dermal exposures were
25    likely. In addition, coexposures to dry-cleaning chemicals occurred. No differences in standard
26    mortality  ratios were seen between the 1,1,2,2-tetrachloroethane and water groups for total
27    mortality, cardiovascular disease, cirrhosis of the liver, or cancer of the digestive and respiratory
28    systems.  The mortality ratio for lymphatic cancers in the 1,1,2,2-tetrachloroethane group was
29    increased relative to controls and the water group, although the number of deaths was small
30    (4 cases observed compared to 0.85 cases expected). No other information was located
31    regarding the carcinogenicity of 1,1,2,2-tetrachloroethane in humans.
32          The only comprehensive animal study that evaluated the carcinogenicity  of 1,1,2,2-tetra-
33    chloroethane was performed by the NCI (1978). Male and female Osborne-Mendel rats  were
34    exposed to TWA doses of 0, 62, or 108 mg/kg-day (males) or 0, 43, or 76 mg/kg-day (females)
35    5 days/week  for 78 weeks, followed by a 32-week observation period during which the rats were
36    not exposed.  No statistically significant increases in tumor incidences  were observed in  rats.
37    However, two hepatocellular carcinomas, which were characterized by NCI (1978) as  rare in
38    Osbourne-Mendel rats, and one neoplastic nodule were observed in the high-dose male rats.  A

      August, 2009                           71           DRAFT - DO NOT CITE OR QUOTE

-------
 1    hepatocellular carcinoma was also observed in a female rat in the control group. NCI (1978)
 2    characterized the carcinogenic results in male rats as "equivocal." Male and female B6C3Fi
 3    mice were exposed to TWA doses of 0, 142, or 284 mg/kg-day 5 days/week for 78 weeks,
 4    followed by a 12-week observation period during which the mice were not exposed.  Statistically
 5    significant, dose-related increases in the incidence of hepatocellular carcinoma were observed in
 6    males (3/36, 13/50, and 44/49 in the control, low-, and high-dose groups, respectively) and
 7    females (1/40, 30/48, and 43/47, respectively). In addition, a decrease in the time to tumor for
 8    the hepatocellular carcinomas was also evident in both sexes of mice.  Lymphomas were also
 9    seen in the male and female mice, but the incidences were not found to be statistically significant.
10    The only other available study observed pulmonary adenomas in female Strain A/St mice given
11    99 mg/kg injections i.p. 3 times/week for 8 weeks (Maronpot et al.,  1986).
12          In vitro studies of the genotoxicity of 1,1,2,2-tetrachloroethane have yielded mixed,
13    though mainly negative, results. Mutagenicity studies in S. typhimurium were predominantly
14    negative, with only 2 of 10 available studies reporting activity (NTP, 2004; Ono et al., 1996;
15    Roldan-Arjona et al., 1991; Milman et al., 1988; Warner et al., 1988; Mitoma et al., 1984;
16    Haworth et al., 1983; Nestmann et al.,  1980; Rosenkranz, 1977; Brem  et al., 1974).  Mixed
17    results were reported for gene conversion, reversion, and recombination in S. cerevisiae
18    (Nestmann and Lee, 1983; Callen et al., 1980), and aneuploidy, but not mitotic cross over, was
19    induced in A. nidulans (Crebelli et al.,  1988). Tests for DNA damage in E. coli were positive
20    (DeMarini and Brooks, 1992; Rosenkranz,  1977; Brem et al., 1974). 1,1,2,2-Tetrachloroethane
21    was not mutagenic in mouse L5178Y lymphoma cells (NTP, 2004) and was negative in tests for
22    DNA damage in other mammalian cells, including induction of DNA repair in primary rat or
23    mouse hepatocytes (Milman et al., 1988; Williams, 1983), induction of chromosomal  aberrations
24    in CHO cells (NTP, 2004; Galloway et al.,  1987),  and induction of cell transformation in
25    BALB/C-3T3 cells (Colacci et al., 1992; Milman et al., 1988; Tu et al., 1985; Arthur Little, Inc.,
26    1983). 1,1,2,2-Tetrachloroethane was positive for induction of SCEs in both BALB/C-3T3
27    (Colacci et al., 1992) and CHO cells (NTP, 2004; Galloway et al., 1987) and for induction of cell
28    transformation in BALB/C-3T3 cells at high (cytotoxic) doses (Colacci et al., 1990).
29          1,1,2,2-Tetrachloroethane also had mixed results for genotoxicity following in vivo
30    exposure. Tests for sex-linked recessive lethal mutations and mitotic recombination in
31    Drosophila were negative (NTP, 2004; Vogel and Nivard, 1993; Woodruff et al., 1985;
32    McGregor, 1980).  Both positive  (Miyagawa et al., 1995) and negative results (Mirsalis et al.,
33    1989) have been reported in mouse hepatocytes tested for UDS, and tests for S-phase DNA
34    induction in hepatocytes were negative in male mice and equivocal in female mice (Mirsalis et
35    al., 1989).  Rat bone marrow cells were negative for chromosomal aberrations in male rats, but
36    positive in female rats (McGregor, 1980).
37          1,1,2,2-Tetrachloroethane showed promoting activity, but limited initiating activity, in rat
38    liver preneoplastic (GGT-positive) foci assays (Milman et al.,  1988; Story et al., 1986).

      August, 2009                            72           DRAFT - DO NOT CITE OR QUOTE

-------
 1    1,1,2,2-Tetrachloroethane initiated, but did not promote, neoplastic transformation in mouse
 2    BALB/c-3t3 cells (Colacci et al., 1996, 1992).
 3
 4    4.7.3. Mode-of-Action Information
 5          The mode of action of the carcinogenic effects of 1,1,2,2-tetrachloroethane is unknown.
 6    Colacci et al. (1987) reported possible covalent binding of radiolabeled 1,1,2,2-tetrachloroethane
 7    to DNA, RNA, and protein in the liver, kidney, lung, and stomach of rats and mice exposed to a
 8    single intravenous dose and analyzed 22 hours postexposure.  However, the conclusion of
 9    covalent binding may be influenced by the presence of radiolabel in the DNA, RNA, and protein
10    that was the result of incorporated radiolabeled carbon into the biomolecules through normal
11    biochemical processes.
12          The mutagenicity data for 1,1,2,2-tetrachloroethane are inconclusive, with in vitro
13    genotoxicity tests generally reporting negative results except for assays of SCE and cell
14    transformation, and in vivo tests of genotoxicity showing a similar pattern.  Several studies have
15    reported increases in the number of hepatocytes in mitosis, but the possible role these effects
16    may have on the carcinogenicity of 1,1,2,2-tetrachloroethane has not been evaluated. The results
17    of rat liver preneoplastic foci  and mouse BALB/C-3T3  cell neoplastic transformation assays
18    suggest that 1,1,2,2-tetrachloroethane may have initiating and promoting activity (Colacci, 1996,
19    1992; Milman et al., 1988; Story et al., 1986), but tumor initiation and promotion studies have
20    not been conducted.
21          Tumor formation by 1,1,2,2-tetrachloroethane may involve metabolism to one or more
22    active compounds, with the predominant pathway leading to the production of dichloroacetic
23    acid (Casciola and Ivanetich,  1984; Halpert and Neal, 1981; Yllner, 1971).  1,1,2,2-Tetrachloro-
24    ethane is metabolized extensively following absorption, at least in part, by cytochrome P450
25    enzymes from the members of the CYP2A, CYP2B, CYP2E, and CYP3A subfamilies (see
26    Section 3.3). Mice are known to metabolize  1,1,2,2-tetrachloroethane to a greater extent than
27    rats, which may, in part, account for the fact that liver tumors occurred in mice at statistically
28    significant levels, but not in rats, following chronic oral exposure.
29          Dichloroacetic acid, which appears to be the main metabolite of 1,1,2,2-tetrachloroethane,
30    induces hepatocellular carcinomas in both sexes of F344 rats and B6C3Fi mice (DeAngelo et al.,
31    1999; DeAngelo et al., 1996; Pereira, 1996; Pereira and Phelps, 1996; Ferreira-Gonzalez et al., 1995;
32    Richmond et al., 1995; Daniel et  al., 1992; DeAngelo et al., 1991; U.S. EPA, 1991b; Bull et al.,
33    1990; Herren-Freund et al., 1987). Dichloroacetic acid is recognized as hepatocarcinogenic in
34    both sexes of two rodent species
35          In addition, 1,1,2,2-tetrachloroethane may be metabolized to form free radicals, which
36    may, in turn, covalently bind to macromolecules, including DNA. Formation of free radicals
37    during 1,1,2,2-tetrachloroethane metabolism  has been demonstrated in spin-trapping experiments
38    (Tomasi et al., 1984).  Both nuclear and microsomal forms of cytochrome P450 enzymes have

      August, 2009                           73           DRAFT - DO NOT CITE OR QUOTE

-------
 1    been implicated in this process, as increased metabolism and covalent binding of metabolites
 2    following pretreatment with phenobarbital (Casciola and Ivanetich, 1984; Halpert, 1982), xylene
 3    (Halpert, 1982), or ethanol (Sato et al., 1980) have been reported.  The presence of covalently
 4    bound label has been reported following inhalation (Dow Chemical Company, 1988), oral
 5    (Mitoma et al., 1985), and intravenous (Eriksson and Brittebo, 1991) administration of
 6    radiolabeled 1,1,2,2-tetrachloroethane.
 7          In summary, only limited data are  available regarding the possible mode(s) of action of
 8    1,1,2,2-tetrachloroethane carcinogenicity.  Metabolism to one or more active compounds may
 9    play a role in tumor development.  Results of genotoxicity studies  of 1,1,2,2-tetrachloroethane
10    are mixed and provide inconclusive evidence for establishing a mutagenic mode of action.
11          No other data are available to inform the mode of action of carcinogenicity for
12    1,1,2,2-tetrachloroethane.
13
14    4.8. SUSCEPTIBLE POPULATIONS AND LIFE STAGES
15    4.8.1.  Possible Childhood Susceptibility
16          Studies in humans and laboratory animals have not thoroughly examined the effect of
17    1,1,2,2-tetrachloroethane exposure on the  immature organism.  The Gulati rat study (Gulati et al.,
18    1991b) demonstrated that fetuses exposed in utero can be adversely affected.  At scheduled
19    sacrifice, average fetal weights were statistically significantly decreased in all dose groups
20    except the 34 mg/kg-day group. In the Gulati mouse study (Gulati et al., 1991a), complete litter
21    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
22    4,575 mg/kg-day dose groups, respectively.  The limited data evaluating the effect of
23    1,1,2,2-tetrachloroethane on the developing organism have not indicated effects on the offspring
24    at levels that did not also cause maternal effects.
25
26    4.8.2.  Possible Gender Differences
27          Studies directly evaluating sex-related differences in toxicity following exposure to
28    1,1,2,2-tetrachloroethane are not available. Some toxicity studies which evaluated both sexes in
29    the same study showed close concordance between sexes with often no more than one dose
30    distinguishing between response levels for a given effect. Men normally have a smaller volume
31    of body fat than women, even accounting  for average size differences, contributing to differential
32    disposition of organic solvents between sexes (Sato and Nakajima, 1987).  Rats have pronounced
33    sex-specific differences in CYPs, primarily involving the CYP2C family which is not found in
34    humans, but humans have not demonstrated sex-specific isoforms  of CYP450 (Mugford and
35    Kedderis, 1998).  Humans have differences in CYP 3A4 activity related to estrogen and
36    progesterone, but these differences are regulated by the hormones  at the level of gene expression
37    (Harris et al., 1995).  Other differences may occur at the Phase 2 level attributable to
38    conjugation. Overall, no consistent differences have been reported between women and men in

      August, 2009                            74           DRAFT - DO NOT CITE OR QUOTE

-------
 1    the handling of xenobiotics such as 1,1,2,2-tetrachloroethane by CYP isoforms (Shimada et al.,
 2    1994).  These distinctions make it difficult to predict from the animal data gender-relevant
 3    differences for human exposure to 1,1,2,2-tetrachloroethane.
 4
 5    4.8.3. Other Susceptible Populations
 6           As metabolism is believed to play an important role in the toxicity of 1,1,2,2-tetrachloro-
 7    ethane, particularly in the liver, individuals with elevated levels of cytochrome P450 enzymes
 8    may have an increased susceptibility to the compound. Halpert (1982) reported an increase in in
 9    vitro metabolite formation and in covalently bound metabolites following pretreatment with
10    xylene  or phenobarbital, both of which increased cytochrome P450 activity.  Sato et al. (1980)
11    similarly reported an increased metabolism of 1,1,2,2-tetrachloroethane in rats following ethanol
12    pretreatment.  Since 1,1,2,2-tetrachloroethane has been demonstrated to inhibit cytochrome P450
13    enzymes (Paolini et al., 1992; Halpert, 1982), presumably through a suicide inhibition
14    mechanism, it is also possible that people coexposed to chemicals that are inactivated by
15    cytochrome P450 enzymes will be more susceptible to those compounds.
16           In addition,  studies of human GST-zeta polymorphic variants show different enzymatic
17    activities toward and inhibition by dichloroacetic acid that could affect the metabolism of
18    1,1,2,2-tetrachloroethane (Lantum et al., 2002;  Blackburn et al., 2001, 2000; Tzeng et al., 2000).
19    Dichloroacetic acid may covalently bind to GST-zeta (Anderson et al., 1999), irreversibly
20    inhibiting one of two stereochemically different conjugates, thus inhibiting its own metabolism
21    and leading to an increase in unmetabolized dichloroacetic acid as the dose and duration of
22    exposure increases (U.S.  EPA, 2003).  GST zeta is a hepatic enzyme that also functions in the
23    pathway for tyrosine catabolism. Populations,  or single individuals, may be more sensitive to
24    1,1,2,2-tetrachloroethane toxicity depending on which GST-zeta variant they possess.
25
26
      August, 2009                           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—with Rationale and Justification
 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 levels of both male and female rats were also affected. For example, increases in serum
28    ALT and SDH were observed at >80 mg/kg-day in male  rats and >170 mg/kg-day in female rats.
29    In addition, increased cholesterol and ALP were observed in female rats at >80 and 170 mg/kg-
30    day, respectively.  Additional histopathology observed in the liver included a statistically
31    significantly increased incidence of minimal to moderate hepatocyte hypertrophy at > 170 mg/kg-
32    day in females and >200 mg/kg-day in males. Also, increased incidence of necrosis and
33    pigmentation were observed at >80 mg/kg-day and hepatocellular mitotic alterations and foci of
34    cellular alterations were observed at >80 and >170 mg/kg-day in male rats, respectively.  In
35    females, increased incidence of hepatocellular hypertrophy was observed at >80 mg/kg-day and
36    necrosis, pigmentation, and foci of cellular alterations were reported at >170 mg/kg-day. Bile
37    duct hyperplasia, increased bile acids, spleen pigmentation, and spleen atrophy were also
38    observed in both male and female rats at the two highest doses.

      August, 2009                           76           DRAFT - DO NOT CITE OR QUOTE

-------
 1          Evidence of liver effects was also observed in mice by NTP (2004).  A statistically
 2    significant increase in relative liver weights was observed in both male and female rats at
 3    >200 and 80 mg/kg-day, respectively.  Increases in serum ALT, ALP, bile acids, and
 4    5'-nucleotidase (males only) were observed in males and females at >370 and 160 mg/kg-day,
 5    respectively. The study authors also reported an increase in SDH at >200 and 80 mg/kg-day in
 6    male and female mice, respectively. Serum cholesterol levels were statistically significantly
 7    increased in female mice at >160 mg/kg-day. The incidence of hepatocellular necrosis was
 8    statistically significantly increased in male mice at >370 mg/kg-day and in female mice at
 9    >700 mg/kg-day. Hepatocellular hypertrophy was also reported in both sexes at >160-
10    200 mg/kg-day.  A statistically significant increase in incidence of liver pigmentation and bile
11    duct hyperplasia occurred at >300 mg/kg-day in females and >370 mg/kg-day in males.
12          In addition to effects on the liver, NTP (2004) also observed effects associated with
13    reproduction in adult rats and mice following subchronic exposure to 1,1,2,2-tetrachloroethane at
14    dose levels as low as 40 mg/kg-day. In male rats, sperm motility was decreased at >40 mg/kg-
15    day, and higher doses resulted in decreased epididymis weight and increased atrophy of the
16    preputial and prostate gland, seminal vesicle, and testicular germinal epithelium.  In female rats,
17    minimal to mild uterine atrophy was increased at > 170 mg/kg-day and clitoral gland atrophy and
18    ovarian interstitial cell cytoplasmic alterations were increased at 320 mg/kg-day.  Female F344
19    rats in the 170 mg/kg-day group also spent more time in diestrus compared to controls. Male
20    mice had increased incidences of preputial gland atrophy at > 100 mg/kg-day.  Less sensitive
21    effects included decreases in absolute testis weight (>700 mg/kg-day), absolute epididymis, and
22    cauda epididymis weights (1,360 mg/kg-day), and a decrease in epididymal spermatozoal
23    motility  (1,360 mg/kg-day).  The only noted reproductive toxicity parameter in female  mice
24    affected  was a significant increase in the length of the estrous cycle at a dose of 1,400 mg/kg-day.
25          A developmental toxicity study by Gulati et al. (1991a) demonstrated that the developing
26    fetus may be sensitive to 1,1,2,2-tetrachloroethane exposure. Gulati et al. (1991a) exposed
27    pregnant CD Sprague-Dawley rats to 0, 34, 98, 180, 278, or 330 mg/kg-day
28    1,1,2,2-tetrachloroethane from GDs 4 through 20. Small but statistically significant decreases
29    were observed in maternal body weight and average fetal weight at >98 mg/kg-day. No other
30    maternal or fetal effects were reported by the study authors.  In a second study, Gulati et al.
31    (1991b)  exposed pregnant Swiss CD-I mice to 0, 987, 2,120, 2,216, or 4,575 mg/kg-day
32    1,1,2,2-tetrachloroethane from GDs 4 through 17. All animals (9/9) in the high-dose group died
33    prior to the end of the study, precluding calculation of the average dose in this exposure group.
34    Maternal body weights were statistically significantly decreased compared to controls at
35    >2,120 mg/kg-day beginning  on study day 9. Gross hepatic effects such as pale or grey and/or
36    enlarged livers and a prominent lobulated pattern were also reported in dams from all groups
37    except at the low dose.  Complete litter resorption occurred in 1/11, 0/9, 2/8, 1/1, and 1/2 dams in
38    the 0, 987, 2,120, 2,216, and 4,575 mg/kg-day groups, respectively.  No other developmental

      August,  2009                            77           DRAFT - DO NOT CITE OR QUOTE

-------
 1    effects were reported.  Gulati et al. (1991a, b) suggested that the developing fetus may be a target
 2    of 1,1,2,2-tetrachloroethane-induced toxicity. However, these developmental studies were
 3    conducted at doses higher than the subchronic NTP (2004) study, which demonstrated liver
 4    effects at lower doses. Therefore, Gulati et al. (1991a, b) was not selected as the principal study
 5    and the observed reproductive effects were not selected as the critical effect following
 6    subchronic exposure to 1,1,2,2-tetrachloroethane. Nevertheless, potential points of departure
 7    (PODs) based on the observed developmental effects from Gulati et al. (1991a) were provided
 8    for comparison (see Section 5.1.2 and Appendix B).
 9          In consideration of the available studies reporting effects of subchronic oral exposure to
10    1,1,2,2-tetrachloroethane in animals, NTP (2004) was chosen as the principal study for the
11    derivation of the subchronic RfD. This study was conducted in both sexes of two species, used
12    five dose levels and a concurrent control group, measured a wide-range of endpoints and tissues,
13    and provide data that were transparently and completely reported.  NTP (2004) identified the
14    liver as the most sensitive target organ of 1,1,2,2-tetrachloroethane-induced toxicity.
15    Specifically, NTP (2004) identified effects on the liver, including increased  liver weight and
16    increased incidence of hepatocellular vacuolization, at low dose levels.  Other liver effects
17    observed in rats and mice at higher doses included  increased liver weight, increased ALT, ALP,
18    and SDH serum levels, increased bile acid levels, and an increased incidence of hepatocellular
19    vacuolization and necrosis.
20          Based on the available data from the NTP (2004) study, the liver appears to be the most
21    sensitive target organ for 1,1,2,2-tetrachloroethane-induced toxicity. Thus, the observed effects
22    in the  liver were considered in the selection of the critical effect for the derivation of the
23    subchronic RfD.  Specifically, liver effects including increased liver weight, increased ALT,
24    ALP, and SDH serum levels, increased bile acid levels, and an increased incidence of
25    hepatocellular vacuolization were taken into consideration and modeled for  the determination of
26    the critical effect and POD (Section 5.1.1.2 and Appendix B). EPA selected increased liver
27    weight as the critical effect because this effect may represent a sensitive endpoint that occurs
28    early in the process leading to hepatocellular necrosis associated with subchronic oral exposure
29    to 1,1,2,2-tetrachloroethane.  The increase in relative liver weight was selected as the basis for
30    the selection of the POD because this analysis takes into account the substantive, dose-dependent
31    decreases in body weight that were observed in both sexes of rats.  Rats were selected as the
32    representative species because they appeared to be  more sensitive than mice to the hepatotoxic
33    effects of 1,1,2,2-tetrachloroethane.  EPA recognizes that the POD for the increased incidence of
34    hepatocellular vacuolization is approximately an order of magnitude lower than the POD for
35    increased relative liver weight,  and would result in a lower RfD than that derived for increased
36    relative liver weight (See Sections 5.1.1.2 and 5.1.3 for more information).  However, the
37    biological significance of this effect following 1,1,2,2-tetrachloroethane exposure is unclear
38    based  on the following considerations.

      August,  2009                            78           DRAFT - DO NOT CITE OR QUOTE

-------
 1          Vacuoles are defined as cavities bound by a single membrane that serve several
 2    functions; usually providing storage areas for fat, glycogen, secretion precursors, liquid, or debris
 3    (Osol, 1972).  Vacuolization is defined as the process of accumulating vacuoles in a cell or the
 4    state of accumulated vacuoles (Grasso, 2002). This process can be classified as either a normal
 5    physiological response or may reflect an early toxicological process. As a normal physiological
 6    response, vacuolization is associated with the sequestration of materials and fluids taken up by
 7    cells, and also with secretion and digestion of cellular products  (Henics and Wheatley, 1999).  In
 8    addition, Robbins et al.  (1976) characterized vacuolization (i.e., intracellular autophagy) as a
 9    normal cellular functional, homeostatic, and adaptive response.
10          Vacuolization is not only a normal physiological response. Vacuolization has been
11    identified as one of four principal types of chemical-induced injury (the other three being cloudy
12    swelling,  hydropic change, and fatty  change) (Grasso, 2002). It is one of the most common
13    responses of the liver following a chemical exposure, typically  in the accumulation of fat in
14    parenchymal cells, most often in the periportal zone (Plaa and Hewitt, 1998). The ability to
15    detect subtle ultrastructural defects, such  as vacuolization, early in the course of toxicity often
16    permits identification of the initial site of the lesion and thus can provide clues to possible
17    biochemical mechanisms involved in the  pathogenesis of liver injury (Hayes, 2001).
18          The hepatocellular vacuolization reported by NTP (2004) was not observed consistently
19    across species (i.e., reported only in male and female rats); whereas the other observed liver
20    effects were reported in both sexes of both species. In addition, NTP (2004) did not characterize
21    the vacuole content following exposure to 1,1,2,2-tetrachloroethane. The study authors indicated
22    that the severity of the hepatocellular vacuolization was minimal to mild and was concentration
23    independent, and NTP (2004)  did not report the localization of  the vacuolization in the liver.
24    The observed vacuolization in the liver at low doses appeared to diminish as dose increased.
25    Specifically, hepatocellular vacuolization increased in a dose dependant manner from 20 to
26    80 mg/kg-day in male rats. At 80 mg/kg-day, 100% of male rats were affected, and at doses of
27    >80 mg/kg-day, the incidence of vacuolization began to decrease. Concurrent with this decrease
28    in incidence of vacuolization,  an increased incidence of hepatocyte hypertrophy, necrosis, and
29    pigmentation were observed.  In female rats, the incidence of vacuolization was 100% at 40 and
30    80 mg/kg-day followed by a diminished response at the two highest doses.  Necrosis and
31    pigmentation were observed in the females at the two high doses. Thus, the qualitative and
32    quantitative biological relationship between the observed hepatocellular toxicity (i.e., hepato-
33    cellular necrosis) and the increased incidence of hepatocellular  cytoplasmic vacuolization in
34    NTP (2004) is unknown.
35
36    5.1.1.2. Methods of Analysis—Including Models (PBPK, BMD, etc.)
37          Benchmark dose (BMD) modeling was conducted using EPA BMD software version 2 to
38    analyze the hepatotoxic effects associated with subchronic exposure to 1,1,2,2-tetrachloroethane

      August, 2009                           79          DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
(see Appendix B for 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 continuous endpoints, the Benchmark Dose Technical
Guidance Document (U.S. EPA, 2000b) states that a change in the mean response equal to one
standard deviation (1 SD) from the control mean can be used to define the benchmark response
(BMR). A BMR of 1 SD from the control mean was selected for the continuous hepatotoxicity
data. For the dichotomous data, i.e., the incidence of hepatocellular cytoplasmic vacuolization,
an excess risk of 10% was selected as the BMR. The effects modeled include liver weight
changes, serum ALT and SDH, bile acids, hepatocellular cytoplasmic vacuolization, and rat fetal
body weights (see Appendix B for details). Table 5-1 presents the model results for the modeled
toxicological effects.
            Table 5-1. Summary of BMD model results for rats exposed to l,l?2,2-tetra-
            chloroethane in the diet for 14 weeks
Endpoint
Model
BMR
BMD
(mg/kg-d)
BMDL
(mg/kg-d)
Males
Cytoplasmic vacuol.
Relative liver weight
Absolute live weight
ALT
SDH
Bile acids
Multistage
Polynomial
Polynomial
Polynomial
Polynomial
Power
10% extra risk
1 SD
ISO
ISO
1 SD
1 SD
1.7
13
30
47
46
81
1.1
11
23
29
32
66
Females
Cytoplasmic vacuol.
Relative liver weight
Absolute liver weight
ALT
SDH
Bile acids
Multistage
Polynomial
Polynomial
Power
-
-
10% extra risk
1 SD
1 SD
1 SD
-
-
15
24
42
86
-
-
9.1
16
30
76
-
-
Developmental
Rat fetal weight
Linear
5% extra risk
44
32
13
14
15
16
17
18
19
20
21
       Potential PODs were identified by BMD modeling of the NTP (2004) rat liver data
shown in Table 5-1.  All continuous dose-response models available in the EPA's Benchmark
Dose Software (BMDS, version 2) were fit to the liver weight data, while all available
dichotomous models in BMDS (version 2) were fit to the incidence data for hepatocellular
cytoplasmic vacuolization.  In addition, the two highest dose groups were dropped prior to BMD
modeling. Animals in the two highest dose groups exhibited significant decreases in body
weight, and it is unclear whether these decreases in body weight were due to exceeding the
maximum tolerated dose or to lower feed consumption as dose increased (as a result of reduced
      August, 2009
                                       80
DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
palatability). In addition, the relative liver weight responses at the two highest doses were not
monotonically increasing, and thus do little to inform the shape of the dose-response curve in the
region of interest (i.e., at low dose).
       Adequate model fits were obtained for relative liver weight in both sexes of rat.
Table 5-2 presents BMDs and the corresponding lower 95% confidence limits (BMDLs) for the
increase in relative liver weight in male and female rats.
            Table 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
Endpoint
Model
BMR
BMD
(mg/kg-d)
BMDL
(mg/kg-d)
Males
Relative liver weight
1° Polynomial
1 SD
13
11
Females
Relative liver weight
2° Polynomial
1 SD
24
16
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
       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. A BMD of 1.7 mg/kg-day and BMDL of 1.1 mg/kg-day were derived
from the multistage model for the increased incidence of hepatocellular cytoplasmic
vacuolization in male rats. For serum ALT levels in male rats, a BMD of 47 mg/kg-day and a
BMDL of 29 mg/kg-day was derived from the polynomial model. For serum SDH in male rats,
a BMD of 46 mg/kg-day and a BMDL of 32 mg/kg-day was derived from the polynomial model.
The serum ALP data were not amenable to BMDS modeling. For bile acid levels in male rats, a
BMD of 81 mg/kg-day and a BMDL of 66 mg/kg-day was derived from the power model.
BMDS modeling derived a BMD of 79 mg/kg-day and a BMDL of 60 mg/kg-day from a linear
model with a BMR of 5% for decreased rat fetal weight.  Modeling details can be found in
Appendix B.
       The BMDiso of 13 mg/kg-day and BMDLiso of 11 mg/kg-day based on the relative liver
weight effects data in the male rat was selected as the POD for the subchronic RfD. The
observed changes in liver weights, serum liver enzyme levels, and hepatocellular necrosis
combine to support hepatotoxicity as the major toxic effect following 1,1,2,2-tetrachloroethane
exposure.

5.1.1.3.  RfD Derivation—Including Application of Uncertainty Factors (UFs)
       To derive the subchronic RfD, the 11 mg/kg-day BMDLiso for relative liver weight
changes in male rats is divided by a total UF of 300.  The UF of 300 comprises component
factors of 10 for interspecies extrapolation,  10 for interhuman variability, and 3 for database
deficiencies.
      August, 2009
                                       81
DRAFT - DO NOT CITE OR QUOTE

-------
 1          A default 10-fold UF was selected to account for the interspecies variability in
 2    extrapolating from laboratory animals (rats) to humans. No relevant information is available on
 3    the toxicity of 1,1,2,2-tetrachloroethane in humans, and data on toxicokinetic and toxicodynamic
 4    differences between animals and humans in the disposition of ingested 1,1,2,2-tetrachloroethane
 5    are not available, other than poorly-reported anesthetic effects in humans and rodents.
 6          A default 10-fold UF was selected to account for variations in sensitivity within human
 7    populations because there is insufficient information on the degree to which humans of varying
 8    gender, age, health status, or genetic makeup might vary in the disposition of, or response to,
 9    ingested 1,1,2,2-tetrachloroethane. However, studies of human GST-zeta polymorphic variants
10    demonstrate different enzymatic activities toward and inhibition by dichloroacetic acid that could
11    affect the metabolism of 1,1,2,2-tetrachloroethane (Lantum et al., 2002; Blackburn et al., 2001,
12    2000; Tzeng et al., 2000). Populations, or single individuals, may be more sensitive to
13    1,1,2,2-tetrachloroethane toxicity depending on which GST-zeta variant they possess. Animal
14    toxicity studies did not show consistent sex-related differences.
15          A threefold UF was selected to account for deficiencies in the database. The NTP (2004)
16    14-week study provides comprehensive evaluations of systemic toxicity and neurotoxicity in two
17    species.  The NTP (2004) study provides information of effects on  sperm, estrous cycle, and
18    male and female reproductive tissues in rats and mice, but the database lacks a two-generation
19    reproductive toxicity study. Available developmental toxicity  studies provide information on
20    embryo or fetotoxicity in orally exposed rats and mice (Gulati et al., 1991a, b), but the studies
21    did not include skeletal and visceral examinations.
22          A UF for LOAEL-to-NOAEL extrapolation was not used because the current approach is
23    to address this factor as one of the considerations in selecting a BMR for BMD modeling.
24          The subchronic RfD for 1,1,2,2-tetrachloroethane is calculated as follows:
25
26                 Subchronic RfD      =     BMDLiSD-UF
27                                      =     11 mg/kg-day - 300
28                                      =     0.04 mg/kg-day (or 4  x 10"2 mg/kg-day)
29
30    5.1.2. Chronic Oral RfD
31    5.1.2.1.  Choice of Principal Study and Critical Effect - with Rationale and Justification
32          Information on the chronic oral toxicity of 1,1,2,2-tetrachloroethane is limited to a
33    78-week cancer bioassay in rats and mice that were exposed by gavage (NCI, 1978).
34    Interpretation of the rat study may be confounded by high incidences of endemic chronic murine
35    pneumonia, although it is unlikely that this contributed to effects observed in the liver.  Based on
36    an increased incidence of hepatic fatty changes, the NOAEL and LOAEL for liver effects were
37    62 and 108 mg/kg-day, respectively.  In the mouse study, reduced survival and lethal kidney
38    lesions were observed at a dose of 284 mg/kg-day, but high incidences of hepatocellular tumors
      August, 2009                           82           DRAFT - DO NOT CITE OR QUOTE

-------
 1    in all treated groups precluded evaluation of noncancer effects in the liver and identification of a
 2    NOAEL or LOAEL.
 3          The 14-week dietary study in rats and mice (NTP, 2004), used to derive the subchronic
 4    RfD, was also considered for the derivation of the chronic RfD. The subchronic NTP (2004)
 5    study appears to be a more sensitive assay than the chronic NCI (1978) bioassay. The NTP
 6    (2004) study also uses lower dose levels and a wider dose range than the NCI (1978) study, and
 7    thereby provides a better characterization of the dose-response curve in the low-dose region.
 8    Additionally, dietary exposure is a more relevant route of exposure for the general population
 9    exposed to 1,1,2,2-tetrachloroethane in the environment than is gavage exposure. For these
10    reasons, the NTP (2004) subchronic study was selected as the principal study.
11          EPA selected increased liver weight as the critical effect because this effect may
12    represent a sensitive endpoint that occurs early in the process leading to hepatocellular necrosis
13    associated with subchronic oral exposure to 1,1,2,2-tetrachloroethane. The increase in relative
14    liver weight was selected as  the basis for the selection of the POD because this analysis takes
15    into account the substantive, dose-dependent decreases in body weight that were observed in
16    both sexes of rats. Additional  liver effects observed included increased liver weight, increased
17    ALT, ALP, and SDH serum levels, increased serum bile acid levels, and increased incidence of
18    hepatocellular vacuolization and necrosis.
19
20    5.1.2.2. Methods of Analysis—Including Models (PBPK, BMD, etc.)
21          The subchronic BMDLiso of 11 mg/kg-day  based on the increased relative liver weight
22    male rat data was used as the POD for the chronic RfD.  The observed increases in liver weights,
23    serum liver enzyme levels, and incidence of hepatocellular necrosis combine to support
24    hepatotoxicity as the critical effect of toxicity  of 1,1,2,2-tetrachloroethane.
25
26    5.1.2.3. RfD Derivation—Including Application of VFs
27          To derive the chronic RfD, the subchronic BMDLiso of 11 mg/kg-day, based on
28    increased liver weight in male  rats, was divided by a UF of 1,000. The UF of 1,000 comprises
29    component factors of 10 for  interspecies extrapolation, 10 for interhuman variability, 3 for
30    subchronic to chronic duration extrapolation, and 3 for database deficiencies, as explained below.
31          A default 10-fold UF was selected to account for the interspecies variability in
32    extrapolating from laboratory animals (rats) to humans.  No relevant information is available on
33    the toxicity of 1,1,2,2-tetrachloroethane to humans,  and data on toxicokinetic and toxicodynamic
34    differences between animals and humans in the disposition of ingested 1,1,2,2-tetrachloroethane
35    are not available, other than  poorly-reported anesthetic effects in humans and rodents.
36          A default 10-fold UF was selected to account for variations in sensitivity within human
37    populations because there is insufficient information on the degree to which humans of varying
38    gender, age, health  status or  genetic makeup might vary in the disposition of, or response to,

      August, 2009                           83           DRAFT - DO NOT CITE OR QUOTE

-------
 1    ingested 1,1,2,2-tetrachloroethane. However, studies of human GST-zeta polymorphic variants
 2    demonstrate different enzymatic activities toward and inhibition by dichloroacetic acid that could
 3    affect the metabolism of 1,1,2,2-tetrachloroethane (Lantum et al., 2002; Blackburn et al., 2001,
 4    2000; Tzeng et al., 2000).  Populations, or single individuals, may be more sensitive to
 5    1,1,2,2-tetrachloroethane toxicity depending on which GST-zeta variant they possess. Animal
 6    toxicity studies which evaluated both sexes  in the same study did not show consistent sex-related
 7    differences. Developmental toxicity studies in animals are limited in scope, but have not
 8    indicated effects on the offspring at levels that did not also cause maternal effects.
 9          A threefold UF was selected to account for extrapolation from a subchronic exposure
10    duration study to a chronic RfD. The study selected as the principal study was a 14 week study
11    by NTP (2004), a study duration that is minimally past the standard subchronic (90 day) study
12    and falls well  short of a standard lifetime study. In addition, some data are available to inform
13    the nature and extent of effects that would be observed with a longer  duration of exposure to
14    1,1,2,2-tetrachloroethane.  Specifically, the  available chronic cancer bioassay data (NCI,  1978)
15    suggest that liver damage observed in F344 rats following subchronic exposure to 1,1,2,2-tetra-
16    chloroethane (NTP, 2004), e.g., increased liver weight and incidence  of necrosis,  and altered
17    serum enzyme and bile levels, may not progress to more severe effects following chronic
18    exposures.  The chronic cancer bioassay was conducted in Osborne-Mendel rats and did not
19    measure liver enzyme levels. However, NCI (1978) observed minimal alterations in liver
20    pathology, including inflammation, fatty metamorphosis, focal cellular change, and angiectasis
21    in rats, and organized thrombus and nodular hyperplasia in mice. NCI (1978) reported that the
22    study authors  performed complete histological analysis on the liver, but specific endpoints
23    assessed were not included.  The available database does not abrogate all concern associated
24    with using a subchronic study as the basis of the RfD. For these reasons, a threefold UF was
25    used  to account for the extrapolation from subchronic to chronic exposure duration for the
26    derivation of the chronic RfD.
27          A threefold UF was selected to account for deficiencies in the database. The NTP (2004)
28    14-week study provides comprehensive evaluations of systemic toxicity and neurotoxicity in
29    both rats and mice. However, the database is limited by the lack of a two-generation
30    reproductive toxicity study.  The NTP (2004) study provides information on effects on sperm,
31    estrous cycle,  and male and female reproductive tissues in rats and mice, but the database lacks a
32    two-generation reproductive toxicity study.  Available developmental toxicity studies provide
33    information on embryo or fetotoxicity in orally exposed rats and mice (Gulati et al.,  199la, b),
34    but the studies did not include skeletal and visceral examinations.
35          A UF for LOAEL-to-NOAEL extrapolation was not used because the current approach is
36    to address this factor as one of the considerations in selecting a BMR for BMD modeling.
      August, 2009                           84           DRAFT - DO NOT CITE OR QUOTE

-------
 1          The chronic RfD for 1,1,2,2-tetrachloroethane is calculated as follows:
 2
 3                 Chronic RfD =     BMDLiSD-UF
 4                              =     11 mg/kg-day-1,000
 5                              =0.01 mg/kg-day (or 1 x 10"2 mg/kg-day)
 6
 7    5.1.3.  RfD Comparison Information
 8          Figure 5-1 is an exposure-response array that presents NOAELs, LOAELs, and the dose
 9    range tested corresponding to selected health effects. The effects observed in the subchronic and
10    chronic studies were considered candidates for the derivation of the sample subchronic and
11    chronic RfDs.
12          In addition to the increase in relative liver weight and the increased incidence of
13    hepatocellular cytoplasmic vacuolization, changes in absolute liver weight and serum levels of
14    ALT and SDH, bile acid levels, and serum cholesterol levels were considered for comparison.
15    Mean rat fetal weights observed following subchronic or chronic exposure to 1,1,2,2-tetrachloro-
16    ethane were also considered for comparison. Table  5-3 provides a tabular summary of sample
17    PODs and resulting subchronic sample RfDs for these endpoints.  Additionally, Figure 5-2
18    provides a graphical representation of this information.  This figure should be interpreted with
19    caution since the PODs across  studies are not necessarily comparable, nor is the confidence the
20    same in the data sets from which the PODs were derived.  Figure  5-3 provides a graphical
21    representation of the derivation of sample chronic RfDs for sample PODs from the subchronic
22    data.
23
      August, 2009                            85          DRAFT - DO NOT CITE OR QUOTE

-------
OvJU
•inn
OKf)
^-^ £ JU
ns
•o
E*
o
o
Q mn

n
U 1






<











'
1 ^ i






» |
1





'



<

^ I
\



» 1

1 1




>

1 <
1





1
t


* LOAEL
• NOAEL


I
^^ 	 ^; 	 ,
lines represent
the range of
doses tested in
, a given study.

t
I

I I T 1 1 1 I 1 1
>- ^~ m -^ -~~ ^^ ,_!_ ' n' JP. CO — . ,_ "0 ro , __ IE „,
ca o ' CD 5 co ^r ™ -o CD , n " n=- ^ -2 ^T 'oro^1" co!=_ *- ^ ^ _ -«•-—-
3 ~ = CS ^ ^g f^^g <^_ Q^I <"2S ^^g Iro^ ^ TO^ ^ ^. ° °
"5^CO'-^_ S,CM «^,- "0^,^" Z^ "OC^S CD^CM "cDZ^^CD — CD ^ "Q. ~— OcT
OCOCOQ_ =3-,-.- Q* ^O n CD CD "O - — - ,— , CD CO i - o ' - — - CD CO^,^05 >2 •— CO [^~
o "cL.b! 1— "o-^0- •— ' h- coJ^CD O>COCD to "o ,-J cOcoCL O^COCD COC-)T~ cjCJ-^o,
ro o o Z co .g»!- ro^Z S^0"1 roroCN ^m1- 2^!z ro^roCM fD-^r^E-^
CL^=>— ^CD^, -^CT— 2 CD - 2TOZ °"S- CL°- 00^ CO CO OT ^
CDCJCJJO CO^ Cl — CO O NX) >— T^Q o ^O • 	 ._ CD O C^ CD -Q 2 °- CD


1
2
Figure 5-1. Exposure response array for subchronic and chronic oral exposure to 1,1^2,2-tetrachloroethane.
    August, 2009
                                           86
DRAFT - DO NOT CITE OR QUOTE

-------
            Table 5-3. Potential PODs with applied UFs and resulting subchronic RfDs

Effect
Hepatocellular
cytoplasmic
vacuolization (rat)
Relative liver weight
(rat)
Absolute liver
weight (rat)
ALT (rat)
SDH (rat)
Bile acids (rat)
Fetal body weight
(rat)

POD (mg/kg-d)
l.lb


llc

23C

29C
32C
66C
60d


Species
Rat


Rat

Rat

Rat
Rat
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


4 x 10"2

8 x 10"2

0.10
0.10
0.22
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).
     CPOD 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).
     August, 2009
87
DRAFT - DO NOT CITE OR QUOTE

-------
            100
              10
      ftf
     •o
      I
      o>
E     0.1
           0.01
1
2
3
4
                    Hepatocellular
                     cytoplasmic
                    vacuolization -
                    rat (NTP, 2004)
                      w
                     Relative liver
                      weight - rat
                    " (NTP= 2004) "
                                                 I
                                             Absolute liver
                                              weight - rat
                                              (NTP, 2004)
                                                   Serum ALT - rat
                                                     (NTP, 2004)
Serum SDH - rat
  CNTP, 2004)
                                                                        Bile acids - rat
                                                                         (NTP, 2004)
 Fetal body
weight - rat
(Gulati et al,
  1991a)
POD


UFA

UFH

UFD


RfD

Critical
effect
          0.001
Figure 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).
     August, 2009
                                                                                 DRAFT - DO NOT CITE OR QUOTE

-------
       100
         10
 CO        1
•o
 CD
        0.1
      0.01
                Hepatocellular
                 cytoplasmic
                vacuolization -
                rat (NTP, 2004)
                             Relative liver
                              weight - rat
                              (NTP, 2004)
                                               Absolute liver
                                               weight - rat
                                               (NTP, 2004)
Serum ALT - rat
  (NTP, 2004)
                                                                    Serum SDH - rat
                                                                      (NTP, 2004)
                                                                                  Bile acids — rat
                                                                                  (NTP, 2004)
                                                                                                                  POD

                                                                                                                  UFA


                                                                                                                  UFH

                                                                                                                  UFD
 Fetal body
weight - rat
(Gulati et al,
  1991a)
                                                                                                                  RfD

                                                                                                                  Critical
                                                                                                                  effect
1

2
     0.001
       Figure 5-3.  PODs for selected endpoints (with critical effect circled) from Table 5-3 with corresponding applied
       UFs and derived sample chronic inhalation RfVs.
August, 2009
                                                           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          Information on additional oral toxicity assessments for 1,1,2,2-tetrachloroethane can be
 4    found  online at TOXNET (2009).
 5
 6    5.2. INHALATION REFERENCE CONCENTRATION (RfC)
 7    5.2.1.  Choice of Principal Study and Critical Effect—with Rationale and Justification
 8          Information on the inhalation toxicity of 1,1,2,2-tetrachloroethane is limited. In Truffert
 9    et al. (1977), rats were exposed to a presumed concentration of 560 ppm (3,909 mg/m3) for a
10    TWA  duration of 5.1 hours/day, 5 days/week for 15 weeks. Findings included transient
11    histological alterations in the liver, including granular appearance and cytoplasmic vacuolation,
12    which were observed after 9 exposures and were no longer evident after 39 exposures.  Because
13    of the  uncertainty regarding the actual exposure concentration for the single dose, and a lack of
14    incidence and severity data, this report cannot be used to identify a NOAEL or LOAEL or for
15    possible derivation of an RfC.
16          Horiuchi et al. (1962) observed fatty degeneration of the liver and splenic congestion in a
17    single  monkey exposed to a TWA of 1,974 ppm (15,560 mg/m3) 1,1,2,2-tetrachloroethane for
18    2 hours/day, 6 days/week for 9 months. The monkey was weak after approximately seven
19    exposures and had diarrhea and anorexia between the 12th and 15th exposures. Beginning at the
20    15th exposure, the monkey was "almost completely unconscious falling upon his side" for 20-
21    60 minutes after each  exposure.  Also, hematological parameters demonstrated sporadic changes
22    in hematocrit and RBC and WBC counts, but the significance of these findings cannot be
23    determined. This study cannot be utilized to identify a NOAEL or LOAEL due to the use of a
24    single  test animal with no control group.
25          Mellon Institute of Industrial Research (1947) observed an increased incidence of lung
26    lesions and an increase in kidney weight  in rats following a 6-month  exposure to 200 ppm
27    1,1,2,2-tetrachloroethane, but these results were not evaluated because the control animals
28    experienced a high degree of pathological effects in the kidney, liver, and lung, and because of
29    the presence of an endemic lung infection in both controls and treated groups. MIIR (1947) also
30    observed increased serum phosphatase levels and blood urea nitrogen levels in a dog exposed to
31    200 ppm 1,1,2,2-tetrachloroethane,  compared to control values, along with cloudy swelling of
32    the liver and the convoluted tubules of the kidney, and light congestion of the lungs. However,
33    identification of a LOAEL or NOAEL is precluded by poor study reporting, high mortality and
34    lung infection in the rats, and the use of a single treated animal in the dog study.
35          Kulinskaya and Verlinskaya (1972) observed inconsistent changes in acetylcholine levels
36    in Chinchilla rabbits exposed to 10 mg/m3 (1.5 ppm) 1,1,2,2-tetrachloroethane for 3 hours/day,
37    6 days/week for 7-8.5 months. A NOAEL or LOAEL was not identified because the changes in
      August, 2009                            90          DRAFT - DO NOT CITE OR QUOTE

-------
 1    acetylcholine were not consistent across time and incompletely quantified, and the biological
 2    significance of the change is unclear.
 3           Shmuter (1977) observed increases in antibody levels in Chinchilla rabbits at 2 mg/m3
 4    1,1,2,2-tetrachloroethane and decreases in antibody levels at 100 mg/m3.  Exposure to
 5    100 mg/m3 1,1,2,2-tetrachloroethane also resulted in a decrease in the relative content of
 6    antibodies in the y-globulin fraction and an increase in the T and P fractions. This is a poorly
 7    reported study that provides inadequate data, including reporting limitations, toxicological
 8    uncertainty in the endpoints, and inconsistent patterns of response, which preclude the
 9    identification of a NOAEL or LOAEL.
10           Effects following the chronic inhalation toxicity of 1,1,2,2-tetrachloroethane included
11    hematological alterations and increased liver fat content in rats exposed to 1.9 ppm (13.3 mg/m3)
12    4 hours/day for 265 days (Schmidt et al., 1972).  Statistically significant changes included
13    increased leukocyte (89%) and Pi-globulin (12%) levels compared to controls after 110 days,
14    and an increased percentage of segmented nucleated neutrophils (36%), decreased percentage of
15    lymphocytes (17%), and increased liver total fat content (34%) after 265 days.  A statistically
16    significant decrease in y-globulin levels (32%) at 60 days postexposure and a decrease in adrenal
17    ascorbic acid content (a measure of pituitary ACTH activity) were observed at all three time
18    periods (64, 21,  and  13%, respectively). This study is insufficient for identification of a NOAEL
19    or LOAEL for systemic toxicity  because most of the observed effects occurred at a single dose or
20    time point, or there was a reversal of the effect at the next dose or time point.  A reproductive
21    assessment in the Schmidt et al.  (1972) study was sufficient for identification of a NOAEL for
22    the single dose tested, 1.9 ppm (13.3 mg/m3), for reproductive effects in male rats, including
23    percentage of mated females having offspring, littering interval, time to 50% littered, total
24    number of pups, pups per litter, average birth weight, postnatal survival on days 1, 2, 7, 14, 21,
25    and 84, sex ratio, and average body weight on postnatal day 84. However, macroscopic
26    malformations or significant group differences in the other indices were not observed at
27    13.3 mg/m3. The lack of information on the reproductive toxicity precludes utilizing the selected
28    NOAEL in the derivation of the  RfC.
29           In addition, effects of chronic exposure to 1,1,2,2-tetrachloroethane included alterations
30    in serum acetylcholinesterase activity in rabbits exposed to 1.5 ppm (10 mg/m3) 1,1,2,2-tetra-
31    chloroethane 3 hours/day, 6 days/week for 7-8.5 months (Kulinskaya and Verlinskaya, 1972)
32    and immunological alterations in rabbits exposed to 0.3-14.6 ppm (2-100 mg/m3) 3 hours/day,
33    6 days/week, for 8-10 months (Shmuter, 1977). These studies are inadequate for identification
34    of NOAELs or LOAELs for systemic toxicity due to inadequate study reporting.
35           The inhalation toxicity database lacks a well-conducted study that demonstrates a dose-
36    related toxicological effect following subchronic and/or chronic exposure to 1,1,2,2-tetrachloro-
37    ethane. Therefore, an inhalation RfC was not derived.
38

      August, 2009                           91           DRAFT - DO NOT CITE OR QUOTE

-------
 1    5.2.2. Methods of Analysis—Including Models (PBPK, BMD, etc.)
 2          A route-to-route extrapolation using the computational technique of Chiu and White
 3    (2006), as described in Section 3.5, was considered.  However, U.S. EPA (1994b) recommends
 4    not conducting a route-to-route extrapolation from oral data when a first-pass effect by the liver
 5    or respiratory tract is expected, or a potential for a portal-of-entry effect in the respiratory tract is
 6    indicated following analysis of short-term inhalation, dermal irritation, in vitro studies, or
 7    evaluation of the physical/chemical properties. In the case of 1,1,2,2-tetrachloroethane, a first-
 8    pass effect by the liver is expected. In addition, the presence of tissue-bound metabolites in the
 9    epithelial linings in the upper respiratory tract may demonstrate a first-pass effect by the
10    respiratory tract (Eriksson and Brittebo, 1991). Lehmann et al. (1936) observed irritation of the
11    mucous membranes of two humans following inhalation of 146 ppm (1,003 mg/m3) for
12    30 minutes or 336 ppm (2,308 mg/m3) for 10 minutes, indicating the potential for portal-of-entry
13    effects in the respiratory system.
14
15    5.2.3. Previous RfC Assessment
16          An inhalation assessment for  1,1,2,2-tetrachloroethane was not previously available on
17    IRIS.
18          Information on additional inhalation toxicity assessments for 1,1,2,2-tetrachloroethane
19    can be found online at TOXNET (2009).
20
21    5.3. UNCERTAINTIES IN THE INHALATION REFERENCE CONCENTRATION
22    (RfC) AND ORAL REFERENCE DOSE (RfD)
23          The following discussion identifies some uncertainties associated with the RfD for
24    1,1,2,2-tetrachloroethane. As presented earlier (Sections 5.1.2 and 5.1.3;  5.2.2 and 5.2.3), EPA
25    standard practices and RfC and RfD guidance (U.S. EPA, 1994b) were followed in applying an
26    UF approach to a POD, a BMDLiso for the subchronic and chronic RfDs. Factors accounting
27    for uncertainties associated with a number of steps in the analyses were adopted to account for
28    extrapolating from an animal bioassay to human exposure, a diverse human population of
29    varying susceptibilities, and to account for database deficiencies.  These extrapolations are
30    carried out with standard approaches given the lack of extensive experimental and human data on
31    1,1,2,2-tetrachloroethane to inform individual steps.
32          An adequate range of animal toxicology data is available for the hazard assessment of
33    1,1,2,2-tetrachloroethane, as described in Section 4.  Included in these studies are short-term and
34    long-term bioassays and a developmental toxicity bioassay in rats and mice, as well as numerous
35    supporting genotoxicity and metabolism studies.  Toxicity associated with oral exposure to
36    1,1,2,2-tetrachloroethane is observed in the liver, kidney, and developing  organism, including
37    decreased fetal body weight and increased number of litter resorptions.
      August, 2009                           92           DRAFT - DO NOT CITE OR QUOTE

-------
 1          Consideration of the available dose-response data to determine an estimate of oral
 2    exposure that is likely to be without an appreciable risk of adverse health effects over a lifetime
 3    led to the selection of the 14-week oral dietary study in rats (NTP, 2004) and increased liver
 4    weight in males as the principal study and critical effect, respectively, for deriving the
 5    subchronic and chronic RfDs for 1,1,2,2-tetrachloroethane. The NTP (2004) data demonstrate
 6    hepatocellular damage, including increased liver weight, increased serum liver enzyme levels,
 7    and increased incidence of hepatic necrosis. Increased liver weight was chosen as the critical
 8    effect because it may represent a sensitive indicator of 1,1,2,2,-tetrachloroethane-induced
 9    hepatoxicity and occurs at a dose lower than the observed overt liver necrosis. The increase in
10    relative liver weight was selected as the basis for the selection of the POD because this analysis
11    takes into account the substantive, dose-dependent decreases in body weight that were observed
12    in both sexes of rats.  The dose-response relationships between oral exposure to  1,1,2,2-tetra-
13    chloroethane and fetal body weight in rats and mice are also suitable for deriving an RfD, but are
14    associated with BMDLs that are less sensitive than the selected critical effect and corresponding
15    BMDL.
16          For comparison purposes, Figure 5-2 presents the PODs, applied UFs, and derived RfDs
17    for the additional endpoints that were modeled using EPA BMD software, version 1.4.1.  The
18    additional endpoints included increased absolute liver weight, changes in serum  ALT and SDH,
19    increased bile acids, and increased incidence of hepatocellular necrosis, all of which  support the
20    liver as the target of 1,1,2,2-tetrachloroethane-induced toxicity following oral exposure. A
21    decrease in rat fetal weight was also modeled. The change in serum ALP was modeled, but a
22    model with adequate fit was not available.
23          The selection of the BMD model for the quantitation of the RfD does not lead to
24    significant uncertainty in estimating the POD, since benchmark effect levels were within the
25    range of experimental data. However, the selected model, the polynomial model, does not
26    represent all  possible models one might fit, and other models could be selected to yield more
27    extreme results, both higher and lower than those included in this assessment.
28          Extrapolating from animals to humans embodies further issues and uncertainties.  An
29    effect and its magnitude associated with the concentration at the POD in rodents are extrapolated
30    to human response.  Pharmacokinetic models are useful in examining species differences in
31    pharmacokinetic processing, however, dosimetric adjustment using pharmacokinetic modeling
32    was not possible for the toxicity observed following oral and inhalation exposure to 1,1,2,2-tetra-
33    chloroethane. Information was unavailable to quantitatively assess toxicokinetic or
34    toxicodynamic differences between animals and humans, so the 10-fold UF was used to account
35    for uncertainty in extrapolating from laboratory animals to humans in the derivation of the RfD.
36          Heterogeneity among humans is another uncertainty associated with extrapolating from
37    animals to humans.  Uncertainty related to human variation needs to be considered; also,
38    uncertainties in extrapolating from a subpopulation, say of one sex or a narrow range of life

      August, 2009                           93           DRAFT - DO NOT CITE  OR QUOTE

-------
 1    stages typical of occupational epidemiologic studies, to a larger, more diverse population need to
 2    be addressed. In the absence of 1,1,2,2-tetrachloroethane-specific data on human variation, a
 3    factor of 10 was used to account for uncertainty associated with human variation in the
 4    derivation of the RfD.  Human variation may be larger or smaller; however, 1,1,2,2-tetrachloro-
 5    ethane-specific data to examine the potential magnitude of over- or under-estimation are
 6    unavailable.
 7          Extrapolating from subchronic PODs to derive chronic reference values is also an
 8    uncertainty encountered in this assessment.  A threefold UF was selected to account for
 9    extrapolation from a subchronic exposure duration study to a chronic RfD. Based on the
10    available data for 1,1,2,2-tetrachloroethane, the toxicity observed in the liver does not appear to
11    increase over time.  The use of data  from a subchronic study to derive a chronic RfD becomes a
12    concern when the damage, in this case hepatoxicity, has the potential to accumulate;  however, if
13    the progression of the effect is not apparent, a reduced UF may be considered (U.S. EPA, 1994b).
14    Specifically, liver damage observed in F344 rats following subchronic exposure to 1,1,2,2-tetra-
15    chloroethane (NTP, 2004), e.g., increased incidence of necrosis or altered serum enzyme and bile
16    levels, did not progress to more severe effects such as cirrhosis or major liver disease following
17    chronic exposures (NCI,  1978). NCI (1978) observed minimal alterations in liver pathology,
18    including inflammation, fatty metamorphosis, focal cellular change, and angiectasis in rats, and
19    organized thrombus and nodular hyperplasia in mice.  Therefore, the available database does not
20    abrogate all concern associated with using a subchronic study as the basis of the RfD, but
21    supports the utilization of a database UF of 3.
22          Data gaps have been identified that are associated with uncertainties in database
23    deficiencies specific to the developmental and reproductive toxicity of 1,1,2,2-tetrachloroethane
24    following oral exposure.  The developing fetus may be a target of toxicity, and the absence of a
25    study specifically evaluating the full range of developmental toxicity endpoints represents an
26    area of uncertainty or gap in the database. The database of inhalation studies is of particular
27    concern due to the paucity of studies, especially subchronic and chronic studies, a multi-
28    generational  reproductive study, and a developmental toxicity study.
29
30    5.4. CANCER ASSESSMENT
31          Under U.S. EPA's Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005a),
32    1,1,2,2-tetrachloroethane is "likely to be carcinogenic to humans" based on data from an oral
33    cancer bioassay in male and female  Osborne-Mendel rats and B6C3Fi mice (NCI, 1978)
34    demonstrating an increase in the incidence of hepatocellular  carcinomas in both sexes of mice.
35    In this study, the incidence of hepatocellular carcinomas was statistically significantly increased
36    in both sexes of B6C3Fi mice at 142 (13/50 males; 30/48 females) and 284 mg/kg-day
37    (44/49 males; 43/47 females), with incidences in the male and female controls of 3/36 and 1/40,
38    respectively. NCI (1978) also demonstrated a decrease in the time to tumor in both sexes of

      August, 2009                            94           DRAFT - DO NOT CITE OR QUOTE

-------
 1    mice.  Male rats exhibited an increased incidence in hepatocellular carcinomas, characterized as
 2    rare tumors, but the increased incidence was not statistically significantly different from controls.
 3    NCI (1978) has characterized the carcinogenic results in male rats as "equivocal." In addition,
 4    the predominant metabolic pathway for 1,1,2,2-tetrachloroethane appears to involve production
 5    of dichloroacetic acid (Casciola and Ivanetich, 1984; Halpert and Neal, 1981; Yllner, 1971).
 6    Dichloroacetic acid was identified as the major urinary metabolite in mice treated with
 7    1,1,2,2-tetrachloroethane by i.p. injection (Yllner et al., 1971) and in in vitro systems with rat
 8    liver microsomal and nuclear cytochrome P450 (Casciola and Ivanetich, 1984; Halpert, 1982;
 9    Halpert and Neal, 1981).
10          The epidemiological human data available are inadequate for evaluation for cancer risk
11    (IARC, 1999). There are a limited number of positive results from genotoxicity studies which
12    suggest that 1,1,2,2-tetrachloroethane treatment in animals can result in UDS  (Miyagawa  et al.,
13    1995), chromosomal aberrations (McGregor, 1980), SCE (NTP, 2004; Colacci et al., 1992), and
14    micronucleus formation (NTP, 2004).  The ability of 1,1,2,2,-tetrachloroethane to alkylate
15    enzymatically purified hepatic DNA was observed following a single oral dose of 150 mg/kg of
16    1,1,2,2-tetrachloroethane in B6C3Fi mice (Dow Chemical Company, 1988).  1,1,2,2-Tetra-
17    chloroethane may have tumor initiating and promoting activity in mammalian cells (Colacci et
18    al., 1996, 1992; Milman et al., 1988; Story  et al., 1986).
19
20    5.4.1.  Choice of Study/Data—with Rationale and Justification
21          The only carcinogenicity bioassay for 1,1,2,2-tetrachloroethane is a chronic gavage study
22    in Osborne-Mendel rats and B6C3Fi mice performed by NCI (1978).  This study was conducted
23    in both sexes in two species  with an adequate number of animals per dose group, with
24    examination of appropriate toxicological endpoints in both sexes of rats and mice. Selection of
25    doses was aided by range-finding toxicity tests.  The rat study did not identify statistically
26    significant increases in tumor incidences in males or females.  Three rare liver tumors in high-
27    dose male rats were noted.
28          The mouse study identified statistically significant, dose-related increases in the
29    incidences of hepatocellular carcinomas in  both sexes. Based on these increases in
30    hepatocellular carcinomas, NCI (1978) concluded that orally administered 1,1,2,2-tetrachloro-
31    ethane is a liver carcinogen in male and female B6C3Fi mice. NCI (1978) stated that there was
32    no evidence for carcinogenicity of 1,1,2,2-tetrachloroethane in Osborne-Mendel rats (NCI, 1978).
33    The tumor data in mice from the NCI study was used for dose-response analysis for oral
34    exposure.
35
36    5.4.2.  Dose-response Data
37          Data on the incidences of hepatocellular carcinomas in male and female mice from the
38    NCI (1978) study were used for cancer dose-response assessment. These data are shown  in

      August, 2009                           95           DRAFT - DO NOT CITE OR QUOTE

-------
 1    Table 5-4.  The control data were pooled from vehicle control groups. The cancer bioassay for
 2    1,1,2,2-tetrachloroethane demonstrated evidence of increased incidence of tumors in both sexes
 3    of one species.
 4
             Table 5-4. Incidences of hepatocellular carcinomas in B6C3Fi mice used for
             dose-response assessment of l,l?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.
      bPooled 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).
 5
 6    5.4.3. Dose Adjustments and Extrapolation Method(s)
 7          Conversion of the doses in the NCI (1978) mouse study to human equivalent doses
 8    (HEDs) to be used for dose-response modeling was accomplished in three steps. The mice were
 9    treated with 1,1,2,2-tetrachloroethane by gavage 5 days/week for 78 weeks and then observed
10    untreated for 12 weeks for a total study duration of 90 weeks. Because the reported TWA doses
11    were for a 5 day/week, 78 week exposure, they were duration-adjusted to account for the partial
12    week exposure (by multiplying by 5 days/7 days) and untreated observation period (by
13    multiplying by 78 weeks/90 weeks). These duration-adjusted animal doses were then converted
14    to HEDs by adjusting for differences in body weight and lifespan between humans and mice.  In
15    accordance with the U.S. EPA (2005a) Guidelines for Carcinogen Risk Assessment., a factor of
16    BW3/4 was used for cross-species scaling.  Because the study duration (90 weeks) was less than
17    the animal lifespan (104 weeks), the scaled dose was then multiplied by the cubed ratio of
18    experimental duration to animal lifespan to complete the extrapolation to a lifetime exposure in
19    humans. The equation and data used to calculate the HEDs are presented below, and the
20    calculated HEDs are presented in Table 5-5.
21
22                             HED = Dose* x (W/70 kg)1/4 x (Le/L)3
23          Where:
24          Dose = average daily animal dose (* TWA converted for 5/7 days, 78/90 weeks)
25          W = average  animal body weight (0.030 kg for male and female B6C3Fi mice [U.S. EPA,
26              1988]).
27          70 kg = reference human body weight (U.S. EPA, 1988)
28          Le = duration of experiment (90 weeks)
29          L = reference mouse lifespan (104 weeks) (U.S. EPA, 1988)
      August, 2009                            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
 2
 3          The mode of action of 1,1,2,2-tetrachloroethane carcinogen!city is unknown. It appears
 4    that metabolism to one or more active compounds is likely to play a role in the development of
 5    the observed liver tumors, but insufficient data preclude proposing a specific mode of action.
 6    Dichloroacetic acid, a metabolite of 1,1,2,2-tetrachloroethane, induces hepatocellular carcinomas
 7    in male and female B6C3Fi mice and F344 rats.  Trichloroethylene (NTP, 1990; NCI, 1976) and
 8    tetrachloroethylene (NTP, 1996; NCI, 1977), also metabolites of 1,1,2,2-tetrachloroethane, have
 9    also been shown to be hepatocarcinogens in rodents.
10          Results of genotoxicity and mutagenicity studies of 1,1,2,2-tetrachloroethane are mixed
11    and insufficient for informing whether 1,1,2,2-tetrachloroethane carcinogenicity is associated
12    with a mutagenic mode of action.  Given that the mechanistic and other information available on
13    cancer risk from exposure to 1,1,2,2-tetrachloroethane is sparse and that the existing data do not
14    inform the mode of action of carcinogenicity, a linear low-dose extrapolation was conducted as a
15    default option for the derivation of the oral slope factor.
16          Dose-response modeling was performed to obtain a POD for quantitative assessment of
17    cancer risk. The data sets for hepatocellular carcinoma in both sexes of mice were modeled for
18    determination of the POD. In accordance  with the U.S. EPA (2005a) cancer guidelines, the
19    BMDLio (lower bound on dose estimated to produce a 10% increase in tumor incidence over
20    background) was estimated by applying the multistage cancer model in the EPA BMDS
21    (version 1.4.1) for the dichotomous incidence data, and selecting the results of the model that
22    best characterizes the cancer incidences. The BMD modeling of the male mouse data did not
23    achieve adequate model fit for any of the dichotomous models; thus, a cancer slope factor was
24    not derived from the male data.  The 1° multistage model was selected for the derivation of the
25    cancer slope factor from the female data because this model provided adequate model fit and the
26    lowest Akaike's Information Criterion (AIC) when compared to the results of the 2° multistage
27    model. In addition, the 2° multistage model had insufficient degrees of freedom to test the
28    goodness-of-fit.  The BMDL of 0.63 mg/kg-day from the modeling of the tumor incidence data
29    in female mice is selected as the POD for use in calculation of an oral slope factor (Table 5-6).
30    Details of the BMD modeling are presented in Appendix C.
31
      August, 2009                            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
BMD
(mg/kg-d)11
0.79
BMDL
(mg/kg-d)a
0.63
       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 of
 7    0.16 (mg/kg-day)"1 was calculated by dividing the human equivalent BMDLio of 0.63 mg/kg-day
 8    into 0.1 (10%) (Appendix C).  The oral slope factor was derived by linear extrapolation to the
 9    origin from the POD of 0.63 mg/kg-day and represents an upper-bound estimate. The slope of
10    the linear extrapolation from the central estimate (i.e., BMD) is 0.1/0.79 mg/kg-day or
11    O.lStmg/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
      August, 2009                           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 | or t slope
factor an unknown
extent
Alternatives could t
or J, slope factor by an
unknown extent
Alternatives could J,
or t slope factor (e.g.,
3.5-fold| [scaling by
BW] or t twofold
(scaling by BW2/3])
J, slope factor if MLE
used rather than lower
bound on POD
Alternatives could t
or J, slope factor by an
unknown extent
Human risk could J, or
t, depending on
relative sensitivity
Human relevance of
mouse tumor data
could J, 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 t or J,
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.
    August, 2009
                                       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.
25          Choice  of species/gender. The oral slope factor for 1,1,2,2-tetrachloroethane was
26    quantified using the tumor incidence data for female mice. The hepatocelluar carcinoma data in
27    male mice demonstrated tumorigenicity, but the data in male mice did not achieve adequate
28    model fit for any of the dichotomous models when BMD modeled.  The male and female  rat
29    tumor incidence data were not suitable for deriving low-dose quantitative risk estimates, and NCI
30    described the rat strain as relatively insensitive to the carcinogenic effects of chlorinated organic
31    compounds.
32          Relevance to  humans. The  oral slope factor is derived from the incidence of
33    hepatocellular carcinomas in female mice. Using liver tumors in B6C3Fi mice as the model for
34    human carcinogenesis is a concern  because of the prevalence of and susceptibility to developing
35    liver tumors in this strain of mice.
36          In addition, the genotoxicity and mutagenicity studies provide limited evidence of a
37    mutagenic mode of action, with 1,1,2,2-tetrachloroethane displaying equivocal results of
      August, 2009                           100          DRAFT - DO NOT CITE OR QUOTE

-------
 1   mutagenic activity.  In addition, there are inadequate data to support any mode of action
 2   hypothesis.
 3          Human population variability.  The extent of inter-individual variability in animals for
 4   1,1,2,2-tetrachloroethane metabolism has not been characterized. A separate issue is that the
 5   human variability in response to 1,1,2,2-tetrachloroethane is also unknown. This lack of
 6   understanding about potential differences in metabolism and susceptibility across exposed
 7   animal and human populations thus represents a source of uncertainty.
 8
 9   5.4.6. Previous Cancer Assessment
10          In the previous IRIS assessment, posted to the IRIS database in 1987, 1,1,2,2-tetrachloro-
11   ethane was characterized as "Classification — C; possible human carcinogen" based on the
12   increased incidence of hepatocellular carcinomas in mice observed in the NCI (1978) bioassay
13   (U.S. EPA, 1987). An oral slope factor of 0.2  (mg/kg-day)"1 was derived using the increased
14   incidence of hepatocellular carcinomas in female mice (NCI, 1978) and a linear multistage
15   extrapolation method.
16          Information on additional cancer assessments for 1,1,2,2-tetrachloroethane can be found
17   online at TOXNET  (2009).
18
19
     August, 2009                           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
      August, 2009                           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

      August, 2009                           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) 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 male rat liver weight and rat fetal body weight data.
14          The BMDiso of 13 mg/kg-day and BMDLiso of 11 mg/kg-day based on the relative liver
15    weight effects seen in the male rat represents a reasonable POD for the derivation of the RfD.
16    To derive the subchronic RfD, the 11 mg/kg-day BMDLiso based on male rat liver weight was
17    divided by a total UF of 300, yielding a subchronic RfD of 0.04 mg/kg-day.  The UF of
18    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.
      August, 2009                           104           DRAFT - DO NOT CITE OR QUOTE

-------
      100
        10
01
    E      0.1
      0.01
    0.001
               Hepatocellular
                cytoplasm ic
               vacuolization -
              rat (NTP, 2004)
                                Relative liver
                                 weight - rat
                                 (NTP, 2004)
                                        Absolute liver
                                         weight - rat
                                         (NTP, 20(14)
                                                         Serum ALT-rat  Serum SDH - rat
                                                          (NTP, 2004)    ^TR 2004)
                                                                               Bile acids - rat
                                                                                (NTP, 2004)
                                                                                                              POD
    Fetal body
    weight - rat
   (Gulati et al,
      1991a)
RfD

Critical
effect
i
2
3
4
5
       Figure 6-1. PODs for selected endpoints (with critical effect circled) with corresponding applied UFs and
       derived sample subchronic inhalation RfVs.
August, 2009
                                                          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-high.  Confidence
19    in the principal study (NTP, 2004) is high.  Confidence in the database is medium. Reflecting
20    high confidence in the principal study and medium confidence in the database, confidence in the
21    subchronic RfD is medium-high.
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 0.01 mg/kg-day was calculated by dividing the subchronic BMDLiso
32    of 11 mg/kg-day for increased relative liver weight by a total UF of 1,000: 10 for interspecies
33    extrapolation, 10 for interhuman variability, 3  for subchronic to chronic duration extrapolation,
34    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

      August,  2009                           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.
    August, 2009                          107          DRAFT - DO NOT CITE OR QUOTE

-------
                 100
                  10
           >*       A
           CO       1
                  0.1
                0.01
              0.001
                         Hepatocellular
                          cytoplasmic
                         vacuolization -
                         rat (NTP, 2004)
                                                       If
                                                                                 /^w;;
                                      Relative liver
                                       weight - rat
                                       (NTP, 2004)
                                       Absolute liver
                                        weight - rat
                                        (NTP, 2004)
Serum ALT - rat
  (NTP, 2004)
                                                                             Serum SDH - rat
                                                                              (NTP, 2004)
                                                                              Bile acids - rat
                                                                               (NTP, 2004)
                                                                                                          POD

                                                                                                          UFA


                                                                                                          UFH

                                                                                                          UFD
 Fetal body
weight - rat
(Gulati et al.,
  1991a)
                                                                                                          RED

                                                                                                          Critical
                                                                                                          effect
1

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 inhalation RfVs.
     August, 2009
                                               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., 1991a), 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

      August,  2009                           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/m3) for 30 minutes or 336 ppm (2,308 mg/m3) 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 turn origeni city 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

      August, 2009                           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 EPA BMDS
29    (version 1.4.1) 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 0.63 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 B.
      August, 2009                           111          DRAFT - DO NOT CITE OR QUOTE

-------
1          In accordance with the U.S. EPA (2005a) guidelines, an oral slope factor of 0.16 (mg/kg-
2    day)"1 is calculated by dividing the human equivalent BMDLio of 0.63 mg/kg-day into 0.1 (10%)
3    (Appendix B).
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
    August, 2009                           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.

ATSDR (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, LAP; 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.
August, 2009                                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. J R 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 B6C3Fj 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. BrMed 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.
August, 2009                                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. (1991a) 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.

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 49:442^145.

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.

Horiuchi, K; Horiguchi, S; Hashimoto, K; et al. (1962) Studies on the industrial tetrachloroethane poisoning. Osaka
City Med J 8:29-38.
August, 2009                                115            DRAFT - DO NOT CITE OR QUOTE

-------
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 databank). (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
Carcinog 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. BrMed 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.

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.
August, 2009                               116            DRAFT - DO NOT CITE OR QUOTE

-------
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; Mitoma, C; Tyson, C; et al. (1984) Comparative pharmacokinetics/metabolism, carcinogenicity and
mutagenicity of chlorinated ethanes and ethylenes (meeting abstract).  Arbeteoch Halsa 29:19.

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 Ser No. 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 Carcinog Tech Rep Ser 27:1-86.: 1-86; 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) inF344/N andB6C3Fl 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/N rats.  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.
August, 2009                               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. MutatRes 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, JF. (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. FundamAppl 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 and Health; PB87163531.

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.

Rosenkranz, HS. (1977) Mutagenicity of halogenated alkanes and their derivatives.  Environ Health Perspect 21:79-
84.
August, 2009                                118            DRAFT - DO NOT CITE OR QUOTE

-------
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 3 8:261-263.

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.

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.
August, 2009                                119            DRAFT - DO NOT CITE OR QUOTE

-------
TSI Mason Laboratories. (1993c) 14 Day pilot gavage toxicity study of 1,1,2,2-tetrachloroethane inB6C3Fl 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 inB6C3Fl 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.htrn (accessed August 4,
2009).

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).
August, 2009                                120            DRAFT - DO NOT CITE OR QUOTE

-------
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-03/001B. 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= 158363 (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.
August, 2009                               121            DRAFT - DO NOT CITE OR QUOTE

-------
1       APPENDIX A. SUMMARY OF EXTERNAL PEER REVIEW AND PUBLIC
2                       COMMENTS AND DISPOSITION
3
4
   August, 2009                    A-1        DRAFT - DO NOT CITE OR QUOTE

-------
 1    APPENDIX B.  BENCHMARK DOSE MODELING RESULTS FOR THE DERIVATION
 2                                        OF THE RFD
 3
 4   Dichotomous data
 5   Hepatocellular cytoplasmic vacuolization
 6          All available dichotomous models in the EPA's BMDS (version 2) were fit to the
 7   incidence of hepatocytocellular cytoplasmic vacuolization in male and female rats administered
 8    1,1,2,2-tetrachloroethane in the diet for 14 weeks (Table B-l). BMDs and their BMDLs
 9   associated with an extra risk of 10% were estimated by each model. In addition, the two highest
10   dose groups were dropped prior to BMD modeling. Two reasons exist for dropping these two
11   highest doses, one biological and the other statistical. First, animals in the two highest dose
12   groups exhibited significant decreases in body weight, and it is unclear whether these decreases
13   in body weight were due to exceeding the maximum tolerated dose or to lower feed consumption
14   as dose increased (as a result of reduced palatability).  Second, the relative liver weight responses
15   at the two highest doses were not monotonically increasing, and thus do little to inform the shape
16   of the dose-response curve in the region of interest (i.e., at low dose).
17
     August, 2009                          B-l          DRAFT - DO NOT CITE OR QUOTE

-------
             Table B-l. BMD modeling results based on incidence of hepatocellular
             vacuolization in male rats exposed to l,l?2,2-tetrachloroethane in the diet for
             14 weeks
Fitted dichotomous
model3
Gamma
Logistic
Log-logistic
Multistage (1°)
Probit
Log-probit
Quantal-linear
Weibull
X2 Goodness-of-fit test
/7-valueb
0.95
0.29
0.88
0.99
0.28
0.92
0.95
0.95
AICC
22.87
25.51
23.09
20.89
25.71
22.98
22.86
22.86
BMD10d
(mg/kg-d)
2.5
6.8
6.2
1.7
6.45
5.5
2.3
2.3
BMDL10e
(mg/kg-d)
1.1
3.7
0.31
1.1
3.73
1.8
1.1
1.1
      aAll dichotomous dose-response models were fit using BMDS, version 2. The "best-fit" model is highlighted in
      boldface type.
      V-Value from the %2 goodness-of-fit test for the selected model. Values <0.1 suggest that the model exhibits a
      significant lack of fit, and a different model should be chosen.
      °Value useful for evaluating model fit. For those models exhibiting adequate fit, lower values of the AIC suggest
      better model fit.
      dBMD10 = BMD at 10% extra risk.
      eBMDL10 = 95% lower confidence limit on the BMD at 10% extra risk.
      Source: NTP (2004).
 1
 2          The models fit to the incidence of hepatocytocellular cytoplasmic vacuolization were
 3    assessed by the %2 goodness-of-fit test statistic (p > 0.1) and AIC. In comparing models that
 4    exhibited adequate fit, a better fit is indicated by a lower AIC value (U.S. EPA, 2000c). The
 5    multistage (1°) model best fit the incidence of hepatocellular cytoplasmic vacuolization in male
 6    rats (Table B-l, Figure  B-l). The BMD modeling results for hepatocytocellular cytoplasmic
 7    vacuolization in female rats provided several models that fit the data; however, these models do
 8    not inform the dose response between the 20 and 40 mg/kg-day dose group and, combined with a
 9    perfect model fit, may lead to the introduction of model uncertainty.  Therefore, the multistage
10    (3°) model was selected to represent hepatocytocellular cytoplasmic vacuolization in female rats
11    because this model provided adequate fit and the most sensitive BMDLio, and because of the
12    uncertainty provided  by the models that provided perfect model fits but were uninformative with
13    regard to the dose-response of the data.
14
      August, 2009
B-2
DRAFT - DO NOT CITE OR QUOTE

-------
            Table B-2. BMD modeling results based on incidence of hepatocelluar
            vacuolization in female rats exposed to l,l?2,2-tetrachloroethane in the diet
            for 14 weeks
Fitted dichotomous
model3
Gamma
Logistic
Log-logistic
Multistage (3°)
Probit
Log-probit
Quantal-linear
Weibull
X2 Goodness-of-fit test
/7-valueb
0.67
1.00
1.00
0.22
1.00
1.00
1.00
1.00
AICC
5.00
4
2.08
9.85
4
2
2.00
2.00
BMD10d
(mg/kg-d)
20.6
29.4
25.0
14.5
28.7
26.4
30.7
30.7
BMDL10e
(mg/kg-d)
17.0
19.4
19.5
9.1
19.4
19.6
19.2
19.2
     aAll dichotomous dose-response models were fit using BMDS, version 2. The "best-fit" models are highlighted in
     boldface type.
     V-Value from the %2 goodness-of-fit test for the selected model.  Values <0.1 suggest that the model exhibits a
     significant lack of fit, and a different model should be chosen.
     °Value useful for evaluating model fit. For those models exhibiting adequate fit, lower values of the AIC suggest
     better model fit.
     dBMD10 = BMD at 10% extra risk.
     eBMDL10 = 95% lower confidence limit on the BMD at 10% extra risk.

     Source: NTP (2004).
2
3
4
5
6
       T3
       £
       O
           0.8
           0.4


           0.2


            0
                               Multistage Model with 0.95 Confidence Level
                  Multistage
             EMDLIBMD
                         10
                     20
30
 40
dose
50
60
70
80
         14:1208/272008
Fit of the multistage model to the incidence of hepatocellular cytoplasmic
vacuoliztation in male rats administered 1,1,2,2-tetrachloroethane in the diet
for 14 weeks (NTP, 2004).
     August, 2009
                                  B-3
                    DRAFT - DO NOT CITE OR QUOTE

-------
 2   ===================================================================
 3           Multistage Model.  (Version:  2.8;   Date:  02/20/2007)
 4           Input Data File: M:\TETRACHLOROETHANE DOSE-RESPONSE
 5   MODELING\NONCANCER\MALE_RAT_HEPATOCYTE_VACUOLIZATION.(d)
 6           Gnuplot Plotting File:   M:\TETRACHLOROETHANE DOSE-RESPONSE
 7   MODELING\NONCANCER\MALE_RAT_HEPATOCYTE_VACUOLIZATION.plt
 8                                              Wed Aug 27 14:12:36 2008
 9    ==================================================================
10
11    HMDS MODEL RUN
19   	
13
14      The form of the probability  function is:
15
16      P[response] = background  +  (1-background)*[1-EXP(
17                    -betal*doseAl)]
18
19      The parameter betas are restricted to  be  positive
20
21
22      Dependent variable = Response
23      Independent variable = Dose
24
25    Total number of observations =  4
26    Total number of records with missing values = 0
27    Total number of parameters  in model = 2
28    Total number of specified parameters = 0
29    Degree of polynomial = 1
30
31
32    Maximum number of iterations =  250
33    Relative Function Convergence has been set  to:  le-008
34    Parameter Convergence has been  set  to:  le-008
35
36
37
38                     Default Initial  Parameter  Values
39                        Background  =             0
40                           Beta(l)  =  1.28571e+018
41
42
43              Asymptotic Correlation  Matrix  of  Parameter Estimates
44
45
46
47
48
49                   Beta(l)
50
51      Beta(l)            1
52
53
54
55
56
57                                                  95.0% Wald Confidence Interval
Do         Variable       Estimate       Std.  Err.    Lower Conf. Limit  Upper Conf.  Limit
3y       Background
60          Beta(l)
61
62   >
     August, 2009                        B-4         DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
       Model
     Full model
   Fitted model
0.9818
  Reduced model


           AIC:
      Analysis of Deviance Table


Log(likelihood)   # Param's  Deviance  Test d.f.
     -9.35947         4
     -9.44611         1      0.173273      3
     -25.8979


      20.8922
     1
33.0768
                                 P-value
<.0001
                                  Goodness  of  Fit

Dose
0.0000
20.0000
40.0000
80.0000

Est. Prob.
0.0000
0.7034
0.9120
0.9923

Expected
0.000
7.034
9.120
9.923

Observed
0
7
9
10

Size
10
10
10
10
Scaled
Residual
0.000
-0.024
-0.134
0.279
 ChiA2 = 0.10
 d.f.  = 3
P-value = 0.9922
   Benchmark Dose Computation
Specified effect =


Risk Type


Confidence level =


             BMD =


            BMDL =


            BMDU =
            0.1


      Extra risk


           0.95


        1.73382


        1.11682


        2.71595
Taken together, (1.11682, 2.71595)  is a 90
interval for the BMD
                             % two-sided confidence
August, 2009
                B-5
           DRAFT - DO NOT CITE OR QUOTE

-------
                               Multistage Model with 0.95 Confidence Level
 4
 5
 6
 1
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
           0.8
           0.6
       o
       K   0.4
       ro
           0.2
                  Multistage
                    BMDL
                        BMD
                        10
                           20
30
 40

dose
50
60
70
80
         16:2608/272008
      Fit of the multistage model to the incidence of hepatocellular cytoplasmic
      vacuoliztation in female rats administered 1,1^2,2-tetrachloroethane in the
      diet for 14 weeks (NTP, 2004).
        Multistage Model.  (Version:  2.8;  Date: 02/20/2007)
        Input Data File: M:\TETRACHLOROETHANE DOSE-RESPONSE
MODELING\NONCANCER\FEMALE_RAT_HEPATOCYTE_VACUOLIZATION.(d)
        Gnuplot  Plotting File:   M:\TETRACHLOROETHANE DOSE-RESPONSE
MODELING\NONCANCER\FEMALE_RAT_HEPATOCYTE_VACUOLIZATION.plt
                                           Wed Aug 27 16:26:45  2008
 HMDS MODEL RUN
   The form of the probability function is:


   P[response] = background + (1-background)*[1-EXP(
                 -betal*doseAl-beta2*doseA2-beta3*doseA3)


   The parameter betas  are  restricted to be positive
   Dependent variable  =  Response
   Independent variable  = Dose
 Total number of  observations  = 4
 Total number of  records  with  missing values
           = 0
     August, 2009
                                   B-6
             DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
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
 Total number of parameters  in model = 4
 Total number of specified parameters = 0
 Degree of polynomial  =  3
 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 =            1
                         Beta(l)  = 1.42857e+018
                         Beta(2)  =            0
                         Beta(3)  =            0
           Asymptotic  Correlation Matrix of Parameter Estimates
                Beta(3)
   Beta(3)
                                  Parameter Estimates
                                 Std. Err.
   Indicates that this value is not calculated.
                            95.0% Wald Confidence Interval
                          Lower Conf.  Limit   Upper Conf. Limit
       Model
     Full model
   Fitted model
0.04924
  Reduced model


           AIC:
      Analysis of Deviance  Table


Log(likelihood)  # Param's   Deviance  Test d.f.
            0          4
     -3.92436          1        7.84872      3
     -27.7259


      9.84872
1
55.4518
                             P-value
<.0001
     Dose
              Est.  Prob.
                Goodness   of   Fit


          Expected    Observed     Size
                          Scaled
                        Residual
0.0000
20.0000
40.0000
80.0000
0.0000
0.2402
0.8889
1.0000
0.000
2.402
8.889
10.000
0
0
10
10
10
10
10
10
0.000
-1.778
1.118
0.000
August, 2009
                 B-7
      DRAFT - DO NOT CITE OR QUOTE

-------
 1    ChiA2 = 4.41      d.f.  = 3        P-value = 0.2204
 2
 3
 4      Benchmark Dose Computation
 5
 6   Specified effect =            0.1
 7
 8   Risk Type        =      Extra risk
 9
10   Confidence level =           0.95
11
12                BMD =        14.5321
13
14               BMDL =        9.14516
15
16               BMDU =        18.0805
17
18   Taken together,  (9.14516, 18.0805)  is a 90     % two-sided confidence
19   interval for the BMD
20
     August, 2009                       B-8         DRAFT - DO NOT CITE OR QUOTE

-------
 1   Continuous data
 2          Available continuous models in the EPA's BMDS (version 2) were fit to the effects
 3   observed in male and female rats including increased liver weights and changes in serum enzyme
 4   (ALT, SDH) and bile acid levels.  In addition, the two highest dose groups were dropped prior to
 5   BMD modeling. Two reasons exist for dropping these two highest doses,  one biological and the
 6   other statistical. First, animals in the two highest dose groups exhibited significant decreases in
 7   body weight, and it is unclear whether these decreases in body weight were due to exceeding the
 8   maximum tolerated dose or to lower feed consumption as dose increased (as a result of reduced
 9   palatability).  Second, the relative liver weight responses at the two highest doses were not
10   monotonically increasing, and thus do little to inform the shape of the dose-response curve in the
11   region of interest (i.e., at low dose). The BMDs and their BMDLs are estimates of the doses
12   associated with a change of 1 SD from the control.  Among those models providing adequate fit
13   to the means (%2/?-value >0.1), the one with the lowest AIC was selected for deriving the POD.
14   If the null hypothesis for constant variance was rejected and the nonhomogeneous variance
15   model did not provide an adequate fit to the variances, the data set was considered not suitable
16   for BMD modeling.
17
18
     August, 2009                           B-9          DRAFT - DO NOT CITE OR QUOTE

-------
1
2
Absolute liver weight
              Table B-3.  Summary of BMD modeling results based on mean absolute
              liver weights in male rats administered l,l?2,2-tetrachloroethane in the diet
              for 14 weeks
Fitted
dose-
response
model3
Linear
2°
Polynomial
("best-fit")
Power
Hill

Variance
model
employed
Constant
Constant


Constant
Constant
Homogeneit
yof
variance
test p-valueb
0.426
0.426


0.426
0.426
/>-Value
for test of
adequacy
of variance
model0
0.426
0.426


0.426
0.426

Goodness-
of-fit test
p-valued
0.297
0.129


0.297
NA



AICe
55.8
57.7


55.8
57.4


BMD1SDf
(mg/kg-d)
30.3
26.6


30.3
30.9


BMDL1SDg
(mg/kg-d)
22.9
15.2


22.9
19.8
     aAll continuous dose-response models were fit using BMDS, version 2. Because the two highest doses were
     deemed to have exceeded the maximum tolerated dose (MTD), these two dose groups were dropped prior to fitting
     the dose-response model. The "best-fit" model(s) is highlighted in boldface type.
     V-Value from the homogeneity of variance test. Values <0.1 suggest variances are nonhomogeneous, and thus a
     nonconstant variance model should be fit to the data.
     c/>-Value from the test of the adequacy of the variance model. Values <0.1 suggest that the variance model fitted to
     the data is inadequate. The only variance model available in BMDS models variance as an exponential power
     function of the log of the mean (i.e., Var(i) = exp(log a x log (mean(i))p). If variances are constant, the results of
     the homogeneity of variance test and the test for the adequacy of the variance model are the same.
     d/>-Value from the goodness-of-fit test.  Values <0.1 suggest that the selected model exhibits significant lack of fit,
     and a different model should be chosen.
     eThis value is defined as an estimate of the expected, relative distance between the fitted model and the unknown
     true model and is used to assess model fit. In comparing models fit to the same data, those with lower AIC values
     are preferred.
     fBMD1SD = BMD at a BMR, where the BMR is defined as being 1 SD from the control mean.
     gBMDL1SD = 95% lower confidence limit on the BMD at the BMR.

     Source: NTP (2004).
     August, 2009
                                            B-10
DRAFT - DO NOT CITE OR QUOTE

-------
             Table B-4.  Summary of BMD modeling results based on mean absolute liver
             weights in female rats administered l,l?2,2-tetrachloroethane in the diet for
             14 weeks

Fitted dose-
response
model3
Linear
2°
Polynomial
("best-fit")
Power
Hill

Variance
model
employed
Nonconstant

Nonconstant

Nonconstant
Nonconstant

Homogeneity
of variance
test/7-valueb
0.095

0.095

0.095
0.095
p- Value for
test of
adequacy of
variance
model0
0.553

0.553

0.553
0.553

Goodness-of-
fittest/7-valued
0.232

0.322

0.305
NA

AICe
-25.0

-24.9

-24.8
-22.8

BMD1SDf
(mg/kg-d)
42.2

57.2

58.8
58.8

BMDL1SDg
(mg/kg-d)
29.9

34.9

34.6
failed
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
aAll continuous dose-response models were fit using BMDS, version 2. Because the two highest doses were
deemed to have exceeded the maximum tolerated dose (MTD), these two dose groups were dropped prior to fitting
the dose-response model. The "best-fit" model(s) is highlighted in boldface type.
V-Value from the homogeneity of variance test. Values <0.1 suggest variances are nonhomogeneous, and thus a
nonconstant variance model should be fit to the data.
c/>-Value from the test of the adequacy of the variance model. Values <0.1 suggest that the variance model fitted to
the data is inadequate. The only variance model available in BMDS models variance as an exponential power
function of the log of the mean (i.e., Var(i) = exp(log a x log (mean(i))p).  If variances are constant, the results of
the homogeneity  of variance test and the test for the adequacy of the variance model are the same.
d/>-Value from the goodness-of-fit test.  Values <0.1 suggest that the selected model exhibits significant lack of fit,
and a different model should be chosen.
eThis value is defined as an estimate of the expected, relative distance between the fitted model and the unknown
true model and is used to assess model fit.  In comparing models fit to the same  data, those with lower AIC values
are preferred.
fBMD1SD  = BMD at a BMR, where the BMR is defined as being 1 SD from the control mean.
gBMDL1SD = 95% lower confidence limit on the BMD at the BMR.

Source: NTP (2004).


       For the increase in absolute liver weight in males, the linear and power models all

provided adequate  fits to the means when the constant variance model was applied. The AICs

for the linear and power models were equivalent.  Thus, the linear and power models were

selected  for deriving a potential POD from this dataset.

       The increase in absolute liver weight in female rats was modeled using nonconstant

variance. The  linear, 2° polynomial, and power models provided adequate model fits and

indistinguishable AIC values.  The linear model was selected for deriving a potential POD from

this dataset because this model provided the most sensitive BMDLiso
      August, 2009
                                         B-ll
DRAFT - DO NOT CITE OR QUOTE

-------
                                    Linear Model with 0.95 Confidence Level
 4
 5
 6

 8
 9
10
11
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
       o
       o.
       in
       
-------
1
2
3
4
5
6
7
0
o
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
31
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
69
70
71
72
73
74
75
76

Asymptotic Correlation Matrix of Parameter

( *** xhe model parameter (s) -rho

Estimates


have been estimated at a boundary point, or have been specified by
the user,
and do not appear in the correlation
alpha beta 0 beta 1

alpha 1 -3.9e-010 -1.4e-010

beta_0 -3.9e-010 1 -0.76

beta 1 -1.46-010 -0.76 1



Parameter Estimates

95.
Variable Estimate Std. Err. Lower
alpha 1.27907 0.286008
beta 0 12.62 0.277027
beta_l 0.0372857 0.00604522



Table of Data and Estimated Values of Interest

Dose N Obs Mean Est Mean Obs Std Dev


0 10 12 . 7 12 .6 0.82
20 10 13 13.4 1.11
40 10 14.5 14.1 1.39
80 10 15.5 15.6 1.23



Model Descriptions for likelihoods calculated


Model Al : Yij = Mu(i) + e(ij)
Var{e(ij)} = Sigma*2

Model A2 : Yij = Mu(i) + e(ij)
Var{e(ij)} = Sigma (i)*2

Model A3: Yij = Mu(i) + e(ij)
Var{e(ij)} = Sigma*2
Model A3 uses any fixed variance parameters that
were specified by the user

Model R: Yi = Mu + e(i)
Var{e(i)} = Sigma*2


Likelihoods of Interest

Model Log (likelihood) # Param' s
Al -23.706964 5
A2 -22.315060 8
A3 -23.706964 5
fitted -24.922604 3
R -38.289882 2


Explanation of Tests

Test 1: Do responses and/or variances differ among
(A2 vs. R)
Test 2: Are Variances Homogeneous? (Al vs A2)

matrix )












0% Wald Confidence Interval
Conf. Limit Upper Conf. Limit
0.718501 1.83963
12.077 13.163
0.0254373 0.0491341





Est Std Dev Scaled Res.


1.13 0.224
1.13 -1.02
1.13 1.09
1.13 -0.288























AIC
57.413929
60.630119
57.413929
55. 845207
80. 579765




Dose levels?


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.)
August, 2009
B-13
DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6

 8
 9
10
11
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
   Test
   Test 1
                     Tests of Interest
-2*log(Likelihood Ratio)   Test df
   Test
   Test
   Test 4
            31.9496

            2 . 78381

            2 . 78381

            2.43128
 p-value


<.0001
0.4262
0.4262
0.2965
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.  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


             BMD =        30.3322
            BMDL =
                          22.8795
                                  Linear Model with 0.95 Confidence Level
              7.8
              7.6
              7.4
              7.2
              6.8
              6.6
              6.4
                                      Linear
                              BMD Lower Bound
                                    BMDL
                                                BMD
                          10
                                20
                                       30
                                              40

                                             dose
                                                    50
                                                           60
                                                                  70
                                                                        80
  11:28 09/03 2008
       Fit of the linear model to mean absolute liver weights in female rats

       administered 1,1^2,2-tetrachloroethane in the diet for 14 weeks (NTP, 2004).
      August, 2009
                                      B-14
                                         DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2    ====================================================================
 3            Polynomial Model.  (Version: 2.13;  Date: 04/08/2008)
 4            Input Data File: C:\USEPA\BMDS2\Temp\tmp9C.(d)
 5            Gnuplot Plotting File:  C:\USEPA\BMDS2\Temp\tmp9C.plt
 6                                                  Wed Sep 03 11:28:48 2008
 g

 9     HMDS  Model  Run
10    	
11
12       The form  of the  response function is:

14       Y[dose] = beta_0 +  beta_l*dose  +  beta_2*dose*2 +  ...

16
17       Dependent variable  =  mean_abs_lvr_wt
18       Independent variable  = DOSE
19       Signs of  the  polynomial  coefficients  are not  restricted
20       The variance  is  to  be modeled as  Var(i) =  exp(lalpha  + log(mean(i))  *  rho)
21
22       Total number  of  dose  groups = 4
23       Total number  of  records  with missing  values  = 0
24       Maximum number of iterations =  250
25       Relative  Function Convergence has been set to: le-008
26       Parameter Convergence has  been  set  to:  le-008
27
28
29
30                     Default Initial  Parameter Values
31                            lalpha =      -1.66258
32                                rho =             0
33                            beta_0 =         6.784
34                            beta 1 =     0.0119571
35
36
37              Asymptotic  Correlation  Matrix of Parameter Estimates
38
39                    lalpha          rho      beta  0        beta 1
40                                                  -
41        lalpha            1            -1         -0.1         0.12
42
43           rho          -1             1         0.1         -0.12
44
45        beta 0         -0.1          0.1            1         -0.86
46
47        beta 1         0.12         -0.12       -0.86             1
48
49
50
51                                    Parameter Estimates
 2
                                                        95.0%  Wald Confidence Interval
            Variable        Estimate        Std.  Err.      Lower Conf. Limit   Upper Conf. Limit
             lalpha         16.6489         9.49635           -1.96361            35.2614
                rho         -9.3669         4.81385           -18.8019          0.0680672
             beta_0         6.74956        0.115438             6.5233            6.97581
             beta_l       0.0127595      0.00197835         0.00888203           0.016637

60
61
62        Table of Data  and Estimated Values  of Interest
63
64     Dose        N   Obs Mean      Est  Mean   Obs  Std Dev  Est Std Dev  Scaled Res.
65    	    ---   	
66
67        0    10       6.84
68       20    10       7.03
69       40    10       7.14
70       80    10         7.8
71
72
73
74     Model Descriptions for  likelihoods  calculated

76
      August, 2009                          B-15         DRAFT - DO NOT CITE OR QUOTE
6.75
7
7.26
7.77
0.54
0 .38
0 .51
0.25
0.539
0 .453
0 .383
0.279
0.531
0. 176
-0 .99
0.337

-------
 1
 2
 3
 4
 5
 6

 8
 9
10
11
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
69
 Model Al:         Yij  = Mu(i)  + e(ij)
           Var{e(ij)}  = Sigma*2


 Model A2 :         Yij  = Mu(i)  + e(ij)
           Var{e(ij)}  = Sigma(i)*2


 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)}
            Model

             Al

             A2

             A3

         fitted

              R
            Likelihoods  of  Interest


            Log(likelihood)    #  Param's      AIC

               15.358711            5      -20.717421

               18.541301            8      -21.082602

               17.948052            6      -23.896104

               16.487734            4      -24.975469

                3.999367            2       -3.998733
                   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. 1
   Test


   Test 1
   Test 2
   Test 3
   Test 4
                     Tests of Interest
-2*log(Likelihood Ratio)   Test  df
            29 .0839

            6.36518

             1.1865

            2.92064
  p-value


 <.0001

0.09513

 0.5525

 0.2322
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


             BMD =        42.2238


            BMDL =        29.9031
      August, 2009
                                      B-16
                                        DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
Relative liver weight
       For the increase in relative liver weight in male rats, the 1° polynomial and power models
fit did not adequately fit the serum SDH data for male rats based on the results of the
homogeneity of variance test (Table B-5). A nonconstant variance model was also run but did
not provide an adequate model fit according to the homogeneity of variance test and the
goodness-of-fit test. However, both models achieved adequate fit according to the goodness-of-
fit/>-value.  Even though the variances were not constant, they were not appreciably variable to
discourage use of either the polynomial or the  power model to represent the data. The
1° polynomial power model will be used to represent the increase in relative liver weight in male
rats, although either the  1° polynomial or the power model  could be used in this capacity.
             Table B-5. Summary of BMD modeling results based on mean relative liver
             weights in male rats administered l,l?2,2-tetrachloroethane in the diet for
             14 weeks
Fitted dose-
response model3
1° Polynomial
("best-fit")
Power
Hill
Variance
model
Constant
Nonconstant
Constant
Nonconstant
Constant
Nonconstant
Homogeneity
of variance
test/7-valueb
0.070
0.070
0.070
0.070
0.070
0.070
/7-Value for test
of adequacy of
variance model0
-
0.078
-
0.078
-
0.077
Goodness-
of-fit test
/7-valued
0.151
0.088
0.151
0.088
NAg
NA
BMD1SDe
(mg/kg-d)
13.1
11.0
13.1
11.0
19.2
17.3
BMDL1SDf
(mg/kg-d)
10.8
7.8
10.8
7.8
12.2
Failed
12
13
14
15
16
17
aAll continuous dose-response models were fit using BMDS, version 2. Because the two highest doses were
deemed to have exceeded the maximum tolerated dose (MTD), these two dose groups were dropped prior to fitting
the dose-response model. The "best-fit" model(s) is highlighted in boldface type.
V-Value from the homogeneity of variance test.  Values <0.1 suggest variances are nonhomogeneous, and thus a
nonconstant variance model should be fit to the data.
c/>-Value from the test of the adequacy of the variance model. Values <0.1 suggest that the variance model fitted to
the data is inadequate. The only variance model available in BMDS models variance as an exponential power
function of the log of the mean (i.e., Var(i) = exp(log a x log (mean(i))p). If variances are constant, the results of
the homogeneity of variance test and the test for the adequacy of the variance model are the same.
d/>-Value from the goodness-of-fit test. Values <0.1 suggest that the selected model exhibits significant lack of fit,
and a different model should be chosen.
eBMD1SD = BMD at a BMR, where the BMR is defined as being 1 SD from the control mean.
fBMDL1SD = 95% lower confidence limit on the BMD at the BMR.
Insufficient degrees of freedom to test model fit.
Source: NTP (2004).

       For the increase in relative liver weight  in female rats, the 2° polynomial and power
models provided adequate fits, according to the goodness-of-fit/>-value, to the data when the
constant variance model was applied (Table B-6).  The polynomial model provided a lower
BMDLiso and was selected to represent the increase in relative liver in female rats.
      August, 2009
                                        B-17
DRAFT - DO NOT CITE OR QUOTE

-------
        Table B-6.  Summary of BMD modeling results based on mean relative liver
        weights in female rats administered l,l?2,2-tetrachloroethane in the diet for
        14 weeks
Fitted dose-
response model3
2° Polynomial
("best-fit")
Power
Hill
Variance
model
Constant
Constant
Constant
Homogeneity
of variance
test/7-valueb
0.11
0.11
0.11
/7-Value for test
of adequacy of
variance model0
-
-
-
Goodness-
of-fit test
/7-valued
0.22
0.15
NAg
BMD1SDe
(mg/kg-d)
23.6
25.3
26.0
BMDL1SDf
(mg/kg-d)
15.7
17.1
17.6
aAll continuous dose-response models were fit using BMDS, version 2. Because the two highest doses were
deemed to have exceeded the maximum tolerated dose (MTD), these two dose groups were dropped prior to fitting
the dose-response model. The "best-fit" model(s) is highlighted in boldface type.
V-Value from the homogeneity of variance test. Values <0.1 suggest variances are nonhomogeneous, and thus a
nonconstant variance model should be fit to the data.
c/>-Value from the test of the adequacy of the variance model. Values <0.1 suggest that the variance model fitted to
the data is inadequate. The only variance model available in BMDS models variance as an exponential power
function of the log of the mean (i.e., Var(i) = exp(log a x log (mean(i))p). If variances are constant, the results of
the homogeneity of variance test and the test for the adequacy of the variance model are the same.
d/>-Value from the goodness-of-fit test. Values <0.1 suggest that the selected model exhibits significant lack of fit,
and a different model should be chosen.
eBMDiSD = BMD at a BMR, where the BMR is defined as being 1 SD from the control mean.
fBMDL1SD = 95% lower confidence limit on the BMD at the BMR.
Insufficient degrees of freedom to test model fit.

Source: NTP (2004).
August, 2009
B-18
DRAFT - DO NOT CITE OR QUOTE

-------
                           Linear 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
31
32
33
  Q.
  W
  0
  CD
  0
     48
     46
     44
     40
     38
     36
     34
           Linear
               BMD
                           BMD
                  10
                               20
30
 40

dose
50
60
70
80
    11:5801/172008
      Fit of the 1° polynomial model to mean relative liver weights in male rats
      administered l,l?2,2-tetrachloroethane in the diet for 14 weeks (NTP, 2004).
        Polynomial Model.  (Version:  2.12;   Date:  02/20/2007)
        Input Data File:  G:\TETRACHLOROETHANE DOSE-RESPONSE
MODELING\NONCANCER\MALE_RAT_REL_LIVER_WEIGHT.(d)
        Gnuplot Plotting  File:   G:\TETRACHLOROETHANE DOSE-RESPONSE
MODELING\NONCANCER\MALE_RAT_REL_LIVER_WEIGHT.pit
                                           Thu Jan 17 11:58:37 2008
 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
   Signs of the polynomial  coefficients  are not restricted
   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
August, 2009
                                        B-19
             DRAFT - DO NOT CITE OR QUOTE

-------
1
2
O
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
49
50
51
52
53
54
55
56
57
58
59
60
61
62

Default Initial Parameter Values
alpha = 3.40667
rho = 0 Specified
beta 0 = 34.646
beta 1 = 0.139757

Asymptotic Correlation Matrix of Parameter Estimates

( *** The model parameter(s) -rho
have been estimated at a boundary point, or have been specified by
and do not appear in the correlation matrix )

alpha beta 0 beta 1

alpha 1 -2.6e-012 -9.1e-012

beta 0 -2.6e-012 1 -0.76

beta 1 -9.1e-012 -0.76 1



Parameter Estimates

95.0% Wald Confidence Inte










the user,














rval
Variable Estimate Std. Err. Lower Conf . Limit Upper Conf . Limit
alpha 3.37042 0.75365 1.8933 4
beta 0 34.646 0.449695 33.7646 3
beta 1 0.139757 0.00981315 0.120524 0.



Table of Data and Estimated Values of Interest

Dose N Obs Mean Est Mean Obs Std Dev Est Std Dev
Res .


-

0 10 34.8 34.6 1.33 1.84
20 10 36.7 37.4 1.39 1.84
40 10 41 40.2 2.69 1.84
80 10 45.6 45.8 1.64 1.84



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

. 84755
5.5274
158991





Scaled





0.248
-1.24
1.37
-0.373

















August, 2009                      B-20        DRAFT - DO NOT CITE OR QUOTE

-------
1
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
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
Model R: Yi = Mu + e(i)
Var{e(i)} = SigmaA2




Likelihoods of Interest

Model Log (likelihood)
Al -42.407525
A2 -38.879991
A3 -42.407525
fitted -44.300775
R -80.370389


Explanation of Tests


# Param's AIC
5 94.815049
8 93.759982
5 94.815049
3 94.601550
2 164.740779




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

Tests of Interest

Test -2*log (Likelihood Ratio) Test

Test 1 82.9808 6
Test 2 7.05507 3
Test 3 7.05507 3
Test 4 3.7865 2

The p-value for Test 1 is less than .05.
and Test 2 will be the same.)



df p-value

<.0001
0.07016
0.07016
0.1506

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.
non-homogeneous variance model

The p-value for Test 3 is less than .1.
different variance model

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

Confidence level = 0.95

BMD = 13.1362


BMDL = 10.7581




Consider running a


You may want to consider a


The model chosen seems








deviations from the control mean









August, 2009
B-21
DRAFT - DO NOT CITE OR QUOTE

-------
                              Polynomial Model with 0.95 Confidence Level
 4
 5
 6
 1
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
          44
          42
       0
       w

       I  40
       0
       a:

       I  38
          36
          34
                Polynomial
                                   BMD
                       10
                          20
30
 40

dose
50
60
70
80
         11:4201/232008
      Fit of the 2° polynomial model to mean relative liver weights in female rats
      administered 1,1^2,2-tetrachloroethane in the diet for 14 weeks (NTP, 2004).
        Polynomial Model.  (Version:  2.12;   Date:  02/20/2007)
        Input Data File:  G:\TETRACHLOROETHANE DOSE-RESPONSE
MODELING\NONCANCER\FEMALE_RAT_REL_LIVER_WEIGHT.(d)
        Gnuplot Plotting  File:   G:\TETRACHLOROETHANE DOSE-RESPONSE
MODELING\NONCANCER\FEMALE_RAT_REL_LIVER_WEIGHT.plt
                                           Wed Jan 23 11:42:39 2008
 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
   Signs of the polynomial coefficients are not restricted
   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
     August, 2009
                                   B-22
             DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
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
   Parameter Convergence has been set to: le-008
                   Default Initial Parameter Values
                           alpha =
                             rho =
                          beta_0 =
                          beta_l =
                          beta 2 =
                               1.93677
                                     0
                               35.2192
                             0.0343205
                           0.000966477
                                Specified
           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_l


    beta 2
         alpha


             1


     -2.1e-010


      1.4e-010


      8.2e-010
            beta_0


         -2.1e-010


                  1


             -0.74


              0.59
  beta_l


1.4e-010


   -0.74


       1


   -0.96
  beta_2


8.2e-010


    0.59


   -0.96


       1
 Dose
Res .
      Variable
        alpha
       beta 0
       beta_l
       beta 2
     Table of Data  and Estimated Values of Interest


            N     Obs  Mean     Est Mean   Obs Std  Dev  Est Std Dev   Scaled
    0
   20
   40
   80
10
10
10
10
35.1
36.7
37.8
44.2
35
36
38
44
.2
.3
.1
.2
1.
1.
1.
0.
77
14
61
85
1
1
1
1
.35
.35
.35
.35
-0
0
-0
0
.351
.935
.701
.117
 Model Descriptions  for likelihoods calculated
 Model Al:         Yij
           Var{e(ij)}


 Model A2:         Yij
           Var{e(ij)}
               Mu (i)  +  e (ij )
               SigmaA2


               Mu (i)  +  e (ij )
               Sigma(i)A2
August, 2009
                          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

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 Log (likelihood) #
Al -31.113274
A2 -28.050020
A3 -31.113274
fitted -31.878683
R -72.394938


Explanation of Tests


Param's AIC
5 72.226548
8 72.100041
5 72.226548
4 71.757366
2 148.789876




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 88.6898 6
Test 2 6.12651 3
Test 3 6.12651 3
Test 4 1.53082 1

The p-value for Test 1 is less than .05.
(A3 vs. fitted)
and Test 2 will be the same.)



df p-value

<.0001
0.1056
0.1056
0.216

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


Benchmark Dose Computation

Specified effect = 1

Risk Type = Estimated standard

Confidence level = 0.95


A homogeneous variance



The modeled variance appears


The model chosen seems







deviations from the control mean


August, 2009
B-24
DRAFT - DO NOT CITE OR QUOTE

-------
1
2               BMD =         23.569
3
4
5              BMDL =        15.6757
6
7
    August, 2009                      B-25        DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
Serum ALT
       The polynomial (2°) and power models, using constant variance, provided adequate fits
for the serum ALT data in both male and female rats (Tables B-7 and B-8). For male rats, the
polynomial model provided a lower BMDLiso and was selected to represent the changes in
serum ALT levels.  The polynomial model was selected to represent the data in female rats as the
model provided the lowest AIC.
        Table B-7.  Summary of BMD modeling results based on mean serum ALT
        levels in male rats administered l,l?2,2-tetrachloroethane in the diet for
        14 weeks

Fitted dose-
response
model3
2° Polynomial
("best-fit")
Power
Hill

Variance
model
employed
Constant
Constant
Constant

Homogeneity
of variance
test/7-valueb
0.43
0.43
0.43
p- Value for
test of
adequacy of
variance
model0
0.43
0.43
0.43

Goodness-
of-fit test
^-valued
0.97
0.94
"NA"h

AICe
202.2
202.2
204.2

BMD1SDf
(mg/kg-d)
46.6
46.6
46.1

BMDL1SDg
(mg/kg-d)
29.1
29.5
29.5
aAll continuous dose-response models were fit using BMDS, version 2.  Because the two highest doses were
deemed to have exceeded the maximum tolerated dose (MTD), these two dose groups were dropped prior to fitting
the dose-response model. The "best-fit" model(s) is highlighted in boldface type.
V-Value from the homogeneity of variance test. Values <0.1 suggest variances are nonhomogeneous, and thus a
nonconstant variance model should be fit to the data.
c/>-Value from the test of the adequacy of the variance model. Values <0.1 suggest that the variance model fitted to
the data is inadequate.  The only variance model available in BMDS models variance as an exponential power
function of the log of the mean (i.e., Var(i) = exp(log a x log (mean(i))p). If variances are constant, the results of
the homogeneity of variance test and the test for the adequacy of the variance model are the same.
d/>-Value from the goodness-of-fit test.  Values <0.1 suggest that the selected model exhibits significant lack of fit,
and a different model should be chosen.
eThis value is defined as an estimate of the expected, relative distance between the fitted model and the unknown
true model and is used to assess model fit.  In comparing models fit to the same data, those with lower AIC values
are preferred.
fBMD1SD  = BMD at a BMR, where the BMR is defined as being 1 SD from the control mean.
gBMDL1SD = 95% lower confidence limit on the BMD at the BMR.
Inadequate degrees of freedom remained for testing the adequacy of model fit. Therefore, the results from this
model should not be used for identifying a potential POD.

Source: NTP (2004).
August, 2009
                                               B-26
DRAFT - DO NOT CITE OR QUOTE

-------
        Table B-8.  Summary of BMD modeling results based on mean serum ALT
        levels in female rats administered l,l?2,2-tetrachloroethane in the diet for
        14 weeks

Fitted dose-
response
model3
2° Polynomial
("best-fit")
Power
Hill

Variance
model
employed
Constant

Constant
Constant

Homogeneity
of variance
test/7-valueb
0.14

0.14
0.14
p- Value for
test of
adequacy of
variance
model0
0.14

0.14
0.14

Goodness-
of-fit test
/7-valued
0.96

0.10
0.03

AICe
183.0

185.5
187.5

BMD1SDf
(mg/kg-d)
86.3

79.8
79.7

BMDL1SDg
(mg/kg-d)
76.1

69.6
45.9
"All continuous dose-response models were fit using BMDS, version 2. Because the two highest doses were
deemed to have exceeded the maximum tolerated dose (MTD), these two dose groups were dropped prior to fitting
the dose-response model. The "best-fit" model(s) is highlighted in boldface type.
V-Value from the homogeneity of variance test.  Values <0.1 suggest variances are nonhomogeneous, and thus a
nonconstant variance model should be fit to the data.
c/>-Value from the test of the adequacy of the variance model. Values <0.1 suggest that the variance model fitted to
the data is inadequate. The only variance model available in BMDS models variance as an exponential power
function of the log of the mean (i.e., Var(i) = exp(log a x log (mean(i))p). If variances are constant, the results of
the homogeneity of variance test and the test for the adequacy of the variance model are the same.
d/>-Value from the goodness-of-fit test.  Values <0.1 suggest that the selected model exhibits significant lack of fit,
and a different model should be chosen.
eThis value is defined as an estimate of  the expected, relative distance between the fitted model and the unknown
true model and is used to assess model fit.  In comparing models fit to the same data, those with lower AIC values
are preferred.
fBMD1SD = BMD at a BMR, where the  BMR is defined as being 1 SD from the control mean.
gBMDL1SD = 95% lower confidence limit on the BMD at the BMR.

Source: NTP (2004).
August, 2009
B-27
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
30
31
32
33
     75
     70
  |  65
  o
  Q.
  o>  60
  a:
     55

     50

     45
           Polynomial
                             BMD
                                                     BMP
                  10
                               20
30
 40

dose
50
60
70
80
    12:1008/132008
      Fit of a 2° polynomial model to mean serum ALT in male rats administered
      1,1,2,2-tetrachloroethane in the diet for 14 weeks (NTP, 2004)
        Polynomial Model.  (Version:  2.12;   Date:  02/20/2007)
        Input Data File: M:\TETRACHLOROETHANE  DOSE-RESPONSE
MODELING\NONCANCER\MALE_RAT_SERUM_ALT.(d)
        Gnuplot Plotting File:  M:\TETRACHLOROETHANE DOSE-RESPONSE
MODELING\NONCANCER\MALE_RAT_SERUM_ALT.pit
                                           Wed  Aug 13 12:10:36 2008
 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
   Signs of the polynomial  coefficients  are  not  restricted
   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
August, 2009
                                       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

Default Initial Parameter Values
alpha = 52.4718
rho = 0 Specified
beta 0 = 47.9727
beta 1 = -0.0143182
beta 2 = 0.00346591


Asymptotic Correlation Matrix of Parameter Estimates

( *** The model parameter(s) -rho
have been estimated at a boundary point, or have been specified b
and do not appear in the correlation matrix )

alpha beta 0 beta 1 beta 2

alpha 1 9.4e-014 -1.7e-014 -2.5e-014

beta 0 9.4e-014 1 -0.74 0.59

beta 1 -1.7e-014 -0.74 1 -0.96

beta 2 -2.5e-014 0.59 -0.96 1



Parameter Estimates

95.0% Wald Confidence Int












y the user,
















erval
Variable Estimate Std. Err. Lower Conf. Limit Upper Conf. Limit
alpha 47.2269 10.5603 26.5292
beta 0 47.9727 2.08238 43.8913
beta 1 -0 0143182 0 132169 -0 273364 0
67 . 9247
52.0541
. ?447?8
beta 2 0.00346591 0.0015323 0.000462665 0.00646915



Table of Data and Estimated Values of Interest

Dose N Obs Mean Est Mean Obs Std Dev Est Std Dev
Res .


-

0 10 48 48 6.32 6.87
20 10 49 49.1 6.32 6.87
40 10 53 52.9 6.32 6.87
80 10 69 69 9.49 6.87


Model Descriptions for likelihoods calculated


Model Al: Yij = Mu(i) + e(ij)
Va r { e ( i j ) } = Si gma A 2

Model A2: Yij = Mu(i) + e(ij)
Var{e(ij)} = Sigma (i)A2






Scaled





0.0125
-0.0335
0.0251
-0.00418











August, 2009
B-29
DRAFT - DO NOT CITE OR QUOTE

-------
1
2
O
4
5
6
7
Q
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
54
55
56
57
58
59
60
61
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 Log (likelihood) #
Al -97.098317
A2 -95.706752
A3 -97.098317
fitted -97.099279
R -115.483425

Explanation of Tests


Param's AIC
5 204.196634
8 207.413504
5 204.196634
4 202.198559
2 234.966851



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 39.5533 6
Test 2 2.78313 3
Test 3 2.78313 3
Test 4 0.00192499 1

The p-value for Test 1 is less than .05.
(A3 vs. fitted)
and Test 2 will be the same.)



df p-value

<.0001
0.4263
0.4263
0.965

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

Benchmark Dose Computation

Specified effect = 1

Risk Type = Estimated standard

Confidence level = 0.95

BMD = 46. 642

BMDL = 29. 1438



A homogeneous variance



The modeled variance appears


The model chosen seems






deviations from the control mean







August, 2009
B-30
DRAFT - DO NOT CITE OR QUOTE

-------
                              Polynomial Model with 0.95 Confidence Level
 4
 5
 6
 1
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
          55
          50
       CD
       w
       o
       Q.
       W
       
-------
1
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
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
T"\J
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
Parameter Convergence has been set to: le-008


Default Initial Parameter Values
alpha = 32.4532
rho = 0 Specified
beta 0 = 46.0273
beta 1 = -0.285682
beta 2 = 0.00403409


Asymptotic Correlation Matrix of Parameter Estimates

( *** The model parameter(s) -rho
have been estimated at a boundary point, or have been specified by
and do not appear in the correlation matrix )

alpha beta 0 beta 1 beta 2

alpha 1 9.2e-010 -1.9e-010 -2.8e-010

beta 0 9.2e-010 1 -0.74 0.59

beta 1 -1.9e-010 -0.74 1 -0.96

beta 2 -2.8e-010 0.59 -0.96 1



Parameter Estimates

95.0% Wald Confidence Inter
Variable Estimate Std. Err. Lower Conf. Limit Upper Conf
alpha 29.2102 6.53159 16.4085 42
beta 0 46.0273 1.63769 42.8175 49














the user,
















val
. Limit
.0118
.2371
beta 1 -0.285682 0.103944 -0.489409 -0.0819545
beta 2 0.00403409 0.00120508 0.00167218 0.006396



Table of Data and Estimated Values of Interest

Dose N Obs Mean Est Mean Obs Std Dev Est Std Dev
Res .

_

0 10 46 46 6.32 5.4
20 10 42 41.9 3.16 5.4
40 10 41 41.1 6.32 5.4
80 10 49 49 6.32 5.4


Model Descriptions for likelihoods calculated


Model Al: Yij = Mu(i) + e(ij)
Va r { e ( i j ) } = Si gma A 2

Model A2 : Yij = Mu(i) + e(ij)
Var{e(ij)} = Sigma (i)A2





Scaled




-0.016
0.0426
-0.0319
0.00532










August, 2009
B-32
DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
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 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 Log (likelihood) #
Al -87.488771
A2 -84.710086
A3 -87.488771
fitted -87.490327
R -93.504675


Explanation of Tests


Param's AIC
5 184.977541
8 185.420172
5 184.977541
4 182.980654
2 191.009351




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 17.5892 6
Test 2 5.55737 3
Test 3 5.55737 3
Test 4 0.00311236 1

The p-value for Test 1 is less than .05.
(A3 vs. fitted)
and Test 2 will be the same.)



df p-value

0.007345
0.1352
0.1352
0.9555

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



Benchmark Dose Computation

Specified effect = 1

Risk Type = Estimated standard



A homogeneous variance



The modeled variance appears


The model chosen seems








deviations from the control mean

August, 2009
B-33
DRAFT - DO NOT CITE OR QUOTE

-------
1   Confidence  level =          0.95
2
3                BMD =        86.3349
4
5
6               BMDL =        76.1405
7
    August, 2009                       B-34        DRAFT - DO NOT CITE OR QUOTE

-------
1    Serum SDH
2           The constant variance models, as well as the nonconstant variance models, for the
3    polynomial (1°) and power models did not adequately fit the serum SDH data for male rats based
4    on the results of the homogeneity of variance test and the test of the adequacy of the variance
5    model (Table B-9).  Even though the variances were not constant, they were not appreciably
6    variable to discourage use of either model to represent the data.  In addition, both models
7    achieved adequate fit according to the goodness-of-fit/?-value.  Therefore, both the polynomial
8    (1°) and power model were selected to represent the serum SDH data in male rats.
9
            Table 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
Fitted dose-
response
model3
1° Polynomial
("best-fit")
Power
Hill
Variance
model
employed
Constant
Constant
Constant
Homogeneity
of variance
test/7-valueb
0.04
0.04
0.04
^j-Value for test
of adequacy of
variance model0
0.03
0.03
0.03
Goodness-
of-fit test
/7-valued
0.26
0.26
0.10
AICe
158.9
158.9
160.9
BMD1SDf
(mg/kg-d)
45.6
45.6
45.4
BMDL1SDg
(mg/kg-d)
31.6
31.6
13.9
     aAll continuous dose-response models were fit using BMDS, version 2. Because the two highest doses were
     deemed to have exceeded the maximum tolerated dose (MTD), these two dose groups were dropped prior to fitting
     the dose-response model. The "best-fit" model(s) is highlighted in boldface type.
     V-Value from the homogeneity of variance test. Values <0.1 suggest variances are nonhomogeneous, and thus a
     nonconstant variance model should be fit to the data.
     c/>-Value from the test of the adequacy of the variance model. Values <0.1 suggest that the variance model fitted to
     the data is inadequate. The only variance model available in BMDS models variance as an exponential power
     function of the log of the mean (i.e., Var(i) = exp(log a x log (mean(i))p). If variances are constant, the results of
     the homogeneity of variance test and the test for the adequacy of the variance  model are the same.
     d/>-Value from the goodness-of-fit test.  Values <0.1 suggest that the selected model exhibits significant lack of fit,
     and a different model should be chosen.
     This value is defined as an estimate of  the expected, relative distance between the fitted model and the unknown
     true model and is used to assess model fit. In comparing models fit to the same data, those with lower AIC values
     are preferred.
     fBMD1SD = BMD at a BMR, where the  BMR is defined as being 1 SD from the control mean.
     8BMDL1SD = 95% lower confidence limit on the BMD at the BMR.

     Source: NTP (2004).
     August, 2009
B-35
DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
        The BMD modeling results for the female SDH data did not achieve adequate fit
according to the goodness-of-fit/>-value and were, ultimately, attempting to fit a negative dose-
response curve (Table B-10). Therefore, a BMDL was not selected from the female SDH data.


        Table B-10. Summary of BMD modeling results on mean serum SDH levels
        in female rats administered l,l?2,2-tetrachloroethane in the diet for 14 weeks
Fitted dose-
response
model3
2° Polynomial
("best-fit")
Power
Hill
Variance
model
employed
Nonconstant
Nonconstant
Nonconstant
Homogeneity
of variance
test/7-valueb
0.04
0.04
0.04
p- Value for test
of adequacy of
variance model0
0.39
0.39
0.39
Goodness-
of-fit test
/7-valued
0.08
0.07
0.02
AICe
156.6
157.0
159.0
BMD1SDf
(mg/kg-d)
98.4
82.7
83.0
BMDL1SDg
(mg/kg-d)
85.4
79.3
"Failed"
     aAll continuous dose-response models were fit using BMDS, version 2.  Because the two highest doses were
     deemed to have exceeded the maximum tolerated dose (MTD), these two dose groups were dropped prior to fitting
     the dose-response model. The "best-fit" model(s) is highlighted in boldface type.
     V-Value from the homogeneity of variance test. Values <0.1 suggest variances are nonhomogeneous, and thus a
     nonconstant variance model should be fit to the data.
     c/>-Value from the test of the adequacy of the variance model.  Values <0.1 suggest that the variance model fitted to
     the data is inadequate.  The only variance model available in BMDS models variance as an exponential power
     function of the log of the mean (i.e., Var(i) = exp(log a x log (mean(i))p). If variances are constant, the results of
     the homogeneity of variance test and the test for the adequacy of the variance model are the same.
     d/>-Value from the goodness-of-fit test.  Values <0.1 suggest that the selected model exhibits significant lack of fit,
     and a different model should be chosen.
     This value is defined as an estimate of the expected, relative distance between the fitted model and the unknown
     true model and is used to assess model fit. In comparing models fit to the same data, those with lower AIC values
     are preferred.
     fBMD1SD  = BMD at a BMR, where the  BMR is defined as being 1 SD from the control mean.
     8BMDL1SD = 95% lower confidence limit on the BMD at the BMR.

     Source: NTP (2004).
     August, 2009
                                          B-36
DRAFT - DO NOT CITE OR QUOTE

-------
                           Linear 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
31
32
33
     34 r   Linear
     32
     30
  I 28
  w
  0

  126
  0
     24
     22
     20
                                    BMP
             BMQ
                  10
                               20
30
 40

dose
50
60
70
80
    15:5308/132008
      Fit of 1° polynomial model to mean serum SDH in male rats administered
      1,1,2,2-tetrachloroethane in the diet for 14 weeks (NTP, 2004)
        Polynomial Model.  (Version:  2.12;   Date:  02/20/2007)
        Input Data File: M:\TETRACHLOROETHANE  DOSE-RESPONSE
MODELING\NONCANCER\MALE_RAT_SERUM_SDH.(d)
        Gnuplot Plotting File:  M:\TETRACHLOROETHANE DOSE-RESPONSE
MODELING\NONCANCER\MALE_RAT_SERUM_SDH.pit
                                          Wed  Aug 13 15:53:27 2008
 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
   Signs of the polynomial coefficients  are  not  restricted
   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
August, 2009
                                       B-37
             DRAFT - DO NOT CITE OR QUOTE

-------
1
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
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


Default Initial Parameter Values
alpha = 17.4748
rho = 0 Specified
beta 0 = 23.6
beta 1 = 0.09


Asymptotic Correlation Matrix of Parameter Estimates

( *** The model parameter(s) -rho
have been estimated at a boundary point, or have been specified by
and do not appear in the correlation matrix )

alpha beta 0 beta 1

alpha 1 -1.7e-010 -1.3e-012

beta 0 -1.7e-010 1 -0.76

beta 1 -1.3e-012 -0.76 1



Parameter Estimates

95.0% Wald Confidence Inter
Variable Estimate Std. Err. Lower Conf. Limit Upper Conf
alpha 16.8273 3.7627 9.45256 24
beta 0 23.6 1.00481 21.6306 25












the user,














val
. Limit
.2021
.5694
beta 1 0.09 0.0219267 0.0470244 0.132976



Table of Data and Estimated Values of Interest

Dose N Obs Mean Est Mean Obs Std Dev Est Std Dev
Res .


-

0 10 23 23.6 3.16 4.1
20 10 27 25.4 3.16 4.1
40 10 26 27.2 6.32 4.1
80 10 31 30.8 3.16 4.1



Model Descriptions for likelihoods calculated


Model Al: Yij = Mu(i) + e(ij)
Va r { e ( i j ) } = Si gma A 2

Model A2: Yij = Mu(i) + e(ij)
Var{e(ij)} = Sigma (i)A2

Model A3: Yij = Mu(i) + e(ij)
Va r { e ( i j ) } = Si gma A 2
Model A3 uses any fixed variance parameters that





Scaled





-0.463
1.23
-0.925
0.154















August, 2009
B-38
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
49
50
51
52
53
54
55
56
57
58
59
were specified by the user

Model R: Yi = Mu + e(i)
Var{e(i)} = SigmaA2








Likelihoods of Interest
Model Log (likelihood)
Al -75.107987
A2 -70.847143
A3 -75.107987
fitted -76.460075
R -83.489967


Explanation of Tests

# Param's AIC
5 160.215973
8 157.694285
5 160.215973
3 158.920150
2 170.979934




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

Tests of Interest

Test -2*log (Likelihood Ratio) Test

Test 1 25.2856 6
Test 2 8.52169 3
Test 3 8.52169 3
Test 4 2.70418 2

The p-value for Test 1 is less than .05.
and Test 2 will be the same.)



df p-value

0.0003023
0.03638
0.03638
0.2587

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.
non-homogeneous variance model

The p-value for Test 3 is less than .1.
different variance model

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

Confidence level = 0.95

BMD = 45.579

BMDL = 31.6105


Consider running a


You may want to consider a


The model chosen seems







deviations from the control mean






60
61
     August, 2009
B-39
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
30
31
32
     34

     32


  $  3°

  I  28
  
-------
1
2
O
4
5
6
7
8
9
10
11
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
T- /
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63

Default Initial Parameter Values
alpha = 17.4748
rho = 0 Specified
control = 23
slope = 0.723981
power = 0.5


Asymptotic Correlation Matrix of Parameter Estimates

( *** The model parameter (s) -rho -power
have been estimated at a boundary point, or have been specified by
and do not appear in the correlation matrix )

alpha control slope

alpha 1 4e-010 -3.5e-010

control 4e-010 1 -0.76

slope -3.5e-010 -0.76 1



Parameter Estimates

95.0% Wald Confidence Inter
Variable Estimate Std. Err. Lower Conf . Limit Upper Conf
alpha 16.8273 3.7627 9.45256 24
control 23.6 1.00481 21.6306 25
slope 0.09 0.0219267 0.0470244 0.1
power 1 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
Res .


-

0 10 23 23.6 3.16 4.1
20 10 27 25.4 3.16 4.1
40 10 26 27.2 6.32 4.1
80 10 31 30.8 3.16 4.1



Model Descriptions for likelihoods calculated


Model Al: Yij = Mu(i) + e(ij)
Var{e (ij ) } = SigmaA2

Model A2: Yij = Mu(i) + e(ij)












the user,














val
. Limit
. 2021
.5694
32976










Scaled





-0.463
1.23
-0.925
0.154










August, 2009                      B-41        DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
           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 + e(i)
            Var{e (i) } = SigmaA2
                       Likelihoods of Interest
Log (likelihood)
-75.107987
-70.847143
-75.107987
-76.460075
-83.489967
# Param' s
5
8
5
3
2
AIC
160.215973
157.694285
160.215973
158.920150
170.979934
 Test 1:


 Test 2:
 Test 3:
 Test 4:
 (Note:
            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
              25.2856
              8.52169
              8.52169
              2.70418
6
3
3
2
    p-value


0.0003023
  0.03638
  0.03638
   0.2587
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.
non-homogeneous variance model


The p-value for Test 3 is less than .1.
different variance model
                                Consider running a
                                You may want to consider a
The p-value for Test 4 is greater than .1.  The model chosen seems
to adequately describe the data
               Benchmark Dose Computation


Specified effect =             1


Risk Type        =     Estimated standard deviations from the control mean
August, 2009
                         B-42
      DRAFT - DO NOT CITE OR QUOTE

-------
1   Confidence  level =          0.95
2
3                BMD = 45.579
4
5
6               BMDL = 31.6105
7
    August, 2009                       B-43        DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
Bile acids
        The polynomial (1°), power, and Hill models all adequately fit the bile acid data in male
rats (Table B-l 1).  The power model, however, has a lower AIC and is therefore selected to
represent the dose-response for the increase in bile acids in male rats.


        Table  B-ll. Summary of BMD modeling  results based on mean serum bile
        acids in male rats administered l,l?2,2-tetrachloroethane in the diet for
        14 weeks
Fitted dose-
response
model3
1° Polynomial
("best-fit")
Power
Hill
Variance
model
employed
Constant
Constant
Constant
Homogeneity
of variance
test/7-valueb
0.57
0.57
0.57
^j-Value for test
of adequacy of
variance model0
0.57
0.57
0.57
Goodness-
of-fit test
/7-valued
0.31
0.87
"NA"h
AICe
226.5
224.4
228.4
BMD1SDf
(mg/kg-d)
105.4
80.7
80.9
BMDL1SDg
(mg/kg-d)
54.2
65.5
44.5
 6
 7
 8
 9
10
11
12
aAll continuous dose-response models were fit using BMDS, version 2. Because the two highest doses were
deemed to have exceeded the maximum tolerated dose (MTD), these two dose groups were dropped prior to fitting
the dose-response model. The "best-fit" model(s) is highlighted in boldface type.
V-Value from the homogeneity of variance test. Values <0.1 suggest variances are nonhomogeneous, and thus a
nonconstant variance model should be fit to the data.
c/>-Value from the test of the adequacy of the variance model. Values <0.1 suggest that the variance model fitted to
the data is inadequate.  The only variance model available in BMDS models variance as an exponential power
function of the log of the mean (i.e., Var(i) = exp(log a x log (mean(i))p). If variances are constant, the results of
the homogeneity of variance test and the test for the adequacy of the variance model are the same.
d/>-Value from the goodness-of-fit test. Values <0.1 suggest that the selected model exhibits significant lack of fit,
and a different model should be chosen.
This value is defined as an estimate of the expected, relative distance between the fitted model and the unknown
true model and is used to assess model fit. In comparing models fit to the same data, those with lower AIC values
are preferred.
fBMD1SD  = BMD at a BMR, where the BMR is defined as being 1 SD from the control mean.
gBMDL1SD = 95% lower confidence limit on the BMD at the BMR.
Inadequate degrees of freedom remained for testing the adequacy of model fit. Therefore, the results from this
model should not be used for identifying a potential POD.

Source: NTP (2004).


       For the bile acid data in  female rats, the polynomial, power, and Hill models adequately

fit a decreasing function of dose for the dose-response curve up to 80 mg/kg-day (Table B-12).

Thus, the dose-response  curve of the models does not represent the increase in bile acids that is

observed over the course of the  dosing regimen for female rats. A model and resulting

         o is therefore not selected for the female data.
      August, 2009
                                          B-44
DRAFT - DO NOT CITE OR QUOTE

-------
        Table B-12. Summary of BMD modeling results based on mean serum bile
        acids in female rats administered l,l?2,2-tetrachloroethane in the diet for
        14 weeks
Fitted dose-
response
model3
1° Polynomial
("best-fit")
Power
Hill
Variance
model
employed
Constant
Constant
Constant
Homogeneity
of variance
test^-valueb
0.30
0.30
0.30
p- Value for test
of adequacy of
variance model0
0.30
0.30
0.30
Goodness-
of-fit test
/7-valued
0.40
0.21
0.23
AICe
277.7
279.4
279.3
BMD1SDf
(mg/kg-d)
384.1
124.0
"Failed"
BMDL1SDg
(mg/kg-d)
87.7
80.7
"Failed"
aAll continuous dose-response models were fit using BMDS, version 2. Because the two highest doses were
deemed to have exceeded the maximum tolerated dose (MTD), these two dose groups were dropped prior to fitting
the dose-response model. The "best-fit" model(s) is highlighted in boldface type.
V-Value from the homogeneity of variance test.  Values <0.1 suggest variances are nonhomogeneous, and thus a
nonconstant variance model should be fit to the data.
c/>-Value from the test of the adequacy of the variance model. Values <0.1 suggest that the variance model fitted to
the data is inadequate. The only variance model available in BMDS models variance as an exponential power
function of the log of the mean (i.e., Var(i) = exp(log a x log (mean(i))p). If variances are constant, the results of
the homogeneity of variance test and the test for the adequacy of the variance model are the same.
d/>-Value from the goodness-of-fit test.  Values <0.1 suggest that the selected model exhibits significant lack of fit,
and a different model should be chosen.
eThis value is defined as an estimate of  the expected, relative distance between the fitted model and the unknown
true model and is used to assess model fit.  In comparing models fit to the same data, those with lower AIC values
are preferred.
fBMDiSD = BMD at a BMR, where the  BMR is defined as being 1 SD from the control mean.
gBMDL1SD = 95% lower confidence limit on the BMD at the BMR.

Source: NTP (2004).
August, 2009
B-45
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
30
31
32
33
  o
  Q.
  W
  0
  ro
  CD
     45
     40
     35
     30
     25
     20
         Power
                                                                          BM
                  10
                               20
30
 40

dose
50
60
70
80
    16:4308/132008
      Fit of the power model to mean serum bile acids in male rats administered
      1,1,2,2-tetrachloroethane in the diet for 14 weeks (NTP, 2004)
        Power Model.  (Version: 2.14;   Date:  02/20/2007)
        Input Data File: M:\TETRACHLOROETHANE  DOSE-RESPONSE
MODELING\NONCANCER\MALE_RAT_BILE_ACIDS.(d)
        Gnuplot Plotting File:  M:\TETRACHLOROETHANE DOSE-RESPONSE
MODELING\NONCANCER\MALE_RAT_BILE_ACIDS.pit
                                          Wed  Aug  13 16:42:59 2008
 HMDS MODEL RUN



   The form of the response  function  is:


   Y[dose] = control + slope * doseApower
   Dependent variable = MEAN
   Independent variable = Dose
   rho is set to 0
   The power is restricted to be greater  than  or  equal  to 1
   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
August, 2009
                                       B-46
             DRAFT - DO NOT CITE OR QUOTE

-------
1
2
O
4
5
6
7
8
9
10
11
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
T- /
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63

Default Initial Parameter Values
alpha = 95.4952
rho = 0 Specified
control = 27.2
slope = 0.000207459
power = 2.42899


Asymptotic Correlation Matrix of Parameter Estimates

( *** The model parameter (s) -rho -power
have been estimated at a boundary point, or have been specified by
and do not appear in the correlation matrix )

alpha control slope

alpha 1 3.2e-008 -2.1e-008

control -9.4e-009 1 -0.5

slope 7e-009 -0.5 1



Parameter Estimates

95.0% Wald Confidence Inter
Variable Estimate Std. Err. Lower Conf . Limit Upper Conf
alpha 86.5274 19.3481 48.6058 12
control 27.9667 1.69831 24.638 31
slope 4.40389e-034 1.8855e-034 7.08372e-035 8.0994
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
Res .


-

0 10 29.2 28 9.17 9.3
20 10 27.5 28 8.54 9.3
40 10 27.2 28 8.54 9.3
80 10 35.9 35.9 12.3 9.3



Model Descriptions for likelihoods calculated


Model Al: Yij = Mu(i) + e(ij)
Var{e (ij ) } = SigmaA2

Model A2: Yij = Mu(i) + e(ij)












the user,














val
. Limit
4.449
. 2953
e-034










Scaled





0.419
-0.159
-0.261
9.82e-007










August, 2009                      B-47        DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
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
           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 + e(i)
            Var{e (i)  } = SigmaA2
                       Likelihoods of Interest
Log (likelihood)
-109.074320
-108.067736
-109.074320
-109.209223
-111.766222
# Param' s
5
8
5
3
2
AIC
228.148640
232.135472
228.148640
224.418447
227.532445
 Test 1:


 Test 2:
 Test 3:
 Test 4:
 (Note:
            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
              7.39697
              2.01317
              2.01317
             0.269807
6
3
3
2
 p-value


0.2857
0.5697
0.5697
0.8738
The p-value for Test 1 is greater than .05.  There may not be a
difference between responses and/or variances among the dose levels
Modelling the data with a dose/response curve may not be appropriate
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
               Benchmark Dose Computation


Specified effect =             1


Risk Type        =     Estimated standard deviations from the control mean
August, 2009
                         B-48
      DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2    Confidence level =          0.95
 3
 4                BMD = 80.7105
 5
 6
 7               BMDL = 65.477
 9
10
11
     August, 2009                       B-49        DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
BMD Analysis Details for Developmental Toxicity
Rat fetal weight means
       Fetuses of pregnant rat dams exposed to 1,1,2,2-tetrachloroethane on GDs 4-20 were
weighed. These fetuses exhibited a dose-responsive decrease in mean body weights (Gulati et
al., 1991a). These means were modeled using BMDS, version 2.1.  All available continuous
models were fit to these data and compared (Table B-13).


       Table B-13.  BMD modeling results  for decreases in mean weights of fetuses
       from rat dams exposed to l,l?2,2-tetrachloroethane in the diet on GDs 4-20
Fitted dose-
response
model3
Linear
2° Polynomial
("best-fit")
Power
Hill
Variance
model
employed
Nonconstant
Nonconstant
Nonconstant
Nonconstant
Homogeneity
of variance
test/7-valueb
O.OOOl
O.0001
O.OOOl
O.OOOl
/7-Value for
test of
adequacy of
variance
model0
0.07
0.07
0.07
0.07
Goodness-
of-fit test
/7-valued
0.1907
0.1113
0.1095
0.2877
AICe
-112.6
-110.7
-110.7
-113.0
BMD/
(mg/kg-d)
43.6
50.1
52.8
151.9
BMDL5g
(mg/kg-d)
31.6
31.7
31.7
43.5
 9
10
11
12
13
14
15
aAll continuous dose-response models were fit using BMDS, version 2.  Because the two highest doses were
deemed to have exceeded the maximum tolerated dose (MTD), these two dose groups were dropped prior to fitting
the dose-response model. The "best-fit" model(s) is highlighted in boldface type.
V-Value from the homogeneity of variance test. Values <0.1 suggest variances are nonhomogeneous, and thus a
nonconstant variance model should be fit to the data.
c/>-Value from the test of the adequacy of the variance model. Values <0.1 suggest that the variance model fitted to
the data is inadequate. The only variance model available in BMDS models variance as an exponential power
function of the log of the mean (i.e., Var(i) = exp(log a x log (mean(i))p). If variances are constant, the results of
the homogeneity  of variance test and the test for the adequacy of the variance model are the same.
d/>-Value from the goodness-of-fit test.  Values <0.1 suggest that the selected model exhibits significant lack of fit,
and a different model should be chosen.
This value is defined as an estimate of  the expected, relative distance between the fitted model and the unknown
true model and is used to assess model fit. In comparing models fit to the same data, those with lower AIC values
are preferred.
fBMDiSD  = BMD at a BMR, where the  BMR is defined as being 1 SD from the control mean.
gBMDL1SD = 95% lower confidence limit on the BMD at the BMR.

Source: Gulati et al. (1991a).


       The linear and Hill models, using nonconstant variance, did not adequately fit the

decreased rat fetal body weight data based on the results of the homogeneity of variance test and

the test of the adequacy of the variance model (Table B-13). Even though the variances were not

constant, they were not appreciably variable to discourage use of either model to  represent the

data. In addition, both models achieved adequate  fit according to the goodness-of-fit/7-value and

provided indistinguishable AIC values. The linear model was selected to represent the decreased

rat fetal body weight because it provided a lower BMDL5 than the Hill model.
      August, 2009
                                         B-50
DRAFT - DO NOT CITE OR QUOTE

-------
                                     Linear Model with 0.95 Confidence Level
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
27
28
29
30
31
32
33
34
35
36
37
              2.4
              2.2
       o
       Q.
       in
       
-------
                               beta_0  =
                               beta  1  =
                                      2.25818
                                  -0.00116202
                 Asymptotic  Correlation Matrix  of  Parameter  Estimates


                       lalpha          rho       beta_0       beta_l


          lalpha             1           -1          0.14         -0.23


             rho            -1             1         -0.14          0.23


          beta_0          0.14         -0.14             1         -0.64


          beta  1         -0.23          0.23         -0.64             1
                                       Parameter  Estimates
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
69
70
71
72
73
74
75
76
Variable
lalpha
rho
beta 0
beta 1
Estimate
7.45781
-14 .8776
2 .25803
-0.0011476
Std. Err.
2 .26957
3 .06463
0.0268149
0.000292836
                                                        95.0% Wald Confidence Interval

                                                      Lower Conf. Limit   Upper Conf.  Limit

                                                            3.00954           11.9061

                                                            -20.8841           -8.87101

                                                            2.20547           2.31058

                                                         -0.00172155       -0.000573654
    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
278
330
9
8
8
9
9
5
2 .28
2 .17
2 .19
1.99
2 .04
1 .81
2.26
2 .22
2 . 15
2.05
1.94
1. 88
0.12
0 .11
0 .08
0.15
0.42
0 .26
0.0973
0 .111
0 .142
0.199
0.302
0 .381
0.677
-1 .25
0. 883
-0.928
1
-0.407
Model Descriptions for likelihoods calculated
Model Al:         Yij = Mu(i)  + e(ij)
          Var{e(ij)} = Sigma*2


Model A2:         Yij = Mu(i)  + e(ij)
          Var{e(ij)} = Sigma(i)*2


Model A3:         Yij = Mu(i)  + e(ij)
          Var{e(ij)} = exp(lalpha + 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)}
                      Likelihoods of Interest


           Model      Log(likelihood)    # Param's      AIC

            Al           51.030436            7     -88.060872

            A2           67.779534           12    -111.559069

            A3           63.367122            8    -110.734244

        fitted           60.309652            4    -112.619303

             R           42.038909            2     -80.077817
                  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)
      August, 2009
                                     B-52
DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6

 8
 9
10
11
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
 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.)
   Test


   Test 1
   Test 2
   Test 3
   Test 4
                     Tests of  Interest
           -2*log(Likelihood Ratio)   Test  df
                       51.4813

                       33 .4982

                       8 . 82482

                       6.11494
10

 5

 4

 4
  p-value


 <.0001

 <.0001

0.06563

 0.1907
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  less  than .1.   You  may want to consider a
different variance model
The p-value for Test 4 is greater than .1.
to adequately describe the data
             Benchmark Dose Computation


Specified effect =          0.05


Risk Type        =     Absolute risk


Confidence level =          0.95


             BMD =        43.5691
                                            The model chosen seems
            BMDL =
                          31.5623
BMDL computation failed for one  or more  point  on  the BMDL curve.
 The BMDL curve will not be plotted
     August, 2009
                                     B-53
                                                   DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
APPENDIX C. BENCHMARK DOSE MODELING RESULTS FOR THE DERIVATION
                          OF THE ORAL SLOPE FACTOR
Hepatocellular Carcinoma
       The EPA's BMDS (version 1.4.1) Multistage-Cancer model was fit to the incidence data
(Table C-l) for hepatocellular carcinomas in B6C3Fi mice exposed via gavage to
1,1,2,2-tetrachloroethane 5 days/week for 78 weeks (NCI, 1978).  Estimated doses (i.e., BMDs
and BMDLs) associated with 10% extra risks were derived. A summary of these BMDs and
BMDLs are presented in Tables C-2 and C-3.
            Table C-l. Data used for dose-response assessment of hepatocellular
            carcinomas in B6C3Fi mice administered l,l?2,2-tetrachloroethane via
            gavage for 78 weeks
Male mice
Female mice
HEDa (mg/kg-d)
Tumor incidence
HED (mg/kg-d)
Tumor incidence
0
3/36
0
1/40
8.4
13/50
8.0
30/48
16.8
44/49
16.1
43/47
12
aHED is based on body weight scaling to the % power.
Source: NCI (1978).

       Table 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
Gender and
species
Female mice
BMR
(% extra risk)
10
BMD
(mg/kg-d)
0.79
BMDL
(mg/kg-d)
0.63
BMDU
(mg/kg-d)
0.16
13
14
15
16
17
18
19
Source: NCI (1978).

       The multistage model did not achieve adequate fit to the hepatocellular carcinoma data in
male mice; therefore, the incidence data was modeled using additional dichotomous models.
The results of the BMD modeling analysis of the incidence of hepatocellular carcinomas in male
rats are presented in Table C-3.  The BMD modeling of the male did not achieve adequate fit for
any of the dichotomous models; thus, a cancer slope factor was not derived from the male data.
     August, 2009
                                     C-l
DRAFT - DO NOT CITE OR QUOTE

-------
             Table C-3. BMD modeling results based on incidence of hepatocellular
             carcinomas in male B6C3Fi mice administered l,l?2,2-tetrachloroethane via
             gavage for 78 weeks
Fitted dichotomous
model3
Multistage (1°)
Multistage (2°)
Gamma
Logistic
Log-Logistic
Probit
Log-Probit
Weibull
/2 Goodness-of-fit
test^-valueb
O.00005
0.02
NAf
0.05
NA
0.02
NA
NA
AICC
134.69
119.97
116.25
117.56
116.25
118.86
116.25
116.25
BMD10d
(mg/kg-d)
1.4
4.2
7.2
4.6
7.2
4.0
7.3
6.8
BMDL10e
(mg/kg-d)
1.1
3.1
5.5
3.5
5.8
3.1
5.9
4.9
Cancer slope
factor
(mg/kg-d)1
0.09
0.03
0.02
0.03
0.02
0.03
0.02
0.02
      aAll dichotomous dose-response models were fit using BMDS, version 1.5.
      V-Value from the %2 goodness-of-fit test for the selected model. Values <0.1 suggest that the model exhibits a
      significant lack of fit, and a different model should be chosen.
      °Value useful for evaluating model fit.  For those models exhibiting adequate fit, lower values of the AIC suggest
      better model fit.
      dBMD10 = BMD at 10% extra risk.
      eBMDL10 = 95% lower confidence limit on the BMD at 10% extra risk.
      f"NA" = insufficient degrees of freedom for evaluating goodness-of-fit.
      Source:  NCI (1978).
 1
 2           The incidence of hepatocellular carcinomas in female mice was modeled using the
 3    multistage (1° and 2°) model. The results of the BMD modeling analysis of the incidence of
 4    hepatocellular carcinomas in female mice are presented in Table C-4.  The 1° multistage model
 5    was selected for the derivation of the cancer slope factor because this model provided adequate
 6    model fit and the lowest AIC when compared to the results of the 2° multistage model. In
 7    addition, the 2° multistage model had insufficient degrees of freedom to test the goodness-of-fit.
 8    The 1°  multistage model analysis of the incidence of hepatocellular carcinomas in female mice
 9    resulted in a BMDLio of 0.63 mg/kg-day and a cancer slope factor of 0.16 (mg/kg-day)"1.
10
      August, 2009
C-2
DRAFT - DO NOT CITE OR QUOTE

-------
            Table C-4.  BMD modeling results based on incidence of hepatocellular
            carcinomas in female B6C3Fi mice administered l,l?2,2-tetrachloroethane
            via gavage for 78 weeks
Fitted dichotomous
model3
Multistage (1°)
Multistage (2°)
/2 Goodness-of-fit
test^-valueb
0.39
NAf
AICC
104.97
106.22
BMD10d
(mg/kg-d)
0.79
1.1
BMDL10e
(mg/kg-d)
0.63
0.65
Cancer slope
factor
(mg/kg-d)1
0.16
0.15
     aAll dichotomous dose-response models were fit using BMDS, version 1.5.
     V-Value from the %2 goodness-of-fit test for the selected model. Values <0.1 suggest that the model exhibits a
     significant lack of fit, and a different model should be chosen.
     °Value useful for evaluating model fit. For those models exhibiting adequate fit, lower values of the AIC suggest
     better model fit.
     dBMD10 = BMD at 10% extra risk.
     eBMDL10 = 95% lower confidence limit on the BMD at 10% extra risk.
     f"NA" = insufficient degrees of freedom for evaluating goodness-of-fit.

     Source: NCI (1978).
                              Multistage Cancer Model with 0.95 Confidence Level
2
3
4
5
6
7
           0.8
       T3
       CD
           0.6
       I   0.4
       ro
           0.2
                        Multistage Cancer
                       Linear extrapolation
                      BMD
                                                      8
                                                     dose
                                                   10
            12
14
16
         15:5609/032008
Fit of the 1° multistage cancer model to the incidence of hepatocellular
carcinomas in female B6C3Fi mice administered l,l?2,2-tetrachloroethane
via gavage for 78 weeks (NCI, 1978)
     August, 2009
                                  C-3
DRAFT - DO NOT CITE OR QUOTE

-------
 2           Multistage  Cancer  Model.  (Version:  1.5;  Date: 02/20/2007)
 3           Input Data  File: M:\TETRACHLOROETHANE DOSE-RESPONSE
 4   MODELING\CANCER\ FEMALE jy[ICE_HEPATOCELLULAR_CARCINOMA.(d)
 5           Gnuplot Plotting File:   M:\TETRACHLOROETHANE DOSE-RESPONSE
 6   MODELING\CANCER\FEMALE_MICE_HEPATOCELLULAR_CARCINOMA.plt
 7                                              Wed Sep 03 15:56:03 2008
 8    ====================================================================
 9
10    HMDS MODEL RUN
j j
12
13      The form of the  probability  function is:
14
15      P[response] = background + (1-background)*[1-EXP(
16                    -betal*doseAl)]
17
18      The parameter betas  are restricted to be positive
19
20
21      Dependent variable = Response
22      Independent variable =  Dose
23
24    Total number of observations =  3
25    Total number of records with missing values = 0
26    Total number of parameters in model = 2
27    Total number of specified parameters = 0
28    Degree of polynomial = 1
29
30
31    Maximum number of  iterations =  250
32    Relative Function  Convergence has  been set to: le-008
33    Parameter Convergence  has been  set to: le-008
34
35
36
37                     Default  Initial Parameter Values
38                        Background  =            0
39                           Beta(l)  =     0.151528
40
41
42              Asymptotic Correlation Matrix of Parameter Estimates
43
44                Background      Beta(l)
45
46   Background            1       -0.54
47
48      Beta(l)        -0.54            1
49
50
51
52                                     Parameter Estimates
53
54                                                 95.0% Wald Confidence  Interval
JJ         Variable        Estimate       Std. Err.    Lower Conf. Limit  Upper Conf. Limit
JO       Background
57          Beta(l)
58
59
60
61
62


     August, 2009                        C-4          DRAFT - DO NOT  CITE OR QUOTE

-------
1
2
3
4
5
6
7
9
10
11
12
13
14
i «
i j
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


Model
Full model
Fitted model
0.3875
Reduced model
AIC:




Dose Est


0.0000 0.
8.0000 0.
16.1000 0.

ChiA2 = 0.72



Benchmark Dose

Specified effect

Risk Type

Confidence level

BMD

BMDL

BMDU

Taken together, (
interval for the

Multistage Cancer

Analysis of Deviance Table

Log (likelihood) # Param's Deviance Test d.f. P-value
-50.1115 3
-50.4848 2 0.746754 1

-92.948 1 85.673 2 <.0001
104.97


Goodness of Fit
Scaled
. Prob. Expected Observed Size Residual


0241 0.964 1 40 0.037
6660 31.967 30 48 -0.602
8872 41.698 43 47 0.600

d.f. = 1 P-value = 0.3948



Computation

0.1

= Extra risk

0.95

0.786171

0.631437

0.989834

0.631437, 0.989834) is a 90 % two-sided confidence
BMD

Slope Factor = 0.158369

43
     August, 2009
C-5
DRAFT - DO NOT CITE OR QUOTE

-------