DRAFT - DO NOT CITE OR QUOTE
EPA/63 5/R-09/007A
www.epa.gov/iris
oEPA
TOXICOLOGICAL REVIEW
OF
HEXACHLOROETHANE
(CAS No. 67-72-1)
In Support of Summary Information on the
Integrated Risk Information System (IRIS)
May 2010
NOTICE
This document is an External Peer 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 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.
11
DRAFT - DO NOT CITE OR QUOTE
-------
CONTENTS—TOXICOLOGICAL REVIEW OF HEXACHLOROETHANE
(CAS No. 67-72-1)
LIST OF TABLES vi
LIST OF FIGURES viii
LIST 01 ABBREVIATIONS AM) ACRONYMS ix
FOREWORD xi
AUTHORS, CONTRIBUTORS, AND REVIEWERS xii
1. INTRODUCTION 1
2. CHEMICAL AM) PHYSICAL IM ORMATION 3
3. TOXICOKINETICS 5
3.1. ABSORPTION 5
3.2. DISTRIBUTION 5
3.3. METABOLISM 8
3.4. ELIMINATION 15
3.5. PHYSIOLOGICALLY BASED PHARMACOKINETIC MODELS 16
4. HAZARD IDENTIFICATION 17
4.1. STUDIES IN HUMANS—EPIDEMIOLOGY, CASE REPORTS, CLINICAL
CONTROLS 17
4.2. SUBCHRONIC AND CHRONIC STUDIES AND CANCER BIOASSAYS IN
AMY1AI.S ORAL AM) INHALATION 19
4.2.1. Oral 19
4.2.1.1. Subchronic Exposure 19
4.2.1.2. Chronic Exposure and Carcinogenicity 23
4.2.2. Inhalation 31
4.2.2.1. Subchronic Exposure 31
4.2.2.2. Chronic Exposure 33
4.3. REPRODUCTIVE/DEVELOPMENTAL STUDIES—ORAL AND
INHALATION 33
4.3.1. Oral 33
4.3.2. Inhalation 36
4.4. OTHER DURATION- OR ENDPOINT-SPECIFIC STUDIES 37
4.4.1. Acute Exposure Studies 37
4.4.1.1. Oral 37
4.4.1.2. Inhalation 38
4.4.2. Short-term Exposure Studies 39
4.4.3. Neurological 41
4.4.3.1. Oral Studies 42
4.4.3.2. Inhalation Studies 43
4.4.4. Immunological 43
4.4.5. Dermatological 44
4.4.6. Eye Irritation 44
4.5. MECHANISTIC DATA AND OTHER STUDIES IN SUPPORT OF THE
MODE OF ACTION 45
in
DRAFT - DO NOT CITE OR QUOTE
-------
4.5.1. Genotoxicity 45
4.5.2. In Vitro and Ex Vivo Studies Using Isolated Target Tissues/Organs or
Cells 51
4.5.3. Structure Activity Relationships 54
4.6. SYNTHESIS OF MAJOR NONCANCER EFFECTS 56
4.6.1. Oral 56
4.6.1.1. Nephrotoxicity 58
4.6.1.2. Hepatotoxicity 60
4.6.1.3. Developmental Toxicity 61
4.6.2. Inhalation 61
4.6.3. Mode-of-Action Information 62
4.7. EVALUATION 01 CARCINOGENICITY 64
4.7.1. Summary of Overall Weight of Evidence 64
4.7.2. Synthesis of Human, Animal, and Other Supporting Evidence 65
4.7.3. Mode-of-Action Information 67
4.7.3.1. Kidney Tumors 67
4.7.3.2. Liver Tumors 80
4.7.3.3. Pheochromocytomas 82
4.8. SUSCEPTIBLE POPULATIONS AND LIFE STAGES 82
4.8.1. Possible Childhood Susceptibility 83
4.8.2. Possible Gender Differences 83
4.8.3. Other 83
5. DOSE-RESPONSE ASSESSMENTS 84
5.1. ORAL REFERENCE DOSE (RID) 84
5.1.1. Choice of Principal Study and Critical Effect—with Rationale and
Justification 84
5.1.2. Methods of Analysis—Including Models 88
5.1.3. RfD Derivation—Including Application of Uncertainty Factors (UFs) 90
5.1.4. RfD Comparison Information 92
5.1.5. Previous RfD Assessment 94
5.2. INHALATION REFERENCE CONCENTRATION (RfC) 94
5.2.1. Choice of Principal Study and Critical Effect—with Rationale and
Justification 94
5.2.2. Methods of Analysis—Including Models 97
5.2.3. RfC Derivation—Including Application of Uncertainty Factors (UFs) 98
5.2.4. RfC Comparison Information 100
5.2.5. Previous RfC Assessment 100
5.3. UNCERTAINTIES IN THE ORAL REFERENCE DOSE AND INHALATION
REFERENCE CONCENTRATION 100
5.4. CANCER ASSESSMENT 102
5.4.1. Choice of Study /Data—with Rationale and Justification 103
5.4.2. Dose-response Data 103
5.4.3. Dose Adjustments and Extrapolation Methods 104
5.4.4. Oral Slope Factor and Inhalation Unit Risk 106
5.4.5. Uncertainties in Cancer Risk Values 107
5.4.5.1. Sources of Uncertainty 109
5.4.6. Previous Cancer Assessment 112
DRAFT - DO NOT CITE OR QUOTE
-------
6. MAJOR CONCLUSIONS IN THE CHARACTERIZATION OF HAZARD AND
DOSE RESPONSE 113
6.1. HUMAN HAZARD POTENTIAL 113
6.2. DOSE RESPONSE 114
6.2.1. Oral Noncancer 114
6.2.2. Inhal ati on Noncancer 115
6.2.3. Cancer 115
7. REFERENCES 118
APPENDIX A: SUMMARY OF EXTERNAL PEER REVIEW AND PUBLIC
COMMENTS AND DISPOSITION A-l
APPENDIX B: BENCHMARK DOSE MODELING OUTPUT B-1
v
DRAFT - DO NOT CITE OR QUOTE
-------
LIST OF TABLES
2-1. Physical properties of HCE 3
3-1. HCE, PERC, and pentachloroethane tissue concentrations in anesthetized sheep 8.5
hours after injection of 500 mg/kg HCE 6
3-2. Time course of HCE concentrations in male rat tissues after 57 days of dietary
exposure to 62 mg/kg-day 7
3-3. HCE concentrations in male and female rat tissues after 110 or 111 days of dietary
exposure 8
3-4. Disposition of HCE in male rats and mice during 48 hours following administration
of an MTD for 4 weeks 10
3-5. Metabolism of HCE measured in rats and mice 11
3-6. Product formation rates and relative ratios of the products formed by CYP450 1A2
metabolism of HCE 14
4-1. Body, kidney, and liver weights of rats exposed to HCE in the diet for 16 weeks 21
4-2. Histopathological results on kidney in rats exposed to HCE in the diet for 16 weeksa 21
4-3. Organ weight to body weight ratios for rats exposed to HCE for 13 weeks 22
4-4. Incidence and severity of nephropathy in male and female rats treated with HCE 24
4-5. Additional kidney effects in HCE-treated rats 26
4-6. Renal tubular hyperplasia and tumor incidences in HCE-treated male rats 26
4-7. Adrenal medullary lesions in HCE-treated male rats 27
4-8. Tumor incidences21 in male rats gavaged with HCE 29
4-9. Tumor incidences in female rats gavaged with HCE 30
4-10. Incidence of hepatocellular carcinomas in mice 31
4-11. Summary of HCE effects on pregnant Wistar rats and their fetuses 35
4-12. Summary of skeletal effects on fetuses from HCE-exposed rats 36
4-13. Summary of acute exposure data in rats, rabbits, and guinea pigs 37
4-14. Summary of toxicity data from male rats exposed to HCE for 21 days 41
vi DRAFT - DO NOT CITE OR QUOTE
-------
4-15. Summary of genotoxicity studies of HCE 46
4-16. Number of enzyme-altered foci in rat liver of the promotion protocol 52
4-17. In vivo covalent binding of [14C]-HCE to DNA, RNA, and proteins from rat and
mouse organs 53
4-18. In vitro binding of [14C]-HCE to calf thymus DNA mediated by microsomal and/or
cytosolic phenobarbital-induced fractions of rat and mouse organs 54
4-19. Oral toxicity studies for HCE 57
4-20. Inhalation toxicity studies with HCE 62
4-21. Nephrotoxic effects characteristic of a2u-globulin nephropathy observed in male and
female rats administered HCE 69
5-1. Incidences of noncancerous kidney and liver effects in rats following oral exposure to
HCE 86
5-2. Summary of the BMD modeling results for the kidney 89
5-3. Potential PODs for nephrotoxicity in male rats with applied UFs and potential
reference values 92
5-4. Noncancerous effects observed in animals exposed to HCE via inhalation 97
5-5. Summary of incidence data in rodents orally exposed to HCE for use in cancer dose-
response assessment 104
5-6. Summary of BMD modeling results for oral cancer assessment of HCE 106
B-l. Dose-response modeling results using BMDS (version 2.0) based on non-cancerous
kidney and liver effects in rats following oral exposure to HCE B-l
B-2. Dose-response modeling results using BMDS (version 2.0) for BMRs of 1, 5, and
10% based on noncancerous kidney and liver effects in rats following oral exposure
to HCE B-5
vii DRAFT - DO NOT CITE OR QUOTE
-------
LIST OF FIGURES
2-1. Structure of HCE 3
3-1. Possible metabolic pathway of HCE 9
5-1. Array of potential PODs with applied UFs and potential reference values for
nephrotoxic effects of studies in Table 5-3 93
viii
DRAFT - DO NOT CITE OR QUOTE
-------
LIST OF ABBREVIATIONS AND ACRONYMS
AIC
Akaike's information criterion
ALD
approximate lethal dosage
ALT
alanine aminotransferase
AST
aspartate aminotransferase
BMD
benchmark dose
BMDL
benchmark dose lower confidence limit
BMDS
Benchmark Dose Software
BMR
benchmark response
BUN
blood urea nitrogen
BW
body weight
CA
chromosomal aberration
CAS
Chemical Abstracts Service
CASRN
Chemical Abstracts Service Registry Number
CBI
covalent binding index
CHO
Chinese hamster ovary
CL
confidence limit
CNS
central nervous system
CPN
chronic progressive nephropathy
CYP450
cytochrome P450
DAF
dosimetric adjustment factor
DEN
diethylnitrosamine
DMSO
dimethylsulfoxide
DNA
deoxyribonucleic acid
FDA
Food and Drug Administration
FEVi.o
forced expiratory volume of 1 second
GD
gestation day
GDH
glutamate dehydrogenase
GGT
y-glutamyl transferase
GSH
glutathione
GST
glutathione-S-transferase
Hb/g-A
animal blood:gas partition coefficient
Hb/g-H
human blood:gas partition coefficient
HCE
hexachl oroethane
HEC
human equivalent concentration
HEP
human equivalent dose
i.p.
intraperitoneal
IRIS
Integrated Risk Information System
IVF
in vitro fertilization
LC50
median lethal concentration
LD50
median lethal dose
LOAEL
lowest-observed-adverse-effect level
MN
micronuclei
MNPCE
micronucleated polychromatic erythrocyte
MTD
maximum tolerated dose
NAG
iV-acetyl-P-D-glucosaminidase
NCI
National Cancer Institute
IX
DRAFT - DO NOT CITE OR QUOTE
-------
NOAEL
no-observed-adverse-effect level
NTP
National Toxicology Program
OCT
ornithine carbamoyl transferase
PBPK
physiologically based pharmacokinetic
PBPK
physiologically based pharmacokinetic
PCNA
proliferating cell nuclear antigen
PERC
tetrachloroethene, tetrachloroethylene, perchloroethylene
POD
point of departure
POD[adj]
duration-adjusted POD
QSAR
quantitative structure-activity relationship
RDS
replicative DNA synthesis
RfC
inhalation reference concentration
RfD
oral reference dose
RGDR
regional gas dose ratio
RNA
ribonucleic acid
SAR
structure activity relationship
SCE
sister chromatid exchange
SD
standard deviation
SDH
sorbitol dehydrogenase
SE
standard error
SGOT
glutamic oxaloacetic transaminase, also known as AST
SGPT
glutamic pyruvic transaminase, also known as ALT
SSD
systemic scleroderma
TCA
trichloroacetic acid
TCE
tri chl oroethy 1 ene
TWA
time-weighted average
UF
uncertainty factor
UFa
interspecies uncertainty factor
UFh
intraspecies uncertainty factor
UFS
subchronic-to-chronic uncertainty factor
UFd
database deficiencies uncertainty factor
U.S. EPA
U.S. Environmental Protection Agency
X
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 chronic exposure to
hexachloroethane. It is not intended to be a comprehensive treatise on the chemical or
toxicological nature of hexachloroethane.
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 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).
XI
DRAFT - DO NOT CITE OR QUOTE
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
CHEMICAL MANAGER/AUTHOR
John Cowden, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC
AUTHORS
Samantha Jones, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC
CONTRIBUTORS
Ted Berner, M.S.
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC
Glinda Cooper, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC
Andrew A. Rooney, Ph.D.
Currently at National Toxicology Program
Center for the Evaluation of Risks to Human Reproduction
National Institute of Environmental Health Sciences
Research Triangle Park, NC
CONTRACTOR SUPPORT
James Kim, Ph.D.
Sciences International, Inc.
Alexandria, VA
REVIEWERS
This document has been provided for review to EPA scientists and interagency reviewers
from other federal agencies and White House offices.
Xll
DRAFT - DO NOT CITE OR QUOTE
-------
INTERNAL EPA REVIEWERS
Ambuja Bale, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
Ghazi Dannan, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
Kate Guyton, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
Maureen Gwinn, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
Jennifer Jinot, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
Channa Keshava, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
Allan Marcus, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
D. Charles Thompson, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
John Whalan
National Center for Environmental Assessment
Office of Research and Development
Xlll
DRAFT - DO NOT CITE OR QUOTE
-------
1. INTRODUCTION
This document presents background information and justification for the Integrated Risk
Information System (IRIS) Summary of the hazard and dose-response assessment of
hexachloroethane (HCE). IRIS Summaries may include oral reference dose (RfD) and inhalation
reference concentration (RfC) values for chronic and other exposure durations, and a
carcinogenicity assessment.
The RfD and RfC, if derived, provide quantitative information for use in risk assessments
for health effects known or assumed to be produced through a nonlinear (presumed threshold)
mode of action. The RfD (expressed in units of mg/kg-day) is defined as an estimate (with
uncertainty spanning perhaps an order of magnitude) of a daily exposure to the human
population (including sensitive subgroups) that is likely to be without an appreciable risk of
deleterious effects during a lifetime. The inhalation RfC (expressed in units of mg/m3) is
analogous to the oral RfD, but provides a continuous inhalation exposure estimate. The
inhalation RfC considers toxic effects for both the respiratory system (portal-of-entry) and for
effects peripheral to the respiratory system (extrarespiratory or systemic effects). Reference
values are generally derived for chronic exposures (up to a lifetime), but may also be derived for
acute (<24 hours), short-term (>24 hours up to 30 days), and subchronic (>30 days up to 10% of
lifetime) exposure durations, all of which are derived based on an assumption of continuous
exposure throughout the duration specified. Unless specified otherwise, the RfD and RfC are
derived for chronic exposure duration.
The carcinogenicity assessment provides information on the carcinogenic hazard
potential of the substance in question and quantitative estimates of risk from oral and inhalation
exposure may be derived. The information includes a weight-of-evidence judgment of the
likelihood that the agent is a human carcinogen and the conditions under which the carcinogenic
effects may be expressed. Quantitative risk estimates may be derived from the application of a
low-dose extrapolation procedure. If derived, the oral slope factor is a plausible upper bound on
the estimate of risk per mg/kg-day of oral exposure. Similarly, an inhalation unit risk is a
plausible upper bound on the estimate of risk per (j,g/m3 air breathed.
Development of these hazard identification and dose-response assessments for HCE has
followed the general guidelines for risk assessment as set forth by the National Research Council
(1983). EPA Guidelines and Risk Assessment Forum Technical Panel Reports that may have
been used in the development of this assessment include the following: Guidelines for the
Health Risk Assessment of Chemical Mixtures (U.S. EPA, 1986a), Guidelines for Mutagenicity
Risk Assessment (U.S. EPA, 1986b), Recommendations for and Documentation of Biological
Values for Use in Risk Assessment (U.S. EPA, 1988), Guidelines for Developmental Toxicity
Risk Assessment (U.S. EPA, 1991a), Interim Policy for Particle Size and Limit Concentration
1
DRAFT - DO NOT CITE OR QUOTE
-------
Issues in Inhalation Toxicity (U.S. EPA, 1994a), Methods for Derivation of Inhalation Reference
Concentrations and Application of Inhalation Dosimetry (U.S. EPA, 1994b), Use of the
Benchmark Dose Approach in Health Risk Assessment (U.S. EPA, 1995), Guidelines for
Reproductive Toxicity Risk Assessment (U.S. EPA, 1996), Guidelines for Neurotoxicity Risk
Assessment (U.S. EPA, 1998), Science Policy Council Handbook. Risk Characterization (U.S.
EPA, 2000a), Benchmark Dose Technical Guidance Document (U.S. EPA, 2000b),
Supplementary Guidance for Conducting Health Risk Assessment of Chemical Mixtures (U.S.
EPA, 2000c), A Review of the Reference Dose and Reference Concentration Processes (U.S.
EPA, 2002), Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005a), Supplemental
Guidance for Assessing Susceptibility from Early-Life Exposure to Carcinogens (U.S. EPA,
2005b), Science Policy Council Handbook: Peer Review (U. S. EPA, 2006a), and A Framework
for Assessing Health Risks of Environmental Exposures to Children (U.S. EPA, 2006b).
The literature search strategy employed for this compound was based on the Chemical
Abstracts Service Registry Number (CASRN) and at least one common name. Any pertinent
scientific information submitted by the public to the IRIS Submission Desk was also considered
in the development of this document. The relevant literature was reviewed through February
2010.
2
DRAFT - DO NOT CITE OR QUOTE
-------
2. CHEMICAL AND PHYSICAL INFORMATION
Hexachloroethane (HCE; CASRN 67-72-1) is a halogenated hydrocarbon consisting of
six chlorines attached to an ethane backbone (Figure 2-1). Synonyms include
1,1,1,2,2,2-hexachloroethane, ethane hexachloride, ethylene hexachloride, perchloroethane,
carbon hexachloride, and carbon trichloride (ChemlDplus Advanced, 2005; ACGIH, 1991).
Certain physical and chemical properties are shown below in Table 2-1 (ACGIH, 2001; ATSDR,
1997a; Budavari, 1989; Howard, 1989; Weast, 1986; Spanggord et al., 1985; Verschueren, 1983;
U.S. EPA, 1982, 1979).
CI CI
CI—c—C—CI
CI CI
Figure 2-1. Structure of HCE.
Table 2-1. Physical properties of HCE
Name
Hexachloroethane
CASRN
67-72-1
Synonyms
1,1,1,2,2,2-hexachloroethane, ethane hexachloride, ethylene hexachloride,
perchloroethane, carbon hexachloride, carbon trichloride
Molecular weight
236.74 g/mol
Molecular formula
C2C16
Melting point
Sublimes without melting
Boiling point
186.8°C
Density
2.091 g/mL at 20°C
Water solubility3
50 mg/L at 22°C; 14 mg/L at 25°C
Log Kow
3.82a, 3.34b, 4.14°
Log Koc
4.3
Vapor pressure
0.5 mmHg at 20°C; 1.0 mmHg at 32.7°C
Henry's law constant
2.8 x 10"3 atm-m3/mol at 20°C
Conversion factor
1 ppm = 9.68 mg/m3; 1 mg/m3= 0.10 ppm
Sources: "Howard (1989); bU.S. EPA (1979); 'Hansch et al. (1995).
HCE was produced in the United States for commercial distribution from 1921 to 1967,
but is currently not commercially distributed (ATSDR, 1997a; IARC, 1979). In the 1970s,
producers of HCE reported that HCE was not distributed, but was used in-house or recycled
(ATSDR, 1997a); distributors in the 1970s imported HCE from France, Spain, and the United
3
DRAFT - DO NOT CITE OR QUOTE
-------
Kingdom (ACGIH, 2001; ATSDR, 1997a). HCE and tetrachloroethane imports combined were
1.5 million pounds in 1989 and 612,000 pounds in 2000 (NTP, 2005). HCE production in 1977
was between 2 and 20 million pounds; more recent information on production of HCE was not
located (NTP, 2005; ATSDR, 1997a). HCE is produced by the chlorination of
tetrachloroethylene (PERC) in the presence of ferric chloride at temperatures of 100-140°C
(ATSDR, 1997a; U.S. EPA, 1991b; Fishbein, 1979; IARC, 1979). HCE is primarily used in the
military for smoke pots, smoke grenades, and pyrotechnic devices (ACGIH, 2001; ATSDR,
1997a; U.S. EPA, 1991b; IARC, 1979). HCE was also identified in the headspace of
chlorine-bleach-containing household products (Odabasi, 2008). In the past, HCE was used as
an antihelminthic for the treatment of sheep flukes, but is no longer used for this purpose since
the U.S. Food and Drug Administration (FDA) withdrew approval for this use in 1971 (ATSDR,
1997a). HCE has also been used as a polymer additive, a moth repellant, a plasticizer for
cellulose esters, and an insecticide solvent, and in metallurgy for refining aluminum alloys
(ATSDR, 1997a; U.S. EPA, 1991b).
4
DRAFT - DO NOT CITE OR QUOTE
-------
3. TOXICOKINETICS
3.1. ABSORPTION
There are no studies that have systematically evaluated HCE absorption in humans by the
oral or inhalation routes of exposure. However, uptake was demonstrated by Younglai et al.
(2002) when HCE was identified in follicular fluid during an analysis for environmental
contaminants in 21 couples undergoing in vitro fertilization (IVF). These data identify the
potential for HCE absorption but not the source or route of exposure. No studies have been
reported that assess the inhalation absorption of HCE in humans. The dermal absorption rate of
HCE has been described as limited (ATSDR, 1997a). Based on physical properties, the
absorption of a saturated HCE solution across human skin was estimated to be
0.023 mg/cm2/hour (Fiserova-Bergerova et al., 1990).
Studies in animals via the oral route of exposure demonstrated that HCE is absorbed and
primarily distributed to fat (Gorzinski et al., 1985; Nolan and Karbowski, 1978; Fowler, 1969).
Fowler (1969) orally administered 500 mg/kg HCE to Scottish Blackface or Cheviot sheep and
found that maximal venous blood concentrations of HCE (10-28 (j,g/mL) were reached at
24 hours after HCE exposure, indicating slow absorption. Jondorf et al. (1957) reported that
rabbits fed [14C]-radiolabeled HCE at 500 mg/kg excreted only 5% of the applied radioactivity in
urine over a period of 3 days (fecal measurements were not conducted). During this 3-day
period, 14-24% of the applied radioactivity was detected in expired air, and the remainder was
present in the tissues and intestinal tract. The amount of HCE absorbed by the rabbits was not
determined; however, based on the amount of radioactivity present in urine and expired air,
approximately 19-29% of the HCE was absorbed. Studies in rats and mice (Mitoma et al., 1985)
using [14C]-radiolabeled HCE (500 mg/kg for rats; 1,000 mg/kg for mice) administered orally,
via corn oil, indicated that the amounts absorbed were 65-71 and 72-88%, respectively, based
on the amount of radiolabel detected in expired air and excreta.
3.2. DISTRIBUTION
There are limited data on the distribution of HCE in humans (Younglai et al., 2002). The
animal studies evaluated (Gorzinski et al., 1985; Nolan and Karbowski, 1978; Fowler, 1969)
consistently demonstrated that HCE is distributed primarily to fat tissue followed by the kidney
and to a lesser extent the liver and the blood (Gorzinski et al., 1985; Nolan and Karbowski,
1978).
Younglai et al. (2002) evaluated the concentrations of various environmental
contaminants in follicular fluid, serum, and seminal plasma of 21 couples undergoing IVF. HCE
was one of the contaminants identified in >50%) of follicular fluid samples, suggesting
postabsorptive distribution to reproductive organs. The average HCE concentration in follicular
5
DRAFT - DO NOT CITE OR QUOTE
-------
fluid was 232 ± 27 pg/mL (mean ± standard error [SE]). HCE was not detected in human female
serum obtained during oocyte retrieval for IVF. This study focused primarily on chemicals such
as pesticides and polychlorinated biphenyls, and the authors could not make any conclusions
with regards to the level of HCE in follicular fluid and its effect on fertility.
Fowler (1969) evaluated the tissue distribution of HCE in sheep. Two sheep were fasted
for 24 hours and then anesthetized with pentobarbitone sodium. An HCE solution (15% w/v in
olive oil) was injected for a total dose of 500 mg/kg directly into the rumen and lower duodenum
(dose was divided). Anesthesia was maintained for 8.5 hours, after which time the sheep were
sacrificed and tissues were taken within 10 minutes of death. Tissues that were evaluated for
HCE include the brain, fat, kidney, liver, and muscle. Bile and blood were also evaluated. HCE
was widely distributed and the highest levels were found in fat of one sheep. Fat from different
sites did not show significant variation in HCE concentration. The second sheep had only trace
amounts of HCE in tissue (see Table 3-1).
Table 3-1. HCE, PERC, and pentachloroethane tissue concentrations in
anesthetized sheep 8.5 hours after injection of 500 mg/kg HCE
Concentration (jig/g)
Sheep 1
Sheep 2
Tissue
HCE
PERC
Pentachloroethane
HCE
PERC
Pentachloroethane
Bile (4 hr)
1.7
0.3
Trace
2.2
0.5
Nil
Blood (6 hr)
0.2
0.4
Trace
0.2
0.2
Nil
Brain
0.2
0.9
0.02
Trace
Trace
Trace
Fat
1.1
2.1
0.02
Trace
0.6
Nil
Kidney
0.1
1.2
Trace
Trace
0.6
Trace
Liver
0.2
0.9
0.01
Trace
2.8
Trace
Muscle
0.04
0.5
0.01
Trace
Trace
Trace
Source: Fowler (1969).
Nolan and Karbowski (1978) studied tissue clearance of HCE in rats. Male F344 rats
were placed on an HCE-containing diet that was calculated to deliver 100 mg/kg-day (later
determined to be 62 mg/kg-day by Gorzinski et al., 1985) for 57 days. After this exposure
period, the rats were returned to an HCE-free control diet and sacrificed (groups of three or four
rats) 0, 3, 6, 13, 22, and 31 days after this change in exposure. Samples of fat, liver, kidney, and
whole blood were collected for HCE analysis. The time-course related tissue HCE
concentrations are presented in Table 3-2. The highest tissue concentrations of HCE were in fat,
which were 3-fold greater than the concentration in the kidney and over 100-fold greater than
blood and liver concentrations. Fat concentrations decreased from 303 ± 50 j_ig/g in a first-order
manner with a half-life of 2.7 days. Concentrations in blood and kidney also decreased in a first-
order manner with half-lives of 2.5 and 2.6 days, respectively. Liver concentrations initially
6
DRAFT - DO NOT CITE OR QUOTE
-------
increased in the first 3 days postexposure, but began to decrease by day 6. The half-life for liver
HCE was 2.3 days (calculated after peak levels were reached at day 3). These same results were
published in a follow-up study by Gorzinski et al. (1985) that included a toxicity assessment.
Table 3-2. Time course of HCE concentrations in male rat tissues after
57 days of dietary exposure to 62 mg/kg-day
HCE tissue concentrations (n = 3 or 4) (mean + SD jig/g tissue)
Days after cessation of HCE exposure
Blood
Liver
Kidney
Fat
0
0.834 + 0.223
0.143 +0.040
81.8 + 5.3
303 + 50
3
0.279 + 0.048
0.399 +0.188
41.0 + 1.4
107.8 + 10.5
6
0.0835 +0.0063
0.303 +0.1563
18.5b
62.45 + 3.043
13
0.015 + 0.005
0.039 +0.023
2.53 + 1.02
6.56 + 0.52
22
0.002 + 0.001
0.001 +0.001
0.194 + 0.171
0.472 + 0.232
31
ND°
ND°
0.026 + 0.006
0.125+0.020
aValues from one of the three rats was consistently low and not used to obtain the mean + standard deviation (SD).
bOne sample was lost and a mean + SD could not be calculated.
°ND: not detected (detection limit of 0.001 (ig/g).
Sources: Gorzinski et al. (1985); Nolan and Karbowski (1978).
Nolan and Karbowski (1978) also evaluated tissue concentrations of HCE in both male
and female rats after an exposure period of 110-111 days (16 weeks) to doses of 3, 30, and
100 mg/kg-day via the diet. The actual doses were approximated as 1, 15, and 62 mg/kg-day
after factoring in volatility of the test material from the food and based on linear nighttime food
consumption rates (Gorzinski et al., 1985). The tissue concentrations are presented in Table 3-3.
Kidney concentrations of HCE were much higher in male rats compared with female rats,
particularly at the highest dose (47-fold greater in males) (Nolan and Karbowski, 1978). Kidney
concentrations of HCE proportionately increased with the doses in males, whereas the increase in
females was dose-dependent but not proportionate. The authors noted that the HCE kidney
concentrations and kidney toxicity were consistently different for the male and female rats.
Consequently, they speculated that the male rats would be 10-30 times more sensitive than
female rats to HCE toxicity, based on the relative HCE concentration measured in the rat kidney
(assuming that toxicity is due to HCE and not a metabolite). Both sexes exhibited comparable
levels (although levels in males were slightly greater) of HCE in blood, liver, and fat;
concentrations in fat were the highest for both sexes. Blood levels of HCE did not correlate well
to either the exposure dose or the dose at the major target organ, the kidney, indicating that blood
levels of HCE may not be a suitable metric for the estimation of exposure to HCE in rats.
7
DRAFT - DO NOT CITE OR QUOTE
-------
Table 3-3. HCE concentrations in male and female rat tissues after 110 or
111 days of dietary exposure
Dose (mg/kg-day)
HCE tissue concentration (n = 3 or 4) (mean + SD, jig/g tissue)
Blood
Liver
Kidney
Fat
1
Male
0.079 + 0.057
0.291+0.213
1.356 +0.286
3.09+0.33
Female
0.067 + 0.039
0.260+0.035
0.369 +0.505
2.59+0.72
15
Male
0.596 + 0.653
1.736 + 1.100
24.33 +5.73
37.90 +6.10
Female
0.162 + 0.049
0.472+0.204
0.688+0.165
45.27+ 11.33
62
Male
0.742 + 0.111
0.713+0.343
95.12 ± 11.56
176.1 + 14.5
Female
0.613+0.231
0.631 +0.262
2.01+0.66
162.1+7.1
Sources: Gorzinski et al. (1985); Nolan and Karbowski (1978).
3.3. METABOLISM
In vitro studies using liver microsomes indicated that the major enzymes involved in
HCE metabolism are phenobarbital-inducible cytochrome P450 (CYP450) isoforms (Salmon et
al., 1985; Town and Leibman, 1984; Nastainczyk et al., 1982, 1981; Salmon et al., 1981);
however, no specific (phenobarbital-inducible) isoforms have been identified. The isoforms
induced by phenobarbital include those from the 2A, 2B, 2C, and 3 A subfamilies. One study
(Yanagita et al., 1997) found some evidence for CYP1A2 involvement in the metabolism of
HCE, although this was not supported by the results from in vitro studies with
3-methylcholanthrene, an inducer of the CYP450 1 subfamily (Nastainczyk et al., 1982, 1981;
Van Dyke and Wineman, 1971). Information regarding the roles of Aroclor 1254-inducible
enzymes other than 1A2 (including CYP 2A6, 2E1, 2C9, 2C19, 2D6, and 3A4) is not available
for HCE.
The metabolism data for HCE are limited because there are only three in vivo studies
available that provide information on metabolites: Mitoma et al. (1985) in rats and mice; Jondorf
et al. (1957) in rabbits; and Fowler (1969) in sheep. Each of these studies tends to support
limited metabolism for HCE. The data from the in vivo and in vitro studies support a conclusion
that metabolism of HCE is incomplete, with excretion of unmetabolized HCE in exhaled air and
possibly in urine. A variety of intermediary metabolites have also been identified in exhaled air
and urine (Fowler, 1969; Jondorf et al., 1957). Figure 3-1 provides a possible metabolic pathway
for HCE derived from the in vivo and in vitro data with ordering of metabolites based on
sequential dechlorination and oxidation state. The HCE metabolism information was
supplemented with data on the metabolism of the PERC (ATSDR, 1997b), trichloroethylene
(TCE; ATSDR, 1997c), and 1,1,2,2-tetrachloroethane (ATSDR, 2008) intermediary metabolites.
8
DRAFT - DO NOT CITE OR QUOTE
-------
CI CI
\ /
c=c
/ \
CI CI
Tetrachloroethylene
ci 4
I
CI H
I I
c—c—c— o
I I
CI H
Trichloroethanol
I
I
CI
I
CI —C —C'
o
\
Cl
Trichloroacetic acid
Cl Cl
I I
ci — c—c—
I I
Cl Cl
Hexachloroethane
Cl
Cl
Cl
OH
1
CC13 - CC12
Cl Cl
I I
H—C—C—H
I I
Cl Cl
1,1,2,2-
Tetrachloroethane
Free radical reactions
Cl Cl
I I
Cl —c—c—H
I I
Cl Cl
Pentachloroethane
2C1-^
Cl
I
Cl
2C1
¦4
I
I
Cl
O
HO
Cl
Cl
Cl
ci ^ \ /
c=c
/ \
Cl H
Trichloroethylene
ci~\
I
HO Cl
H Cl
Dichloroethanol
Dichloroacetic acid
2C1 S
o o v
I
I
ci
HO OH
Oxalic acid
CO,
I
\ O Cl
HO H
Monochloroacetic acid
One carbon pool
Sources: Adapted from ATSDR (1997a); Mitoma et al. (1985); Town and
Leibman (1984); Nastainczyk et al. (1982, 1981); Bonse and Henschler (1976);
Fowler (1969); Jondorf et al. (1957).
Figure 3-1. Possible metabolic pathway of HCE.
9
DRAFT - DO NOT CITE OR QUOTE
-------
Mitoma et al. (1985) examined the distribution of HCE in male Osborne-Mendel rats and
male B6C3Fi mice to evaluate the extent to which radiolabeled compound is metabolized in the
48 hours after administration of 125 or 500 mg/kg to the rats and 250 or 1,000 mg/kg to the
mice. These doses were selected based on the maximum tolerated dose (MTD) and Vi MTD of
HCE; the MTD in rats and mice is 500 mg/kg (2.11 mmol/kg) and 1,000 mg/kg (4.22 mmol/kg),
respectively. Four animals per dose were orally administered unlabeled HCE as a solution in
corn oil 5 days/week for 4 weeks, followed by a single dose of [14C]-radiolabeled HCE. The
48-hour observation period began after administration of the radiolabeled HCE. The animals
were then sacrificed, and urine and feces were collected from the cages. Table 3-4 summarizes
the metabolic disposition data (based on the detection of radiolabel) at the high dose in rats and
mice. The comparable data for the lower doses were not reported.
Table 3-4. Disposition of HCE in male rats and mice during 48 hours
following administration of an MTD for 4 weeks
Rat (500 mg/kg-day)
Mouse (1,000 mg/kg-day)
Percent of administered dose
Expired air
64.55 + 6.67
71.51+5.09
C02
2.37+0.76
1.84 + 0.94
Excreta
6.33+2.39
16.21+3.76
Carcass
20.02 + 3.70
5.90 + 1.60
Recovery
93.28 + 6.23
95.47 + 23.95
Total metabolism (C02 + excreta + carcass)
28.72
23.95
Source: Mitoma et al. (1985).
Recovery of the radiolabel was >90% for both rats and mice. Total metabolism was
calculated by the authors as the sum of the radiolabel present in carbon dioxide, excreta, and the
carcass. This is an assumption by the authors and is not an accurate estimate of metabolism
since actual metabolites were not quantified. Data on the extent of metabolism for the
radiolabeled material are presented in Table 3-5. Based on the mass balance between dose and
the estimate for the sum of the metabolites, 30% of the parent compound was metabolized by
both the rats and mice. This is consistent with the 60-70%> of the high dose that was reported to
be present unchanged in exhaled air. However, this assumes that all of the exhaled radiolabel
that was not identified as carbon dioxide was the unmetabolized parent compound. The major
urinary metabolites, determined qualitatively by high performance liquid chromatography, were
trichloroethanol and trichloroacetic acid (TCA) for both rats and mice. Trichloroethanol and
TCA were also qualitatively considered the major urinary metabolites for other halogenated
hydrocarbon compounds, including PERC, that were evaluated by Mitoma et al. (1985).
10
DRAFT - DO NOT CITE OR QUOTE
-------
Table 3-5. Metabolism of HCE measured in rats and mice
Species
Dose (mmol/kg)
Metabolism (mmol/kg)
Percent metabolized"
Rat
0.53
0.16
30
2.11
0.60
28
Mouse
1.05
0.32
30
4.22
1.01
24
aPercent metabolism was calculated from the dose and the reported sum of the metabolites. This calculation is
likely an underestimation of metabolism since the exhaled air was likely to include some volatile metabolites based
on the data from Jondorf et al. (1957).
Source: Mitoma et al. (1985).
Jondorf et al. (1957) reported that rabbits fed [14C]-radiolabeled HCE at 500 mg/kg (route
of administration not reported by study authors) excreted only 5% of the applied radioactivity in
urine over 3 days (72 hours), indicating slow metabolism. This is consistent with the results in
mice and rats reported by Mitoma et al. (1985) in which approximately 2-4% of the label was
found in urine after 48 hours. During this 3-day period, 14-24% of the radioactivity was
detected in expired air (a lower percentage than seen for rats at a comparable dose by Mitoma et
al., 1985), and the remainder was present in tissues and the intestinal tract. However, the authors
did not have the capability of quantifying HCE in tissues. Reported urinary metabolites include
trichloroethanol (1.3%), dichloroethanol (0.4%), TCA (1.3%), dichloroacetic acid (0.8%),
monochloroacetic acid (0.7%), and oxalic acid (0.1%). The expired air contained HCE, carbon
dioxide, PERC, and 1,1,2,2-tetrachloroethane (TCE was not found). Quantitative data on the
volatile metabolites in exhaled air were not reported.
The only other metabolite data come from the work of Fowler (1969) in sheep. HCE was
administered to four Scottish Blackface and six Cheviot cross sheep at three dose levels: 0 (two
sheep), 500 (six sheep), 750 (one sheep), and 1,000 (one sheep) mg/kg. Two HCE metabolites,
PERC and pentachloroethane, were detected in sheep blood 24 hours after oral HCE
administration by drenching bottle. Following administration of 500 mg/kg, blood
measurements were 10-28 |ag/mL for HCE, 0.6-1.1 |ig/mL for PERC, and 0.06-0.5 |ig/mL for
pentachloroethane. Blood concentrations of HCE, PERC, and pentachloroethane were 2.3-
2.6 times greater than the corresponding concentrations in erythrocytes. Data were not reported
for the 750 and 1,000 mg/kg doses. In vitro experiments using fresh liver slices suspended in an
olive oil emulsion confirmed the presence of the metabolites PERC and pentachloroethane.
The metabolites identified in the in vivo studies (Mitoma et al., 1985; Fowler, 1969;
Jondorf et al., 1957) along with the in vitro studies (Town and Leibman, 1984; Nastainczyk et
al., 1982) and ATSDR (1997a) were used in the derivation of Figure 3-1. The proposed
metabolic pathway is based on limited information; therefore, it is likely that intermediate
11
DRAFT - DO NOT CITE OR QUOTE
-------
chemical reactions are not captured in the figure, which presents the formation of the various
metabolites as single-step reactions.
The in vivo data on HCE metabolism are supported by in vitro studies of hepatic
metabolism using liver microsomes. Nastainczyk et al. (1982, 1981) reported two studies that
provide evidence that HCE is metabolized by phenobarbital-inducible CYP450 isoforms that
catalyze their reductive dechlorination with NADPH, cytochrome bs, and NADH as electron
donors. HCE metabolism was measured using liver microsomes from male Sprague-Dawley rats
that were either pretreated with phenobarbital or 3-methylcholanthrene, or were not pretreated.
Only phenobarbital-induced rat liver microsomes demonstrated an increase in HCE metabolism
(27.0 ±1.1 nmol/mg protein/minute [mean + standard deviation or SD] compared with 8.0 ±
1.2 nmol/mg protein/minute for controls). Oxidation of NADPH (under anaerobic conditions)
with an oxidation rate of 35 ± 2 nmol/mg protein/minute (mean ± SD) provided support for
reductive dehalogenation mediated by CYP450. Carbon monoxide inhibited the NADPH
oxidation rate, further indicating that CYP450 enzymes were involved in the reaction. The major
HCE metabolite of this reductive process was PERC. Nastainzcyk et al. (1982) determined that
the stoichiometry of the reaction was represented by the following equation:
Nastainczyk et al. (1982, 1981) proposed that since CYP450 is a one electron donor, the
two electrons would be transferred sequentially. The first electron reduction would result in a
carbon radical; the second electron reduction would result in a carbanion. From the carbanion,
three possible stabilization reactions are possible: (1) protonation by a hydrogen atom from the
milieu, forming pentachloroethane; (2) a-elimination of chloride to form the carbene, which
could be stabilized by the reduced CYP450; or (3) P-elimination of chloride to form PERC,
which is the major HCE metabolite. Nastainczyk et al. (1982) found that the products of
reductive dechlorination of HCE were 99.5% PERC and 0.5% pentachloroethane at
physiological pHs. At a higher pH (8.4-8.8), the ratio of pentachloroethane (one electron
reduction) to PERC (two electron reduction) increased since transfer of the second electron can
occur via cytochrome b5, which is influenced by pH. These reaction outcomes were proposed by
the authors to also apply to other polyhalogenated hydrocarbons.
To provide additional support for the reaction being catalyzed by CYP450, Nastainczyk
et al. (1982, 1981) inhibited CYP450 using carbon monoxide, metyrapone (CYP450 3A
inhibitor), or a-naphthoflavone (CYP450 1A and CYP450 IB inhibitor) (see Omiecinski et al.,
1999 for review). In vitro metabolism of HCE by phenobarbital-induced rat liver microsomes
was inhibited >99% when carbon monoxide was added to the incubation mixture. Metyrapone at
a concentration of 10"4 M inhibited PERC formation by 46 ± 10% (mean ± SD) and
NADPH + H+ + C13C—CCI3
(HCE)
+ NADP+ + C12C=CC12 + 2 H+ + 2 CI
(PERC)
12 DRAFT - DO NOT CITE OR QUOTE
-------
pentachloroethane formation by 41 ± 8%. Treatment with 10"3 M metyrapone inhibited HCE
metabolism to a greater extent, reducing PERC and pentachloroethane formation 66 + 8 and
79 ± 10%, respectively. a-Naphthoflavone (10"4 M) did not inhibit HCE metabolism as
effectively as metyrapone, inhibiting PERC formation by 13 ± 2% and pentachloroethane
formation by 26 ± 4%. These data indicate that CYP450 3 A isoforms are involved in HCE
metabolism and a-naphthoflavone does not inhibit the primary CYP450 involved in the
metabolism of HCE. Since metyrapone did not completely inhibit HCE metabolism by
phenobarbital-induced liver microsomes, the remainder of HCE metabolism may be accounted
for by the CYP450 2A and 2B subfamilies whose inhibition was not evaluated in this study.
Town and Leibman (1984) prepared liver microsomes from phenobarbital-induced male
Holtzman rats to study the rate of metabolism of HCE to PERC. The formation of PERC was
favored in a low oxygen environment at observed metabolism rates of 50.2 ± 0.45, 1.25 ± 0.25,
and 0 nmol/minute/mg protein in atmospheres of N2, air, and O2, respectively. When any part of
the NADPH-generating system, such as NADP+, glucose 6-phosphate, and glucose 6-phosphate
dehydrogenase, was omitted from the experiment, the metabolism of HCE to PERC was
inhibited (>91%). In addition, the use of carbon monoxide as a monooxygenase inhibitor
arrested HCE metabolism. Enzymes responsible for metabolism of HCE to PERC were located
in the microsomes, rather than the cytosol, of phenobarbital-treated rat livers. Formation of
malondialdehyde and conjugated dienes was statistically, significantly increased following
treatment with HCE (8 mM), indicating lipid peroxidation. The authors suggested the
involvement of a free radical. The Km and Vmax for the enzymatic formation of PERC from HCE
were 1.20 mM and 52.0 nmol/minute/mg, respectively. Phenobarbital-induced liver microsomes
from ICR mice were also studied and yielded Km and Vmax values of 3.34 mM and 30.2 nmol/
minute/mg, respectively. PERC formation was not detected in liver microsomes from
phenobarbital-induced New Zealand White rabbits, suggesting that HCE metabolism resulting in
the formation of PERC did not occur. These results support the hypothesis that rat liver
metabolism of HCE (reductive dehalogenation) occurs by CYP450. The report identifies PERC
as a metabolite of HCE; however, the metabolite was not quantitatively measured.
Salmon et al. (1981) used Aroclor 1254-induced Sprague-Dawley rats to quantify the
dechlorination of HCE. In this case, dechlorination was measured by the release of radioactive
CP from the [36Cl]-radiolabeled HCE substrate during incubation with liver microsomes from
induced rats. The Km and Vmax were determined as 2.37 mM and 0.91 nmol/minute/mg protein,
respectively. A control group of noninduced rats was not included.
Salmon et al. (1985) reported a follow-up study that used liver microsomes from
noninduced rats (Wistar-derived Alderley Park strain) and a reconstituted CYP450 system from
noninduced and phenobarbital-induced New Zealand White rabbits. Metabolic experiments of
HCE using liver microsomes from noninduced rats yielded a Km of 6.0 [xM and a Ymax of
13
DRAFT - DO NOT CITE OR QUOTE
-------
3.55 nmol NADPH/minute/mg protein (2.41 nmol NADPH/minute/nmol CYP450). These
results are not directly comparable to the previous study (Salmon et al., 1981) because of the use
of a different rat strain. A reconstituted CYP450 system from phenobarbital-induced New
Zealand White rabbits yielded Km and Vmax values of 50 |iM and 2.39 nmol NADPH/minute/
nmol CYP450, respectively (Salmon et al., 1985). Microsomes from rabbits induced with
P-naphthoflavone did not metabolize HCE. These results provide further evidence that the
reductive dechlorination of HCE is catalyzed by phenobarbital-inducible CYP450 isoforms.
Yanagita et al. (1997) used recombinantly-expressed rat CYP450 1A2 in Saccharomyces
cerevisiae to evaluate the in vitro metabolism of several chlorinated ethylenes and ethanes,
including HCE. The metabolism of HCE by wild-type CYP450 1A2 under aerobic conditions
resulted in the formation of PERC (3.7 nmol/2.5 nmol CYP450/hour), pentachloroethane
(0.8 nmol/2.5 nmol CYP450/hour), and TCE (0.6 nmol/2.5 nmol CYP450/hour). CYP450 1A2
is a major hepatic CYP450 enzyme, but is not a phenobarbital-inducible isoform; the major
phenobarbital-inducible CYP450 enzymes are the 2A and 2B subfamilies. A follow-up study
(Yanagita et al., 1998) that examined NADPH oxidation rates under anaerobic conditions found
that CYP450 1A2 wild type had a Vmax of 1.3 mol/mol CYP450/minute, a Km of 0.25 mM, and
an NADPH oxidation rate of 1.4 mol/mol CYP450/minute. Product formation rates and relative
ratios of the products formed by metabolism of HCE from the Yanagita et al., (1998) study are
shown in Table 3-6.
Table 3-6. Product formation rates and relative ratios of the products
formed by CYP450 1A2 metabolism of HCE
CYP450 1A2
Product formation (nmol/nmol CYP450/minute)
Ratio of PERC:
pentachloroethane + TCE
PERC
Pentachloroethane
TCE
Wild type
0.68
0.10
0.0034
6.6
Source: Yanagita et al. (1998).
Beurskens et al. (1991) used HCE as a reference compound to examine the metabolism of
three hexachlorocyclohexane isomers. Liver microsomes from male Wistar rats that were
induced with phenobarbital converted HCE to PERC and pentachloroethane at an initial
dechlorination rate of 12 nmol/minute/nmol CYP450 under anaerobic conditions.
Van Dyke (1977) and Van Dyke and Wineman (1971) evaluated the dechlorination
mechanisms of HCE and chlorinated olefins (alkenes) by using rat liver microsomes (a source of
CYP450 enzymes). An initial study with HCE and other chlorinated ethanes found that the
optimal configuration for dechlorination was a dichloromethyl group. HCE demonstrated a
considerable amount of dechlorination (3.9%) in this in vitro study; however, the authors
determined that HCE was unstable in aqueous solution and that this dechlorination was
nonenzymatic based on the evidence of dechlorination in the absence of NADP.
14
DRAFT - DO NOT CITE OR QUOTE
-------
Gargas and Andersen (1989) and Gargas et al. (1988) determined kinetic constants for
HCE metabolism in the rat using exhalation rates and a physiologically based pharmacokinetic
(PBPK) inhalation model described by Ramsey and Andersen (1984) for styrene. The Vmax
(scaled to a 1-kg rat) was 1.97 ± 0.05 mg/hour, or 8.3 [j,mol/hour. The Km was 0.80 mg/L, or
3.38 [xM.
3.4. ELIMINATION
No studies are available that evaluated the elimination of HCE in humans. Animal
studies indicated that the major routes of HCE elimination are either fecal or by expired air
(Mitoma et al., 1985; Fowler, 1969; Jondorf et al., 1957). The sheep studies (Fowler, 1969)
indicate that orally administered HCE is eliminated by the fecal route without absorption and
metabolism while the rodent studies (Mitoma et al., 1985) provided evidence that HCE is
absorbed and eliminated by exhalation. It is unknown why there is a discrepancy between the
studies in sheep and rodents.
Rabbits fed [14C]-radiolabeled HCE at 0.5 g/kg (Jondorf et al., 1957) eliminated 14-24%
of the radioactivity in expired air during a 3-day period following exposure. Only 5% of the
radiolabel was detected in urine. Fecal measurements were not conducted.
Fowler (1969) orally administered HCE to Scottish Blackface and Cheviot cross sheep.
Two Cheviot cross sheep were administered a single dose of 0.5 g/kg HCE and were confined to
metabolism cages; urine and feces were collected over a period of 4 days for HCE analysis.
More than 80% of the total fecal excretion of HCE occurred in the first 24 hours, and only small
amounts were detected in the urine. To assess bile concentrations of HCE, two Scottish
Blackface sheep were fasted for 24 hours and anaesthetized with pentobarbitone sodium. The
hepatic duct was cannulated to collect bile; HCE was injected at a dose of 0.5 g/kg (15% w/v in
olive oil) into the rumen and lower duodenum. Bile was collected continuously, with 2 mL
retained every 30 minutes for analysis. HCE was detected in bile of anaesthetized sheep at
15 minutes, compared with 27 minutes for blood; at maximum, HCE was 8-10-fold greater in
bile.
Mitoma et al. (1985) evaluated excretion of HCE in Osborne-Mendel rats and B6C3Fi
mice following 4 weeks of administration of an MTD (500 mg/kg-day in rats, 1,000 mg/kg-day
in mice). Excretion of radiolabel was monitored for 48 hours following administration of a
tracer dose of [14C]-HCE. The findings are presented in Table 3-4. Most of the radiolabel was
detected in expired air, indicating this to be a major route of elimination. The authors did not
investigate whether the exhaled material was parent compound or volatile metabolite, and
assumed that it was the parent compound. A low percentage of the exhaled radioactivity was in
the form of CO2, with rats exhaling slightly more than mice. The amount of radioactivity in the
excreta, on the other hand, was lower in rats than in mice (Table 3-4). The excreta contained
6.3 and 16.2% of the radiolabel in rats and mice, respectively.
15
DRAFT - DO NOT CITE OR QUOTE
-------
3.5. PHYSIOLOGICALLY BASED PHARMACOKINETIC MODELS
No physiologically based pharmacokinetic (PBPK) models for HCE have been developed
specifically for mammalian species. Models for waterborne chloroethanes have been reported
for rainbow trout and channel catfish; however, these are outside the scope of this toxicological
review and are not described.
Gargas and Andersen (1989) and Gargas et al. (1988) determined kinetic constants for
HCE metabolism in the rat using exhalation rates and a PBPK inhalation model described by
Ramsey and Andersen (1984) for styrene. These reports by Gargas and Andersen (1989) and
Gargas et al. (1988) do not describe a PBPK model for HCE, only kinetic constants for
metabolism by inhalation. During these breath chamber experiments, fur deposition (fur
loading) was observed to occur. At an exposure concentration of 53.3 ppm HCE at 6 hours, the
chemical mass in body tissues was 7.29 mg and the chemical mass on fur was 0.6 mg (7.6% of
total chemical mass).
16
DRAFT - DO NOT CITE OR QUOTE
-------
4. HAZARD IDENTIFICATION
4.1. STUDIES IN HUMANS—EPIDEMIOLOGY, CASE REPORTS, CLINICAL
CONTROLS
There are few published studies relating to the toxicology of HCE in humans. Case
reports of pneumonitis (Allen et al., 1992) and pneumonitis with evidence of liver abnormalities
(Loh et al., 2008, 2006) have been described in soldiers exposed to smoke bombs containing
HCE and zinc oxide. However, the smoke produced by this incineration is primarily zinc
oxychloride and zinc chloride and it is not likely that these effects are a result of HCE. Some
aluminum production processes involve the use of HCE in tablet or powder form, resulting in
exposures to fumes containing hexachlorobenzene, octachlorostyrene, dioxins, dibenzofurans,
and other organochlorinated compounds. A case report of a hepatocellular carcinoma (Selden et
al., 1989) and limited data concerning some clinical serologic measures (Selden et al., 1999,
1997) in aluminum foundry workers involved in this process are available, but these data are not
directly relevant to the question of health effects of HCE in other settings. No epidemiologic
studies of the carcinogenicity of HCE were included in a 1985 review of cancer epidemiology
with respect to halogenated alkanes and alkenes (Axelson, 1985). A study of Swedish workers
involved in smoke bomb production has provided some information pertaining to exposure levels
and symptoms and clinical parameters relating primarily to liver and pulmonary function (Selden
et al., 1994, 1993).
Two separate studies were conducted on a small population of Swedish workers
occupationally exposed to HCE while producing military white smoke munitions. The first
study reported on biological exposure monitoring (Selden et al., 1993) and the second study
described health effects resulting from HCE exposure (Selden et al., 1994). The smoke
formulation was approximately 60% HCE, 30% titanium dioxide, 8% aluminum powder, 2%
cryolite, and a trace of zinc stearate. At the time this study was conducted in 1989, no HCE dust
was found in the air sample filters, but the integrated results of personal and stationary charcoal
tube samples revealed approximate HCE concentrations by location of 10-30 mg/m3 (milling/
mixing), 5-25 mg/m3 (pressing), <5 mg/m3 (assembly room), and nondetectable (storage room)
(Selden et al., 1993).
In the first study (Selden et al., 1993), the exposed group consisted of 12 people (six men
and six women) ranging in age from 23 to 57 (mean, 31.4 years; median, 30 years) (Selden et al.,
1993). The principal control group (n = 12) consisted of assembly line workers from the same
company who were unexposed to chlorinated hydrocarbons, but had some exposure to glass fiber
dust. They were matched to the exposure group by sex and age (± 5 years), except in the case of
one exposed male subject where only a younger control could be found. This latter-exposed
male subject was excluded from the analysis of health effects (Selden et al., 1994). A second
17
DRAFT - DO NOT CITE OR QUOTE
-------
control group of formerly HCE-exposed workers (3 males, 10 females; age range, 31-57 years;
mean, 43.6 years) was used in the biological exposure monitoring study.
Blood samples were collected for analysis of HCE concentration. For the exposed group,
samples were drawn 5 weeks into a temporary production break (the "baseline" period), and the
second samples were drawn 5 months later, after production had been underway for 5 weeks (the
"production" period). Analyses of blood plasma HCE indicated that all values for both control
groups (n = 25) were below the limit of detection (<0.02 [ig/L).
Exposed subjects were stratified into three subgroups (n = 4) of perceived exposure (low,
medium, or high) based on information pertaining to work tasks, presence at work, and use of
protective equipment. At baseline, the HCE concentrations in 10 of the samples from exposed
workers were in the range of <0.02-0.06 |ig/L, 1 sample was 0.15 (J,g/L, and 1 sample was
0.52 (J,g/L. The last sample was from an individual who had remained in an HCE-contaminated
area during the baseline period. Plasma HCE levels in the production period increased by nearly
100-fold over that of the baseline samples (mean of 7.30 ± 6.04 [^g/L in the production sample
compared with 0.08 ± 0.14 [j,g/L in the baseline samples, p < 0.01). Although the magnitude of
individual increases varied considerably, there was a significant (p < 0.05) linear trend for values
in the low-, medium-, and high-exposure subgroups (means of 3.99, 7.14, and 10.75 (J,g/L,
respectively). These results demonstrate that a considerable increase in plasma HCE can occur
after a relatively brief occupational exposure, even though workers used fairly sophisticated
personal protective equipment.
As noted above, 11 of the subjects from the first study (Selden et al., 1993) and their
11 age- and sex-matched controls were included in the second health effects study (Selden et al.,
1994). Data pertaining to 15 clinical symptoms (including headaches, sleep quality, palpations,
difficulty concentrating, tension/restlessness, frequency of coughing, watery eyes/runny nose,
itching/other skin problems, shortness of breath/chest discomfort, and general health) were
obtained from self-administered questionnaires for the exposed workers and the company
controls. Similar data had been obtained in a previous study of 130 metal shop workers, and
these workers were used as a second, "historical" comparison group in the analysis of the
symptom data. Whole blood and serum samples from the 11 exposed and 11 matched company
controls were analyzed for routine clinical parameters. Spot urine samples were analyzed for
hemoglobin, protein, and glucose. Lung function was assessed by measuring vital capacity and
1-second forced expiratory volume (FEVi).
The matched company controls reported more symptoms of ill health than exposed
subjects, although the differences were not statistically significant. Although not statistically
significant, the exposed group reported a higher prevalence of "dry skin/dry mucous
membranes" (3/11 or 27%) than the matched controls (1/9, 9%) or historical controls (13/130,
10%), and a higher prevalence of "itching/other skin problems" (3/11, 27%) than the historical
controls (16/130, 12%). The prevalence of "itching/other skin problems" in the matched controls
18
DRAFT - DO NOT CITE OR QUOTE
-------
(3/11, 27%) was the same as that in the exposed group. These symptoms centered on the wrist
and neck areas, and the authors suggested that this could reflect exposure to HCE through joints
in the protective equipment, or could possibly be a "traumiterative effect of the equipment
itself." Clinical examination revealed no dermatological or respiratory mucous membrane
abnormalities in either group. The authors noted that a previous unpublished study of the plant
workers (but with primarily different workers) had also found dermatologic complaints in up to
90% of the exposed workers.
All of the spot urine tests were normal, and there was no evidence of an effect of HCE
exposure on pulmonary function as measured by vital capacity and FEVi. Exposed subjects had
significantly higher levels of serum creatinine, urate, and bilirubin than controls (p < 0.05),
although the group means were still in the normal range. One exposed subject had a marginally
elevated level of serum alanine aminotransferase (ALT) (70.5 U/L versus <41.1 U/L reference),
while one control subject displayed increased levels of serum ALT and aspartate
aminotransferase (AST) (67.6 and 186.4 U/L, respectively; 41.1 U/L reference for each). The
control individual's values returned to normal after 8 months, while the exposed subject's serum
ALT value worsened to 87.6 U/L 4 months later (Selden et al., 1994). Available data pertaining
to these liver function tests from 1982, when exposure levels at the worksite were higher than in
the current study, did not show elevations in these liver enzymes in this individual at that time.
Within the exposed group, there was no correlation between plasma HCE concentrations and the
clinical chemistry parameters, although the authors do not discuss the power limitations of this
exposure-response analysis (Selden et al., 1993). In summary, these studies demonstrated HCE
exposure in the smoke bomb production workers, but the health effects study is too small to
reach definitive conclusions. The interpretation of small differences in clinical parameters,
within the normal range, is uncertain. Based on the available data, the possible
dermatologic/mucosal effects and hepatic effects are the areas in most need of additional
research.
4.2. SUBCHRONIC AND CHRONIC STUDIES AND CANCER BIOASSAYS IN
ANIMALS—ORAL AND INHALATION
4.2.1. Oral
4.2.1.1. Subchronic Exposure
Two subchronic toxicity assays for HCE were reported (NTP, 1989; Gorzinski et al.,
1985, 1980). The Gorzinski et al. (1985, 1980) study (16 weeks) reported histopathological
evaluations that found kidney degeneration in males, kidney degeneration in females, and
minimal hepatic effects. The NTP (1989) study (13 weeks) reported kidney effects in male rats
such as degeneration and necrosis of renal tubular epithelium, hyaline droplet formation, and
tubular regeneration and tubular casts. Female rats in this study exhibited a dose-response
increase in the incidence of hepatocellular necrosis of the centrilobular area. The NTP (1989)
19
DRAFT - DO NOT CITE OR QUOTE
-------
study suggested that male rats may be more susceptible to kidney effects, whereas female rats
may be more susceptible to liver effects.
Gorzinski et al. (1980) conducted a 16-week toxicity study in male and female F344 rats.
Ten rats/sex/dose were exposed via the diet, formulated to deliver doses of 3, 30, or
100 mg HCE/kg-day (purity 99.4%). However, due to sublimation of HCE from the feed, the
actual doses were reported as 1.3, 20, or 82 mg/kg-day and later, based on feeding and diurnal
eating patterns, were determined to be 1, 15, or 62 mg/kg-day, respectively (Gorzinski et al.,
1985). Gorzinski et al. (1980) is a Research and Development Report by Dow Chemical and is
not publicly available. The data for this study were published in the peer-reviewed literature by
Gorzinski et al. (1985) and are presented in detail below.
Gorzinski et al. (1985) fed 1, 15, or 62 mg/kg-day HCE (purity 99.4%) to F344 rats
(10 rats/sex/dose) for 16 weeks. As described in Section 3.2, HCE concentrations in male
kidneys were proportionately increased with administered dose, while the increases in females
were not proportionate. At the high dose, male rats displayed statistically significant increases in
absolute and relative kidney weights accompanied by macroscopically observed alterations.
Male rats displayed slight hypertrophy and/or dilation of proximal convoluted tubules of the
kidneys at incidences of 0/10, 1/10, 7/10, and 10/10 for the 0, 1, 15, and 62 mg/kg-day dose
groups, respectively. The increased incidence of slight hypertrophy and/or dilation of proximal
convoluted tubules was statistically significant in males at the 15 and 62 mg/kg-day doses. Male
rats displayed atrophy and degeneration of renal tubules at incidences of 1/10, 2/10, 7/10, and
10/10 for the 0, 1, 15, and 62 mg/kg-day dose groups, respectively. The increased incidence of
atrophy and degeneration of renal tubules was statistically significant in males at the 15 and
62 mg/kg-day doses. Female rats did not display hypertrophy and/or dilation of proximal
convoluted tubules of the kidneys at any dose, but did exhibit atrophy and degeneration of
proximal tubules (1/10, 1/10, 2/10, and 6/10 at the 0, 1, 15, and 62 mg/kg-day doses,
respectively). However, the increased incidence of atrophy and degeneration of proximal tubules
was only statistically significant in females at the 62 mg/kg-day dose. Male rats of the
62 mg/kg-day group exhibited statistically significant increases in absolute and relative liver
weights; histopathology revealed a slight swelling of the hepatocytes in this group. Although
female rats exhibited a statistically significant increase in relative liver weight at the high dose,
there was no evidence of hepatotoxicity in the histopathological examination. The data for liver
and kidney weights are presented in Table 4-1 and the data for the kidney effects are presented in
Table 4-2.
20
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-1. Body, kidney, and liver weights of rats exposed to HCE in the
diet for 16 weeks
Sex
Dose level
(mg/kg-day)
Fasted body
weight (g)
Liver
Kidney
Absolute (g)
Relative (g/100 g
body weight)
Absolute (g)
Relative (g/100 g
body weight)
Male3
0
314.4 + 12.4
8.32 + 0.27
2.65+0.06
2.28+0.08
0.73+0.04
1
328.0 + 7.2
8.46 + 0.22
2.58+0.07
2.31+0.09
0.70+0.02
15
329.0 +24.4
8.69 + 0.80
2.64+0.09
2.40+0.15
0.73+0.01
62
324.2 + 10.0
8.98 + 0.54b
2.77 + 0.12b
2.51 +0.12b
0.77 + 0.02b
Female3
0
176.7 + 6.9
4.65+0.26
2.63+0.06
1.40 + 0.08
0.79+0.03
1
174.0 + 7.9
4.74 + 0.22
2.73+0.11
1.38+0.05
0.79+0.03
15
176.7 + 4.6
4.79 + 0.21
2.69+0.09
1.39+0.06
0.79+0.04
62
170.8 + 5.1
4.71+0.23
2.76 + 0.10b
1.39+0.05
0.81+0.02
aData are presented as means + SD of 10 rats/sex.
Statistically significant from control using Dunnett's test (p = 0.05).
Source: Gorzinski et al. (1985).
Table 4-2. Histopathological results on kidney in rats exposed to HCE in the
diet for 16 weeks3
Organ
Effect
Sex
Dose (mg/kg-day)
0
1
15
62
Kidney
Slight hypertrophy and/or dilation of proximal
convoluted tubules
Male
0
1
7b
10b
Female
0
0
0
0
Atrophy and degeneration of renal tubules0
Male
1
2
7b
10b
Female
1
1
2
6b
aData are presented as number of positive observations for 10 rats/sex/dose.
bEPA determined statistical significance from control using Fisher's Exact Test (p = 0.05).
°Graded as slight in 1 of 10 male control rats and very slight in 1 of 10 control female rats. Severity of nephropathy
was not reported for HCE-exposed rats.
Source: Gorzinski et al. (1985).
The authors concluded that the no-observed-effect level for both male and female rats
was 1 mg/kg-day. EPA considered 1 mg/kg-day as the male no-observed-adverse-effect level
(NOAEL) and 15 mg/kg-day as the lowest-observed-adverse-effect level (LOAEL), based on
renal tubule toxicity in male rats. For female rats, EPA considered the NOAEL as 15 mg/kg-day
and the LOAEL as 62 mg/kg-day, based on renal tubule toxicity.
NTP (1989) conducted a 13-week study of HCE oral toxicity in F344/N rats. Groups of
10 rats/sex/dose were administered 0, 47, 94, 188, 375, or 750 mg/kg (purity >99%) by corn oil
gavage, 5 days/week for 13 weeks. The time-weighted average (TWA) doses were 0, 34, 67,
134, 268, and 536 mg/kg-day, respectively. In the 536 mg/kg-day group, 5/10 male rats (only
21
DRAFT - DO NOT CITE OR QUOTE
-------
the 5 males that died were examined microscopically) and 2/10 female rats died before the end of
the study. Mean body weights of 536 mg/kg-day male and female rats were decreased 19 and
4%, respectively, compared with controls. Statistically significant increases in liver weights
were noted at doses of >67 mg/kg-day (females) and >134 mg/kg-day (males), and in kidney
weights at doses of >268 mg/kg-day (females) and >67 mg/kg-day (males). Organ weight to
body weight ratios (mg/g) generally increased in a dose-related manner for both male and female
rats exposed to HCE (Table 4-3).
Table 4-3. Organ weight to body weight ratios for rats exposed to HCE for
13 weeks
HCE dose by gavage (mg/kg-day)
0
34
67
134
268
536
Male3
Number
10
10
10
10
9
5
Body weight
340 +7.6
349 + 8.8
343 + 5.9
348 + 5.9
319 + 4.0
262 + 13.5
Liver
35.8 + 0.61
37.3+0.37
36.0 + 0.71
39.1 +0.62b
42.5 +0.74b
46.3 +0.95b
Brain
6.0+0.30
5.7 + 0.17
5.7+0.10
5.8 + 0.23
6.3+0.21
7.2 + 0.3 lb
Heart
2.8+0.04
2.8 + 0.04
2.9+0.07
3.2 + 0.17°
3.3 +0.18b
3.2 + 0.10°
Kidney
3.0+0.05
3.8 + 0.37
4.1 +0.27c
4.7 + 0.44b
5.2 + 0.35b
4.7 + 0.28b
Lung
4.2+0.21
4.6 + 0.40
4.4+0.48
3.9 + 0.22
3.9+0.15
4.9 + 0.50
Right testis
4.2+0.05
4.8 + 0.38
4.3+0.10
4.4 + 0.17
4.7+0.05
5.3 +0.21b
Thymus
0.8+0.04
0.8 + 0.06
0.6+0.02
0.8 + 0.10
0.7+0.04
0.6 + 0.06
Female"
Number
10
10
10
10
10
8
Body weight
206 + 3.7
210 + 3.9
208 + 2.6
200 + 2.9
203 +4.3
189 + 3.8
Liver
32.2 + 0.56
33.4+0.63
34.3 +0.39c
36.3 +0.44b
42.0 + 0.60b
52.4 + 0.88b
Brain
8.7+0.17
8.6 + 0.14
8.6+0.10
9.0 + 0.14
9.0+0.15
9.5 +0.17b
Heart
2.9+0.04
3.0 + 0.05
3.0+0.03
3.0 + 0.04
3.1+0.07
3.4 + 0.07b
Kidney
3.1+0.04
3.2 + 0.05
3.2+0.07
3.2 + 0.06
3.6 + 0.05b
4.1 +0.10b
Lung
4.2+0.09
4.1+0.09
4.2+0.10
4.1+0.06
4.2+0.08
4.5 + 0.13
Thymus
1.1+0.05
1.1+0.05
1.1+0.04
1.0 + 0.06
1.1+0.07
0.8 + 0.05b
aData are presented as mean + SE in mg/g, except for body weight in grams.
Statistically different from controls, p < 0,01
Statistically different from controls, p < 0.05
Source: NTP (1989).
Kidney effects (characterized by hyaline droplet formation, tubular regeneration, and
tubular casts), similar to the toxicity noted in the 16-day study also conducted by NTP (1989),
were observed in 90% of 34 mg/kg-day males and in males from all other HCE dose groups
(incidence data only reported for the 34 mg/kg-day dose group). NTP (1989) reported that the
22
DRAFT - DO NOT CITE OR QUOTE
-------
severity of these effects increased with dose (data not presented by NTP). These kidney effects
were not observed in any of the treated females. At the 536 mg/kg-day dose, 5/10 males died.
Kidneys from these five animals were examined microscopically and revealed papillary necrosis,
degeneration, and necrosis of the renal tubular epithelium. Hepatocellular necrosis of the
centrilobular area was observed in 2/5 males and 8/10 females at the 536 mg/kg dose, 1/10 males
and 4/10 females at the 268 mg/kg-day dose, and 2/10 females at the 134 mg/kg-day dose.
Additionally, males of the 536 mg/kg-day dose group exhibited hemorrhagic necrosis of the
urinary bladder. EPA considered the female rat NOAEL as 67 mg/kg-day and the LOAEL as
134 mg/kg-day, based on hepatocellular necrosis. A NOAEL could not be identified for male
rats since kidney effects were observed in >90% of the male rats at all tested doses (compared to
none of the controls). EPA considered the LOAEL for male rats as 34 mg/kg-day (lowest dose
tested), based on kidney lesions.
4.2.1.2. Chronic Exposure and Carcinogenicity
The National Toxicology Program (NTP) and National Cancer Institute (NCI) conducted
two chronic toxicity/carcinogenicity bioassays in rats and one in mice. Increased incidences of
renal tubular hyperplasia, renal adenoma or carcinoma, adrenal medulla hyperplasia,
pheochromocytomas, and malignant pheochromocytomas were noted in male F344/N rats;
female rats did not develop HCE-related tumors (NTP, 1989). Osborne-Mendel rats of both
sexes in the NCI (1978) study exhibited tumor types that have been previously identified as
spontaneous lesions in this strain, and do not provide evidence of carcinogenicity. B6C3Fi mice
of both sexes exhibited hepatocellular carcinomas, although only male mice demonstrated a dose
response with tumor incidence (NCI, 1978). Based on the body of evidence accumulated by
these studies, NTP and NCI concluded that there was evidence of HCE carcinogenicity in male
F344 rats and mice of both sexes, respectively, but there was no evidence of carcinogenicity in
female F344 or male and female Osborne-Mendel rats (NTP, 1989; NCI, 1978).
NTP (1989) conducted a chronic toxicity/carcinogenicity bioassay in F344/N rats.
Groups of 50 male rats/dose were administered 0, 10, or 20 mg/kg-day (TWA doses of 0, 7, or
14 mg/kg-day, respectively, after adjusting for continuous exposure) of HCE (purity >99%) by
corn oil gavage, 5 days/week for 103 weeks. Groups of 50 female rats/dose were administered 0,
80, or 160 mg HCE/kg by corn oil gavage, 5 days/week for 103 weeks (TWA doses of 0, 57, or
114 mg/kg-day, respectively, after adjusting for continuous exposure). These sex-specific doses
were selected based on the results of the 13-week study conducted by NTP (1989) that
demonstrated kidney lesions in male rats at the lower doses and liver lesions in female rats at the
higher doses. All animals were necropsied.
Mean body weights of the 14 mg/kg-day male rats were 5—6% lower than controls after
week 81. Mean body weights of the 114 mg/kg-day female rats were 5—9% lower between
weeks 41 and 101. Nephropathy, characterized by tubular cell degeneration and regeneration,
23
DRAFT - DO NOT CITE OR QUOTE
-------
tubular dilatation and atrophy, glomerulosclerosis, interstitial fibrosis, and chronic inflammation,
was observed in both treated and control rats. Incidences of male nephropathy were 48/50 in
controls, 48/50 in the 7 mg/kg-day dose group, and 47/50 in the 14 mg/kg-day dose group. The
mean severity scores for nephropathy in male rats increased with dose (2.34 ± 0.14, 2.62 ±0.15,
and 2.68 ± 0.16 in the 0, 7, and 14 mg/kg-day groups, respectively), with the 14 mg/kg-day
group being statistically significantly higher than the control group. While the mean severity
scores did not show more than a 15% increase over control in the high-dose group, an
examination of the various grades of severity revealed more moderate and marked nephrotoxicity
in treated male rats compared with controls, which predominantly exhibited mild nephropathy
(Table 4-4).
Incidences of female nephropathy were 22/50 for controls, 42/50 in the 57 mg/kg-day
dose group, and 44/49 in the 114 mg/kg-day dose group. The severity scores for nephropathy in
female rats were statistically significantly increased in both treated groups: 0.72 ±0.13 (mean ±
SE) in controls, 1.38 ± 0.11 in the 57 mg/kg-day group, and 1.69 ± 0.12 in the 114 mg/kg-day
group. Examination of the various grades of severity showed mild and moderate nephropathy in
treated females compared with controls, which predominantly presented less than minimally
severe nephropathy. Females did not exhibit marked nephropathy in the control or treated
groups (Table 4-4).
Table 4-4. Incidence and severity of nephropathy in male and female rats
treated with HCE
Dose (mg/kg-day)
Severity
0
7
14
0
57
114
Male
Female
None (0)
2
2
3
28
8
5
Minimal (1)
4
3
4
10
17
12
Mild (2)
26
21
13
10
23
25
Moderate (3)
11
10
16
2
2
7
Marked (4)
7
14
14
0
0
0
Total incidence (minimal to marked)
48
48
47
22
42b
44b
Total number of rats
50
50
50
50
50
49
Overall severity0
2.34 + 0.14
2.62+0.15
2.68 + 0.16a
0.72 + 0.13
1.38 + 0.1 lb
1.69 + 0.12b
aAuthors reported as statistically significantly different from controls, p < 0,05.
bAuthors reported as statistically significantly different from controls, p < 0.01,
°Mean + SE.
Source: NTP(1989).
In light of these variations in severity, EPA considered the responses observed in both the
control and treated male rats associated with more severe (moderate and marked severity)
24
DRAFT - DO NOT CITE OR QUOTE
-------
nephropathy to better distinguish the HCE-related effects. Incidences of male nephropathy (that
were of moderate or marked severity) were 18/50, 24/50, and 30/50 in the control, 7, and
14 mg/kg-day dose groups, respectively. Similar to the male rats, the incidence of nephropathy
associated with the more severe (mild and moderate) responses was considered in the females
rats. Therefore, incidences of female nephropathy (that were of mild or moderate severity) were
12/50, 25/50, and 32/50 in the control, 57, and 114 mg/kg-day dose groups, respectively.
Additional kidney effects were noted in male rats (presented in Table 4-5). Linear
mineralization of the renal papillae was increased in a dose-dependent manner: 15/50 (30%) and
32/50 (64%) in the 7 and 14 mg/kg-day dose groups, respectively, compared with 2/50 (4%) in
controls. Hyperplasia of the pelvic transitional epithelium was increased in treated rats (14% in
7 and 14 mg/kg-day HCE dose groups) compared to 0% of control rats. Nonneoplastic lesions
such as casts (4%), cytomegaly (4%), chronic inflammation (4%), and focal necrosis (2%) were
observed in some of the male rats administered 14 mg/kg-day. An increased incidence of renal
tubule pigmentation was noted in 4/50 (8%) of the 7 mg/kg-day dose group and 5/50 (10%) of
the 14 mg/kg-day dose group, compared with 1/50 (2%) in the controls. Regeneration of the
renal tubule was observed in three males administered 14 mg/kg-day HCE.
Additional kidney effects in female rats included linear mineralization of the renal
papillae, although the incidence was not dose-dependent: 14/50 (28%) in vehicle controls,
22/50 (44%) in the 57 mg/kg-day dose, and 13/50 (26%) in the 114 mg/kg-day dose. Female rats
also exhibited casts (4% at 114 mg/kg-day) and chronic inflammation (2% at both 57 and
114 mg/kg-day). Pigmentation of the renal tubule was present in 4, 4, and 6% of control, 57, and
114 mg/kg-day females, respectively. Renal tubule regeneration was observed in treated females
(but not controls); 4% of the 57 mg/kg-day dose group and 2% of the 114 mg/kg-day dose group.
Only male rats demonstrated an increase in hyperplasia of the pelvic transitional epithelium and
a dose-dependent increase in incidences of mineralization along the renal papillae.
25
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-5. Additional kidney effects in HCE-treated rats
HCE Dose (mg/kg-day)
Males
Females
Vehicle
control
7
14
Vehicle
control
57
114
Renal tubule pigmentation
1/50 (2%)
4/50 (8%)
5/50 (10%)
2/50 (4%)
2/50 (4%)
3/50 (6%)
Linear mineralization of renal papillae
2/50 (4%)
15/50
(30 %)a
32/50
(64%)a
14/50 (28%)
22/50 (44%)
13/50
(26%)
Hyperplasia of the pelvic transitional
epithelium
0/50 (0%)
7/50
(14%)a
7/50
(14%)a
Not
observed
Not
observed
Not
observed
"EPA determined statistical significance using Fisher's exact test, p < 0.05.
Source: NTP(1989).
EPA considered the male LOAEL as 7 mg/kg-day based on increased incidence of
moderate or marked nephropathy (Table 4-4), hyperplasia of the pelvic transitional epithelium
(Table 4-5), increased incidence of renal tubule pigmentation (Table 4-5), and linear
mineralization of the renal papillae (Table 4-5). EPA considered 57 mg/kg-day the female
LOAEL, based on dose-related increases in incidence and severity (minimal to moderate)
nephropathy. The male and female NOAELs could not be established as toxic effects were
observed at the lowest doses tested.
Renal tubular hyperplasia was observed at an increased incidence in treated male rats:
4/50 (8%) in the 7 mg/kg-day dose and 11/50 (22%; statistically significantly higher than
controls) in the 14 mg/kg-day dose, compared with 2/50 (4%) for control (Table 4-6). Only one
female rat, administered 57 mg/kg-day, exhibited renal hyperplasia. Dose-related increases in
the incidence of combined renal adenomas and carcinomas were observed in males rats
administered HCE at doses of 7 (4%) and 14 mg/kg-day (14%, statistically significantly higher
than controls) compared with controls (2%). No HCE-related tumors were observed in female
rats. NTP concluded that these data provided evidence of carcinogenicity in male rats based on a
comparison with the historical controls in the study laboratory (1/300; 0.3 ± 0.8%) and in NTP
studies (10/1,943; 0.5 ± 0.9%).
Table 4-6. Renal tubular hyperplasia and tumor incidences in HCE-treated
male rats
Vehicle control
7 mg/kg-day HCE
14 mg/kg-day HCE
Hyperplasia
2/50 (4%)
4/50 (8%)
11/50 (22%)a
Adenoma
1/50 (2%)
2/50 (4%)
4/50 (8%)
Carcinoma
0/50 (0%)
0/50 (0%)
3/50 (6%)
Adenoma or carcinoma
1/50 (2%)
2/50 (4%)
7/50 (14%)a
aSignificantly different from vehicle controls, p < 0,01,
Source: NTP (1989).
26
DRAFT - DO NOT CITE OR QUOTE
-------
This study demonstrates specificity for HCE-induced renal effects in male rats. The
males of both dose groups were administered 8 times less HCE than the corresponding females.
However, the treated male rats demonstrated more severe nephropathy than the treated female
rats. NTP (1989) also observed more severe nephropathy in control male rats (i.e., mild
nephropathy) than in control females (i.e., miminal nephropathy). Male rats, but not female rats,
also exhibited renal hyperplasia and tumors. NTP (1989) indicated that the renal hyperplasia and
tumors observed in the HCE-treated male rats represented a morphologic continuum.
Effects in the adrenal gland were also noted in HCE-treated rats. Hyperplasia of the
adrenal medulla was reported in 9 and 20% of male rats administered 7 and 14 mg/kg-day HCE,
respectively, compared with 12% of controls. Female rats in the control (10%) and
114 mg/kg-day (15%) groups exhibited hyperplasia of the adrenal medulla; this effect was not
observed in the 57 mg/kg-day dose group.
Adrenal medullary lesions were observed in male rats, but not female rats (Table 4-7).
Pheochromocytoma incidences were statistically significantly increased in the 7 mg/kg-day
group (26/45, 58%). The increase of pheochromocytomas in the 14 mg/kg-day group (19/49,
39%) was not statistically significant compared with controls (14/50, 28%). There were no
statistically significant differences in the incidences of malignant pheochromocytomas and
complex pheochromocytomas (defined as pheochromocytomas containing nervous tissue in
addition to the typical adrenal medullary cells) between controls and treated male rats. The
combined incidence of all three types of pheochromocytomas was statistically significantly
increased in males treated with 7 mg/kg-day HCE (62%) but not in males treated with
14 mg/kg-day HCE (43%) when compared with vehicle controls (30%) and historical controls in
the study laboratory (75/300; 25 ± 7%) and in NTP studies (543/1,937; 28 ± 11%). NTP
concluded that the increased incidences of pheochromocytomas in male rats were possibly
treatment-rel ated.
Table 4-7. Adrenal medullary lesions in HCE-treated male rats
Control
7 mg/kg-day
14 mg/kg-day
Focal hyperplasia
6/50 (12%)
4/45 (9%)
10/49 (20%)
Pheochromocytoma
14/50 (28%)
26/45 (58%)a
19/49 (39%)
Complex pheochromocytoma
0/50
0/45
2/49 (4%)
Malignant pheochromocytoma
1/50 (2%)
2/45 (4%)
1/49 (2%)
Combined pheochromocytoma
15/50 (30%)
28/45 (62%)a
21/49 (43%)
aSignificantly different from vehicle controls, p < 0,01,
Source: NTP (1989).
27
DRAFT - DO NOT CITE OR QUOTE
-------
NCI (1978; Weisburger, 1977) conducted a chronic toxicity/carcinogenicity bioassay in
Osborne-Mendel rats. HCE (purity >98%) at doses of 0, 250, or 500 mg/kg-day was
administered by corn oil gavage to 50 rats/sex/dose for 5 days/week for 78 weeks. Following
termination of exposure, animals were observed for 33-34 weeks for a total duration of 111-
112 weeks. Twenty rats/sex were used for the untreated and vehicle controls. Starting in
week 23, rats treated began a 5-week cyclic rotation that involved 1 week without exposure
followed by dosing for 4 weeks. After adjustment from 5 days/week for 78 weeks, with the
5-week cyclic rotation for part of the time, to continuous exposure over the standard 2 years for a
chronic bioassay, the TWA doses were 113 and 227 mg/kg-day.
Mortality was accelerated in the HCE-treated rats (NCI reported a statistically significant
association between increased dose and mortality). The 113 and 227 mg/kg-day males exhibited
survival rates of 24/50 (48%) and 19/50 (38%), respectively, compared with 14/20 (70%) in the
untreated controls and 11/20 (55%) in vehicle controls (seven rats in the vehicle control group
were sacrificed in week 60). Mortality in the treated groups occurred early in the bioassay.
Approximately 20% of the high- and low-dose males died by weeks 15 and 45, respectively,
compared with 90 weeks until 20% mortality for the controls. Survival rates for the female rats
were 14/20 {10%) for both the untreated and vehicle controls, and 27/50 (54%) and 24/50 (48%)
for the 113 and 227 mg/kg-day dose groups, respectively. Mortality also occurred early in the
bioassay for the female rats. Approximately 20% of the high- and low-dose females died by
weeks 25 and 30, respectively, compared with 110 weeks until 20% mortality for the controls.
Chronic inflammatory kidney lesions were observed in both control and treated rats:
male rats exhibited incidences of 15/20 {15%) in untreated controls, 14/20 {10%) in vehicle
controls, 32/49 (65%) in the 113 mg/kg-day dose group, and 25/50 (50%) in the 227 mg/kg-day
dose group; female rats exhibited incidences of 8/20 (40%) in untreated controls, 4/20 (20%) in
vehicle controls, 18/50 (36%) in the 113 mg/kg-day dose group, and 20/49 (41%) in the
227 mg/kg-day dose group. Tubular nephropathy (characterized by degeneration, necrosis, and
the presence of large hyperchromatic regenerative epithelial cells) was observed in 45 and 66%
of males and 18 and 59% of females in the 113 and 227 mg/kg-day dose groups, respectively.
These effects were not observed in the untreated or vehicle controls. EPA considered the
LOAEL as 113 mg/kg-day (lowest dose tested), based on a dose-related increase in the incidence
of nephropathy in both males and females. The NOAEL could not be identified.
Tumor types exhibited by male rats surviving at least 52 weeks included kidney tubular
cell adenoma, pituitary chromophobe adenoma, thyroid follicular cell adenoma or carcinoma,
and testicular interstitial cell tumors (Table 4-8). Due to the high mortality in the 227 mg/kg-day
males, statistical analyses of male rat tumors were based only on those rats surviving at least
52 weeks. Increased incidences of kidney tubular cell adenoma (4/37) and pituitary
chromophobe adenoma (4/32) were observed in the male rats of the 113 mg/kg-day dose group
but not in the 227 mg/kg-day group. Male vehicle controls did not exhibit kidney tubular cell
28
DRAFT - DO NOT CITE OR QUOTE
-------
adenomas, although 11% (2/18) exhibited pituitary chromophobe adenomas. Thyroid follicular
cell adenoma or carcinoma were observed in 11, 8, and 18% in vehicle control, 113, and
227 mg/kg-day males, respectively; high-dose males also demonstrated the shortest time to first
tumor of 60 weeks, compared with vehicle control (111 weeks) and low-dose males (92 weeks).
Testicular interstitial cell tumors were not observed in vehicle control or 113 mg/kg-day males,
but were observed in 10% of 227 mg/kg-day males.
Table 4-8. Tumor incidences3 in male rats gavaged with HCE
Tumor type
Vehicle control
113 mg/kg-day
227 mg/kg-day
Kidney tubular cell adenoma
0/18 (0%)
4/37(11%)
0/29 (0%)
Weeks to first tumor
-
86
-
Pituitary chromophobe adenoma
2/18 (11%)
4/32 (13%)
0/24 (0%)
Weeks to first tumor
105
104
-
Thyroid follicular cell adenoma or carcinoma
2/18 (11%)
3/36 (8%)
5/28 (18%)
Weeks to first tumor
111
92
60
Testis interstitial cell tumor
0/18 (0%)
0/36 (0%)
3/29 (10%)
Weeks to first tumor
-
-
109
aDue to early accelerated mortality, the statistical analyses for the incidences of tumors are based on animals
surviving at least 52 weeks.
Source: NCI (1978).
Tumor types exhibited by female rats included kidney hamartoma (nonneoplastic
overgrowth), pituitary chromophobe adenoma, thyroid follicular cell adenoma or carcinoma,
mammary gland fibroadenoma, and ovary granulose cell tumors (Table 4-9). Females
administered 227 mg/kg-day HCE had an incidence of 6% for kidney hamartoma, while none of
these tumors were observed in the vehicle control or 113 mg/kg-day female rats. The increased
incidences of the remaining tumor types observed in female rats were not dose-dependent.
Incidences of pituitary chromophobe adenomas, thyroid follicular cell adenoma or carcinomas,
and mammary gland fibroadenomas were lower in HCE-treated animals than in controls. Ovary
granulose cell tumors were increased in the low-dose group, compared to controls, although none
of the female rats in the high-dose group exhibited this tumor. NCI (1978) noted that all of these
tumor types had been encountered previously as spontaneous lesions in the Osborne-Mendel rat,
and the authors reported that no statistical differences in frequencies were observed between
treated and control rats. NCI concluded that there was no evidence of carcinogenicity in this rat
study.
29
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-9. Tumor incidences in female rats gavaged with HCE
Tumor type
Vehicle control
113 mg/kg-day
227 mg/kg-day
Kidney hamartoma
0/20 (0%)
0/50 (0%)
3/49 (6%)
Weeks to first tumor
-
-
112
Pituitary chromophobe adenoma
7/20 (35%)
15/50 (30%)
6/46 (13%)
Weeks to first tumor
89
89
112
Thyroid follicular cell adenoma or carcinoma
2/20 (10%)
3/47 (6%)
3/47 (6%)
Weeks to first tumor
111
112
109
Mammary gland fibroadenoma
6/20 (30%)
13/50 (26%)
9/50 (18%)
Weeks to first tumor
106
57
94
Ovary granulosa cell tumor
1/20 (5%)
4/48 (8%)
0/49 (0%)
Weeks to first tumor
111
111
-
Source: NCI (1978).
NCI (1978; Weisburger, 1977) conducted a chronic oral study in 50 B6C3Fi mice/sex/
dose administered 0, 500, or 1,000 mg/kg-day HCE (purity >98%) via corn oil gavage for
5 days/week for 78 weeks. Following exposure termination, animals were observed for 12-
13 weeks for a total duration of 90-91 weeks. Twenty mice/sex were included as untreated and
vehicle controls. Starting in week 9, the doses were increased to 600 and 1,200 mg/kg-day; no
explanation was provided for this change in dose. After adjustment from 5 days/week for
78 weeks to continuous exposure, the TWA doses were 360 and 722 mg/kg-day. Survival rates
were unexpectedly low in males, particularly in the control and low-dose groups: 25 and 5% in
the vehicle and untreated control groups and 14 and 58% in the 360 and 722 mg/kg-day dose
group, respectively. NCI (1978) did not suggest a reason why more high-dose male mice
survived compared with the low-dose and control males. Individual animal data were not
available to make survival adjustments to the tumor incidence data discussed below. Survival
rates in females were 80 and 85% in vehicle and untreated control groups and 80 and 68% in the
360 and 722 mg/kg-day dose groups, respectively. As a result of the low survival rates in the
vehicle and untreated male control groups, NCI compared tumor incidences in the dosed males
and females to the pooled vehicle control data derived from concurrently run bioassays for
several other chemicals. Animals were all of the same strain, housed in the same room,
intubated with corn oil, tested concurrently for at least 1 year, and examined by the same
pathologists.
Chronic inflammation of the kidney was observed in control and treated male mice: 67,
80, 66, and 18% of untreated controls, pooled vehicle controls, low dose, and high dose,
respectively. Female mice in the pooled vehicle control group (15%) and 722 mg/kg-day (2%),
but not the untreated control and 360 mg/kg-day dose groups, exhibited chronic kidney
inflammation. Tubular nephropathy (characterized by degeneration of convoluted tubule
epithelium at the junction of the cortex and medulla, enlarged dark staining regenerative tubular
30
DRAFT - DO NOT CITE OR QUOTE
-------
epithelium, and infiltration of inflammatory cells, fibrosis, and calcium deposition) was not
observed in untreated or pooled vehicle controls of either sex, but was observed in mice treated
with HCE: 49/50 and 47/49 in males and 50/50 and 45/49 in females in the 360 and 722 mg/kg-
day dose groups, respectively. Information on the severity of these effects at the different dose
levels was not presented. No other HCE-related nonneoplastic effects were observed. EPA
considered 360 mg/kg-day as the LOAEL for this study based on tubular nephropathy. EPA
considered that a NOAEL was not established.
Increases in the incidence of hepatocellular carcinomas were observed in male and
female mice exposed to HCE (Table 4-10). Hepatocellular adenomas were not noted in the
report. NCI (1978) reported statistically significant increases in the incidence of hepatocellular
carcinomas in 30 and 63% of 360 and 722 mg/kg-day males, compared with 10 and 15% of
pooled vehicle and matched vehicle controls, respectively. Female mice also demonstrated an
increased tumor response, 40 and 31% of 360 and 722 mg/kg-day females compared with 3 and
10%) of pooled vehicle and matched vehicle controls, respectively. Although the increases in
HCE-treated females were not dose-dependent, a higher incidence of hepatocellular carcinomas
was observed at the low dose (20/50) compared with the high dose (15/49). NCI concluded that
HCE was carcinogenic in both sexes of B6C3Fi mice (1978).
Table 4-10. Incidence of hepatocellular carcinomas in mice
Pooled vehicle control3
Matched vehicle control
360 mg/kg-day
722 mg/kg-day
Males
6/60 (10%)
3/20 (15%)
15/50 (30%)b
31/49 (63%)°
Females
2/60 (3%)
2/20 (10%)
20/50 (40%)°
15/49 (31%)°
"As a result of the exceptionally low survival rates in the vehicle and untreated control groups, NCI used the pooled
vehicle control data derived from concurrently run bioassays for several other chemicals. Animals were all of the
same strain and housed in the same room. Incidences reported were not adjusted for survival.
Statistically significant, p = 0.008.
Statistically significant, p < 0.001.
Source: NCI (1978).
4.2.2. Inhalation
4.2.2.1. Subchronic Exposure
Only one study is available in the peer-reviewed literature that evaluated the subchronic
(Weeks et al., 1979) inhalation toxicity of HCE. Weeks et al. (1979) exposed Sprague-Dawley
rats, Beagle dogs, Hartley guinea pigs, and Coturnix japonica (Japanese quail) to HCE for 6
weeks. The effects observed in these species include neurotoxicity, reduced body weight gain,
increased organ weights, and some evidence of respiratory tract irritation.
Weeks et al. (1979) exposed Sprague-Dawley rats (25/sex/concentration) to control air,
15, 48, or 260 ppm HCE (145, 465, and 2,517 mg/m3, respectively; purity 99.8%) for
31
DRAFT - DO NOT CITE OR QUOTE
-------
6 hours/day, 5 days/week for 6 weeks. Postexposure observation was carried out for 12 weeks.
An oxygen consumption test was also conducted. The authors reported that in the 2,517 mg/m3
group, body weight gain of male rats, but not the nonpregnant female rats, was reduced
beginning in the third week of exposure (although quantitative information was not reported).
All rats in the 2,517 mg/m3 group exhibited tremors, ruffled pelt, and red exudates around the
eyes following the fourth week of exposure. The authors reported that in the male rats, relative
kidney, spleen, and testes weights were significantly increased; in the female rats, only relative
liver weights were significantly increased (although quantitative information was not reported).
One male and one female rat died during the fourth week. During the observation period,
treatment-related effects disappeared. No gross changes were evident at necropsy; however,
after sacrifice, male and female rats of the 2,517 mg/m3group had a higher incidence and severity
of mycoplasma-related lesions in nasal turbinates, trachea, and lung compared with controls.
The authors concluded that these lesions were related to potentiation of an endemic mycoplasia
infection rather than a direct effect of HCE exposure. However, no data were presented
demonstrating the presence of mycoplasia in the lung. There were no histopathological
differences observed between control and exposed rats sacrificed 12 weeks postexposure. No
treatment-related effects were observed in the rats exposed to 145 and 465 mg/m3 HCE.
In the oxygen consumption test, male rats (5/concentration) were tested prior to and
following exposure to 145, 465, or 2,517 mg/m3HCE for 15 minutes, 3 days/week for the
duration of the study (6 weeks). The 2,517 mg/m3 rats exhibited significantly decreased mean
rates of consumption prior to (15%) and after (13%) HCE exposure. The authors suggested that
this decrease in oxygen consumption, while nonspecific, is indicative of an alteration in basal
metabolic rate. No histopathological effects were observed at this concentration. EPA
considered 465 mg/m3 the NOAEL and 2,517 mg/m3 the LOAEL, based on reduced body weight
gain, and increased organ weights.
Weeks et al. (1979) also exposed male Sprague-Dawley rats (15/concentration) exposed
to 15, 48, or 260 ppm HCE (145, 465, or 2,517 mg/m3) for 6 hours/day, 5 days/week for 6 weeks
and examined them for behavioral changes related to learned and unlearned responses (described
in detail in Section 4.4.3.2). Similar to the other treated rats, body weight gain was reduced.
Final mean body weight gain in male rats was reduced 2, 5, and 10% (statistically significant) in
the 145, 465, and 2,517 mg/m3 dose groups, respectively, compared with controls. Additionally,
relative lung, liver, kidney, and testes weights were increased (quantitative information not
reported) compared with controls.
Weeks et al. (1979) also exposed four male Beagle dogs/concentration to control air, 15,
48, or 260 ppm HCE (145, 465, and 2,517 mg/m3, respectively; purity 99.8%) for 6 hours/day,
5 days/week for 6 weeks. Postexposure observation was carried out for 12 weeks. Blood
samples were evaluated for blood chemistry parameters. In addition, the dogs underwent
pulmonary function tests prior to and following exposure. One dog died within 5 hours of
32
DRAFT - DO NOT CITE OR QUOTE
-------
exposure to 2,517 mg/m3. The remaining animals in the 2,517 mg/m3 group exhibited signs of
neurotoxicity consisting of tremors, ataxia, hypersalivation, head bobbing, and facial
fasciculations. No blood parameters were significantly affected and no exposure-related
histopathological lesions were observed following necropsy on dogs sacrificed 12 weeks
postexposture. Dogs evaluated for pulmonary functions while anesthetized did not display any
significant effects. The HCE-exposed dogs did not display any treatment-related toxicity at
12 weeks postexposure. EPA considered 465 mg/m3 the NOAEL and 2,517 mg/m3 the LOAEL,
based on neurotoxic effects.
Weeks et al. (1979) also exposed male Hartley guinea pigs (10/concentration) to control
air, 15, 48, or 260 ppm HCE (145, 465, and 2,517 mg/m3, respectively; purity 99.8%) for
6 hours/day, 5 days/week for 6 weeks. Postexposure observation was carried out for 12 weeks.
Guinea pigs were also evaluated for sensitization potential following inhalation exposure to
HCE. Two guinea pigs died during each of the fourth and fifth weeks, resulting in four total
deaths. Guinea pigs of the 2,517 mg/m3 group displayed reductions in body weight beginning at
the second week of exposure and significantly increased liver to body weight ratios (quantitative
information was not reported). No treatment-related effects were observed in the other exposure
groups. EPA considered the NOAEL as 465 mg/m3 and the LOAEL as 2,517 mg/m3, based on
decreased body weight and significantly increased relative liver weight.
Weeks et al. (1979) also exposed male and female quail (C. japonica, 20/concentration)
to control air, 15, 48, or 260 ppm HCE (145, 465, and 2,517 mg/m3, respectively; purity 99.8%)
for 6 hours/day, 5 days/week for 6 weeks. Postexposure observation was carried out for
12 weeks. The only observed effect was excess mucus in nasal turbinates in 2/10 quail in the
2,517 mg/m3 group after 6 weeks. The authors considered the excess mucus to be transient based
on the lack of any inflammation or histopathological effects. Although the study authors
considered the excess mucus to be a transient effect, EPA notes that the lack of inflammation and
histopathological effects does not preclude the presence of more sensitive indicators of immune
response (e.g., antibodies or other immune signaling chemicals) unable to be detected with
methods available to the study authors. EPA considered 2,517 mg/m3 (highest exposure
concentration) as the NOAEL, while the LOAEL could not be established from this study.
4.2.2.2. Chronic Exposure
No inhalation chronic exposure studies were identified.
4.3. REPRODUCTIVE/DEVELOPMENTAL STUDIES—ORAL AND INHALATION
4.3.1. Oral
Weeks et al. (1979) exposed 22 pregnant Sprague-Dawley rats/dose to 50, 100, or
500 mg/kg HCE (purity 99.8%) by gavage on gestation days (GDs) 6-16. Gavage controls
received corn oil and positive controls received 250 mg/kg aspirin. Dams orally administered
33
DRAFT - DO NOT CITE OR QUOTE
-------
500 mg/kg HCE displayed tremors on GDs 15 and 16. Body weight gain of the 500 mg/kg dams
was significantly lower than controls beginning on GD 8. Rats in the 500 mg/kg group exhibited
an increased incidence of mucopurulent nasal exudates compared with controls. Approximately
70% of the orally exposed 500 mg/kg group had upper respiratory tract irritation; 20% had
subclinical pneumonitis, compared with 10% in controls.
The aspirin-positive control group produced fetuses with lower body weights and
malformations such as hydrocephalus, spina bifida, and cranioschesis. None of the fetuses
exhibited any significant skeletal or soft tissue anomalies, although fetuses from dams gavaged
with 500 mg/kg HCE displayed significantly lower gestation indices, lower numbers of viable
fetuses/dam, and higher fetal resorption rates compared with controls (data not shown). EPA
considered the maternal NOAEL and LOAEL as 100 and 500 mg/kg, respectively, based on
neurological effects (tremors) and body weight decreases. EPA considered the developmental
NOAEL and LOAEL to be the same as the maternal values, based on decreased viability and
increased resorption rates.
Shimizu et al. (1992) evaluated the teratogenicity of HCE (purity not specified) in
pregnant Wistar rats at doses of 0, 56, 167, or 500 mg/kg administered by gavage during GDs 7-
17 (20-21 rats/dose). The dams of the 500 mg/kg dose group exhibited significantly decreased
weight gain after the second day of HCE treatment (8th day of pregnancy); dams in the
167 mg/kg dose group displayed significantly decreased weight gain after the fourth day of
treatment (10th day of pregnancy), but not after the treatment ended on the 18th day of pregnancy.
Food intake was also significantly decreased in the 500 and 167 mg/kg dose groups after the
second and third days, respectively, of HCE treatment; however, intake was normal when
treatment ended. Dams in both the 167 and 500 mg/kg dose groups exhibited decreased motor
activity (incidence and method of analysis not reported); dams in the 500 mg/kg dose group also
exhibited piloerection and subcutaneous hemorrhage. These effects decreased or disappeared
when HCE exposure ended. An autopsy performed on dams on GD 20, 3 days post-HCE
exposure, revealed three rats with whitening of the liver in the 500 mg/kg dose group. The
significance of this observation is unknown. No deaths occurred in any of the dose groups.
There were no significant differences between the HCE treatment and control groups
with respect to the numbers of corpora lutea, implants, or live fetuses (Table 4-11). There was
no significant difference in the incidence of dead or resorbed fetuses, except for a significant
increase during the late stage of pregnancy in the 500 mg/kg dose group (6.4% versus none in the
control). Fetuses in the 500 mg/kg dose group also displayed significantly decreased body
weight; 2.5 ± 0.57 (mean ± SD) and 2.3 ± 0.45 g in male and female fetuses, compared with
3.3 ± 0.20 and 3.1 ± 0.24 g in male and female controls, respectively.
34
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-11. Summary of HCE effects on pregnant Wistar rats and their
fetuses
Dose (mg/kg)
0
56
167
500
Number of dams
20
20
20
21
% of dead or resorbed fetuses
8.7
9.2
7.0
14.7
Early stage
8.7
00
00
6.1
13.1
Late stage
0.4
0.9
6.4a
Body weight of live fetuses (g)b
Male
3.3+0.20
3.3+0.17
3.2+0.21
2.5 +0.573
Female
3.1+0.24
3.0+0.20
2.9+0.17
2.3 +0.453
"Significantly different from control, p < 0,01,
bValues are mean + SD.
Source: Shimizu et al. (1992).
The investigators (Shimizu et al., 1992) examined the fetuses for external anomalies and
found one case of acaudate in the 500 mg/kg dose group. Other anomalies included two fetuses
with subcutaneous hemorrhage in the 167 and 500 mg/kg dose groups and one case of hyposarca
in the 500 mg/kg dose group. No skeletal malformations were observed in any group, although a
statistically significant increase in skeletal variations was observed in the 500 mg/kg (60.3%)
group compared with controls (1.3%). Skeletal variations were significantly increased in the
500 mg/kg group (2 cases in the lumbar rib and 78 cases in the rudimentary lumbar rib) and
nonsignificantly increased in the 167 mg/kg group (6 cases in the rudimentary lumbar rib)
compared with controls (2 cases in the rudimentary lumbar rib) (Table 4-12). The degree of
ossification (including numbers of sternebrae, proximal and middle phalanges, and sacral and
caudal vertebrae) was significantly decreased in the 500 mg/kg dose group. No visceral
malformations were observed and no significant differences in visceral anomalies were noted.
The authors concluded that there was no indication of teratological effects in rats for dose levels
of HCE below 500 mg/kg. Shimizu et al. (1992) established a NOAEL of 56 mg/kg for dams
and 167 mg/kg for fetuses. EPA considered the LOAEL for dams as 167 mg/kg-day, based on
decreased motor activity and significantly decreased body weight. EPA considered the LOAEL
for fetuses as 500 mg/kg, based on significantly increased skeletal variations and significantly
decreased ossification and fetal body weight.
35
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-12. Summary of skeletal effects on fetuses from HCE-exposed rats
Dose (mg/kg)
0
56
167
500
Number of fetuses examined
136
136
136
137
Percent of fetal variations
1.3
0
3.8
60.3a
Number of fetuses with variations
Lumbar rib
0
0
0
2
Rudimentary lumbar rib
2
0
6
78
Ossification13
Number of sternebrae
4.7+0.07
4.5+0.08
4.5+0.08
3.4+0.273
Number of proximal and middle
phalanges
Fore limb
3.2+0.05
3.1+0.04
3.1+0.04
2.9+0.113
Hind limb
4.0+0.01
4.0+0.01
4.0+0.01
3.4+0.233
Number of sacral and caudal vertebrae
6.9+0.06
6.9+0.08
7.0+0.04
5.7+0.373
aSignificantly different from control, p < 0.01,
bAs reported by Shimizu et al. (1992) the litter was used as the statistical unit for calculation of fetal values; thus,
these values represent the means + SD of litter means within each group.
Source: Shimizu et al. (1992).
4.3.2. Inhalation
Weeks et al. (1979) exposed 22 pregnant Sprague-Dawley rats/concentration to control
air, 15, 48, or 260 ppm HCE (145, 465, and 2,517 mg/m3, respectively; purity 99.8%) by
inhalation on GDs 6-16. Dams in the 2,517 mg/m3 group displayed tremors during GDs 12-16.
Body weight gain of the dams was significantly lower than controls beginning on GD 8 for the
2,517 mg/m3 group, and beginning on GD 14 for the 465 mg/m3 group. Rats in the 465 and
2,517 mg/m3 groups exhibited an increased incidence of mucopurulent nasal exudates compared
with controls. Inflammatory exudate was observed in the lumen of the nasal turbinates of 85%
of the 465 mg/m3 group and 100% of the 2,517 mg/m3 group. The authors attributed the
increased exudate to an endemic mycoplasia infection.
Fetuses of HCE-treated dams did not exhibit any significant skeletal or soft tissue
anomalies. EPA considered the NOAEL for the dams as 465 mg/m3 and the LOAEL as 2,517
mg/m3, based on neurological effects (tremors). EPA considered 2,517 mg/m3 (highest
concentration tested) as a developmental NOAEL, based on the lack of treatment-related effects,
while a developmental LOAEL could not be established from this study.
36
DRAFT - DO NOT CITE OR QUOTE
-------
4.4. OTHER DURATION- OR ENDPOINT-SPECIFIC STUDIES
4.4.1. Acute Exposure Studies
4.4.1.1. Oral
Several studies evaluated acute toxicity of HCE in animal species and reported lethal
dose concentrations. Oral lethal doses ranged from 2,332 to 8,640 mg/kg in rats, >1,000 mg/kg
in male rabbits, and 4,970 mg/kg in guinea pigs (Kinkead and Wolfe, 1992; Weeks et al., 1979).
According to the Hodge and Sterner Scale, these lethal doses place HCE in low toxicity range
(Hodge and Sterner, 1949). Reynolds (1972) administered a single dose of HCE (purity not
specified) at 26 mmol/kg (6,155 mg/kg) by gavage in mineral oil to male rats and reported that
liver function (assessed by microsomal protein concentration, antipyrine demethylase activity,
NADP-neotetrazolium reductase activity, glucose 6-phosphatase activity, and conjugated diene
concentration in microsomal lipids) was unaffected 2 hours after exposure. Kinkead and Wolfe
(1992) determined that the oral median lethal dose (LD50) for HCE (purity not specified) in male
and female Sprague-Dawley rats (5 rats/sex/dose) was 4,489 mg/kg (95% confidence limit [CL],
2,332-8,640 mg/kg). A study in sheep that was conducted at high doses (500-1,000 mg/kg)
found reduced hepatic function (Fowler, 1969).
Weeks et al. (1979) and Weeks and Thomasino (1978) determined acute oral toxicity
values for Sprague-Dawley rats, New Zealand White rabbits, and Hartley guinea pigs by
administering a single dose of HCE (99.8% purity) dissolved in corn oil (50% w/v) or
methylcellulose (5% w/v) via gavage. Approximate lethal dosages (ALD) or LD50 values were
calculated after a 14-day observation period (Table 4-13). All LD50 values were >1,000 mg/kg.
Table 4-13. Summary of acute exposure data in rats, rabbits, and guinea
pigs
Lethal value
Species
Treatment
Diluent
mg/kg
95% CL
Slope
Rabbit, male
Oral ALD
Methylcellulose
>1,000
Rat, male
Intraperitoneal
(i.p.) ALD
Corn oil
2,900
Rat, male
Oral ALD
Corn oil
4,900
Rat, female
Oral LD50
Corn oil
4,460
3,900-5,110
9.3
Methylcellulose
7,080
6,240-8,040
19.9
Rat, male
Oral LD50
Corn oil
5,160
4,250-6,270
6.1
Methylcellulose
7,690
6,380-9,250
8.5
Guinea pig, male
Oral LD50
Corn oil
4,970
4,030-6,150
4.7
Rabbit, male
Dermal LD50
Water paste
>32,000
Sources: Weeks et al. (1979); Weeks and Thomasino (1978).
37
DRAFT - DO NOT CITE OR QUOTE
-------
Fowler (1969) orally administered a single dose of HCE (purity not specified) through a
drenching bottle to Scottish Blackface and Cheviot cross sheep at three dose levels: 500 (six
sheep), 750 (one sheep), and 1,000 mg/kg (one sheep). Hepatotoxicity was assessed by
measurement of plasma enzyme activities and bromsulphthalein dye clearance tests, which are
widely-used indices of hepatic function in sheep. Plasma activities of glutamate dehydrogenase
(GDH), sorbitol dehydrogenase (SDH), ornithine carbamoyl transferase (OCT), and AST were
determined daily until they reached stable levels. Increases in these enzymes are indicative of
hepatic damage. HCE exposure resulted in a 3-6-fold increase in GDH, with the exception of
one sheep that exhibited a 55-fold increase. SDH was increased 3-6-fold and OCT was
increased 2-10-fold. GDH, SDH, and OCT levels peaked at 48 hours and returned to normal
within 4-5 days. AST increased only slightly. Bromsulphthalein dye clearance tests found a
reduction in transfer from liver cells to bile at 72 hours after HCE exposure, indicating reduced
hepatic function.
4.4.1.2. Inhalation
Median lethal concentration (LC50) values for HCE have not been reported. One study is
available in the peer-reviewed literature that evaluated acute inhalation exposure to HCE (Weeks
and Thomasino 1978). Six male rats/concentration (strain not specified, although one table in
the report indicated strain as Sprague-Dawley) were exposed to 260 or 5,900 ppm HCE (2,500 or
57,000 mg/m3) for 8 hours and to 1,000 ppm HCE (17,000 mg/m3) for 6 hours. Postexposure
observation was carried out for 14 days. Male rats exposed for 8 hours to 2,500 mg/m3 HCE
displayed no toxic signs during exposure or for 14 days thereafter. Body weight gain was
slightly, but not statistically significantly, reduced over the 14-day exposure period. Male rats
exposed for 8 hours to 57,000 mg/m3 HCE displayed severe toxic signs including death. At 6
hours, one rat had a staggered gait. At 8 hours, 2/6 rats were dead. The surviving rats showed
statistically significant reductions in mean body weight on exposure days 0 (7%), 1 (21%), 3
(19%), 7 (15%>), and 14 (15%>), compared with controls. Necropsy did not reveal any gross
exposure-related lesions. Microscopy revealed that two of the four surviving rats had minimally
to moderately severe subacute diffuse interstitial pneumonitis and vascular congestion.
Additionally, a purulent exudate of the nasal turbinates was observed in one control and one
treated rat. The authors concluded that this effect was not exposure-related, but rather was
indicative of a low-grade endemic upper respiratory disease. The male rats exposed for 6 hours
to 17,000 mg/m3 showed slight reductions in body weight gain on postexposure days 1 (5%) and
3 (4%>) and body weights similar to controls for the remaining 11 days of the postexposure
period. Two of the six rats demonstrated a staggered gait. No exposure-related gross or
histopathological changes were observed in tissues and organs.
38
DRAFT - DO NOT CITE OR QUOTE
-------
4.4.2. Short-term Exposure Studies
Several studies evaluated short-term toxicity of HCE in animal species. A 12-day study
in male New Zealand White rabbits found liver degeneration and necrosis, as well as tubular
nephrosis in the kidney, indicating that both the liver and kidney are potential target tissues for
HCE-induced toxicity (Weeks et al., 1979). Short-term toxicity assays in rats (16 and 21 days)
demonstrated kidney effects in males (NTP, 1996, 1989) but not females (NTP, 1989).
Weeks et al. (1979) conducted a 12-day study of HCE in male New Zealand White
rabbits. Five rabbits/dose were administered a daily oral dose via a stomach tube of 100, 320, or
1,000 mg/kg HCE (purity 99.8%) suspended in 5% aqueous methylcellulose. Blood was drawn
from the central ear artery of the rabbits on treatment days 1, 4, 8, and 12, and on day 4
following termination of dosing. Serum was analyzed for the following parameters: glutamic
oxaloacetic transaminase (SGOT; also known as AST), glutamic pyruvic transaminase (SGPT;
also known as ALT), blood urea nitrogen (BUN), alkaline phosphatase, bilirubin, total protein,
potassium, and sodium. On the fourth day following the termination of dosing, rabbits were
necropsied and the following tissues were examined: eye, brain, lung, kidney, liver, spleen,
heart, stomach, pancreas, large intestine, skeletal muscle, bone, urinary bladder, small intestine,
and testes.
The 1,000 mg/kg dose group exhibited significantly reduced body weight (beginning on
treatment day 7) and increased relative liver and kidney weights. The 320 mg/kg dose group
exhibited significantly reduced body weight beginning on day 10. The 100 mg/kg dose group
did not display any changes. The 320 and 1,000 mg/kg dose groups displayed liver degeneration
and necrosis, including fatty degeneration, coagulation necrosis, hemorrhage, ballooning
degeneration, eosinophilic changes, and hemosiderin-laden macrophages and giant cells. These
effects were not observed in controls or rabbits of the 100 mg/kg dose group. Liver lesions
increased in severity in a dose-related manner in which the effects were more severe in the
1,000 mg/kg group compared with the 320 mg/kg group. Tubular nephrosis of the convoluted
tubules in the corticomedullary region of the kidney was also observed in the rabbits of the
320 and 1,000 mg/kg dose groups. These animals also exhibited tubular nephrocalcinosis of a
minimal degree. The only blood chemistry parameters that were affected were significantly
decreased potassium and glucose levels in the 320 and 1,000 mg/kg groups. EPA considered the
NOAEL as 100 mg/kg and the LOAEL as 320 mg/kg, based on dose-related increases in severity
of liver and kidney lesions.
The NTP (1989) conducted a 16-day study of oral HCE toxicity in F344/N rats. Groups
of five rats/sex/dose were administered 0, 187, 375, 750, 1,500, or 3,000 mg HCE/kg (purity
>99%) for 12 doses over 16 days by corn oil gavage. TWA doses were 0, 140, 281, 563, 1,125,
and 2,250 mg/kg-day, respectively. Necropsy was performed on all rats; all organs and tissues
were examined for grossly visible lesions and histopathology. All rats of the 1,125 and
2,250 mg/kg-day dose groups and 1/5 males and 2/5 females from the 563 mg/kg-day dose group
39
DRAFT - DO NOT CITE OR QUOTE
-------
died before the end of the study. Final mean body weights (statistical analyses were not
reported) were decreased by 25% in males of the 563 mg/kg-day dose group; female body
weights were decreased by 37% in the 563 mg/kg-day dose group. Microscopic observations of
the kidneys revealed hyaline droplet formation in the cytoplasm of renal tubular epithelium in all
treated males, and tubular cell regeneration and eosinophilic granular casts of cell debris in
tubule lumina of male rats administered 140 and 281 mg/kg-day. EPA considered
140 mg/kg-day (lowest dose tested) a male rat LOAEL based on kidney tubule lesions, while a
NOAEL could not be established for male rats. EPA considered the female rat LOAEL as
563 mg/kg-day, based on a dose-related decrease in body weight, and the female rat NOAEL as
281 mg/kg-day.
NTP (1996) conducted a 21-day study of oral HCE toxicity in male F344/N rats. Groups
of five rats/dose were administered 0.62 or 1.24 mmol HCE/kg-day (146 or 293 mg/kg-day,
respectively; purity 100%) by corn oil gavage. Necropsies were performed on all rats; the right
kidney, liver, and right testis were weighed and underwent histopathological evaluation. Urine
samples were collected during an overnight period that began 4 days before the end of the study.
Urinalysis included measurements of volume, specific gravity, creatinine, glucose, total protein,
AST, y-glutamyl transferase (GGT), and /V-acetyl-P-D-glucosaminidase (NAG). A Mallory-
Heidenhain stain was used for kidney sections to evaluate protein droplets, particularly hyaline
droplet formation. Cell proliferation analyses were performed on kidney sections and were
scored by a labeling index indicating the percentage of proximal and distal tubule epithelial cells
in S-phase.
Results from the measured endpoints/parameters are summarized in Table 4-14.
Absolute and relative kidney weights were significantly increased in both dose groups; absolute
and relative (significant at high dose) liver weights were increased in both dose groups. Rats of
the 293 mg/kg-day group also exhibited significantly lower urinary creatinine and specific
gravity, while glucose and urine volumes were greater than controls. AST and NAG activities
were significantly higher than in controls. Nephropathy, consisting of hyaline droplet
accumulation, was observed in the male rats in addition to increased incidences of tubule
regeneration (3/5 and 4/5 for 146 and 293 mg/kg-day, respectively) and granular casts (4/5 and
3/5 for 146 and 293 mg/kg-day, respectively). The mean proliferating cell nuclear antigen
(PCNA) labeling index was significantly increased 5.7- and 9.2-fold, compared with controls, in
the 146 and 293 mg/kg-day dose groups. EPA did not identify a NOAEL because effects
(including increased absolute and relative kidney weight, increased AST and NAG activity,
increased PCNA labeling index, and nephropathy) were observed at the low dose level. EPA
considered 146 mg/kg-day a LOAEL based on statistically significant increases in kidney lesions
and urinalysis parameters.
40
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-14. Summary of toxicity data from male rats exposed to HCE for
21 days
Vehicle control
146 mg/kg-day HCE
293 mg/kg-day HCE
Right kidney weight3
Absolute (g)
1.009+0.025
1.157+ 0.011b
1.250+0.022b
Relative (mg/g)
3.19+0.04
3.77 + 0.06b
4.07 + 0.05b
Liver weight3
Absolute (g)
11.041 +0.291
11.959 + 0.178
13.479 +0.390
Relative (mg/g)
34.82 + 0.60
39.01+0.92
43.84 + 0.64b
Right testis weight3
Absolute (g)
1.412+0.037
1.409 + 0.023
1.430 +0.016
Relative (mg/g)
4.47+0.09
4.60 + 0.11
4.66+0.05
Urinalysis
Creatinine (mg/dL)
143.22 + 18.12
79.56 + 11.01
54.48+ 3.06b
Glucose (|ig/mg creatinine)
169 + 3
344 + 30
446 + 23b
Protein (mU/mg creatinine)
1,322 + 59
1,748 +257
2,980 + 103
AST (mU/mg creatinine)
6 + 1
40 + 6°
66 + 5b
GGT (mU/mg creatinine)
1,456 + 47
1,547 + 66
1,897+73
NAG (mU/mg creatinine)
11+0
23 +2C
36+ lb
Volume (mL/16 h)
4.2+0.8
7.5+9
10.6+ l.lb
Specific gravity (g/mL)
1.038 +0.005
1.024 + 0.003
1.020+0.001b
PCNA labeling index (mean + SE)
0.13+0.02
0.74 + 0.19°
1.2 + 0.2°
aData are mean + SE.
bSignificantly different from control (p < 0.01).
Significantly different from control (p < 0.05).
Source: NTP(1996).
4.4.3. Neurological
Neurological endpoints for HCE toxicity have been evaluated in several studies. The
studies listed below provide evidence that HCE produces neurological effects; however, it is
unknown if the central nervous system (CNS) effects are due to the parent compound or the
metabolites. Sheep exposed to high doses of HCE (500-1,000 mg/kg) developed facial muscle
tremors (Fowler, 1969; Southcott, 1951), and a staggering uncoordinated gait (Southcott, 1951).
Sprague-Dawley rats evaluated for HCE-induced effects on avoidance latency (i.e., learned
behavior) and spontaneous motor activity (i.e., unlearned behavior) exhibited slight, but not
statistically significant, behavioral effects at 2,517 mg/m3. Male and female rats also exhibited
tremors and ruffled pelt at 2,517 mg/m3 (Weeks et al., 1979). Beagle dogs developed signs of
neurotoxicity such as tremors, ataxia, hypersalivation, and head bobbing following exposure to
2,517 mg/m3 HCE. Dogs showed similar signs of neurotoxicity intermittently throughout the
41
DRAFT - DO NOT CITE OR QUOTE
-------
HCE exposures, with signs disappearing overnight. During an observation period of 12 weeks
following exposure, these symptoms were not observed (Weeks et al., 1979).
4.4.3.1. Oral Studies
Fowler (1969) orally administered a single dose of HCE (purity not specified) to Scottish
Blackface and Cheviot cross sheep at three dose levels: 500 (10 sheep), 750 (1 sheep), and
1,000 mg/kg (1 sheep). Slight facial muscle tremors were noted in three sheep between 1 and
4 hours after dosages of 500-1,000 mg/kg HCE. The HCE dose level for the individual sheep
exhibiting facial tremors was not specified by the authors. Fowler (1969) also examined two
sheep administered 0.3 mL/kg PERC and two sheep administered 0.3 mL/kg pentachloroethane,
two proposed major metabolites of HCE. The sheep exposed to PERC exhibited no effects
following exposure, while the sheep exposed to pentachloroethane exhibited narcosis. One
pentachloroethane-exposed animal was recumbent within 30 minutes of exposure, exhibiting
flaccid limbs, depression of normal reflexes, and labial tremors. The sheep regained normal
posture 9 hours postexposure and appeared normal 72 hours postexposure. The second
pentachloroethane-treated sheep became recumbent within 20 minutes of exposure and exhibited
labial tremors. However, unlike the first sheep, this animal appeared normal 1.5 hours
postexposure. EPA considered the LOAEL as 500 mg/kg (lowest dose tested), based on
neurotoxic effects (tremors), while a NOAEL could not be established from these data.
Southcott (1951) treated 30 Merino Wethers sheep suffering from liver fluke infections
with 15 g HCE-bentonite dispersible powder (13.5 g HCE, 445 mg/kg; 15 sheep) or 30 g
HCE-bentonite (27 g HCE, 906 mg/kg; 15 sheep). The purity of the HCE was not specified.
One day after treatment, two sheep died and nine others were unable to rise and stand. One of
the severely affected sheep (i.e., unable to rise and stand) was from the 445 mg/kg HCE group
and the other eight were from the 906 mg/kg group. Some severely affected animals (two from
the 445 mg/kg group) could walk if placed on their feet, but displayed a staggering,
uncoordinated gait and fell again. The lips, face, neck, and forelegs were afflicted by fine
muscular tremors that were observed in most of the animals. EPA considered the LOAEL as
445 mg/kg (lowest dose tested), based on neurological effects consisting of tremors, staggering,
uncoordinated gait, and inability to stand, while a NOAEL could not be established from this
study.
As described in Section 4.3.1, Shimizu et al. (1992) reported decreased motor activity
(incidence and method of analysis not reported) in pregnant Wistar rats (20-21 rats/dose) at
doses of 167 and 500 mg/kg administered by gavage during GDs 7-17. These effects decreased
or disappeared when HCE exposure ended. Weeks et al. (1979) exposed 22 pregnant
Sprague-Dawley rats/dose to 50, 100, or 500 mg/kg HCE by corn oil gavage on GDs 6-16.
Dams orally administered 500 mg/kg HCE displayed tremors on GDs 15 and 16.
42
DRAFT - DO NOT CITE OR QUOTE
-------
4.4.3.2. Inhalation Studies
Weeks et al. (1979) exposed male Sprague-Dawley rats (15/concentration) to air, 15, 48,
or 260 ppm HCE (145, 465, or 2,517 mg/m3, respectively; purity 99.8%) for 6 hours/day,
5 days/week for 6 weeks. Learned behavior endpoints, evaluated using an avoidance latency
task by measuring the time it took the rats to avoid foot shock by escaping into a safe
compartment and unlearned behavior endpoints (spontaneous motor activity; evaluated by
photobeam interruptions) were measured in the animals. The avoidance latency task was
conducted prior to exposure, 1 day into exposure, after 3 weeks of exposure, and after 6 weeks of
exposure. Spontaneous motor activity was tested after 3 and 6 weeks of exposure.
Avoidance latency was slightly, but not significantly, increased in the 465 and
2,517 mg/m3 groups at 6 weeks (median 3.9 and 3.3 seconds, respectively) compared with
control (median 2.2 seconds). Spontaneous motor activity counts were slightly, but not
significantly, increased in the HCE-treated rats (mean ± SD): 231 ±77 for 145 mg/m3, 183 ±
109 for 465 mg/m3, and 201 ± 102 for 2,517 mg/m3, compared with control rats (163 ± 74).
Weeks et al. (1979) concluded that the rats did not display obvious signs of behavioral toxicity.
However, tremors and a ruffled pelt were noted in a separate experiment in male and female rats
exposed to 2,517 mg/m3 HCE during the fourth week of exposure. Tremors and lack of
grooming are indicators of neurobehavioral effects (Kulig et al., 1996). The investigators
sacrificed the rats 12 weeks after the last exposure and reported that all measurable changes (e.g.,
brain histopathology, body weights) were comparable to controls.
Weeks et al. (1979) also exposed 22 pregnant Sprague-Dawley rats/concentration and
4 Beagle dogs/concentration to 145, 465, and 2,517 mg/m3 HCE by inhalation. Rat dams in the
2,517 mg/m3 group displayed tremors during GDs 12-16. Dogs in the 2,517 mg/m3 exposure
group developed tremors, ataxia, hypersalivation, and displayed severe head bobbing, facial
muscular fasciculations, and held their eyelids closed during exposure. One dog experienced
convulsions and died within 5 hours after initial exposure. The surviving dogs exhibited less
severe symptoms during exposure, but recovered overnight after removal from exposure.
4.4.4. Immunological
Ten male Hartley guinea pigs/dose were exposed by inhalation to control air or three
concentrations of HCE (purity 99.8%): 15, 48, or 260 ppm (145, 465, and 2,517 mg/m3,
respectively; Weeks et al., 1979). Exposures were conducted for 6 hours/day, 5 days/week for
3 weeks. Following exposure, animals were allowed to rest for 2 weeks. The guinea pigs were
then challenged with a single intradermal injection of 0.1% HCE in saline. A sensitization
response was not produced.
43
DRAFT - DO NOT CITE OR QUOTE
-------
4.4.5. Dermatological
Yamakage and Ishikawa (1982) examined certain patients suffering from systemic
scleroderma (SSD) for potential exposure to solvents. These patients also presented with
localized scleroderma with bilateral distribution of multiple skin lesions reminiscent of those
observed in several cases of occupational or agent-induced scleroderma. Of nine such patients,
seven had had significant subchronic/chronic exposure (5-44 years), while an eighth had had a
significant acute exposure (2 weeks). The solvents involved were reported as "variable and
mostly unidentified." As an experimental follow-up, groups of ddY mice received daily
intraperitoneal (i.p.) injections for 17 days with 1 of 10 experimental solvents, as well as with
0.9% saline to mitigate treatment lethality. For HCE, 17 mice were injected daily with 0.01 mL
of HCE (purity not specified) and 0.1 mL of 0.9% saline. Along with naphtha ("Esso No. 5")
and n-hexane, HCE was found, by double-blind histological examination and electron
microscopy, to be a significant inducer of sclerodermatous changes in skin taken from the
animals' backs, near the forelimbs. HCE treatment resulted in evident dermal sclerosis in five
mice, slight fibrosis in one mouse, and no change in nine mice; two mice died. PERC, a primary
metabolite of HCE, was similarly tested in 10 mice. Injections of 0.005 mL (+ saline) resulted in
evident dermal sclerosis in one mouse, slight fibrosis in two, no change in six, and death in one.
Even though this experimental route of exposure is generally irrelevant to humans, the skin
lesions produced by HCE were "fundamentally similar" to those produced by control reference
solvents that have been implicated in human occupational SSD. Thus, this study provides
indirect evidence that suggesting that HCE may be capable of inducing SSD-type conditions in
humans.
Weeks and Thomasino (1978) conducted two dermal studies in male New Zealand White
rabbits. A single 24-hour application of 500 mg of technical dry HCE to intact and abraded skin
of six rabbits did not result in primary irritation of intact or abraded skin when assessed at
24 hours, 72 hours, or 7 days after exposure. HCE was placed in Irritation Category IV (no
irritation). In the second study, HCE was applied as a paste in 0.5 mL of distilled water. Intact
skin displayed no edema and barely perceptible erythema at 24 hours. Abraded skin displayed
barely perceptible erythema in one rabbit with moderate to slight erythema reactions. HCE was
placed in Irritation Category III (mild or slight irritation).
4.4.6. Eye Irritation
Weeks and Thomasino (1978) applied a single, 24-hour dose of 100 mg dry HCE to one
eye of six male New Zealand White rabbits. Moderate corneal damage, iritis, and conjunctivitis
was observed in 5/6 rabbits 24, 48, and 72 hours after exposure. No effects were observed
7 days after exposure. HCE was placed in Irritation Category II for eye effects (corneal opacity
reversible within 7 days or persisting for 7 days).
44
DRAFT - DO NOT CITE OR QUOTE
-------
4.5. MECHANISTIC DATA AND OTHER STUDIES IN SUPPORT OF THE MODE OF
ACTION
4.5.1. Genotoxicity
In vivo genotoxicity studies have not been performed in humans exposed to HCE. In
vivo exposure to animals resulted in predominantly negative results. Similarly, in vitro
genotoxicity studies conducted in microorganisms, cultured mammalian cells, and insects
(Table 4-15) were largely negative both in the presence and absence of exogenous metabolic
activation. Ashby and Tennant (1988) examined genotoxic carcinogenesis in a set of
222 chemicals tested in rodents by NCI/NTP; HCE did not induce mutagenicity in Salmonella
typhimurium reverse mutation tester strains. NTP's technical report on the toxicity and
carcinogenicity of HCE in F344/N rats concluded that HCE (purity >99%) was not significantly
genotoxic, and that the increased incidence of tumors occurred through a mechanism other than
one involving the induction of mutations (NTP, 1989). In an examination of putative
"nongenotoxic" carcinogens on the basis of their reported mutagenicity per se (the ability to
induce alterations in deoxyribonucleic acid (DNA) structure or content, i.e., gene mutation,
chromosomal aberrations [CAs], or aneuploidy), HCE was categorized as having insufficient
data for evaluation (Jackson et al., 1993). Studies conducted by Lohman and Lohman (2000)
considering DNA damage, recombination, gene mutation, sister chromatid exchange (SCE),
micronuclei (MN), CA, aneuploidy, and cell transformation as endpoints indicate that the genetic
activity profile for HCE is predominantly negative. However, some positive findings have been
reported in assays for gene conversion, somatic mutation/recombination, DNA adducts, and
SCEs.
45
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-15. Summary of genotoxicity studies of HCE
Test system
Genetic endpoint
Strain/cells
Results
Reference
Comments
In vitro tests
Bacterial
Gene reversion/
S. typhimurium
TA98, TA100, TA1535,
TA1537, TA1538
- (±S9)
Simmon and
Kauhanen
(1978)
TA98, TA100, TA1535,
TA1537, TA1538
- (±S9)
Weeks et al.
(1979)
TA98, TA100, TA1535,
TA1537
- (±S9)
Haworth et al.
(1983)
Liquid
preincubation
protocol
TA98, TA100, TA1535,
TA1537
- (±S9)
Milman et al.
(1988)
Forward mutations
BA13
- (±S9)
Roldan-Aijona
et al. (1991)
Liquid
preincubation
protocol
SOS test
TA1535/pSK1002
- (±S9)
Nakamura et al.
(1987)
umu test;
Liquid
preincubation
protocol
Mammalian
CAs
Chinese hamster ovary
(CHO)
- (±S9)
Galloway et al.
(1987)
SCEs
CHO
- (-S9),
+ (+S9)a
Galloway et al.
(1987)
HCE
precipitation at
doses causing
positive results
MN
AHH-1
-
Doherty et al.
(1996)
Human cell line
MCL-5
-
Doherty et al.
(1996)
Human cell line
h2El
-
Doherty et al.
(1996)
Human cell line
Cell
transformation
BALB/C-3T3
-
Milman et al.
(1988)
DNA adduct
formation
(nonhuman)
Wistar rats, calf thymus
DNA
+ DNA binding
in liver, kidney,
lung, and
stomach
Lattanzi et al.
(1988)
DNA adducts
not identified
BALB/c mice, calf
thymus DNA
+ DNA binding
in liver, kidney,
lung, and
stomach
Lattanzi et al.
(1988)
DNA adducts
not identified
Fungi
Mitotic
recombination
S. cerevisiae D3
- (±S9)
Simmon and
Kauhanen
(1978)
S. cerevisiae D4
- (±S9)
Weeks et al.
(1979)
S. cerevisiae D7
- (±S9)
Bronzetti et al.
(1990, 1989)
Aneuploidy
Aspergillus nidulans PI
diploid
Crebelli et al.
(1995,1992,
1988)
46 DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-15. Summary of genotoxicity studies of HCE
Test system
Genetic endpoint
Strain/cells
Results
Reference
Comments
In vivo tests
Rat
Rat liver foci
Osborne-Mendel
- (initiation)
+ (promotion)
Milman et al.
(1988)
Initiation or
promotion
protocols
DNA adduct
formation
(nonhuman)
Wistar rats
Weakly + DNA
binding in liver
Lattanzi et al.
(1988)
Adducts not
identified
Mice
Micronucleus
induction
CD-I mice
-
Crebelli et al.
(1999)
Replicative DNA
synthesis (RDS)
B6C3Fi mice
+
Yoshikawa
(1996);
Miyagawa et al.
(1995)
Hepatic cell
proliferation
BALB/c mice
Moderately +
DNA binding in
liver
Lattanzi et al.
(1988)
Adducts not
identified
Human
lymphocytes
Isolated human
lymphocytes
+ (+S9)
Tafazoli et al.
(1998)
DNA strand
breaks
Human lymphocyte
cultures
-
Tafazoli et al.
(1998)
Comet assay
Drosophila
Mitotic
recombination
Drosophila
Weakly +
Vogel and
Nivard (1993)
Eye mosaic
assay
Using the standard Ames assay for reversion of S. typhimurium histidine tester strains
(TA1535, TA1537, TA1538, TA98, and TA100), Simmon and Kauhanen (1978) found HCE to
be nonmutagenic at concentrations of 5,000 or 10,000 jag HCE/plate (purity not specified), both
in the absence and presence of an exogenous Aroclor 1254-stimulated rat liver S9 metabolic
activation system. HCE was reported to be slightly toxic at the 10,000 [j,g/plate concentration in
the absence of the S9 mix. Weeks et al. (1979) also reported negative results using the same
tester strains, test protocol, solvent, and metabolic activation system over a concentration range
of 0.1-500 jag HCE/plate (purity 99.8%). Further, as a part of NTP's mutagenicity screening
program, HCE was dissolved in dimethylsulfoxide (DMSO) and tested in two independent trials
in two separate laboratories over a collective concentration range of 1-10,000 [j,g/plate. HCE
was negative for induction of reverse mutation in S. typhimurium (tester strains TA1535,
TA1537, TA98, and TA100), with and without S9 metabolic activation (NTP, 1989; Haworth et
al., 1983). Finally, HCE (purity >97%) was reported to be negative in several Ames tester
strains, both with and without S9 from the Aroclor 1254-induced livers of both sexes of Osborne
Mendel rats and B6C3Fi mice (Milman et al., 1988).
Using a different S. typhimurium indicator strain, BA13, in a liquid preincubation
protocol of the Ara test, Roldan-Arjona et al. (1991) found HCE to be negative. This bacterial
assay examines the ability of an agent to induce forward mutations from L-arabinose sensitivity
to resistance, and theoretically might be expected to detect a broader range of mutagens than
47
DRAFT - DO NOT CITE OR QUOTE
-------
reverse-mutation assays. HCE (purity 98%) was dissolved in DMSO and tested over a
concentration range of 1.5-30.0 [jmol/plate (355-7,102 |ig/plate), both with and without rat liver
S9 metabolic activation. Of the 16 chemicals tested in this study, HCE was the only one that did
not demonstrate any toxicity, which the authors speculated was probably related to its low
solubility in water. HCE (purity not specified) was negative when assayed in the umu test using
S. typhimurium tester strain TA1535/pSK1002 (Nakamura et al., 1987). This study also
employed a liquid preincubation protocol, and was conducted both with and without rat liver S9
metabolic activation up to a concentration of 42 [j,g/mL (the solvent, either water or DMSO, was
not specified for individual test agents). Although the available data indicate that HCE is not
mutagenic to Salmonella, Legator and Harper (1988) suggested that this may be related to
inadequate reductive dechlorination (i.e., if HCE is activated by metabolic pathways not present
in the in vitro system used).
HCE was assayed for its ability to induce mitotic recombination in tester strain D3 of the
yeast S. cerevisiae (Simmon and Kauhanen, 1978). No significant activity over a concentration
range of 0.1-5.0% HCE (1-50 mg/mL; purity not specified), with or without exogenous rat liver
S9 metabolic activation, was observed. In addition, negative findings for HCE were reported by
Weeks et al. (1979) using the S. cerevisiae D4 strain.
Bronzetti et al. (1989) evaluated HCE (purity not specified) for mitotic gene conversion
at the trp locus and reverse point mutation at the ilv locus in the S. cerevisiae D7 tester strain.
Two-hour liquid suspension exposures were conducted both on a logarithmic growth phase
culture having high levels of CYP450 metabolizing enzymes and on stationary growth phase
cultures either with or without exogenous liver S9 mix. Exposures were from 5 to 12.5 mM
(1.2-3.0 mg/mL) and were reportedly limited by solubility. HCE was inactive for both gene
conversion and reverse mutation in stationary cultures with or without S9, and for reverse
mutation in the logarithmic culture. However, statistically significant (p < 0.05-0.001) increases
in revertant frequency of more than twofold over background were observed at every
concentration (Bronzetti et al., 1989).
The ability of various halogenated hydrocarbons to induce aneuploidy in the PI diploid
strain of the mold Aspergillus nidulans has been reported (Crebelli et al., 1995, 1992, 1988).
Liquid suspension exposures (3 hours) to concentrations of 0.0025-0.04%) HCE (0.005-
0.84 mg/mL; purity >98%) resulted in survival rates of 100-48%). Exposure to these
concentrations did not induce mitotic malsegregation of chromosomes.
A number of studies have evaluated the effects of in vivo and in vitro HCE exposures on
various cytogenetic endpoints in higher organisms (Crebelli et al., 1999; Tafazoli et al., 1998;
Doherty et al., 1996; Vogel and Nivard, 1993; NTP, 1989; Galloway et al., 1987). Crebelli et al.
(1999) utilized the mouse bone marrow micronucleus test to investigate the in vivo induction of
micronucleated polychromatic erythrocytes (MNPCEs) by 10 aliphatic halogenated
hydrocarbons, including HCE. CD-I mice (5/sex/concentration) were injected i.p. with HCE
48
DRAFT - DO NOT CITE OR QUOTE
-------
doses of 2,000 or 4,000 mg/kg (purity >98%), representing approximately 40 and 70-80% of the
LD50, respectively. Animals were sacrificed and bone marrow cells were harvested at 24 and
48 hours post-treatment. At least 5,000 polychromatic erythrocytes/animal were analyzed. HCE
treatment induced clinical signs of general toxicity, but no significant increases in the frequency
of MNPCEs were noted in any treated group.
Vogel and Nivard (1993) utilized a Drosophila eye mosaic assay to monitor genetic
damage in somatic cells, predominantly interchromosomal mitotic recombination, caused by the
exposure of larvae to various chemicals. In the case of HCE (3% ethanol solvent; purity not
specified), adult flies of the C-l cross were permitted to lay eggs for 3 days on food
supplemented with 10 mM HCE. Examination for light spots in the normally colored eyes of the
resulting flies revealed what the authors classified as a weak positive response for HCE—a
reproducible increase of not more than a doubling of the spontaneous frequency at a dose
associated with toxicity. The authors suggested that the effect was unspecific and not likely
related to genotoxicity.
HCE was evaluated for its ability to induce MN and DNA damage in isolated human
lymphocytes from two donors (Tafazoli et al., 1998). Lymphocytes were exposed for 3 hours in
the presence of exogenous metabolic activation (S9 mix), or for 48 hours in the absence of S9.
Results using cells from one donor ("A") were reported for HCE (purity >99%) for exposures of
0.05-1.00 mM (0.012-0.24 mg/mL) in the presence of S9. Neither toxicity nor MN induction
was evident. Cells from the other donor ("D") were exposed to higher HCE concentrations of 1-
16 mM (a saturating concentration; 0.24-3.79 mg/mL), both with and without S9. Toxicity
(measured as a significant decrease in the relative division index) was still not observed, but
statistically positive results for percent cells with MN were recorded at HCE concentrations of
1 and 8 mM (0.24 and 1.89 mg/mL, respectively) in the absence of S9 (12 and 11%, respectively,
versus a control value of 5.5%,p < 0.05), and at 1 mM (0.24 mg/mL) in the presence of S9
(19.8%) versus a control value of 9%,p< 0.01). In the second part of the study, lymphocyte
cultures exposed to test agents for 3 hours with and without S9 were assessed for DNA damage
(breaks, alkali-labile sites) using the Comet assay. HCE did not affect the measured DNA
damage parameters (tail length, fraction of total cellular DNA in the tail, and tail moment).
Doherty et al. (1996) examined in vitro induction of MN by HCE in three human cells
lines with metabolic competence; lymphoblastoid AHH-1 (native CYP1A1 activity), MCL-5
(transfected with cDNAs encoding human CYP1A2, 2A6, 3A4, 2E1, and microsomal epoxide
hydrolase), and h2El (with cDNA for human CYP2E1). Exponentially growing cultures were
exposed for approximately one cell cycle (18 hours for AHH-1, 24 hours for MCL-5 and h2El)
to 0, 0.01, 0.05, or 0.1 mM HCE (purity not specified; 0, 0.002, 0.012, or 0.024 mg/mL,
respectively), and then processed for scoring of kinetochore-positive and -negative MN. No MN
formation was observed in any of the three cell lines in response to HCE exposure. However,
49
DRAFT - DO NOT CITE OR QUOTE
-------
MN induction was enhanced by exposure to an HCE metabolite, PERC, in h2El and MCL-
5 cells.
Induction of CAs and SCEs in cultured Chinese hamster ovary (CHO) cells exposed to
HCE was investigated as part of an NTP screening program for genotoxicity (NTP, 1989;
Galloway et al., 1987). Concentrations for analysis were selected based on observations of cell
confluence and mitotic cell availability. HCE concentrations (purity >99%) ranged from 10 to
1,000 [j,g/mL (0.01-1.0 mg/mL). For both endpoints, linear regression was used to test for
dose-response trends. For individual doses, induction of CA was considered significant if
p values (adjusted by Dunnett's method to correct for multiple dose comparisons) relative to
controls were <0.05, while increases of SCEs/chromosome >20% over control values were
considered significant. For CAs, the durations of exposure were 8-10 hours in the absence of S9
metabolic activation and 2 hours in the presence of S9. For induction of SCEs, exposure
durations were 26 hours without S9 and 2 hours with S9 (followed by 24-hour incubation
without HCE). CAs were not observed in response to HCE exposure without S9. In the
presence of S9, the first study (0.15-0.50 mg/mL HCE) did not induce CAs; however, the second
study (0.20-0.40 mg/mL HCE) was judged equivocal due to a positive response at the low dose
(15.0% cells with CA versus 5.0% for the DMSO control). HCE (0.010-0.33 mg/mL) did not
induce SCE in the absence of S9; however, positive results were obtained in the presence of S9
(0.10-1.0 and 0.40-1.0 mg/mL HCE).
In vitro cell transformation studies were conducted to understand the effect of HCE in the
process of chemical carcinogenesis. In the absence of exogenous metabolic activation, a 3-day
exposure to concentrations of HCE (purity >97%) from 0.16 to 100.0 |ig/mL (0.00016-
0.100 mg/mL) failed to induce morphological cell transformation in BALB/c-3T3 cells, as
measured by the incidence of Type III foci (characterized by the authors as an aggregation of
multilayered, densely stained cells that are randomly oriented and exhibit a criss-cross array at
the edge of the focus) (Milman et al., 1988; Tu et al., 1985). Milman et al. (1988) also examined
the capacity of HCE to initiate and promote tumors in a rat liver foci assay. To assess initiation
potential, 24 hours after partial hepatectomy, 10 young adult male Osborne-Mendel rats received
the MTD of HCE in corn oil by gavage. Six days later, the animals received a 0.05% dietary
exposure to the tumor promoter phenobarbital for 7 weeks. Following sacrifice, livers were
examined histopathologically for foci containing GGT, a putative preneoplastic indicator. To
assess promotion potential, animals were initiated 24 hours after partial hepatectomy with an i.p.
injection of 30 mg of the tumor initiator, diethylnitrosamine (DEN). Six days later, the animals
received the MTD of HCE in corn oil by gavage, 5 days/week for 7 weeks. The animals were
sacrificed and their livers were examined for the presence of GGT-positive foci. In these assays,
HCE failed to demonstrate any initiating activity, but did show significant (p < 0.05) promoting
capability (4.38 ± 1.04 GGT+ foci/cm2, versus 1.77 ± 0.49 for the corn oil control).
50
DRAFT - DO NOT CITE OR QUOTE
-------
Yoshikawa and colleagues reported on the activity of HCE and other putative
nongenotoxic (i.e., Ames-negative) mouse hepatocarcinogens in an in vivo-in vitro hepatocyte
replicative DNA synthesis (RDS) assay (Yoshikawa, 1996; Miyagawa et al., 1995). Groups of
4-5 male B6C3Fi mice were exposed to single gavage doses of 0, 1,000, or 2,000 mg/kg HCE
(purity not specified). The hepatocytes were prepared at 24, 39, or 48 hours after exposure. The
1,000 mg/kg HCE-treated hepatocytes prepared 39 hours after exposure yielded a positive mean
RDS response of 1.21 ± 0.46% (the investigators noted that an RDS incidence rate of 0.4% for
any dose group was considered a positive response for the chemical). The remaining HCE
groups were negative with mean responses of 0.15-0.35%), while the solvent control mean was
0.26 ±0.17%.
4.5.2. In Vitro and Ex Vivo Studies Using Isolated Target Tissues/Organs or Cells
A study using a rat liver foci assay (Milman et al., 1988, Story et al., 1986) found that
HCE was a tumor promoter rather than an initiator. In vitro and in vivo assays were conducted
to assess the ability of HCE to bind to DNA, ribonucleic acid (RNA), and protein in several
mouse and rat tissues (Lattanzi et al., 1988). This study reported that binding of radiolabeled
carbon to DNA, RNA, and protein was observed following [14C]-HCE administration in both in
vitro and in vivo assays in mice and rats (Lattanzi et al., 1988), suggesting that either HCE or its
metabolites may bind to these macromolecules. The role of this binding in mediating HCE-
induced toxicity was not further evaluated.
Story et al. (1986) and Milman et al. (1988) conducted a rat liver foci assay to assess the
initiation and promotion potential of HCE, along with eight other chlorinated aliphatics. Male
Osborne-Mendel rats (10 rats/group) were given partial hepatectomies and then administered the
initiation protocol or the promotion protocol. In the initiation protocol, the rats were
administered by gavage the MTD of 2.1 mmol/kg (497 mg/kg) HCE (purity 98%>), followed
6 days later with 7 weeks of phenobarbital in the diet at 0.05%>. Control rats were administered
by gavage either corn oil (negative control) or 30 mg/kg DEN (positive control), followed by the
phenobarbital treatment. In the promotion protocol, rats were dosed with 30 mg/kg DEN by i.p.
injection, followed 6 days later with the MTD of 497 mg/kg HCE, 5 days/week for 7 weeks.
Phenobarbital was administered (in the same manner as HCE) as a positive control. Control rats
were either administered DEN or water, followed by corn oil for the promotion phase. Livers
were removed and stained for GGT activity. Results from the initiation protocol were negative,
with only a small number of GGT+ foci (1.0 foci/cm2 at most). However, initiation with DEN
followed by HCE or phenobarbital resulted in statistically significant increases in GGT+ foci
(Table 4-16). Absolute and relative liver weights were increased by HCE in the promotion
protocol. These results indicate that HCE is not an initiator in the rat liver foci assay, but is
capable of promotion.
51
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-16. Number of enzyme-altered foci in rat liver of the promotion
protocol
Promotion treatment
Total number of foci/cm2
+ DEN initiation
- DEN initiation
HCE
4.4 + 1.0a
0.1+0.2
Phenobarbital
3.9 + 1.03
0.3+0.2
Corn oil
1.7+0.5
0.2+0.2
aStatistically different from DEN + corn oil control group, p < 0.05
Sources: Milman et al. (1988); Story et al. (1986).
Lattanzi et al. (1988) conducted in vivo and in vitro assays to assess the binding of
[14C]-HCE (specific activity 14.6 mCi/mmol, radiochemical purity 98%) to nucleic acids in
various organs from mice and rats following metabolic activation. For the in vivo studies,
6 male Wistar rats and 12 male BALB/c mice were injected i.p. with 127 j_iCi/kg HCE (purity
98%). The animals were fasted and sacrificed 22 hours after injection. The organs (liver,
kidney, lung, and stomach) were removed, pooled, and processed to obtain DNA, RNA, and
proteins. The in vitro studies examined microsomal and cytosolic fractions from these same
organs. The incubation mixture included 2.5 jaCi [14C]-HCE, 1.5 mg calf thymus DNA or
polyribonucleotide, 2 mg microsomal proteins (plus 2 mg NADPH), and/or 6 mg of cytosolic
proteins (plus 9.2 mg glutathione [GSH]). Coenzymes were not utilized with the controls.
Measures for binding to macromolecules were determined by the presence of radiolabeled
carbon from [14C]-HCE in the DNA, RNA, and protein. The presence of radiolabeled carbon
may indicate HCE binding directly to the macromolecules or incorporation of radiolabeled
carbon from intermediate metabolites into these macromolecules.
In vivo binding data for HCE are presented in Table 4-17. Binding to macromolecules
was interpreted by the presence of radiolabeled carbon; however, HCE-specific metabolites were
not measured. In both rats and mice, binding values (in pmol HCE/mg) for RNA were
consistently much greater than those for DNA or protein. Greater binding to RNA was observed
in the kidneys of rats and mice (5-28 times greater) compared with the binding measured in the
livers, lungs, and stomachs. DNA exhibited the lowest amount of HCE binding. Species
differences were evident for all three macromolecule types (DNA, RNA, and protein) with the
mouse exhibiting much higher levels (9 times greater) of covalent binding for DNA in the liver
than the rat. The binding was 2 and 3 times greater for mice than rats with RNA and protein,
respectively, from the liver. The binding to DNA was similar between species, but slightly
greater in mice, for the kidney, lung, and stomach analyses. According to classifications
reported by Lutz (1986, 1979), the covalent binding index (CBI) values calculated on rat and
mouse liver indicate weak (rat) to moderate (mice) oncogenic potency in HCE-treated rodents.
52
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-17. In vivo covalent binding of [14C]-HCE to DNA, RNA, and
proteins from rat and mouse organs
(pmol/mg)
Liver3
Kidney3
Lung3
Stomach3
Rat
Mouse
Rat
Mouse
Rat
Mouse
Rat
Mouse
DNA
(CBIb)
0.43 +0.05c
(15.1)b
3.92 + 0.20d
(140)b
0.42
0.50
0.14
0.35
0.26
0.37
RNA
46.59+ 7.23c
108.08 + 21.57d
232.94
564.98
15.55
60.10
8.33
21.04
Protein
4.94 + 1.14°
14.99 + 0.83d
2.59
4.91
0.89
3.42
0.80
2.41
aData are from pooled organs from 6 male Wistar rats or 12 male BALB/c mice, except for liver (see indices).
bCBI calculated according to Lutz (1986, 1979), as cited in Lattanzi et al. (1988). Classification of CBI values for
oncogenic potency: strong, in the thousands; moderate, in the hundreds; weak, in the tens; and below one for
nongenotoxic oncogens.
cMean + SE of six individual values.
dMean + SE of four values, each obtained from three pooled livers.
Source: Lattanzi et al. (1988).
In vitro binding data for HCE are presented in Table 4-18. Liver microsomes from rats
and mice catalyzed HCE binding to DNA at comparable levels. Kidney microsomes from rats
and mice produced statistically significantly greater amounts of HCE binding to DNA when
compared with controls. Kidney microsomes from mice had a threefold increase in HCE binding
to DNA when compared to controls, while kidney microsomes from rats had a twofold increase
in HCE binding to DNA when compared to controls. Microsomes from lung and stomach in
both species did not display increased DNA binding activity over corresponding controls in the
absence of coenzymes. Cytosolic fractions from all organs in mice and rats exhibited higher
levels of HCE binding to DNA than microsomal fractions. Mouse liver cytosols produced much
greater levels of HCE binding to DNA than rat liver cytosols. When both microsomal and
cytosolic fractions were in the incubation mixture, HCE binding to DNA was decreased for liver
and kidney. SKF 525-A, a nonspecific CYP450 inhibitor, caused a 50.5% decrease in HCE
binding to DNA (data not included in report). Lattanzi et al. (1988) stated that addition of GSH
to the microsomal fractions also resulted in inhibition of HCE binding to DNA (authors did not
include data in report). When microsomal and cytosolic fractions were heat-inactivated, HCE
binding to DNA was similar to control, providing further support that HCE binding to DNA is
enzymatically catalyzed. This study provided evidence that HCE is metabolized by microsomal
CYP450 enzymes and cytosolic GSH transferases, and that DNA binding may be increased
following HCE metabolism.
53
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-18. In vitro binding of [14C]-HCE to calf thymus DNA mediated by
microsomal and/or cytosolic phenobarbital-induced fractions of rat and
mouse organs
Microsomes + NADPH
Cytosol + GSH
Microsomes + cytosol
(+ NADPH, + GSH)
Rat
Mouse
Rat
Mouse
Rat
Mouse
Liver
Standard3
90.83 + 5.31b
105.39+ 7.80b
195.51 ± 2144°
346.17 + 18.91b
95.06+ 6.29c
133.44 +2.42a
Controls3
55.19 + 4.90
46.96+4.19
92.96 + 26.07
128.56 + 8.92
52.85 + 12.93
99.84 + 8.06
Kidney
Standard
395.84+ 78.58c
78.86 +6.85c
246.85 +35.39c
251.42+ 45.38c
247.99+ 3.40b
ND
Controls
136.26 + 9.04
39.12 + 5.34
88.82 + 30.91
81.91+9.93
144.61 + 12.86
ND
Lung
Standard
125.60 + 22.37
87.37 +7.90
126.65 + 16.84b
168.52 ± 19 41b
234.26 +28.35b
ND
Controls
121.13 + 16.54
86.10 + 3.27
40.23 +7.34
60.44 + 21.90
56.27 + 5.32
ND
Stomach
Standard
94.41 + 14.38
47.67 + 17.00
289.58 +31.19b
228.74 + 20.42b
76.79 + 5.34b
ND
Controls
93.20 + 15.24
47.12 + 11.20
130.51+4.01
51.52+6.20
44.77+2.28
ND
aData (total DNA binding in pmol/mg) are reported as mean + SE of three values; ND, not determined. Controls
were conducted in the absence of coenzymes.
Statistically different from control, p < 0,01,
cStatistically different from control, p < 0,05.
Source: Lattanzi et al. (1988).
4.5.3. Structure Activity Relationships
Several studies were conducted with the objective of defining structure activity
relationships (SARs) of halogenated hydrocarbons and toxicity. NTP (1996) defined a group of
chlorinated ethanes that resulted in hyaline droplet nephropathy in male F344/N rats and a group
of halogenated ethanes that resulted in renal toxicity in the absence of hyaline droplet
nephropathy. In a series of studies, Crebelli et al. (1995, 1992, 1988) evaluated chlorinated and
halogenated hydrocarbons for their ability to induce chromosome malsegregation, lethality, and
mitotic growth arrest in the mold A. nidulans.
NTP (1996) conducted a 21-day oral toxicity study with halogenated ethanes in male
F344/N rats. Chemicals under investigation were 1,1,1,2-tetrachloroethane, 1,1,2,2-tetrachloro-
ethane, pentachloroethane, 1,1,2,2-tetrachloro-1,2-difluoroethane, 1,1,1 -trichloro-2,2,2-trifluoro-
ethane, l,2-dichloro-l,l-difluoroethane, 1,1,1-trichloroethane, 1,1,1,2-tetrabromoethane,
1,1,2,2-tetrabromoethane, pentabromoethane, and HCE (purity >98%). Groups of five male
rats/dose were administered 0.62 or 1.24 mmol/kg-day of the halogenated ethane (for HCE,
146 and 293 mg/kg-day, respectively). Increased kidney weights and evidence of renal toxicity
were observed in many of the rats administered halogenated ethanes; however, this was not
54
DRAFT - DO NOT CITE OR QUOTE
-------
always coincident with hyaline droplet nephropathy. Hyaline droplet nephropathy (assessed by
Mallory-Heidenhain staining, which allows for greater sensitivity in evaluating hyaline droplets
within the tubules of the kidney) was only observed in rats administered pentachloroethane,
1,1,1,2-tetrachloroethane, and HCE. RDS, indicated by PCNA labeling index, was increased in
male rats administered chemicals that induced hyaline droplet nephropathy (pentachloroethane,
1,1,1,2-tetrachloroethane, and HCE) as well as pentabromoethane and 1,1,2,2-tetrachloroethane,
compared with control rats. The increase in cell proliferation in the kidneys (as measured by the
PCNA labeling index) observed with some of the halogenated ethanes that did not induce hyaline
droplet nephropathy suggests the contribution of another toxic mechanism. NTP (1996)
concluded that the capacity to induce hyaline droplet nephropathy in male rats was restricted to
ethanes with four or more halogens, and only the chlorinated (compared with the fluorinated and
brominated) ethanes were active. This study also predicted that if hyaline droplet nephropathy is
the determining factor in the induction of renal tubule cell neoplasia, then chemicals such as
bromo- or chlorofluoroethanes would be negative for kidney neoplasia in 2-year cancer
bioassays of male rats.
Crebelli et al. (1988) evaluated three chloromethanes and eight chlorinated ethanes
(including HCE) for the induction of chromosome malsegregation in A. nidulans. Although 8 of
the 11 compounds tested provided positive results including the 3 chloromethanes and 5 out of
8 chlorinated ethanes, HCE was negative for chromosome malsegregation induction. Analyses
of relationships between biological and chemical variables indicate that the ability of a chemical
to induce chromosome malsegregation was not related to any of the chemical descriptors
examined, including molecular weight, melting point, boiling point, refractive index,
octanol/water partition coefficient, and the free energy of binding to biological receptors.
Because of the similarity of the chemical descriptors between the positive chlorinated ethanes,
aside from 1,1,1-trichloroethane which was negative, the authors argue against a previous
hypothesis that nonspecific interactions with hydrophobic cellular structures is the mechanism of
aneuploidy induction (Onfelt, 1987).
Crebelli et al. (1992) evaluated the ability of 24 chlorinated aliphatic hydrocarbons to
induce chromosome malsegregation, lethality, and mitotic growth arrest in the mold, A. nidulans.
Data were combined with previous data on 11 related compounds (Crebelli et al., 1988) to
generate a database for quantitative structure-activity relationship (QSAR) analysis. Physico-
chemical descriptors and electronic parameters for each chemical were included in the analysis.
Out of the 24 chemicals, 19 were negative for the induction of chromosome malsegregation;
5 chemicals produced reproducible increases in the frequency of euploid whole chromosome
segregants. HCE was negative for the induction of chromosome malsegregation. QSAR
analyses on these 35 chlorinated aliphatic hydrocarbons indicate that toxicity, such as the
induction of lethality, is primarily related to steric factors (the spatial orientation of reactive
centers within a molecule) and measures of the volume occupied by an atom or functional group
55
DRAFT - DO NOT CITE OR QUOTE
-------
(molar refractivity). Measures of molar refractivity are a function of temperature, index of
refraction, and atmospheric pressure. Mitotic growth arrest was also primarily related to molar
refractivity. However, aneugenic activity was related to both molar refractivity and electronic
factors, such as the ease in accepting electrons (described by density and the energy of the lowest
unoccupied molecular orbital).
These QSAR studies (Crebelli et al., 1992, 1988) were expanded to include 20 additional
halogenated hydrocarbons (Crebelli et al., 1995). Chemicals in this study were also assayed for
lipid peroxidation in rat liver microsomes, and the authors reported that a partial coincidence was
found between the ability of a chemical to initiate lipid peroxidation and to disturb chromosome
segregation at mitosis. This updated study concluded that electronic and structural parameters
that determine the ease of homolitic cleavage of the carbon-halogen bond play a primary role in
the peroxidative properties of haloalkanes.
4.6. SYNTHESIS OF MAJOR NONCANCER EFFECTS
4.6.1. Oral
Table 4-19 summarizes the oral toxicity studies that have been reported in laboratory
animals. The primary noncancer effects observed in these studies include decreased body weight
or body weight gain, increased absolute and relative kidney weights, increased absolute and
relative liver weights, various effects associated with renal tubule toxicity in the kidney, and
hepatocellular necrosis. Developmental studies in rats did not consistently demonstrate fetal
effects, especially in those cases where maternal toxicity was absent.
56
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-19. Oral toxicity studies for HCE
Species
Dose
(mg/kg-d)/
duration
NOAEL
(mg/kg-d)
LOAEL
(mg/kg-d)
Effect
Reference
F344/N rats,
male (5/dose)
0, 100, 320 or
1,000 by oral;
12 d
100
320
Increased liver and kidney
weights; liver degeneration
and necrosis; tubular
nephrosis and
nephrocalcinosis
NTP (1996)
F344/N rats
(5/sex/dose)
0, 34, 67, 134,
268, or 536 by
gavage; 16 d
Male: not
established
Female: 67
Male: 34
Female: 563
Male: kidney effects in all
dose groups
Female: decreased body
weight
NTP (1989)
F344 rats
(10/sex/dose)
0, 146, or 62 by
gavage; 21 d
Not established
146
Increased kidney weight,
nephropathy (hyaline
droplets, tubule regeneration,
granular casts); effects on
urinalysis parameters
NTP (1996)
F344/N rats
(10/sex/dose)
0, 113, or 536 by
gavage; 13 wks
Not established
113
Tubular nephropathy in both
sexes
NTP (1989)
F344 rats
(10/sex/dose)
0, 360, or 62 by
diet; 16 wks
Male: 1
Female: 15
360
Tubular nephropathy in both
sexes
Gorzinski et
al. (1985)
F344/N rats
(50/sex/dose)
0, 7, or 227 by
gavage; 78 wks
Not established
113
Tubular nephropathy in both
sexes
NTP (1989)
B6C3Fi mice
(50/sex/dose)
0, 360, or 722 by
gavage; 91 wks
Not established
360
Tubular nephropathy in both
sexes
Weeks et al.
(1979)
F344/N rats
(21/dose)
0, 7, or 500 by
gavage; 103 wks
Not established
Male: 7
Female: 57
Male: tubular nephropathy;
renal tubular hyperplasia
Female: tubular nephropathy
NTP (1989)
Pregnant
Sprague-Dawley
rats (22/dose)
0, 50, 100, or 500
by gavage on
GDs 6-16
Maternal: 100
Maternal: 500
Maternal: body weight
decreased; increased mucus
in nasal turbinates;
subclinical pneumonitis
Fetal: no elfects
Weeks et al.
(1979)
Pregnant Wistar
rats (21/dose)
0, 56, 167, or 500
by gavage on
GDs 7-17
Maternal: 56
Developmental:
167
Maternal: 167
Developmental:
500
Maternal: decreased weight
gain and motor activity
Fetal: reduced body weight
increased incidence of
skeletal variations; decreased
ossification
Shimizu et
al. (1992)
Acute and short-term toxicity tests in animals reported liver necrosis and tubular
nephrosis in male rabbits (Kinkead and Wolfe, 1992; Weeks et al., 1979; Weeks and Thomasino,
1978), and evidence of kidney effects such as nephropathy with hyaline droplet formation and
tubular cell regeneration in male rats (NTP, 1996, 1989). Female rats in short-term toxicity tests
displayed only decreased body weights at the LOAEL of 563 mg/kg-day with a NOAEL of 281
mg/kg-day (NTP, 1989). Oral LD50 values in rats ranged from 4,460 to 7,690 mg/kg (Weeks et
al., 1979).
57
DRAFT - DO NOT CITE OR QUOTE
-------
4.6.1.1. Nephrotoxicity
Two short-term studies in F344 rats (NTP, 1996; 1989; 21- and 16-day studies,
respectively) reported nephrotoxic effects at all administered doses in male rats. The formation
of hyaline droplets accompanied by cell regeneration and eosinophilic granular casts was
observed in the renal tubules of male rats administered 140-563 mg/kg-day HCE (NTP, 1989).
Female rats did not exhibit any renal toxicity. In the 21-day study by NTP (1996), male rats
exhibited increased absolute and relative kidney weights, tubular regeneration and granular casts,
and increased labeling index in kidneys at doses of 146 and 293 mg/kg-day HCE. Tubular
nephrosis, and to a minimal degree, tubular nephrocalcinosis were observed in the kidney of
male New Zealand White rabbits administered 320 and 1,000 mg/kg-day (but not 100 mg/kg-
day) HCE (Weeks et al., 1979). Compared with rabbits, the rats were more sensitive to renal
effects induced by HCE. A gender-specific response was demonstrated in the male rats (NTP,
1989). However, the use of only male rats (NTP, 1996) and male rabbits (Weeks et al., 1979) in
the other two studies makes it difficult to evaluate if the observed renal effects were gender-
specific.
Subchronic exposure (13 weeks) resulted in kidney effects including hyaline droplet
formation, tubular regeneration, and tubular casts in male F344/N rats administered HCE doses
of 34-536 mg/kg-day (NTP, 1989). Males in the 536 mg/kg-day dose group also exhibited renal
papillary necrosis and degeneration and necrosis of renal tubule epithelium. Female rats did not
display these kidney effects. These results suggest a sex-specific difference in HCE toxicity.
Another study (Gorzinski et al., 1985) in F344 rats reported slight hypertrophy and dilation of
the renal tubules in males and renal tubule atrophy and degeneration in male and female rats.
Evidence of kidney effects in female rats consisted of very slight renal tubular atrophy and
degeneration observed histopathologically at the highest dose tested. EPA considered the
NOAEL and LOAEL for male rats as 7 and 15 mg/kg-day, respectively, while the corresponding
values in the females were 15 and 62 mg/kg-day, indicating greater sensitivity of the males to the
renal effects of HCE. These data and tissue distribution information (see Section 3.2
Distribution) show that the male kidney accumulated higher HCE concentrations than the female
kidney, indicating that the kidney is the primary target organ following oral exposure to HCE
and that there are potential gender differences in the distribution and metabolism of HCE.
Consequently, male rats are likely more sensitive to the nephrotoxicity of HCE than female rats.
Additionally, Gorzinski et al. (1985) is the only study of either short-term or subchronic duration
to report renal effects in female rats.
Chronic toxicity tests were conducted by NTP on F344/N rats and by NCI on Osborne-
Mendel rats and B6C3Fi mice (NTP, 1989; NCI, 1978). NTP (1989) administered much lower
doses of HCE (7 and 14 mg/kg-day in males; 57 and 114 mg/kg-day in females) to the F344 rats
compared with the Osborne-Mendel rats (113 and 227 mg/kg-day) in the NCI (1978) study. In
the NTP (1989) chronic study, nephropathy (characterized as tubular cell degeneration and
58
DRAFT - DO NOT CITE OR QUOTE
-------
regeneration, dilation and atrophy, glomerulosclerosis, interstitial fibrosis, and chronic
inflammation) was observed in both male and female rats. In the case of the male rats, the
response was roughly equivalent across the control and treated groups, with nephropathy in more
than 94% of animals. The high incidence of nephropathy observed in control rats was likely a
result of a spontaneous syndrome known as chronic progressive nephropathy (CPN) that is
associated with aged rats, especially F344 and Osborne-Mendel strains (see Section 4.7.3.2.1 for
additional discussion). To examine the effects of chronic HCE exposure separate from CPN, the
nephropathy incidence in terms of severity was evaluated. The severity was increased in the
treated male rats compared with the controls. In considering severity, the increases in incidences
of nephropathy in males (that were of moderate or marked severity) were 18/50 (36%), 24/50
(48%>), and 30/50 (60%>) in the control, 7, and 14 mg/kg-day dose groups, respectively. In
females, both the incidence (44% of controls and approximately 84% of treated) and severity of
nephropathy were dose-related. When considering the severity, incidences of nephropathy in
females (that were of mild or moderate severity) were 12/50 (24%), 25/50 (50%), and 32/50
(64%) in the control, 57, and 114 mg/kg-day dose groups, respectively.
Dose-related increases (30 and 64% at 7 and 14 mg/kg-day, respectively) in linear
mineralization of the renal papillae and treatment-related increases (14% at 7 and 14 mg/kg-day)
in hyperplasia of pelvic transitional epithelium in the kidney were observed in the male rats. In
females, an increased incidence of mineralization was only noted at the low dose (44% at
57 mg/kg-day compared with 28% in controls). The low dose for the females was 8 times
greater than that for the males, yet the signs of nephropathy were more severe in the males.
In the NCI (1978) study, Osborne-Mendel rats of both sexes displayed chronic
inflammatory kidney lesions in both control and treated groups, although tubular nephropathy
(characterized by degeneration, necrosis, and the presence of large hyperchromatic regenerative
epithelial cells) was observed only in the HCE-exposed male and female rats. There were
dose-related increases in incidences of nephropathy in males (45 and 66%, respectively) and
females (15 and 59%, respectively) administered 113 and 227 mg/kg-day HCE. The chronic
toxicity test in B6C3Fi mice (NCI, 1978) is the only study conducted in this species. Male mice
experienced low survival in the control and 360 mg/kg-day (low-dose) groups. Chronic kidney
inflammation was observed in 67 and 80% of males in the vehicle and untreated control groups,
respectively, as well as in 66 and 18% of the 360 and 722 mg/kg-day HCE males, respectively.
The report did not provide an explanation for the large response in the control and low-dose mice
and the relatively small response in the high-dose group. Female mice exhibited chronic kidney
inflammation only in vehicle controls (15%) and the high-dose group (2%). Tubular
nephropathy was observed in both dose groups of both sexes at high incidences (92-100%), and
was characterized by degeneration of convoluted tubule epithelium with some hyaline casts.
Enlarged dark staining regenerative tubular epithelium was also observed, with the kidney
exhibiting infiltration of inflammatory cells, fibrosis, and calcium deposition. The response in
59
DRAFT - DO NOT CITE OR QUOTE
-------
the treated male and female mice compared with the absence of nephropathy in the controls
suggests that the doses used in this study were too high.
The available information for HCE-induced nephropathy in rats, mice, male rabbits, and
sheep indicates that the male rat is the most sensitive sex/species to the renal toxicity of HCE.
Limited, if any, information is available for species other than the rat; however, the doses that
elicited toxic responses in mice (NCI, 1978), male rabbits (Weeks et al., 1979), and sheep
(Fowler, 1969) were at least 45-fold greater than the lowest dose (7 mg/kg-day; NTP, 1989) that
induced a statistically significant response in rats.
4.6.1.2. Hepatotoxicity
Short-term studies in rats (NTP, 1996), male rabbits (Weeks et al., 1979), and sheep
(Fowler, 1969) reported hepatotoxicity at doses approaching >300 mg/kg-day. Male F344 rats
exhibited significantly increased relative liver weights at the highest dose of 293 mg/kg-day.
AST and NAG serum activities were also significantly higher than in controls. These effects
were not observed at 146 mg/kg-day HCE (NTP, 1996). Liver degeneration and necrosis,
including fatty degeneration, coagulation necrosis, hemorrhage, ballooning degeneration,
eosinophilic changes, and hemosiderin-laden macrophages and giant cells were observed in male
New Zealand White rabbits administered 320 and 1,000 mg/kg-day HCE (but not
100 mg/kg-day), increasing in severity with increasing dose. Sheep given single oral doses of
500-1,000 mg/kg of HCE exhibited plasma levels of GDH, SDH, and OCT that were increased
twofold or more than levels in controls, indicating reduced hepatic function.
Effects in the liver of animals treated with HCE were observed in male and female rats in
two subchronic studies (NTP, 1989; Gorzinski et al., 1985). Liver weight increased in a
dose-related fashion from the lowest dose (34 mg/kg-day) to the highest (536 mg/kg-day).
Females were more sensitive than males; severity and statistical significance increased in
females at doses lower than those eliciting toxicity in male rats. Hepatocellular necrosis was
noted in females at doses ranging from 134 to 156 mg/kg-day and in males at the two highest
doses, 268 and 536 mg/kg-day (NTP, 1989). Gorzinski et al. (1985) reported slight swelling of
hepatocytes in control and treated males, although there were dose-related increases in
incidences of swelling at the two highest doses (15 and 62 mg/kg-day). Other than a statistically
significant increase (5%) in liver weight at 62 mg/kg-day HCE, the females were not affected.
This is in contrast to the hepatocellular effects noted in female rats in the NTP study (NTP,
1989). However, the highest dose used by Gorzinski et al. (1985), 62 mg/kg-day, is below the
67 mg/kg-day NOAEL for females of the NTP (1989) study, indicating that sufficient doses may
not have been reached in the Gorzinski et al. (1985) study to cause hepatotoxicity in female rats.
There were no liver effects observed in the animals administered HCE for chronic
durations. The range of doses in the subchronic assay (0, 34, 67, 134, 268, and 536 mg/kg-day
on F344 rats; NTP, 1989) encompassed the doses used in the chronic assays for female F344 rats
60
DRAFT - DO NOT CITE OR QUOTE
-------
(57 and 114 mg/kg-day; NTP, 1989) and Osborne-Mendel rats (113 and 227 mg/kg-day; NCI,
1978). Hepatocellular necrosis was observed in female rats in the subchronic study, but not the
chronic study. The LOAEL for female F344/N rat hepatocellular necrosis, 134 mg/kg-day, in
the subchronic study (NTP, 1989) occurred at a dose that exceeded the highest dose of the
chronic study (NTP, 1989), suggesting that a sufficiently high dose may have not been achieved
to elicit hepatocellular necrosis despite the longer exposure period. The NCI (1978) study in
Osborne-Mendel rats was conducted with doses above the LOAEL for hepatocellular necrosis in
female F344/N rats (NTP, 1989), but hepatocellular effects were not observed. Osborne-Mendel
rats may not be as sensitive to HCE-induced hepatotoxicity as F344/N rats. The only study in
mice (NCI, 1978; chronic) did not report any hepatotoxic effects other than the development of
hepatocellular tumors.
HCE-induced liver effects were only observed in animals in short-term and subchronic
studies. Female rats exhibited a greater sensitivity to liver effects as evidenced by the effects
observed at lower doses compared with males (NTP, 1989). The implications of the slight
swelling of hepatocytes in the absence of other histopathological effects at 15 and 62 mg/kg-day
in male rats (Gorzinski et al., 1985) are unknown. Rabbits (males) and sheep demonstrated
hepatic effects at doses at least fourfold greater than the lowest dose (67 mg/kg-day) that induced
a statistically significant response in female rats.
4.6.1.3. Developmental Toxicity
Two developmental studies in rats indicated that HCE induced teratogenicity in the
presence of maternal toxicity (Shimizu et al., 1992; Weeks et al., 1979). In the Shimizu et al.
(1992) study, maternal rats gavaged with 167 and 500 mg/kg HCE displayed decreased motor
activity. At the high dose, dams also exhibited piloerection and subcutaneous hemorrhage.
Fetuses of the 500 mg/kg dose displayed decreased body weight, skeletal variations such as
rudimentary lumbar ribs, and ossification effects, but no skeletal malformations were observed.
The NOAEL for this study was 56 mg/kg for the dams and 167 mg/kg for the fetuses. In Weeks
et al. (1979), maternal rats gavaged with 500 mg/kg HCE displayed pulmonary effects such as
increased incidence of mucopurulent nasal exudates, upper respiratory tract irritation, and
subclinical pneumonitis. The fetuses did not exhibit any skeletal or soft tissue anomalies. The
maternal LOAEL and NOAEL were 500 and 100 mg/kg, respectively.
4.6.2. Inhalation
Inhalation toxicity has only been evaluated in a single 6-week repeated exposure study in
multiple species performed by Weeks et al. (1979). There is some uncertainty regarding the
exposure to HCE vapor because HCE would remain a vapor only when surrounded by heated air.
However, as soon as the hot HCE vapor was mixed with room temperature air, most (but not all)
vapor in the airstream would condense into fine particles (a solid aerosol). The data from this
61
DRAFT - DO NOT CITE OR QUOTE
-------
study are summarized in Table 4-20. The study authors reported NOAELs and LOAELs for
Beagle dogs, guinea pigs, and rats of 48 ppm (465 mg/m3) and 260 ppm (2,517 mg/m3),
respectively. Neurological effects, such as tremors and ataxia, were observed in Beagle dogs and
in pregnant and nonpregnant Sprague-Dawley rats. Rats and guinea pigs exhibited reduced body
weight gain and increased relative liver weight. Male rats also displayed increased relative
spleen and testes weights. Behavioral tests were conducted in male Sprague-Dawley rats at the
same exposure concentrations, and no significant effects were observed. Overall, the
information on the inhalation toxicity of HCE is limited.
Table 4-20. Inhalation toxicity studies with HCE
Species
Concentration
(mg/m3)/durationa
NOAEL
(mg/m3)
LOAEL
(mg/m3)
Effect
Reference
Male Beagle dogs
(4/concentration)
0, 145, 465, or
2,517; 6 wks
465
2,517
Tremors, ataxia,
hypersalivation,
head bobbing, facial
muscular
fasciculations
Weeks et al.
(1979)
Male Hartley guinea
pigs (10/concentration)
0, 145, 465, or
2,517; 6 wks
465
2,517
Reduced body
weight, increased
relative liver weight
Weeks et al.
(1979)
Sprague-Dawley rats
(25/sex/concentration)
0, 145, 465, or
2,517; 6 wks
465
2,517
Males: reduced
body weight gain,
increased relative
kidney, spleen, and
testes weights
Females: increased
relative liver weight
Weeks et al.
(1979)
C. Japonica (Japanese
quail)
(20/concentration)
0, 145, 465, or
2,517; 6 wks
2,517
Not established
No effects
Weeks et al.
(1979)
Pregnant Sprague-
Dawley rats
(22/concentration)
0, 145, 465, or
2,517; GDs 6-16
Maternal: 465
Developmental:
2,517
Maternal: 2,517
Developmental:
Not established
Maternal: tremors,
decreased body
weight
Fetal: no effects
Weeks et al.
(1979)
Male Sprague-Dawley
rats (15/concentration)
0, 145, 465, or
2,517; 6 wks
465
2,517
Behavioral tests:
avoidance latency
and spontaneous
motor activity
Weeks et al.
(1979)
a145, 465, and 2,517 mg/m3 correspond to concentrations reported by Weeks et al. (1979) as 15, 48, and 260 ppm,
respectively.
4.6.3. Mode-of-Action Information
Reports on HCE-induced human health effects are limited and confounded by coexposure
to multiple solvents or other toxicants (e.g., HCE-zinc oxide smoke). Studies that observed
substantial HCE exposure in smoke bomb production workers were too small to provide
definitive conclusions on health effects.
62
DRAFT - DO NOT CITE OR QUOTE
-------
Animal studies suggest that HCE is primarily metabolized to PERC and
pentachloroethane by CYP450 enzymes of the liver, with likely subsequent metabolism to TCE.
Metabolites identified in the urine include TCA, trichloroethanol, oxalic acid, dichloroethanol,
dichloroacetic acid, and monochloroacetic acid. However, only 5% of a radiolabeled compound
was measured in the urine, indicating that all of the urinary metabolites account for a small
percentage of the dose. It is unknown whether unchanged HCE or its metabolites are responsible
for the liver and kidney toxicities observed in animal studies. Only one study attempted to assess
the extent of HCE metabolism in rats and mice and estimated that 24-29% of administered HCE
is metabolized (Mitoma et al., 1985). This study did not quantify actual metabolite
concentrations, so these estimations are of questionable accuracy.
The mode of action for HCE-induced kidney toxicity is unknown. HCE-induced
nephropathy has been observed in both sexes of rats and mice. Specifically, short-term assays in
male rats showed nephropathy characterized by hyaline droplet accumulation and increased
incidences of tubule regeneration and granular casts (NTP, 1996, 1989). Cell proliferation of
kidney sections using PCNA labeling analysis was also increased (NTP, 1996). Subchronic and
chronic animal bioassays confirmed these renal effects (NTP, 1989; Gorzinski et al., 1985; NCI,
1978). Chronic inflammatory kidney lesions and tubular nephropathy were observed in rats, and
tubular nephropathy was also observed in mice (NCI, 1978).
Some data suggest that an a2u-globulin mode of action could contribute to HCE-induced
nephropathy. However, there is insufficient evidence to conclude that the kidney effects
observed following HCE exposure (NTP, 1989) are related to an a2U-globulin mode of action for
the following reasons: (1) the lack of a2u-globulin immunohistochemical data for HCE-induced
nephrotoxicity and carcinogenicity, (2) the hyaline droplet accumulation is caused by excessive
protein load that may not be exclusively related to a2U-globulin accumulation, and (3) the
existence of renal toxicity in female rats and male and female mice indicates that the nephrotoxic
effects are not limited to an a2u-globulin-induced sequence of lesions.
It is also possible that advanced CPN, an age-related renal disease of laboratory rodents
that occurs spontaneously, may contribute to the observed nephrotoxicity following HCE
exposure. However, changes in the severity of the nephropathy were observed to be greater in
male rats exposed to HCE compared with controls, indicating that HCE exposure exacerbated
effects in the kidney. Additionally, HCE-exposed male rats demonstrated dose-dependent
increases in incidences of mineralization of the renal papillae and hyperplasia of pelvic
transitional epithelium. Neither of these effects increased in a dose-related manner in the
controls or the HCE-exposed female rats, suggesting that CPN is not solely responsible for the
nephropathy observed by NTP (1989).
The liver has been demonstrated to be a target organ in several animal species. Sheep
(Fowler, 1969) and male rabbits (Weeks et al., 1979) exhibited hepatotoxicity characterized by
clinical chemistry parameters that indicated reduced hepatic function and showed
63
DRAFT - DO NOT CITE OR QUOTE
-------
histopathological findings including hepatocellular necrosis. Subchronic studies showed
statistically significant decreases in relative and absolute liver weight (Gorzinski et al., 1985) and
statistically significant increases in relative liver weight and hepatocellular necrosis (NTP, 1989)
in female F344/N rats. Studies of TCA (a potential metabolite of HCE) indicate that free radical
generation may play a role in mediating toxicity, particularly in the liver. However, no data are
available demonstrating generation of free radicals following exposure to HCE, and it is
unknown whether unchanged HCE or its metabolites are responsible for the liver and kidney
toxicities observed in animal studies. Town and Leibman (1984) reported lipid peroxidation (as
indicated by a statistically significant increase in the formation of malondialdehyde and
conjugated dienes) following treatment with HCE (8 mM). The authors suggested the
involvement of a free radical. However, this mode of action has not been explored or further
addressed in the literature for HCE.
The presence of radiolabeled carbon measured by in vivo binding studies suggested that
HCE can bind to DNA, RNA, and protein (Lattanzi et al., 1988). Binding to macromolecules
was interpreted by the presence of radiolabeled carbon; however, radiolabeled carbon may have
been incorporated into these macromolecules from intermediary HCE metabolites. In the rat,
higher levels of DNA, RNA, and protein binding were observed in the kidney and liver
compared with the lung and stomach. The mouse demonstrated the highest levels of DNA and
protein binding in the liver and RNA binding in the liver and kidney. Studies using CYP450
indicate that HCE must be metabolized to reactive intermediates prior to binding to
macromolecules. Therefore, renal toxicity and hepatotoxicity may also involve HCE binding to
DNA, RNA, or protein, resulting in cytotoxicity and contributing to the cytotoxic damage from
radicals.
The neurological effects observed in Beagle dogs (Weeks et al., 1979) and sheep (Fowler,
1969; Southcott, 1951) are commonly observed effects of chlorinated hydrocarbons. These
effects have not been extensively studied for HCE, and data are inadequate to determine a mode
of action.
4.7. EVALUATION OF CARCINOGENICITY
4.7.1. Summary of Overall Weight of Evidence
Under EPA's Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005a), HCE is
"likely to be carcinogenic to humans" based on data from oral cancer bioassays in F344/N rats
and B6C3Fi mice (NTP, 1989; NCI, 1978). No human data are available to assess the
carcinogenic potential of HCE. NTP (1989) reported dose-dependent increases (statistically
significant at the high dose) in the combined incidence of hepatocellular adenoma or carcinoma
and increases (statistically significant at the low dose) in the incidence of pheochromocytomas in
male F344/N rats. Tumors were not observed in the female F344/N rats in the NTP (1989)
study. NCI (1978) observed statistically significant increases in the incidence of hepatocellular
64
DRAFT - DO NOT CITE OR QUOTE
-------
carcinomas in male and female B6C3Fi mice. The male rats demonstrated a statistically
significantly increased tumor response for hepatocellular carcinomas that was dose-related. The
female mice displayed a statistically significantly elevated incidence of hepatocellular
carcinomas at both doses, although no dose-related increase in tumor response was evident. The
Osborne-Mendel rats in the NCI (1978) study did not provide consistent evidence of
carcinogenicity. HCE was shown to be a promoter, but not an initiator, in an Osborne-Mendel
rat liver foci assay (Milman et al., 1988; Story et al., 1986). Binding of radiolabeled carbon to
DNA, RNA, and protein following administration of [14C]-HCE was observed in both in vitro
and in vivo assays in mice and rats (Lattanzi et al., 1988).
U.S. EPA's Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005a) indicate that
for tumors occurring at a site other than the initial point of contact, the weight of evidence for
carcinogenic potential may apply to all routes of exposure that have not been adequately tested at
sufficient doses. An exception occurs when there is convincing information (e.g., toxicokinetic
data) that absorption does not occur by other routes. Information available on the carcinogenic
effects of HCE via the oral route demonstrates that tumors occur in tissues remote from the site
of absorption. Information on the carcinogenic effects of HCE via the inhalation and dermal
routes in humans or animals is absent. Based on the observance of systemic tumors following
oral exposure, and in the absence of information to indicate otherwise, it is assumed that an
internal dose will be achieved regardless of the route of exposure. Therefore, HCE is "likely to
be carcinogenic to humans" by all routes of exposure.
4.7.2. Synthesis of Human, Animal, and Other Supporting Evidence
There are currently no data from human studies pertaining to the carcinogenicity of HCE.
NTP (1989) conducted a chronic toxicity/carcinogenicity bioassay in F344/N rats. Groups of
50 male rats/dose were administered TWA doses of 7 and 14 mg/kg-day of HCE (purity >99%)
by corn oil gavage, 5 days/week for 103 weeks. Groups of 50 female rats/dose were
administered, by corn oil gavage, 5 days/week for 103 weeks, TWA doses of 57 and
114 mg/kg-day. Male rats exhibited a dose-related, statistically significant increase in the
incidence of combined renal adenomas or carcinomas at the highest dose. Combined renal
adenomas or carcinomas were observed in 2, 4, and 14% of controls, 7, and 14 mg/kg-day males,
respectively. No HCE-related renal tumors were observed in female rats. The combined
incidence of all three types of pheochromocytomas (benign, malignant, and complex
pheochromocytomas) was statistically significantly increased in males treated with 7 mg/kg-day
HCE (62%) and increased in males treated with 14 mg/kg-day (43%) when compared with
vehicle controls (30%) and historical controls in the study laboratory (75/300; 25 ± 7%) and in
NTP studies (543/1,937; 28 ± 11%). No HCE-related adrenal gland tumors were observed in
female rats.
65
DRAFT - DO NOT CITE OR QUOTE
-------
NCI (1978; Weisburger, 1977) conducted a chronic toxicity/carcinogenicity bioassay in
Osborne-Mendel rats. HCE (purity >98%) at doses of 0, 250, or 500 mg/kg-day was
administered by corn oil gavage to 50 rats/sex/dose for 5 days/week for 78 weeks. Following
termination of exposure, rats were observed for 33-34 weeks for a total duration of 111-
112 weeks. Twenty rats/sex were used for the untreated and vehicle controls. Starting in
week 23, rats in the exposure groups began a 5-week cyclic rotation that involved 1 week
without exposure followed by dosing for 4 weeks. After adjustment from 5 days/week for
78 weeks, with the 5-week cyclic rotation for part of the time, to continuous exposure over the
standard 2 years for a chronic bioassay, the TWA doses were 113 and 227 mg/kg-day. Mortality
was increased in the 113 and 227 mg/kg-day males with survival rates of 24/50 (48%) and
19/50 (38%>), respectively, compared with 14/20 {10%) in the untreated controls. Survival rates
for the female rats were 14/20 {10%) for both the untreated and vehicle controls, and
27/50 (54%>) and 24/50 (48%>) for the 113 and 227 mg/kg-day dose groups, respectively.
All of the tumor types observed had been encountered previously as spontaneous lesions
in the Osborne-Mendel rat, and no statistical differences in frequencies were observed between
treated and control rats. NCI concluded that there was no evidence of carcinogenicity in this rat
study. Notably, the doses used in the Osborne-Mendel rats of the NCI (1978) study were
approximately 16 times greater than those doses administered to F344 male rats by NTP (1989).
In a B6C3Fi mouse study conducted by NCI (1978; Weisburger, 1977), HCE (purity
>98%>) was administered by corn oil gavage at TWA doses of 360 and 722 mg/kg-day for
5 days/week for 78 weeks, followed by 12-13 weeks of an observation period (total 91 weeks).
Survival rates in males were 5/20 (25%), 1/20 (5%), 7/50 (14%>), and 29/50 (58%>) in the vehicle
control, untreated control, and 360 and 722 mg/kg-day dose groups, respectively. Survival rates
in females were 80, 85, 80, and 68%> in vehicle control, untreated control, 360 and 722 mg/kg-
day groups, respectively. Both male and female mice exhibited statistically significantly
increased incidences of hepatocellular carcinomas. The treated males demonstrated an increased
tumor response for hepatocellular carcinomas that was dose-related: 30 and 63%> in the 360 and
722 mg/kg-day dose groups, respectively, compared with 10%> in pooled vehicle controls and
15%o in matched vehicle controls. Females demonstrated an increased tumor response that was
not dose related in that a higher incidence of hepatocellular carcinomas occurred at the low dose
(40%>) compared with the high dose (31%); pooled vehicle and matched vehicle controls had
incidences of 3 and 10%, respectively. NCI concluded that HCE was carcinogenic in both sexes
of B6C3Fi mice.
Evidence of HCE's promotion (following treatment with DEN), but not initiation,
potential was observed in the liver of male Osborne-Mendel rats administered a single gavage
dose of 497 mg/kg HCE (Milman et al., 1988; Story et al., 1986). Lattanzi et al. (1988) reported
in vivo and in vitro binding of HCE to DNA, RNA, and protein in mice and rats. In both rats and
mice administered single i.p. injections of 127 j_iCi/kg [14C]-HCE, in vivo covalent binding of
66
DRAFT - DO NOT CITE OR QUOTE
-------
HCE for RNA was consistently much greater than that for DNA or protein. DNA exhibited the
lowest amount of HCE binding. Species differences were evident for all three macromolecule
types (DNA, RNA, and protein), with the mouse exhibiting much higher levels (9 times greater)
of covalent binding for DNA in the liver than the rat. The binding was 2 and 3 times greater for
mice than rats with RNA and protein, respectively, from the liver. The binding was similar
between species, but slightly greater in mice, for the kidney, lung, and stomach analyses. In
vitro covalent binding to DNA was observed at comparable levels in liver microsomes from both
rats and mice following exposure to HCE. Kidney microsomes from rats and mice produced
statistically significantly greater amounts of DNA binding compared with controls, with greater
amounts of DNA binding from mice (threefold increase) compared with rats (twofold increase).
Microsomes from the lungs and stomachs in both species did not display increased DNA binding
activity over corresponding controls.
4.7.3. Mode-of-Action Information
Hepatocellular and renal adenomas and carcinomas and pheochromocytomas were
observed in rats and mice following oral exposure to HCE (NTP, 1989; NCI, 1978). The
mechanistic data available for HCE is limited; and the mode(s) of carcinogenic action of HCE in
the liver, kidney, and adrenal gland is unknown. However, there are data suggesting that the
induction of kidney tumors in male rats involves the accumulation of a2u-globulin in the kidney
and the induction of liver tumors in male and female mice may involve increased cytotoxicity,
inflammation, and regenerative cell proliferation in the liver, respectively.
4.7.3.1. Kidney Tumors
Description of the Hypothesized Mode of Action
Hypothesized mode of action. Generally, kidney tumors observed in cancer bioassays are
assumed to be relevant for assessment of human carcinogenic potential. However, male rat-
specific kidney tumors that are caused by the accumulation of a2U-globulin are not generally
considered relevant to humans. Accumulation of a2U-globulin in hyaline droplets initiates a
sequence of events that leads to renal nephropathy and, eventually, renal tubular tumor
formation. The phenomenon is unique to the male rat since female rats and other laboratory
mammals administered the same chemicals do not accumulate a2U-globulin in the kidney and do
not subsequently develop renal tubule tumors (Doi et al., 2007; IARC, 1999; U.S. EPA, 1991c).
Some experimental data suggest that development of kidney tumors in male rats
following exposure to HCE may involve an a2U-globulin-mediated mode of action. However, an
analysis of the data as outlined below indicate that there is insufficient evidence to establish the
role of a2U-globulin in HCE-induced kidney tumors. Specifically, the key events leading to
development of kidney tumors in male rats exposed to HCE have not been adequately
characterized. For example, no immunohistochemical data are available that demonstrate the
67
DRAFT - DO NOT CITE OR QUOTE
-------
presence of a2U-globulin in hyaline droplets. Furthermore, reported renal toxicity in female rats
and male and female mice exposed to HCE suggests a mode of action other than a2U-globulin-
associated nephropathy. In the absence of information demonstrating the involvement of
a2u-globulin processes, male rat renal toxicity/tumors are considered relevant for risk assessment
purposes.
Identification of key events.
The the role of a2U-globulin accumulation in the development of renal nephropathy and
carcinogenicity observed following HCE exposure was evaluated using the U.S. EPA (1991c)
Risk Assessment Forum Technical panel report. This report (U.S. EPA, 1991c) provides specific
guidance for evaluating chemical exposure-related male rat renal tubule tumors for the purpose
of risk assessment, based on an examination of the potential involvement of a2U-globulin
accumulation.
The protein, a2U-globulin, is a member of a large superfamily of low-molecular-weight
proteins and was first characterized in male rat urine. It has been detected in various tissues and
fluids of most mammals, including humans. However, the particular isoform of a2U-globulin
commonly detected in male rat urine is considered specific for the male rat; moreover, the urine
and kidney concentrations detected in the mature male rat are several orders of magnitude greater
than in any other age, sex, or species tested (Doi et al., 2007; IARC, 1999; U.S. EPA, 1991c).
The hypothesized mode of action ascribed to a2U-globulin-associated nephropathy is
defined by a progressive sequence of effects in the male rat kidney, often culminating in renal
tumors. The involvement of hyaline droplet accumulation in the early stages of nephropathy
associated with a2U-globulin-binding chemicals is an important difference from the sequence of
events observed with classical carcinogens. The pathological changes that precede the
proliferative sequence for classical renal carcinogens also include early nephrotoxicity (e.g.,
cytotoxicity and cellular necrosis) but no apparent hyaline droplet accumulation. Furthermore,
the nephrotoxicity that can ensue from hyaline droplet accumulation is novel because it is
associated with excessive a2U-globulin accumulation. This a2U-globulin accumulation is
proposed to result from reduced renal catabolism of the a2U-globulin chemical complex and is
thought to initiate a sequence of events leading to chronic proliferation of the renal tubule
epithelium. The histopathological sequence of events in mature male rats consists of the
following (see Table 4-21 summarizing available data on HCE for each step of this sequence):
• Excessive accumulation of hyaline droplets in renal proximal tubules
• Immunohistochemical evidence that a2U-globulin is the protein accumulating in the
hyaline droplets
• Subsequent cytotoxicity and single-cell necrosis of the tubule epithelium;
68
DRAFT - DO NOT CITE OR QUOTE
-------
• Sustained regenerative tubule cell proliferation;
• Development of intralumenal granular casts from sloughed cellular debris associated
with tubule dilatation and papillary mineralization;
• Foci of tubule hyperplasia in the convoluted proximal tubules; and
• Renal tubule tumors.
Table 4-21. Nephrotoxic effects characteristic of a2U-globulin nephropathy
observed in male and female rats administered HCE
Study, dose,
duration, and
sex
NTP, 1989
7 or 14
mg/kg-d (M);
57 or 114
mg/kg-d (F)
103 wks
NCI, 1978
113 or 227
mg/kg-d
104 wks
Gorzinski et
al., 1985
1, 15, or 62
mg/kg-d
16 wks
NTP, 1989
34, 67, 134,
268,
or 536 mg/kg-d
13 wks
NTP, 1996
146 or 293
mg/kg-d
3 wks
NTP, 1989
140, 281, or
563 mg/kg-d
16 d
M
F
M
F
M
F
M
F
M
F
M
F
Accumulation of
hyaline droplets
X
X
NT
X
Accumulation of
a2u-globulin in
hyaline droplets
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
Necrosis/
degeneration
X
X
X
X
X
X
X
NT
Tubular
regeneration
X
X
X
X
X
X
NT
X
Granular
casts/dilatation
X
X
X
X
X
X
X
NT
X
Papillary
mineralization
X
NT
Tubular
hyperplasia
X
X
NT
NT = not tested; X = presence of effect; M = male; F = female
In addition to this histopathological sequence, U.S. EPA (1991c) provides more specific
guidance for evaluating chemically induced male rat renal tubule tumors for the purpose of risk
assessment. To determine the appropriateness of the data for use in risk assessment, chemicals
inducing renal tubule tumors in the male rat are examined in terms of three categories:
• The a2u-globulin sequence of events accounts for the renal tumors.
• Other potential carcinogenic processes account for the renal tumors.
• The a2u-globulin-associated events occur in the presence of other potential
carcinogenic processes, both of which result in renal tumors.
69
DRAFT - DO NOT CITE OR QUOTE
-------
Therefore, it is important to determine whether the a2U-globulin process is involved in
nephrotoxicity and carcinogenicity following HCE exposure and, if so, to what extent
a2u-globulin-associated events, rather than other processes, account for the tumor increase.
Determination of these elements requires a database of bioassay data not only from male
rats, but also from female rats and mice, and such toxicity studies should demonstrate whether or
not a2u-globulin processes are operative. In the absence of sufficient information demonstrating
the involvement of a2u-globulin processes, it should be assumed that any male rat renal
toxicity/tumors are relevant for risk assessment purposes.
As outlined in the U.S. EPA Risk Assessment Forum Technical Panel report (U.S. EPA,
1991c), the following information from studies of male rats is used for demonstrating that the
a2u-globulin process may be a factor in any observed renal effects—an affirmative response in
each of the three categories is desired. The three categories of information and criteria are as
follows:
• Increased number and size of hyaline droplets in the renal proximal tubule cells of
treated male rats. The abnormal accumulation of hyaline droplets in the P2 segment
helps differentiate a2U-globulin inducers from chemicals that produce renal tubule
tumors by other modes of action.
• Accumulating protein in the hyaline droplets is u.2U-globulin. Hyaline droplet
accumulation is a nonspecific response to protein overload; thus, it is necessary to
demonstrate that the protein in the droplet is, in fact, a2U-globulin.
• Additional aspects of the pathological sequence of lesions associated with
u.2u-globulin nephropathy are present. Typical lesions include single-cell necrosis,
exfoliation of epithelial cells into the proximal tubular lumen, formation of granular
casts, linear mineralization of papillary tubules, and tubule hyperplasia. If the
response is mild, not all of these lesions may be observed. However, some elements
consistent with the pathological sequence must be demonstrated to be present.
Experimental Support for the Hypothesized Mode of Action
Strength, consistency, and specificity of association
NTP (1989)—16-day study
In a short-term exposure study, NTP (1989) administered 140, 281, 563, 1,125, or
2,250 mg/kg-day HCE to F344/N rats via gavage for 16 days. All of the surviving HCE-exposed
male rats exhibited hyaline droplets in the cytoplasm of the renal tubular epithelium.
Additionally, male rats exposed to 140 and 281 mg/kg-day HCE demonstrated tubular cell
regeneration and eosinophilic granular casts of cell debris in the tubule lumina at the
corticomedullary junction. NTP (1989) did not report regeneration or granular casts in the
surviving males of the 563 mg/kg-day dose group. NTP (1989) did not report the incidence or
70
DRAFT - DO NOT CITE OR QUOTE
-------
severity of the lesions observed in the treated males. None of the nephrotoxic effects were
observed at any HCE dose in the female rats or in the controls.
NTP (1996)—21-day study
In a second short-term exposure study, NTP (1996) administered 146 or 293 mg/kg-day
HCE by gavage to male F344/N rats for 21 days. Marked hyaline droplet accumulation was
observed and categorized by severity in relation to controls. The hyaline droplet accumulation
exhibited by HCE-exposed male rats was characterized as two severity grades above the control
rats. A Mallory-Heidenhain stain allowed for greater sensitivity in evaluating hyaline droplets
within the tubules of the kidney and further supported the presence of the hyaline droplets in the
kidney tubules. Increased incidence of tubular regeneration (60 and 100% in the 146 and
293 mg/kg-day dose groups, respectively) was also observed in male rats following HCE
exposure. The severity of the tubular lesions was considered mild at both doses. Eosinophilic
granular casts, of minimal to mild severity, were identified in the outer medullary tubules in male
rats exposed to HCE: 80 and 60% in the 146 and 293 mg/kg-day HCE, respectively. There was
a dose-related, statistically significant increase in the PCNA labeling index in HCE-treated male
rats. The percentage of replicating proximal and distal tubule epithelial cells was increased
5.7-fold over controls in the 146 mg/kg-day dose group and 9.2-fold over the controls in
293 mg/kg-day dose group. The nephrotoxic effects reported by NTP (1996) were not noted in
the control animals. Female rats were not included in this study; therefore, gender specificity of
the nephrotoxic effect was not examined.
NTP (1989)—13-week study
In a subchronic exposure study, NTP (1989) administered 34, 67, 134, 268, or
536 mg/kg-day HCE via gavage to F344/N rats for 13 weeks. Male rats from all dose groups
exposed to HCE exhibited exposure-related kidney effects, although incidence data were only
reported for the 34 mg/kg-day dose group. These kidney effects were characterized by hyaline
droplet formation in the renal tubular epithelium, eosinophilic granular casts of cell debris in the
tubular lumina at the corticomedullary region (with associated tubular dilatation), and tubular
cell regeneration. The severity of these lesions increased with HCE exposure dose, although the
severity grades were not reported. Furthermore, as the HCE exposure dose increased, the
animals developed additional lesions. Renal papillary necrosis and renal tubule epithelium
degeneration and necrosis were observed in all 536 mg/kg-day males (only the five male rats that
died before the end of the study were analyzed microscopically).
71
DRAFT - DO NOT CITE OR QUOTE
-------
Urinalysis in male rats administered HCE showed fine and course granules, cellular casts,
and epithelial cells, findings that were consistent with the histopathological changes observed in
the male rats. Kidney weights of HCE-exposed males were increased 27, 37, 57, 73, and 57% in
34, 67, 134, 268, and 536 mg/kg-day males, respectively (increases were statistically significant,
compared with control kidney weights except the low-dose group). Female kidney weight was
increased following HCE exposure: 16 and 32% (statistically significant) in the 268 and
536 mg/kg-day dose groups, respectively. Treated females showed no other HCE-exposure-
related kidney effects.
Gorzinski et al. (1985)—16-week study
Gorzinski et al. (1985) observed dose-related levels of HCE in the kidneys of male F344
rats fed 1, 15, or 62 mg/kg-day HCE for 16 weeks. HCE was also detected in the kidneys of
female rats, although at much lower levels and did not increase proportionally with dose. Renal
tubular atrophy and degeneration was observed in male rats: 20, 70, and 100% in the 1, 15, and
62 mg/kg-day dose groups, respectively. These renal degenerative effects were also noted in
10%) of the male controls, although the authors noted that these lesions were graded as slight.
Slight hypertrophy and/or dilation of the proximal convoluted tubules were noted in 10, 70, and
100%) of the HCE-exposed male rats in the 1,15, and 62 mg/kg-day dose groups, respectively.
Slight hypertrophy and dilation of the proximal convoluted tubules were not observed in the
male control rats. Peritubular fibrosis was also noted in the high-dose group males. Renal
tubular atrophy and degeneration were observed in 10, 20, and 60%> of female rats in the 1,15,
and 62 mg/kg-day dose groups, respectively. These lesions were seen in one female control rat
(10%>), although the authors characterized the severity grade of the lesions as very slight.
Male rat sensitivity was evident in the histopathological changes seen in the
HCE-exposed male rats compared with the female rats. Renal effects were either observed in
more male rats than female rats (statistical analyses were not reported) or did not occur in
females. Additionally, kidney concentrations of HCE were much higher in male rats compared
with female rats. Gorzinski et al. (1985) noted that the differences in HCE concentrations
measured in male rat and female rat kidneys may explain the differences observed in the kidney
effects (i.e., male sensitivity to HCE exposure).
NCI (1978)—78-week study
NCI (1978) conducted a carcinogenicity bioassay in Osborne-Mendel rats administered
113 and 227 mg/kg-day HCE via gavage for 5 days/week for 78 weeks. Chronic inflammatory
kidney lesions were observed in both control and HCE-exposed rats. Male rats exhibited chronic
inflammation in the kidney: 75, 70, 65, and 50%> of untreated control, vehicle control, 113, and
227 mg/kg-day dose groups, respectively. Similarly, female rats showed an incidence of
inflammatory lesions in 40, 20, 36, and 41%> in the untreated control, vehicle control, 113, and
72
DRAFT - DO NOT CITE OR QUOTE
-------
227 mg/kg-day dose groups, respectively. The control and HCE-exposed male rats exhibited
greater sensitivity to the chronic inflammation compared with the female rats. NCI (1978) noted
that these lesions observed in the control and HCE-exposed animals of both sexes were
characteristic of age-related renal lesions. Some renal lesions observed in older rats could be
related to a spontaneous syndrome known as CPN. CPN is associated with aged rats, especially
F344, Sprague-Dawley, and Osborne-Mendel strains. CPN is frequently more severe in males
compared with females. Hard et al. (1993) reported the pathologic features attributed to CPN
including:
• Thickening of tubular and glomerular basement membranes;
• Basophilic segments of proximal convoluted tubules with sporadic mitoses indicative
of tubule cell proliferation;
• Tubular hyaline casts of proteinaceous material originating in the more distal portion
of the nephron, mainly in the medulla, and later plugging a considerable length of the
tubule;
• Focal interstitial aggregations of mononuclear inflammatory cells within areas of
affected tubules;
• Glomerular hyalinization and sclerosis;
• Interstitial fibrosis and scarring;
• Tubular atrophy involving segments of proximal tubule;
• Occasional hyperplastic foci in affected tubules (chronically in advanced cases); and
• Accumulation of protein droplets in sporadic proximal tubules (in some advanced
cases).
Several of the CPN pathological effects are similar to and can obscure the lesions
characteristic of a2U-globulin-related hyaline droplet nephropathy (Hard et al., 1993).
Additionally, renal effects of a2U-globulin accumulation can exacerbate the effects associated
with CPN (U.S. EPA, 1991c). However, Webb et al. (1989) suggested that exacerbated CPN
was one component of the nephropathy resulting from exposure to chemicals that induce
a2u-globulin nephropathy. Male rat sensitivity has been noted with both CPN and a2U-globulin
nephropathy.
With the exception of atrophy of the proximal tubule, tubular cell proliferation, and
hyaline casts of proteinaceous material, the histopathological effects associated with CPN are
distinctive from those of a2U-globulin nephropathy. The urinalysis and serum chemistry of CPN
rats show albuminuria, hypoalbuminemia, and hypocholesterolemia as well as increased
73
DRAFT - DO NOT CITE OR QUOTE
-------
serum creatinine and urea nitrogen levels, whereas these changes in a2U-globulin nephropathy are
minimal (Hard et al., 1993).
NCI (1978) reported tubular nephropathy in HCE-exposed rats, but not in untreated or
vehicle controls. Increased incidence of nephropathy described as tubular degeneration and
necrosis and the presence of large hyperchromatic regenerative epithelial cells was observed in
45 and 66% of male rats exposed to 113 and 227 mg/kg-day HCE, respectively. Female rats also
exhibited tubular nephropathy following HCE exposure: 18 and 59% in the 113 and 227 mg/kg-
day dose groups, respectively. In addition to the tubular nephropathy, observed effects overlying
these lesions included focal pyonephritis, tubular ectasia, cast formation, chronic interstitial
nephritis and fibrosis, and focal glomerulosclerosis. Renal tubular cell adenomas were observed
in four male rats (11% incidence rate) exposed to 113 mg/kg-day HCE. Similar renal tumors
were not observed in males from the high-dose group, males from the vehicle control, males
from the untreated control, or female rats. NCI (1978) concluded that there was no evidence of
HCE-exposure-related carcinogenicity in Osborne-Mendel rats based on the lack of statistical
significance and dose-response in the tumor incidence rate. However, it is possible that the
truncated duration of HCE treatment (78 weeks, cyclical) and the significantly accelerated
mortality in the male rats did not allow enough time for the renal tubule tumors to develop.
According to Goodman et al. (1980), the incidences of spontaneous renal tubule tumors in
control male and female Osborne-Mendel rats (as recorded in the NCI Carcinogenesis Testing
Program) were 0.3 and 0%, respectively. The incidence of renal adenomas (11%, first observed
at 86 weeks; 8 weeks after the treatment period ended) following administration of 113 mg/kg-
day HCE exceeded both the concurrent (0%) and historical (0.3%) controls in males.
NT I1 (1989)—103-week study
NTP (1989) administered 7 or 14 mg/kg-day HCE in corn oil via gavage to male F344/N
rats for 103 weeks. Kidney effects consisting of tubular cell degeneration and atrophy, tubular
dilatation, tubular cell regeneration, glomerulosclerosis, interstitial fibrosis, and chronic
inflammation were observed in >94% of the HCE-exposed male rats. The incidence of
nephropathy in male control rats was 96%. The mean severity of the kidney effects in male rats
increased following HCE exposure: 2.34 ± 0.14, 2.62 ± 0.15, and 2.68 ± 0.16 (statistically
significant) in the control, 7, and 14 mg/kg-day dose groups, respectively. Kidney effect severity
was considered mild for the controls and mild to moderate for the HCE-exposed male rats.
While the mean severity scores do not show more than a 15% increase over control in the high-
dose group, more moderate and marked nephropathy was observed in HCE-exposed male rats
compared with controls. The incidences of severe (moderate or marked) nephropathy in males
were 18/50, 24/50, and 30/50 in the control, 7, and 14 mg/kg-day dose groups, respectively.
Additionally, the male rats exhibited increased incidences in linear mineralization of the renal
papillae: 4, 30, and 64% in the control, 7, and 14 mg/kg-day dose groups, respectively. Pelvic
74
DRAFT - DO NOT CITE OR QUOTE
-------
epithelium hyperplasia was also observed in 14% of male rats exposed to either 7 or 14 mg/kg-
day HCE. These hyperplastic effects were not observed in either the controls or the treated
females.
NTP (1989) administered 57 or 114 mg/kg-day HCE in corn oil via gavage to female
F344/N rats for 103 weeks. The incidences of nephropathy in female rats following chronic
HCE exposure were 44, 84, and 90% for the control, 57, and 114 mg/kg-day dose groups,
respectively. The severity scores for nephrotoxicity in female rats were statistically significantly
increased in both treated groups: 0.72 ± 0.13, 1.38 ±0.11, and 1.69 ± 0.12 in the control, 57, and
114 mg/kg-day dose groups, respectively. The average severity of nephropathy was considered
minimal for the controls and minimal to mild for the HCE-exposed female rats. Examination of
the various grades of nephropathy severity shows more mild and moderate nephrotoxicity in
HCE-exposed females compared with controls. In females, the incidences of severe (mild or
moderate) nephropathy were 12/50, 25/50, and 32/50 in the control, 57, and 114 mg/kg-day dose
groups, respectively (statistical analysis was not reported). Female rats also showed an increase
in linear mineralization at 57 (44%) and 114 mg/kg-day (26%) compared with relatively high
response in the controls (28%). This increase in linear mineralization was not dose-related. The
HCE-exposed male rats also exhibited renal tubular hyperplasia, renal tubule adenomas, and
renal tubule carcinomas. The combined renal adenoma or carcinoma incidences were 2, 4, and
14%) (3, 6, and 24% after adjusting for intercurrent mortality) in the control, 7, and 14 mg/kg-day
dose groups, respectively. There were no HCE-related neoplasms observed in female rats treated
with 57 or 114 mg/kg-day HCE. NTP (1989) noted that the hyperplasia and tumors of the renal
tubules represented a morphologic continuum. The observed hyperplasia incidences were 4, 8,
and 22% of the control, 7, and 14 mg/kg-day dose groups, respectively. The incidence of renal
tubule neoplasia in male rats also exceeded historical controls (0.5%). Female rats did not
exhibit renal tubule hyperplasia.
A sex difference was noted in the observed nephropathy, as males were more sensitive to
HCE-exposure-related nephropathy than females. This sex specificity is apparent for the
nephrotoxicity and grades of nephropathy severity in both control and HCE-treated groups.
Although administered only one-eighth of the dose given to the female rats, the male rats
demonstrated a greater incidence of nephropathy that was more severe and included additional
kidney effects (i.e., increases in incidence of mineralization of the renal papillae and hyperplasia
of pelvic transitional epithelium) compared with the female rats.
With the exceptions of glomerulosclerosis, interstitial fibrosis, and chronic inflammation,
the observed nephrotoxic effects in the male rats are characteristic of a2u-globulin nephropathy.
However, NTP (1989) did not report accumulation of hyaline droplets containing the
a2u-globulin protein in the proximal tubule. It is possible that hyaline droplets were present,
considering that the 16-day and 13-week rats examined by NTP (1989) exhibited hyaline
droplets; however, the hyaline droplets were likely obscured by the prevalence of the other
75
DRAFT - DO NOT CITE OR QUOTE
-------
lesions. Evidence of these effects in almost all of the control males and in treated and control
female rats also complicates the characterization of the mode of action. Considering that
a2u-globulin nephropathy is typically male rat-specific, the appearance of nephrotoxic effects in
the female rats as well as the male and female controls and the identification of other effects not
specifically associated with a2U-globulin (i.e., glomerulosclerosis and interstitial fibrosis) suggest
that the effects may not be the result of a2U-globulin accumulation.
Considering the strain and age of the rats in the chronic (103 weeks) NTP study (1989), it
is also possible that the rats were affected by CPN (i.e., increased incidence of nephrotoxicity in
the control rats). However, changes in severity of the nephropathy that are greater in the HCE-
exposed animals indicate some chemical-related effects. Additionally, HCE-exposed male rats
demonstrated dose-dependent increases in incidence of mineralization of the renal papillae and
hyperplasia of pelvic transitional epithelium. Neither of these effects increased in a dose-related
manner in the controls or the HCE-exposed female rats. Therefore, the treatment-related effects
in male and female rats serve as evidence that CPN is not solely responsible for the nephropathy
observed by NTP (1989).
Limitations in the available studies. These studies describe the effects associated with
HCE exposure using a general, nonspecific term: tubular nephropathy (Weeks et al., 1979; NCI,
1978). This general term does not provide information on the specific histopathological changes
characterizing the nephropathy. Additionally, the reported incidences of effects were grouped
and measured as nephropathy rather than individual effects. Effects described in this way are
difficult to interpret with regards to a2U-globulin nephropathy. One study (NTP, 1996) was
limited in its usefulness because only male rats were exposed and the experimental design sought
to draw conclusions about SARs involved in the induction of hyaline droplet nephropathy of
11 halogenated ethanes. The study focused predominantly on the kidneys and the purpose of the
study was to compare chlorinated ethanes, not to examine the mode of action of HCE. The
divergence in doses used for male and females in the NTP (1989) chronic exposure experiment
highlighted the male sensitivity to HCE-induced nephrotoxicity. However, this study design
made it difficult to otherwise compare the sexes. Additionally, three of the six HCE exposure
studies utilized only two dose groups, limiting the ability to characterize the dose response of
HCE-exposure-related nephropathy.
Summary of evidence for strength, specificity, and consistency. Generally, kidney tumors
observed in cancer bioassays are assumed to be relevant for assessment of human carcinogenic
potential. However when the mode-of-action evidence demonstrates that kidney tumors in male
rats result from an accumulation of a2u-globulin, the tumor data are considered to not be relevant
to humans, and are not suitable for use in risk assessment (IARC, 1999; U.S. EPA, 1991c). The
criteria for demonstrating the a2U-globulin-related mode of action for risk assessment purposes
have been defined (U.S. EPA, 1991c). Three criteria are considered to be desirable: (1) an
increase in hyaline droplets in the renal proximal tubule cells; (2) the determination that the
76
DRAFT - DO NOT CITE OR QUOTE
-------
accumulating protein in the droplets is a2U-globulin; and (3) the presence of additional
pathological lesions associated with a2U-globulin. The key event in the histopathological
sequence for the a2u-globulin-related mode of action is excessive accumulation of hyaline
droplets containing a2U-globulin in renal proximal tubules.
None of the HCE studies performed the necessary immunohistochemical assays to
confirm the presence of a2U-globulin protein within the hyaline droplets observed following
administration of HCE (NTP, 1996, 1989). It is unclear whether HCE is binding to a2u-globulin
or to other proteins during the formation of hyaline droplets, or if another mechanism is
operating. This represents an important data gap considering that the presence of this protein is
essential to identifying the a2U-globulin-related mode of action.
In addition, the data on female rats and mice of both sexes from chronic exposure studies
(NTP, 1989; NCI, 1978) do not support an a2U-globulin-associated mode of action for HCE-
exposure related nephropathy. NCI (1978) reported dose-related nephropathy in female rats that
was not apparent in the controls. Nephropathy was also reported in male and female mice
chronically administered HCE (NCI, 1978). NCI (1978) reported the appearance of renal tubular
effects in almost all (>92%) of the HCE-treated male and female mice following chronic HCE
exposure, but the mice did not develop renal tubule tumors. The presence of kidney effects in
HCE-exposed female rats and male and female mice, which generally do not accumulate the
a2u-globulin protein, suggests a mode of action other than a2u-globulin-associated nephropathy.
Dose-response concordance. The initial key event in the histopathological sequence for
the a2U-globulin-related mode of action is excessive accumulation of hyaline droplets containing
a2u-globulin in renal proximal tubules. The accumulation of a2u-globulin in hyaline droplets
must occur at lower doses than subsequent a2U-globulin-related effects. None of the HCE studies
performed the necessary immunohistochemical assays to confirm the presence of a2U-globulin
protein within the hyaline droplets observed following administration of HCE (NTP, 1996,
1989). Therefore, this key event cannot be demonstrated from the available data.
Most of the effects characterizing the histopathological sequence of events in epithelial
cells of the proximal tubules leading to renal tumors (Doi et al., 2007; IARC, 1999; U.S. EPA,
1991c) increased in incidence with increasing doses of HCE in the short-term and subchronic
exposure studies. Dose-related increases in nephrotoxicity and renal carcinogenicity were noted
in the two chronic HCE exposure studies. The short-term and subchronic exposure studies did
not report evidence of carcinogenicity in rats administered HCE. In the NTP (1989) study, male
rats administered 7 or 14 mg/kg-day HCE for 2 years exhibited a dose-related increased
incidence of renal tubule adenomas and carcinomas. Histopathological effects associated with
a2U-globulin nephropathy (tubular cell degeneration and atrophy, tubular dilatation, and tubular
cell regeneration) were noted in almost all of the treated and untreated animals. A dose-response
relationship was difficult to detect considering the number of animals affected by nephrotoxicity.
However, dose-related increases over controls for toxic kidney effects such as linear
77
DRAFT - DO NOT CITE OR QUOTE
-------
mineralization, severity of nephrotoxicity, and renal tubule hyperplasia were observed. NTP
(1989) did not report interim data; therefore, examinations were performed at study termination.
Consequently, the nephrotoxicity (generally attributed to leading up to the formation of renal
tubular tumors associated with a2U-globulin) is reportedly increased at doses similar to those that
induce tumor formation.
Overall, dose-related kidney effects were noted for almost all of the male rats
administered HCE at doses ranging from 1 to 563 mg/kg-day. Even at the lowest HCE dose
administered in the studies, renal effects were observed in male rats. Animals treated with
greater amounts of HCE exhibited dose-related increases in incidence and severity of effect
when compared with those of the lower dose groups. It is difficult to establish dose-response
concordance between the noncancer nephropathy and the renal tubule tumors reported by NTP
(1989). Renal tubule tumors were observed at 7 mg/kg-day HCE, the lowest dose administered
for a chronic duration, which also induced significant nephropathy in HCE-exposed animals.
The other studies that administered doses within an order of magnitude of 7 mg/kg-day were the
NTP (1989) study (34 or 67 mg/kg-day for 13 weeks) and the Gorzinski et al. (1985) study (1,
15, or 62 mg/kg-day for 16 weeks). Although nephropathy was noted in the shorter duration
studies (NTP, 1996, 1989; Gorzinski et al., 1985), the only evidence of carcinogenicity was from
the chronic exposure studies (NTP, 1989; NCI, 1978).
Temporal relationship. The initial key event in the histopathological sequence for the
a2U-globulin-related mode of action is excessive accumulation of hyaline droplets containing
a2u-globulin in renal proximal tubules. The accumulation of a2U-globulin in hyaline droplets
must occur first in the sequela leading to a2u-globulin-related nephrotoxicity and tumor
formation. None of the HCE studies performed the necessary immunohistochemical assays to
confirm the presence of a2U-globulin protein within the hyaline droplets observed following
administration of HCE (NTP, 1996, 1989). Therefore, this key event and the important temporal
relationship for the accumulation of a2u-globulin cannot be demonstrated from the available data.
Histopathological effects associated with a2U-globulin-related nephropathy were observed
in animals treated with HCE in studies that varied in exposure duration from 16 days to 2 years.
The sequence of histopathological events characteristic of the a2U-globulin-related mode of
action was noted in the chronic exposure study NTP (1989) that reported renal tubule adenomas
and carcinomas. All of the studies (NTP, 1996, 1989; Gorzinski et al., 1985; NCI, 1978) that
administered HCE for shorter durations than the NTP (1989) study reported similar
histopathological changes, although an increase in renal tubule tumors was not observed. It is
unknown if the nephropathy observed by NTP (1989) led to the reported renal tubule tumors
because the animals were only examined at the end of the 103-week study period. A temporal
relationship cannot be distinguished from reported data.
Biological plausibility and coherence. The kidney toxicity and tumor formation that was
observed in rats and mice are biologically plausible effects that could potentially occur in
78
DRAFT - DO NOT CITE OR QUOTE
-------
humans. If the tumor formation in male rats, however, is due to accumulation of a2u-globulin
protein in the renal tubules, then these tumors would not be considered to be relevant to humans.
The sequence of events including accumulation of a2u-globulin protein in the renal tubules of
male rats initiating a sequence of nephrotoxic events leading to renal tubule tumor formation was
evaluated as a hypothesized mode of action for HCE-induced carcinogenicity and nephropathy
(Doi et al., 2007; IARC, 1999; U.S. EPA, 1991c). These a2u-globulin related effects are
typically not observed in female rats or other species due to the absence or minimal presence of
the a2u-globulin protein in these animals (Hard et al., 1993). Evidence of nephrotoxic effects in
female rats in two chronic studies (NTP, 1989; NCI, 1978) and in male and female mice in one
chronic study (NCI, 1978) precludes the conclusion that HCE is acting through an a2u-globulin-
associated mode of carcinogenic action.
Other Possible Modes of Action
There is insufficient evidence to support an a2U-globulin-related mode of action for renal
tumors following HCE exposure. It is possible that advanced CPN may play a role in the
incidence of nephrotoxicity and kidney tumors in male rats. CPN is an age-related renal disease
of laboratory rodents that occurs spontaneously. The observed renal lesions in male rats
following exposure to HCE are effects commonly associated with CPN. Nephropathy (described
as tubular cell degeneration and regeneration, tubular dilatation and atrophy, glomerulosclerosis,
interstitial fibrosis, and chronic inflammation) was also observed in female rats (NTP, 1989), as
well as in male and female mice (NCI, 1978). However, changes in severity of the nephropathy
were observed to be greater in male rats exposed to HCE compared to controls, indicating that
HCE exposure exacerbated effects in the kidney. Additionally, HCE-exposed male rats
demonstrated dose-dependent increases in incidence of mineralization of the renal papillae and
hyperplasia of pelvic transitional epithelium. Neither of these effects increased in a dose-related
manner in the controls or the HCE-exposed female rats. The treatment-related effects in male
and female rats serve as evidence that CPN is not solely responsible for the nephropathy
observed by NTP (1989).
Conclusions about the Hypothesized Mode of Action
Support for the hypothesized mode of action in animals. The mode of action for the
carcinogenic effects of HCE in the kidney is unknown, although there are data to indicate that
a2u-globulin accumulation may play a role in the observed tumors in male rats. Studies
following short-term, subchronic, and chronic exposure of male rats have reported renal lesions
(NTP, 1996, 1989; Gorzinski et al., 1985; NCI, 1978) and formation of renal tubule adenomas
and carcinomas (preceded by hyperplasia) following chronic HCE exposure (NTP, 1989),
suggesting an a2u-globulin-related mode of action. However, the key event in the
79
DRAFT - DO NOT CITE OR QUOTE
-------
histopathological sequence of events demonstrating an a2U-globulin-related mode of action
(excessive accumulation of hyaline droplets containing a2U-globulin in renal proximal tubules)
leading to the development of kidney tumors in male rats exposed to HCE has not been
characterized. None of the HCE studies performed immunohistochemical assays to confirm the
presence of a2U-globulin protein within the hyaline droplets observed following administration of
HCE (NTP, 1996, 1989). It is unknown if HCE is binding to a2U-globulin or to other proteins
during the formation of hyaline droplets. This represents an important data gap. On the other
hand, it is possible that an a2U-globulin-associated mode of action may, in fact, be responsible for
the tumors observed in male rats and that more than one mode of action may be operating to
induce the nephropathy and tumor formation observed across species and sexes.
In addition, data are available that demonstrate kidney effects in female rats and mice of
both sexes from chronic exposure studies (NTP, 1989; NCI, 1978). The NCI (1978) study
reported dose-related nephropathy in female rats that was not apparent in the controls.
Nephropathy was also reported in male and female mice chronically-administered HCE (NCI,
1978). The presence of kidney effects in HCE-exposed male and female mice, which generally
do not accumulate the a2U-globulin protein, suggests a mode of action other than a2U-globulin
nephropathy.
Relevance of the Hypothesized Mode of Action to Humans
Generally, kidney tumors observed in cancer bioassays are assumed to be relevant for
assessment of human carcinogenic potential. However, for male rat kidney tumors, when the
mode-of-action evidence demonstrates that the response is secondary to a2u-globulin
accumulation, the tumor data are not used in the cancer assessment (U.S. EPA, 1991b). There is
insufficient evidence to conclude that the renal adenomas and carcinomas observed in male rats
administered HCE (NTP, 1989) are related to an a2U-globulin mode of action for the following
reasons: (1) there is a lack of a2u-globulin immunohistochemical data for HCE-induced
nephrotoxicity and carcinogenicity; (2) the hyaline droplet accumulation is caused by excessive
protein load that may not be exclusively related to a2U-globulin accumulation; and (3) the
existence of renal toxicity in female rats and male and female mice indicates that the nephrotoxic
effects are not limited to an a2u-globulin-induced sequence of lesions. Therefore, the renal
adenomas and carcinomas observed in male rats administered HCE (NTP, 1989) were
considered relevant to humans.
4.7.3.2. Liver Tumors
Hepatocellular carcinomas were observed in male and female B6C3Fi mice administered
360 or 722 mg/kg-day HCE, via gavage, in a chronic oral bioassay conducted by NCI (1978).
Tumor incidences in males of both dose groups were statistically significantly elevated compared
with control groups, and demonstrated a dose response. Both dose groups of female mice
80
DRAFT - DO NOT CITE OR QUOTE
-------
presented statistically significantly elevated incidences of hepatocellular carcinoma compared
with control groups, but a dose response was not observed. The investigators did not find
nonneoplastic liver effects (such as organized thrombus, inflammation, fibrosis, necrosis,
infarctions, amyloidosis, or hyperplasia) in either sex.
The mode of action for the carcinogenic effects of HCE in the liver is unknown.
Metabolism studies of HCE indicate that the major enzymes involved are phenobarbital-
inducible CYP450s. These are primarily localized in the liver. Although tissue-specific
metabolism of HCE has not been studied extensively, the majority of HCE metabolism is
presumed to occur in the liver. HCE is proposed to metabolize to PERC and pentachloroethane
and is likely subsequently metabolized to TCE. It is possible that the HCE-induced
hepatocellular carcinomas in mice occur as a result of the binding of HCE metabolites to liver
macromolecules and the generation of free radicals during HCE metabolism, causing key events
in the carcinogenic process such as cytotoxicity, inflammation, and regenerative cell
proliferation. However, these potential key events have not been systematically evaluated for
HCE.
In a 13-week study, hepatocellular necrosis of the centrilobular area was observed in rats
(NTP, 1989). It is unknown if this could be considered a key event in the carcinogenic process
because rats in the available studies (NTP, 1989; NCI, 1978) have not displayed hepatocellular
neoplastic endpoints. Although mice demonstrated hepatocellular carcinoma, nonneoplastic
effects such as hepatocellular necrosis were not observed (NCI, 1978). HCE-induced
hepatocellular carcinomas in mice varied in microscopic appearance (NCI, 1978). Some
carcinomas were characterized by well-differentiated hepatic cells with uniform cord
arrangement, while others had anaplastic liver cells with large hyperchromatic nuclei, often with
inclusion bodies and vacuolated pale cytoplasm. Arrangement of neoplastic liver cells also
varied from short stubby cords to nests of cells and occasional pseudo-acinar formations.
Neoplasms in control mice did not vary in appearance from those in HCE-treated mice.
In vivo binding of radiolabeled carbon to DNA, RNA, and protein from liver, kidney,
lung, and stomach following administration of [14C]-HCE was consistently greater in mice
compared with rats (Lattanzi et al., 1988). Binding to macromolecules was interpreted by the
presence of radiolabeled carbon; however, radiolabeled carbon may have been incorporated into
these macromolecules from intermediary HCE metabolites. In vitro binding studies using calf
thymus DNA demonstrated that mouse liver cytosol (induced by phenobarbital) mediated more
extensive DNA binding than rat liver cytosol (Lattanzi et al., 1988). Comparisons of HCE
metabolism rates indicated that mice metabolize HCE at twice the rate of rats (Mitoma et al.,
1985).
Cellular damage leading to cytotoxicity, inflammation, and regenerative cell proliferation
is a possible consequence of this binding in the liver. The binding studies provide a line of
evidence as to why the liver is the major carcinogenic target in the mouse, but not the rat.
81
DRAFT - DO NOT CITE OR QUOTE
-------
Regenerative cell proliferation has been evaluated in the kidney, but not in the liver of
HCE-treated rats (NTP, 1996). RDS in hepatocytes was evaluated in mice treated with HCE
(Yoshikawa, 1996; Miyagawa et al., 1995). This study reported ambiguous results; the lower
HCE dose caused a statistically significant increase in RDS, whereas the higher dose did not
(Yoshikawa, 1996; Miyagawa et al., 1995). Rat liver foci experiments provide support for the
hypothesis that HCE acts as a tumor promoter, not as a tumor initiator (Milman et al., 1988;
Story et al., 1986).
The in vivo binding data suggest that HCE is sequestered in the liver of mice and rats and
metabolic data suggest that mice metabolize HCE at a greater rate compared with rats.
Considering the greater potential for metabolism in mice compared with rats and the proposed
increase in DNA binding following metabolism of HCE (Lattanzi et al., 1988), the increased
incidence of hepatocellular carcinomas in mice, but not rats, may be related to DNA binding.
However, the DNA binding measurements were based solely on the presence of radiolabeled
carbon; specific HCE metabolites were not identified. Therefore, this process does not take into
account the possibility of normal biological mechanisms in which the radiolabeled carbon can be
incorporated into the macromolecules via anabolic processes. All together, while it is possible
that metabolism and binding in mice are involved in the development of liver tumors, the role of
DNA binding in the mode of action for HCE-induced hepatotoxicity and carcinogenesis is not
known and, as such, the mode of action is not known.
4.7.3.3. Pheochromocytomas
Pheochromocytomas are catecholamine-producing neuroendocrine tumors. The
relevance of rodent pheochromocytomas as a model for human cancer risk has been the subject
of discussion in the scientific literature (e.g., Greim et al., 2009; Powers et al., 2008). In humans,
pheochromocytomas are rare and usually benign, but may also present as or develop into a
malignancy (Eisenhofer et al., 2004; Lehnert et al., 2004; Elder et al., 2003; Goldstein et al.,
1999). Hereditary factors in humans have been identified as important in the development of
pheochromocytomas (Eisenhofer et al., 2004). Pheochromocytomas are more common in
laboratory rats, though evidence suggests that certain rat pheochromocytomas may have
similarity to human pheochromocytomas (Powers et al., 2009). Furthermore, mechanisms of
action inducing pheochromocytomas in rats are expected to occur in humans as well (Greim et
al., 2009). Therefore, in the absence of information indicating otherwise, adrenal gland tumors
in rodents are considered relevant to humans.
No studies were identified to determine a mode of action for HCE-induced tumors of the
adrenal gland. Therefore, the mode of action for pheochromocytomas observed following oral
exposure to HCE is unknown.
4.8. SUSCEPTIBLE POPULATIONS AND LIFE STAGES
82
DRAFT - DO NOT CITE OR QUOTE
-------
No studies were located that address the susceptibility of populations or life stages to
HCE-induced toxicity or carcinogenicity in humans.
4.8.1. Possible Childhood Susceptibility
No studies were located that addressed possible childhood susceptibility to HCE-induced
toxicity or carcinogenicity. Although it is unknown if HCE toxicity is mediated by parent
compound or its metabolites, CYP450 enzymes of the 2A, 2B, and 3 A subfamilies and CYP450
1A2 are involved in HCE metabolism. Many drugs reportedly exhibit a higher systemic
clearance in children than in adults (Evans et al., 1989). Blanco et al. (2000) compared liver
microsomal CYP450 activities of humans <10 years old with those >10-60 years old and
concluded that factors other than maximal CYP450 catalytic activities, such as reductions in
hepatic blood flow, hepatic size, and oxygen supply in the elderly, may be responsible for age-
related changes in drug clearance. Studies of fetal and neonatal livers indicate that CYP450
expression is similar to adult levels by a few months of age (Lacroix et al., 1997; Vieira et al.,
1996; Cazeneuve et al., 1994; Treluyer et al., 1991). However, Dome (2004) reported in a
review article that Phase I (including CYP450 activities) and Phase II enzymatic activities are
1.3-1.5-fold higher in children (aged 1-16 years) compared with adults. Therefore, the extent to
which variable age-related expression of CYP450 contributes to childhood susceptibility is
unknown. Considering the substantial portion of HCE that remains as parent compound, the
impact, if any, of age on CYP450 expression and HCE metabolism cannot be assessed.
4.8.2. Possible Gender Differences
Toxicity studies in rats indicate that male rats are more sensitive to HCE-induced
nephrotoxicity than females (NTP, 1989; Gorzinski et al., 1985, 1980; NCI, 1978). Evidence
suggests that female rats are more sensitive to HCE-induced hepatotoxicity. The reasons for
these sex-specific differences are unknown, but may be related to sex-specific differences in
tissue concentrations following HCE administration (i.e., higher concentrations observed in male
rat tissues when compared with female rats, see Table 3-3), sex hormone differences, and/or
gender differences in CYP450 activities. No additional studies were located that addressed
possible gender differences for HCE-induced toxicity or carcinogenicity.
4.8.3. Other
CYP450 enzymes are polymorphic in the human population. Polymorphisms result in
CYP450 enzymes with variant catalytic activity for substrates such as HCE. This could
potentially result in decreased HCE detoxification or increased HCE bioactivation.
Detoxification enzymes such as the glutathione-S-transferase (GST) family are also polymorphic
in the human population, with variant catalytic activities that could affect the detoxification of
HCE.
83
DRAFT - DO NOT CITE OR QUOTE
-------
5. DOSE-RESPONSE ASSESSMENTS
5.1. ORAL REFERENCE DOSE (RfD)
5.1.1. Choice of Principal Study and Critical Effect—with Rationale and Justification
Data on the health effects of oral HCE exposure in humans are not available. The oral
exposure database for HCE includes a 103-week gavage study in F344 rats (NTP, 1989), a
78-week gavage study in Osborne-Mendel rats (NCI, 1978), a 91-week gavage study in B6C3Fi
mice (NCI, 1978), a 16-week feeding study in F344 rats (Gorzinski et al., 1985), and a 13-week
gavage study in F344 rats (NTP, 1989). The short-term study data were not considered in the
selection of the principal study for the derivation of the RfD because the database contains dose-
response data from studies of subchronic and chronic durations. However, short-term studies in
rats (NTP, 1996, 1989) were used to support findings in the subchronic and chronic studies. The
available oral exposure studies identified kidney or liver effects associated with exposure to
HCE. Reported effects include tubular nephropathy (NTP, 1989; NCI, 1978), atrophy and
degeneration of renal tubules (NTP, 1989; Gorzinski et al., 1985), slight hypertrophy and/or
dilation of proximal convoluted renal tubules (Gorzinski et al., 1985), linear mineralization of
renal tubules (NTP, 1989), hyperplasia of the renal pelvic transitional epithelium (NTP, 1989),
and hepatocellular necrosis (NTP, 1989).
In the NTP (1989) chronic study, HCE was administered via gavage at doses of 7 and
14 mg/kg-day in male F344 rats and 57 and 114 mg/kg-day in female F344 rats for 103 weeks.
Nephropathy (characterized by tubular cell degeneration and regeneration, tubular dilatation and
atrophy, glomerulosclerosis, interstitial fibrosis, and chronic inflammation) was observed in
HCE-treated rats of both sexes. Nephropathy was also reported in control rats of both sexes.
Although a high incidence of nephropathy was observed in control rats, the study authors
reported that the incidence of more severe nephropathy increased in dosed rats relative to
controls (NTP, 1989). EPA considered the increase in severity of nephropathy in male rats by
analyzing the incidence of greater than mild nephropathy. EPA determined that the increased
incidence of moderate or marked nephropathy in males was statistically significant at the
14 mg/kg-day dose (see Table 5-1). EPA considered the increased severity of nephropathy in
female rats by analyzing the incidence of nephropathy that was greater than minimal
nephropathy. EPA determined that the increased incidences of mild to moderate nephropathy
were statistically significant in females at the 57 and 114 mg/kg-day doses (see Table 5-1).
Linear mineralization of the renal papillae and hyperplasia of the renal pelvic epithelium were
increased in a dose-dependent, statistically significant manner in the treated male rats. EPA
determined that the increased incidences of linear mineralization of the renal papillae and
hyperplasia of the renal pelvic epithelium were statistically significant in males at the 7 and
14 mg/kg-day doses (see Table 5-1). Considering the increased severity of nephropathy
84
DRAFT - DO NOT CITE OR QUOTE
-------
following HCE exposure and dose-dependent increases in the incidence of mineralization of the
renal papillae and hyperplasia of renal pelvic transitional epithelium in male rats, the
nephropathy observed in the NTP (1989) study was exacerbated by HCE exposure. The NTP
(1989) chronic study did not identify NOAELs for male or female rats as kidney effects were
observed at the lowest doses tested. EPA considered the male rat LOAEL as 7 mg/kg-day based
on increased incidence in moderate or marked tubular nephropathy (characterized by
degeneration, necrosis, and regenerative epithelial cells), hyperplasia of the pelvic transitional
epithelium, and linear mineralization of the renal papillae in the NTP (1989) study. EPA
considered the female rat LOAEL as 57 mg/kg-day, based on dose-related increases in incidence
and severity of nephropathy in the NTP (1989) study.
In the NCI (1978) chronic rat study, HCE was administered via gavage to groups of
50 male and 50 female Osborne-Mendel rats for 5 days/week, cyclically for 66 of the 78 weeks,
followed by an observation period of 33-34 weeks (total of 112 weeks). The TWA doses of
HCE were 113 and 227 mg/kg-day. Tubular nephropathy was observed in all groups of treated
animals, but was not observed in either untreated or vehicle controls. Statistically significant
increases in incidence of tubular nephropathy were observed at 113 and 227 mg/kg-day HCE in
both male and female rats (see Table 5-1). The NCI (1978) study did not identify a NOAEL for
tubular nephropathy in rats. EPA considered the LOAEL as 113 mg/kg-day, based on a dose-
related increase in incidence of nephropathy in both male and female rats.
In the NCI (1978) chronic mouse study, HCE was administered via corn oil gavage to
groups of 50 male and 50 female B6C3Fi mice for 5 days/ week for 78 weeks followed by an
observation period of 12-13 weeks (total of 90 weeks). Starting in week 9, the HCE doses were
increased, though no explanation for the increase was provided. The TWA doses of HCE were
360 and 722 mg/kg-day. Because of low survival rates in the vehicle and untreated male control
groups, NCI (1978) compared tumor incidences in the dosed males and females to the pooled
vehicle control data derived from concurrently run bioassays for several other chemicals. NCI
(1978) reported chronic kidney inflammation (i.e., tubular nephropathy characterized by
degeneration of the convoluted tubule epithelium at the junction of the cortex and medulla and
hyaline casts) in male and female B6C3Fi mice administered 360 and 721 mg/kg-day HCE.
EPA considered the LOAEL for this study as 360 mg/kg-day based on tubular nephropathy,
while a NOAEL could not established from these data.
In the Gorzinski et al. (1985) study, HCE was administered (in feed) to groups of 10 male
and 10 female F344 rats at doses of 0, 1, 15, or 62 mg/kg-day for a period of 16 weeks. Kidney
effects consisted of slight hypertrophy and/or dilation of proximal convoluted renal tubules and
atrophy and degeneration of renal tubules. Slight hypertrophy and/or dilation of the proximal
convoluted renal tubules was not observed in the control rats of either sex or in HCE exposed
female rats. EPA determined that increases in slight hypertrophy and/or dilation of the proximal
convoluted renal tubules were statistically significant in male rats treated with 15 or
85
DRAFT - DO NOT CITE OR QUOTE
-------
62 mg/kg-day HCE (see Table 5-1). Atrophy and degeneration of renal tubules was observed in
both male and female rats. EPA determined that increases in incidences of atrophy and
degeneration of renal tubules were statistically significant in male rats treated with 15 or
62 mg/kg-day HCE and in female rats fed 62 mg/kg-day HCE (see Table 5-1). EPA considered
the male rat LOAEL as 15 mg/kg-day and the male rat NOAEL as 1 mg/kg-day, based on
increased incidence of the renal tubule effects. EPA considered the female rat LOAEL as
62 mg/kg-day and the female rat NOAEL as 15 mg/kg-day, based on increased incidence of
renal tubule effects.
In the NTP (1989) subchronic study, HCE was administered via gavage to groups of
10 male and 10 female F344 rats at TWA doses of 0, 34, 67, 134, 268, and 536 mg/kg-day for
13 weeks. Kidney effects (i.e., hyaline droplet formation, renal tubular regeneration, and renal
tubular casts) were observed in male rats from all HCE exposure groups, though incidence data
were only provided for the 34 mg/kg-day dose group. NTP (1989) reported that the severity of
kidney effects in male rats increased with dose, but no data on severity were presented. No
kidney effects were reported in female F344 rats exposed to HCE. Liver effects were observed
in male and female rats at higher doses of HCE and EPA determined that statistically significant
increases in hepatocellular necrosis were observed in female rats exposed to 268 or
536 mg/kg-day HCE (see Table 5-1).
The incidence of kidney and liver effects from the studies considered for selection as the
principal study are summarized in Table 5-1. As incidence data on kidney effects reported in the
13-week subchronic study (NTP, 1989) were limited to males in the 34 mg/kg-day dose group,
these data are not presented in Table 5-1.
Table 5-1. Incidences of noncancerous kidney and liver effects in rats
following oral exposure to HCE
Study
Duration
(route)
Strain/sex/species
Endpoint
Dose
(mg/kg-day)
Incidence
Kidney Effects
NCI (1978)
78 wks
(gavage)
Osborne-Mendel male
rat
Tubular nephropathy
0
0/20 (0%)
113
22/49a (45%)
227
33/503 (66%)
Osborne-Mendel
female rat
Tubular nephropathy
0
0/20 (0%)
113
9/50a (18%)
227
29/49a (59%)
NTP (1989)
103 wks
(gavage)
F344 male rat
Moderate to marked tubular
nephropathy
0
18/50 (36%)
7
24/50 (48%)
14
30/503 (60%)
F344 female rat
Mild to moderate tubular
nephropathy
0
12/50 (24%)
57
25/503 (50%)
114
32/49a (65%)
86
DRAFT - DO NOT CITE OR QUOTE
-------
Table 5-1. Incidences of noncancerous kidney and liver effects in rats
following oral exposure to HCE
Duration
Dose
Study
(route)
Strain/sex/species
Endpoint
(mg/kg-day)
Incidence
NTP (1989)
103 wks
F344 male rat
Linear mineralization
0
2/50 (4%)
(gavage)
7
15/50a (30%)
14
32/50a (64%)
NTP (1989)
103 wks
F344 male rat
Hyperplasia of the renal pelvic
0
0/50 (0%)
(gavage)
transitional epithelium
7
7/50a (14%)
14
7/50a (14%)
Gorzinski
16 wks
F344 male rat
Slight hypertrophy and/or
0
0/10 (0%)
etal. (1985)
(dietary)
dilation of proximal convoluted
renal tubules
1
1/10 (10%)
15
7/10a (70%)
62
10/10a
(100%)
Gorzinski
16 wks
F344 male rat
Atrophy and degeneration of
0
1/10 (0%)
etal. (1985)
(dietary)
renal tubules
1
2/10 (20%)
15
7/10a (70%)
62
10/10a
(100%)
F344 female rat
Atrophy and degeneration of
0
1/10 (0%)
renal tubules
1
1/10 (10%)
15
2/10 (20%)
62
6/10a (60%)
Liver Effects
NTP (1989)
13 weeks
F344 male rat
Hepatocellular necrosis
0
0/10 (0%)
(gavage)
33.5
0/10 (0%)
67.1
0/10 (0%)
134.3
0/10 (0%)
267.8
1/10 (10%)
535.7
2/5 (40%)
F344 female rat
Hepatocellular necrosis
0
0/10 (0%)
33.5
0/10 (0%)
67.1
0/10 (0%)
134.3
2/10 (20%)
267.8
4/10a (40%)
535.7
8/10a (80%)
"EPA determined statistical significance using Fisher's Exact Test (p < 0.05).
These chronic and subchronic studies in rats and mice indicate that the kidney and liver
are both target organs of HCE oral toxicity in rodents. Given the number of effects reported in
the kidney and the greater sensitivity of these effects in available studies, the kidney is
considered the primary target of oral HCE exposure toxicity in rodents. HCE exposure resulted
in a number of kidney effects: atrophy and degeneration of renal tubules in male and female
87
DRAFT - DO NOT CITE OR QUOTE
-------
F344 rats (Gorzinski et al., 1985), slight hypertrophy and/or dilation of proximal convoluted
renal tubules in male F344 rats (Gorzinski et al., 1985), linear mineralization in male F344 rats
(NTP, 1989), tubular nephropathy in male and female F344 rats (NTP, 1989), hyperplasia of the
renal pelvic transitional epithelium in male F344 rats (NTP, 1989), and tubular nephropathy in
male and female Osborne-Mendel rats (NCI, 1978). Further consideration was given to these
endpoints as candidate critical effects for the determination of the point of departure (POD) for
derivation of the oral RfD.
Although the doses associated with hepatic effects were more than 10-fold higher than
doses associated with kidney effects, data from the NTP (1989) study on incidence of
hepatocellular necrosis from the female rats were also considered as candidate critical effects for
comparison purposes. The data on the male rat liver effects from the NTP (1989) study were not
considered because the incidence of hepatocellular necrosis was not significantly elevated above
controls at any HCE dose. The kidney effects reported in the 13-week subchronic study (NTP,
1989) were not further considered because the lack of the incidence data for the control groups
made it uncertain whether the 34 mg/kg-day HCE dose represented a LOAEL. In addition, the
HCE doses administered were more than fourfold higher than those doses associated with kidney
effects in other subchronic (Gorzinski et al., 1985) and chronic (NTP, 1989) studies. The ability
of the chronic NTP (1989) study to inform the effects observed at the lowest dose tested in the
Gorzinski et al. (1985) study is limited because the lowest dose tested in the chronic exposure
study represented a LOAEL. The chronic study in B6C3Fi mice (NCI, 1978) was not considered
for selection as the principal study because the HCE doses that induced kidney effects were more
than sevenfold higher than doses associated with kidney effects in rats following subchronic
(Gorzinski et al., 1985) or chronic (NTP, 1989; NCI, 1978) exposure.
5.1.2. Methods of Analysis—Including Models
The benchmark dose (BMD) modeling approach (U.S. EPA, 2000b) was employed to
identify the candidate POD for each of the endpoints described above. A benchmark response
(BMR) of 10% extra risk was considered appropriate for derivation under the assumption that it
represents a minimally biologically significant response level. All of the dichotomous dose-
response models available in the EPA benchmark dose software (BMDS), version 2.0, were fit to
the incidence data for kidney effects in male and female rats reported by NTP (1989), NCI
(1978), and Gorzinski et al. (1985), as well as the incidence data for hepatocellular necrosis in
female rats reported by NTP (1989). Details of the BMD dose-response modeling reported in
Table 5-2 are presented in Appendix B (Table B-l). In addition, the BMD and 95% lower bound
confidence limit on the BMD (BMDL) modeling outcomes for a BMR of 5 and 1% are presented
in Appendix B (Table B-2) for comparison with the 10% BMR. From the BMD modeling
analysis results presented in Table B-l, candidate PODs were selected. Table 5-2 summarizes
88
DRAFT - DO NOT CITE OR QUOTE
-------
the BMD modeling results of the available data and the BMR levels and the candidate PODs are
identified for each effect.
Table 5-2. Summary of the BMD modeling results for the kidney
Study
Endpoint
Sex/species
(group size)
Duration
(route)
"Best-fit"
model
BMD
(mg/kg-d)
BMDL10
(mg/kg-d)
Gorzinski et al.
(1985)
Slight hypertrophy
and/or dilation of
proximal convoluted
renal tubules
Male rats
(n = 10)
16 wks
(dietary)
Gamma
Quantal-
linear, and
Weibull
1.22
0.710
Gorzinski et al.
(1985)
Atrophy and
degeneration of renal
tubules
Male rats
(n = 10)
16 wks
(dietary)
Gamma,
Multistage 1°,
and Quantal-
linear
1.34
0.728
Female rats
(n = 10)
Probit
16.10
10.51
NCI (1978)
Tubular
nephropathy
Male rats
(n ~ 50)
78 wks
(gavage)
Gamma,
Multistage 1°,
and Weibull
21.22
16.99
Female rats
(n ~ 50)
Multistage 2°
80.63
41.89
NTP (1989)
Increased severity of
tubular
nephropathy
Male rats
(n ~ 50)
103 wks
(gavage)
Probit 1°
3.81
2.60
Female rats
(n ~ 50)
Gamma,
Quantal-
linear, and
Weibull
15.17
10.72
NTP (1989)
Linear mineralization
Male rats
(n ~ 50)
103 wks
(gavage)
Probit
3.98
3.22
NTP (1989)
Hyperplasia of the
pelvic transitional
epithelium
Male rats
(n ~ 50)
103 wks
(gavage)
LogLogistic
7.05
4.48
The range of candidate PODs (approximately 60-0.6 mg/kg-day) is about 100-fold.
Kidney effects (i.e., tubular nephropathy, linear mineralization of the renal tubules, hyperplasia
of the pelvic transitional epithelium, atrophy and degeneration of renal tubules, and slight
hypertrophy and/or dilation of the proximal convoluted renal tubules) observed in male rats
resulted in lower candidate PODs than comparable effects in female rats.
89
DRAFT - DO NOT CITE OR QUOTE
-------
The most sensitive effect observed in male rats exposed to HCE is slight hypertrophy
and/or dilation of proximal convoluted renal tubules (Gorzinski et al., 1985). However, the
candidate POD for slight hypertrophy and/or dilation of proximal convoluted renal tubules (i.e.,
0.710 mg/kg-day) is nearly identical to the candidate POD for atrophy and degeneration of renal
tubules (i.e., 0.728 mg/kg-day). As tubular nephropathy in the chronic studies (NTP, 1989; NCI,
1978) was characterized as atrophy and degeneration of renal tubules, this endpoint has been
consistently observed following HCE exposure in several studies. Therefore, atrophy and
degeneration of renal tubules was selected as the candidate critical effect for this subchronic
exposure study. As shown in Appendix B, the gamma, multistage 1°, logistic, probit, Weibull
models in BMDS (version 2.0) provided adequate fits to the incidence data for atrophy and
degeneration of renal tubules in male rats from the Gorzinski et al. (1989) 16-week study (Table
B-l), as assessed by a %2goodness-of-fit p-values, as well as BMDio and BMDLio estimates from
these models were within a factor of three of each other, suggesting no appreciable model
dependence. The models with the lowest Akaike's information criterion (AIC; a measure of the
deviance of the model fit that allows for comparison across models for a particular endpoint)
values were for the gamma, multistage 1°, and quantal-linear models; therefore, the model with
the lowest BMDLio was selected. These models had identical BMDio and BMDLio values.
Therefore, the BMDLio of 0.728 mg/kg-day associated with a 10% extra risk for nephropathy in
male rats was selected as the candidate POD for these data.
The tubular nephropathy in male rats observed in the chronic exposure study (NTP, 1989)
resulted in higher PODs than the atrophy and degeneration of renal tubules in male rats observed
following 16 weeks of HCE exposure (Gorzinski et al., 1985). The ability of the chronic NTP
(1989) study to inform the effects observed at the lowest dose tested in the Gorzinski et al.
(1985) study is limited because the lowest dose tested in the chronic exposure study represented
a LOAEL. Therefore, the Gorzinski et al. (1985) study was selected as the principal study and
atrophy and degeneration of renal tubules in male rats was selected as the critical effect. The
BMDLio of 0.728 mg/kg-day was selected as the POD and serves as the basis for the derivation
of the oral RfD for HCE. This endpoint is supported by additional kidney effects associated with
oral exposure to HCE and supports the weight of evidence for HCE-associated nephrotoxicity.
5.1.3. RfD Derivation—Including Application of Uncertainty Factors (UFs)
The derivation of the RfD for atrophy and degeneration of renal tubules in male F344 rats
from the Gorzinski et al. (1985) 16-week toxicity study was calculated from the BMDLio of
0.728 mg/kg-day. The composite UF of 3,000 was comprised of the following:
• A default interspecies UF (UFA) of 10 was applied to account for the variability in
extrapolating from rats to humans. Although the toxicokinetics have been minimally
90
DRAFT - DO NOT CITE OR QUOTE
-------
evaluated in animals, the toxicokinetics of HCE have not been fully characterized in
either rats or humans.
• A default intraspecies UF (UFH) of 10 was applied to adjust for potentially sensitive
human subpopulations in the absence of information on the variability of response to
HCE in the human population. Current information is unavailable to assess human-
to-human variability in HCE toxicokinetics and toxicodynamics.
• The study selected as the principal study was a 16-week study by Gorzinski et al
(1985), a study duration that is minimally past the standard subchronic (90-day) study
and falls well short of a standard lifetime study. Kidney effects were observed in
male rats in the Gorzinski et al. (1985) subchronic study at doses below the range of
exposure tested in the available chronic exposure studies. In addition, the ability of
the available chronic studies to inform the effects observed at the low dose is limited
because the lowest dose tested in the NTP (1989) chronic exposure study represented
a LOAEL. Therefore, there are no data to exclude the possibility that chronic
exposure could increase the severity of the observed kidney effects or could result in
similar effects at lower doses. For these reasons, a sub chronic-to-chronic UF (UFS)
of 10 was used to account for the extrapolation from subchronic-to-chronic exposure
duration.
• An UF for a LOAEL to a NOAEL extrapolation was not applied because the current
approach is to address this extrapolation as one of the considerations in selecting a
BMR for BMD modeling. In this case, a BMR of a 10% increase in the incidence of
renal tubule atrophy and degeneration was selected under an assumption that it
represents a minimal biologically significant change.
• An UF of 3 was applied to account for deficiencies in the HCE toxicity database,
including the lack of a multigenerational reproductive study. The database includes
studies in laboratory animals, including chronic and subchronic dietary exposure
studies and two oral developmental toxicity studies. One of the available oral
developmental toxicity studies demonstrated that HCE exposure decreased gestational
indices and fetal viability, and increased resorptions with maternal toxicity at
500 mg/kg-day (Weeks et al., 1979). The second oral developmental toxicity study
showed maternal toxicity at both the mid- and high doses (167 and 500 mg/kg-day)
with decreased fetal body weight and increased late stage resorptions and skeletal
variations at the high dose (Shimizu et al., 1992). The toxic effects observed in the
developmental toxicity studies were observed at doses higher than those observed to
induce renal toxicity in the subchronic and chronic toxicity studies. Therefore, in
consideration of the oral database for HCE, a database UF of 3 was applied to account
for the lack of a two-generational reproductive study.
Given the UFs established above, the RfD for HCE was calculated employing the
following equation:
91
DRAFT - DO NOT CITE OR QUOTE
-------
RfD = POD - UF
= 0.728 mg/kg-day ^ 3,000
= 2 x 10"4 mg/kg-day
5.1.4. RfD Comparison Information
The predominant noncancer effect of acute, short-term, subchronic, and chronic oral
exposure to HCE is renal toxicity. Table 5-3 presents the potential PODs for nephrotoxicity in
male rats with applied UFs and potential reference values. Figure 5-1 provides a graphical
display of dose-response information from three studies that reported kidney toxicity in male rats
following chronic and subchronic oral exposure to HCE, focusing on potential PODs that could
be considered in deriving the oral RfD. As discussed in Sections 5.1.1 and 5.1.2, among those
studies that demonstrated kidney toxicity, atrophy and degeneration of renal tubules in male
F344 rats from the Gorzinski et al. (1985) study provided the POD for deriving the RfD (see
dotted box in Figure 5-1). Potential reference values that might be derived from other studies are
also presented. Only endpoints observed in male rats are presented because the database for
HCE consistently showed that male rats exhibited greater sensitivity to HCE toxicity compared
with females.
Table 5-3. Potential PODs for nephrotoxicity in male rats with applied UFs
and potential reference values
Potential PODs (mg/kg-day)
Total
UF
ufa
UFh
UFS
ufd
Potential
reference values
(mg/kg-d)
Reference
Tubular nephropathy; BMDL (2-yr)
16.99
300
10
10
1
3
0.0566
NCI (1978)
Hyperplasia of pelvic transitional
epithelium; BMDL (2-yr)
4.48
300
10
10
1
3
0.0149
NTP
(1989)
Linear mineralization; BMDL (2-yr)
3.22
300
10
10
1
3
0.0107
Moderate to marked tubular
nephropathy; BMDL (2-yr)
2.60
300
10
10
1
3
0.0087
Slight hypertrophy and/or dilation of
proximal convoluted renal tubules;
BMDL (16-wk)
0.710
3,000
10
10
10
3
0.0002
Gorzinski
et al.
(1985)
Atrophy and degeneration of renal
tubules; BMDL (16-wk)
0.728
3,000
10
10
10
3
0.0002
92
DRAFT - DO NOT CITE OR QUOTE
-------
Figure 5-1. Array of potential PODs with applied UFs and potential
reference values for nephrotoxic effects of studies in Table 5-3.
100
10
0.1
0.01
0.001
0.0001
£ Point of Departure
(HJ UPA, Interspecies
|xj Ufit Jntrsspecies
HUF*. Subchronic to1
Chrome
SI IJF1X Database
r I <:: ense Dose
Increased
incidence of
moderate to
matked
tubular
nephropathy
BMDL
NIP. 1989
2-yr
Increased
incidence of
hyperplasia of
the pelvic
transitional
epithelium
BMDL
OTP, 1989
2-yr
Increased
incidence
mineralization
BMDL
NIP, 1989
2-yr
Increased
incidence of
tubular
nephropathy
BMDL
NCI, 1978
91-wk
Increased
incidence of
slight
hypertrophy
and/or dilution
of convoluted
renal tubules
BMDL
Goranski et
ml,, 1985
16-wk
Increased
incidence of
atrophy and
degeneration of
tubules
BMDL
Gofianski et
al„ 1985
16-wk
93
DRAFT - DO NOT CITE OR QUOTE
-------
The nephropathy observed by NCI (1978) was similar to that reported by NTP (1989);
however, the animals in the NTP study were exposed to and exhibited effects at a lower range of
doses of HCE than those in the NCI study (Table 5-1). NTP (1989) described tubular
nephropathy characterized by degeneration, necrosis, and regenerative epithelial cells in rats.
Gorzinski et al. (1985) described similar renal effects characterized by atrophy and degeneration
of renal tubules and slight hypertrophy and/or dilation of proximal convoluted tubules. Linear
mineralization of the renal tubules, hyperplasia of the pelvic transitional epithelium, slight
hypertrophy and/or dilation of the proximal convoluted tubules, increased severity of tubular
nephropathy, and atrophy and degeneration or renal tubules were all reported in male rats
exposed to HCE (NTP, 1989; Gorzinski et al., 1985). Additionally, nephropathy was observed
in both male and female rats, whereas linear mineralization was only observed in male rats.
Kidney effects were observed in male rats in the Gorzinski et al. (1985) study at doses below the
range of exposure tested in the NTP (1989) study. In addition, the ability of the chronic studies
to inform the effects observed at the low dose in the Gorzinski et al. (1985) study is limited
because the lowest dose tested in the NTP (1989) chronic exposure study represented a LOAEL.
The potential POD associated with atrophy and degeneration of renal tubules from the Gorzinski
et al. (1985) study was lower than the POD based on increased severity of tubular nephropathy
from NTP (1989). Therefore the POD based on atrophy and degeneration of renal tubules from
the Gorzinski et al. (1985) study was selected to serve as the basis for the derivation of the RfD.
5.1.5. Previous RfD Assessment
In the previous RfD assessment for HCE, completed in 1987, the Gorzinski et al. (1985)
study was employed in deriving the RfD using the NOAEL/LOAEL approach. In this study, the
identified LOAEL for atrophy and degeneration of renal tubules was 15 mg/kg-day, with a
corresponding NOAEL of 1 mg/kg-day. A composite UF of 1,000 was employed to account for
the following three limitations or uncertainties: (1) interspecies extrapolation (UFA = 10);
(2) intraspecies variation (UFH = 10); and (3) subchronic-to-chronic extrapolation (UFS = 10).
An RfD of 1 x 10 3 mg/kg-day was derived. For the current assessment, the atrophy and
degeneration of renal tubules in rats reported by Gorzinski et al. (1985) also served as the basis
for the RfD; however, BMD modeling was used to derive a POD, and an additional UF of 3 for
database deficiencies was applied.
5.2. INHALATION REFERENCE CONCENTRATION (RfC)
5.2.1. Choice of Principal Study and Critical Effect—with Rationale and Justification
The database of inhalation toxicity studies on HCE is limited. Human studies
demonstrated HCE exposure in smoke bomb production workers, but the sample sizes are too
small to reach definitive conclusions regarding health effects and the exposure was likely a
mixture of HCE and zinc oxide. There are no chronic studies available, and only a single
94
DRAFT - DO NOT CITE OR QUOTE
-------
subchronic inhalation study (in four species) that included a developmental toxicity experiment
is available. Weeks et al. (1979) exposed Sprague-Dawley rats, male Beagle dogs, male Hartley
guinea pigs, and Japanese quail to HCE air concentrations of 145, 465, or 2,517 mg/m3 for
6 hours/day, 5 days/week, for 6 weeks. Postexposure observations were carried out for
12 weeks.
As discussed in Section 4.4.3.2, toxic effects observed in treated rats, dogs, and guinea
pigs (the quail did not show signs of toxicity) were at the highest exposure level, 2,517 mg/m3,
except for dams in the 465 mg/m3 exposure group in the developmental study, which exhibited
significantly decreased body weight gain and an increased incidence (85%) of mucopurulent
nasal exudate. This inflammatory exudate was observed in 100% of the dams treated with 2,517
mg/m3. Similar to the dams, male and female rats exposed to 2,517 mg/m3 HCE for 6 weeks
exhibited mucopurulent exudate in the nasal turbinates. Excess mucus in the nasal turbinates
was also observed in 2/10 quail in the 2,517 mg/m3 concentration group. Effects of this nature
were not observed in the 465 or 145 mg/m3 rats and quail or in the treated guinea pigs and dogs.
Weeks et al. (1979) concluded that the excess mucus in two of the 2,517 mg/m3 quails
was a transient effect of the HCE exposure because there was no evidence of inflammatory cells
or tissue damage. The authors attributed the increased incidence of respiratory lesions in rats to
an endemic mycoplasia infection as evidenced by the histopathological observation of an
increased incidence and severity of mycoplasia-related lesions in the nasal turbinates
(mucopurulent exudate), trachea (lymphoid hyperplasia in the lamina propria), and lung
(pneumonitis) of 2,517 mg/m3 male and female rats. Similar lesions characteristic of respiratory
mycoplasmosis in rodents were detected in an oral developmental study in rats that paralleled the
inhalation developmental study described above (both conducted by Weeks et al., 1979).
Irritation of the upper respiratory tract was observed in approximately 70% of the pregnant rats
(20%) diagnosed with subclinical pneumonitis) orally exposed to 500 mg/kg HCE, compared
with 10%) of controls showing irritation and pneumonitis.
The presence of the infection in the rats in both the oral and inhalation studies and in the
controls of the oral study suggests that respiratory tract effects are a potentiation of the
underlying mycoplasia infection rather than a direct result of HCE exposure. Additionally, the
reduced weight gain in the rats could be related to the condition of the infected animals,
considering that mycoplasma-infected rodents generally gain less weight or lose weight
compared with noninfected rodents (Xu et al., 2006; Sandstedt et al., 1997). Reduced weight
gain was also observed in the 2,517 mg/m3 guinea pigs, but mycoplasma infection was not
reported (Weeks et al., 1979). Like rats and mice, guinea pigs can carry the mycoplasma
organism; however, they are not clinically affected (Fox et al., 1984; Holmes, 1984). No data
were presented demonstrating the presence of mycoplasma in the lungs; therefore, the respiratory
tract effects cannot be excluded from consideration as a potential critical effect.
95
DRAFT - DO NOT CITE OR QUOTE
-------
As discussed in Section 4.4.3, neurobehavioral effects were consistently observed in the
rats and dogs exposed to 2,517 mg/m3. The male and female rats in the 6-week study exhibited
tremors and ruffled pelt. The pregnant rats developed tremors on GDs 12-16. Similarly, the
dams exposed to 500 mg/kg HCE in the concurrent oral developmental study by Weeks et al.
(1979) experienced tremors on GDs 15 and 16 of the 11-day exposure period. The HCE-exposed
dogs showed tremors, ataxia, and hypersalivation, severe head bobbing, facial muscular
fasciculations, and closed eyelids. These effects were noted in the dogs throughout the study,
although they disappeared overnight during nonexposure time periods.
Supporting data for the study were reported in an acute study by Weeks and Thomasino
(1978), in which a single 8-hour inhalation exposure to 2,500 or 57,000 mg/m3 HCE and a single
6-hour exposure to 17,000 mg/m3 HCE in male Sprague-Dawley rats resulted in neurological and
lung effects. The male rats exposed to 57,000 mg/m3 HCE had reduced body weight gain
compared with controls over the 14 days postexposure. By 6 hours of exposure, one rat had a
staggered gait. Necropsy did not reveal any gross exposure-related lesions, although microscopy
revealed that two of these rats had subacute diffuse interstitial pneumonitis of minimal to
moderate severity and vascular congestion associated with these lung effects. Following 6 hours
of exposure to 17,000 mg/m3, the six rats in this group showed reduced weight gain compared
with controls and two of these rats exhibited a staggered gait. No exposure-related gross or
histopathological changes were observed in tissues and organs. These effects were not
noticeable 14 days postexposure.
The subchronic inhalation study by Weeks et al. (1979), as the only repeated exposure
study available, was selected as the principal study for the derivation of the RfC. This study
used three concentrations and incorporated a variety of endpoints (toxicological, teratological,
neurological, pulmonary) across a range of species (see Table 5-4). The primary limitation of
Weeks et al. (1979) is the minimal amount of quantitative information provided characterizing
the reported effects. Several experiments only utilized one sex, and additional exposure
concentration(s) between the mid- and high concentration would have allowed for better
characterization of the exposure-response curve. However, this study identified neurotoxicity,
statistically significant decreases in body weight gain, and upper and lower respiratory tract
irritation. The responses were generally observed following exposure to the highest
concentration, and not in the two lower concentrations. Considering the consistent observation
of neurotoxic effects across experiments in rats and dogs, these effects following inhalation
exposure to HCE were selected as the critical effect.
96
DRAFT - DO NOT CITE OR QUOTE
-------
Table 5-4. Noncancerous effects observed in animals exposed to HCE via
inhalation
Species
Dose/duration
NOAEL
(mg/m3)
LOAEL
(mg/m3)
Effect
Sprague-Dawley rats
(25/sex/dose)
0, 145, 465, or
2,517 mg/m3;
6 wks
465 mg/m3
2,517 mg/m3
Males: neurotoxic effects (tremors
and ruffled pelt), reduced body
weight gain, increased relative,
spleen, and testes weights
Females: neurotoxic effects (tremors
and ruffled pelt), increased relative
liver weight
Male Beagle dogs (4/dose)
465 mg/m3
2,517 mg/m3
Tremors, ataxia, hypersalivation,
head bobbing, facial muscular
fasciculations
Male Hartley guinea pigs
(10/dose)
465 mg/m3
2,517 mg/m3
Reduced body weight, increased
relative liver weight
Japanese quail (20/dose)
2,517 mg/m3
Not
established
No effects observed
Pregnant Sprague-Dawley
rats (22/dose)
0, 145, 465,
or 2,517 mg/m3;
GDs 6-16
Maternal:
465 mg/m3
Maternal:
2,517 mg/m3
Maternal: tremors
Developmental: no effects
Source: Weeks et al. (1979).
5.2.2. Methods of Analysis—Including Models
The Weeks et al. (1979) study included three exposure groups (145, 465, and
2,517 mg/m3) plus a control. Neurological effects were observed in male and female Sprague-
Dawley rats, male Beagle dogs, and pregnant Sprague-Dawley rats only at the highest dose
tested. Incidence data were not reported. Application of BMD modeling was precluded because
100% of the high-exposure animals displayed neurological effects. Therefore, a NOAEL served
as the POD. The NOAEL of 465 mg/m3, identified in Weeks et al. (1979), was selected as the
POD for the derivation of the RfC based on effects in male and female rats and male dogs
exposed to HCE for 6 weeks and pregnant rats exposed on GDs 6-16. Although the NOAELs
are the same, the male and female rats exposed to HCE for 6 weeks were selected as the study
animals upon which to base the POD, as the duration of exposure for the dams in the teratology
study was only 11 days and only four male dogs were exposed to HCE in the 6-week study.
The NOAEL is based on intermittent HCE inhalation exposures in male and female rats
for 6 hours/day, 5 days/week. Thus, prior to deriving the RfC, this POD was adjusted for
continuous exposure (24 hours/day, 7 days/week). The duration-adjusted POD (POD[Adj]) is
derived using the following equation (U.S. EPA, 1994b):
POD[adj] = (POD) x (hours of exposure/24 hours) x (days of exposure/7 days)
= (465 mg/m3) x (6/24 hours) x (5/7 days)
= 83.0 mg/m3
97
DRAFT - DO NOT CITE OR QUOTE
-------
The Methods for Derivation of Inhalation Reference Concentrations and Application of
Inhalation Dosimetry (hereafter referred to as the RfC Methodology) recommends converting the
POD[adj] to a human equivalent concentration (HEC) (U.S. EPA, 1994b). The RfC
Methodology separates gases into three categories based on their water solubility and reactivity
with tissues in the respiratory tract. Determining whether HCE is a Category 2 or 3 gas is
difficult because data regarding the inhalation effects are limited. HCE is slightly water soluble
and although HCE has been observed in blood following oral exposures to HCE, it is unknown
whether HCE accumulates in blood following inhalation exposure. Given this limited
information, HCE is likely a Category 2 gas because it is slightly water soluble and causes
effects distal to the site of inhalation exposure (i.e., systemic effects). For Category 2 gases,
HEC values are calculated using methods for Category 1 gases for portal-of-entry effects and
Category 3 methods for systemic effects (U.S. EPA, 1994b). In view of the fact that
neurotoxicity is a systemic effect, the methods for Category 3 gases were used to derive the
HEC.
The RfC Methodology (U.S. EPA, 1994b) suggests that HECs be estimated by applying
to the duration-adjusted exposure level (POD[Adj]), a dosimetric adjustment factor (DAF) that is
specific for the breathing characteristic of the species to be compared. The DAF for a
Category 3 gas is based on the regional gas dose ratio (RGDR), where the RGDR is the ratio of
the animal blood:gas partition coefficient (Hb/g)A and the human blood:gas partition coefficient
(Hb/g)H.
PODjhec] = POD[adj] x (Hb/g)A/(Hb/g)H
However, the human and animal blood partition coefficients for HCE are not known. In
accordance with the RfC Methodology (U.S. EPA, 1994b) when the partition coefficients are
unknown a ratio of 1 is used. This results in a NOAEL[Hec] of 83.0 mg/m3.
POD[hec] = POD[adj] x (Hb/g)A/(Hb/g)H
= 83.0 mg/m3 x 1
= 83.0 mg/m3
5.2.3. RfC Derivation—Including Application of Uncertainty Factors (UFs)
The NOAEL[hec] value of 83 mg/m3 for evidence of neurotoxicity in Sprague-Dawley
rats was used as the POD to derive the RfC for HCE. A composite UF of 3,000 was applied as
follows:
• For animal-to-human interspecies differences (UFA), a UF of 3 was applied to
account for the uncertainty in extrapolating from laboratory animals to humans. This
98
DRAFT - DO NOT CITE OR QUOTE
-------
value is adopted by convention, where an adjustment from an animal-specific
NOAELadj to a NOAELHec has been incorporated. Application of an UF of
10 would depend on two areas of uncertainty (i.e., toxicokinetic and toxicodynamic
uncertainties). In this assessment, the toxicokinetic component associated with HCE
is mostly addressed by the determination of an HEC as described in the RfC
methodology (U.S. EPA, 1994b). The toxicodynamic uncertainty is also accounted
for to a certain degree by the use of the applied dosimetry method and an UF of 3 is
retained to account for uncertainty regarding the toxicodynamic differences between
rats and humans.
• A default intraspecies UF (UFH) of 10 was applied to account for potentially sensitive
human subpopulations in the absence of information on the variability of response to
HCE in the human population. Information is currently unavailable to assess human-
to-human variability in HCE toxicokinetics and toxicodynamics.
• A sub chronic-to-chronic UF (UFS) of 10 was applied to account for the use of the
POD selected following a subchronic duration of exposure to HCE to estimate a
chronic exposure RfC.
• An UF for a LOAEL to a NOAEL extrapolation was not applied because this
assessment utilized a NOAEL as the POD.
• A 10-fold UF was used to account for deficiencies in the toxicity database on
inhalation exposure to HCE. The toxicity data on inhalation exposure to HCE is
limited and largely restricted to one subchronic (6-week) inhalation study (Weeks et
al., 1979) in rats, male dogs, male rabbits, and quail. The same investigators
performed a developmental study and an acute study in rats. Maternal toxicity was
observed at both doses. Fetuses of HCE-treated dams did not exhibit any significant
skeletal or soft tissue anomalies. The toxic effects observed in the dams in the
developmental study were similar to those observed in the rats exposed for 6 weeks,
although additional effects were observed in the rats exposed for a longer duration.
The absence of teratogenic effects does not abrogate concern given the paucity of the
inhalation database for HCE. The database lacks a multigeneration reproductive
toxicity study. In addition, the database lacks studies of neurotoxicity and
developmental neurotoxicity, endpoints of concern based on the available inhalation
data. Therefore, in consideration of the inhalation database for HCE, a database UF
of 10 was applied.
Given the UFs established above, the RfC for HCE was calculated employing the
following equation:
RfC = NOAEL [hec] - UF
= 83 mg/m3 3,000
= 0.028 mg/m3 or 3 x 10"2 mg/m3
99
DRAFT - DO NOT CITE OR QUOTE
-------
5.2.4. RfC Comparison Information
The predominant noncancer effect of subchronic inhalation exposure to HCE is
neurotoxicity. The other effects noted by Weeks et al. (1979) at the same dose level were
decreases in body weight and increases in organ (liver or kidney) weights in male guinea pigs,
male and female rats, and pregnant rats. As discussed in Sections 5.2.1 and 5.2.2, the
neurotoxicity reported in the available inhalation study (Weeks et al., 1979) was selected for the
RfC derivation because of the consistent observation of the neurotoxicity. Based on the lack of
alternative endpoints to be considered for the basis of the RfC, a graphical display of dose-
response information from the subchronic inhalation study was not provided. For the reasons
discussed above and in Section 5.2.1, neurotoxic effects in male and female rats, pregnant rats,
and male dogs reported by Weeks et al. (1979) are considered the most sensitive effects and were
selected to serve as the basis for the derivation of the RfC for HCE.
5.2.5. Previous RfC Assessment
An RfC for HCE was not previously developed by the U.S. EPA. In the 1987 IRIS
Summary, Weeks et al. (1979) was briefly summarized in the Additional Studies/Comments
section for the oral RfD. The IRIS Summary (1987) stated that Weeks et al. (1979) administered
HCE to rats by inhalation at 145, 465, or 2,520 mg/mg3, 6 hours/day during gestation. At the two
highest doses, maternal toxicity was observed, but there was no evidence of fetoxicity or
teratogenicity. No additional discussion was presented in the IRIS Summary (1987) describing
why this study was not used to develop an RfC.
5.3. UNCERTAINTIES IN THE ORAL REFERENCE DOSE AND INHALATION
REFERENCE CONCENTRATION
The following discussion identifies uncertainties associated with the quantification of the
RfD and RfC for HCE. Following EPA practices and guidance (U.S. EPA, 1994b), the UF
approach was applied to the chosen PODs to derive an RfD and RfC (see Sections 5.1.3 and
5.2.3). Factors accounting for uncertainties associated with a number of steps in the analyses
were adopted to account for extrapolating from an animal study to human exposure, a diverse
human population of varying susceptibilities, and database deficiencies.
The oral database includes short-term, subchronic, and chronic studies in rats, and a
chronic study in mice, and developmental studies in rats. Toxicity associated with oral exposure
to HCE is predominantly reported as kidney toxicity, specifically, renal tubule nephropathy. The
inhalation database includes a subchronic study in rats, pregnant rats, male dogs, male guinea
pigs, and quail. Toxicity associated with inhalation exposure to HCE in this study is mainly
neurotoxicity. Critical data gaps have been identified in Section 4 and uncertainties associated
with data deficiencies are more fully discussed below.
100
DRAFT - DO NOT CITE OR QUOTE
-------
After consideration of the candidate PODs, the RfD of 2 x 10"4 mg/kg-day was derived
from a BMDLio of 0.728 mg/kg-day, which was based on the observation of atrophy and
degeneration of renal tubules in male F344 rats from the Gorzinski et al. (1985) 16-week toxicity
study. The dose-response relationships for oral exposure to HCE and nephropathy in other
studies of rats are also available for deriving an RfD, but are associated with higher
NOAELs/LOAELs that are less sensitive than the selected critical effect and corresponding
POD. The derived RfD was quantified using a BMDLio for the POD. The selection of the BMD
model for the quantitation of the RfD does not lead to significant uncertainty in estimating the
POD since benchmark effect levels were within the range of experimental data. However, the
selected models do not represent all possible models one might fit, and other models could be
selected to yield different results, both higher and lower than those included in this assessment.
Uncertainty exists in the selection of the BMR level utilized in the BMD modeling of the critical
effect (atrophy and degeneration of renal tubules in male F344 rats) to estimate the POD. In the
absence of information to identify the level of change in atrophy and degeneration of renal
tubules in male F344 rats related to a biologically significant change, a BMR of 10% was
selected for the modeling of the increased incidence to represent a minimally biologically
significant change.
The RfC was derived from a NOAEL[Hec] value of 83 mg/m3 for evidence of
neurotoxicity in Sprague-Dawley rats from a subchronic (6-week) inhalation study by Weeks et
al. (1979). A POD based on a NOAEL or LOAEL is, in part, a reflection of the particular
exposure concentration or dose at which a study was conducted. It lacks characterization of the
dose-response curve and for this reason is less informative than a POD obtained from benchmark
dose-response modeling. The subchronic inhalation study in rats (Weeks et al., 1979) was
selected as the principal study and neurotoxicity was identified as the critical effect. A NOAEL
of 465 mg/m3 was selected to serve as the POD and the basis for derivation of the RfC.
Extrapolating from animals to humans adds further uncertainty. The effect and its
magnitude at the POD in rats are extrapolated to a human response. Pharmacokinetic models are
useful for examining species differences in pharmacokinetic processing; however, dosimetric
adjustment using pharmacokinetic modeling was not available for oral exposure to HCE.
Information was unavailable to quantitatively assess toxicokinetic or toxicodynamic differences
between animals and humans, so a 10-fold UF was used to account for uncertainty in
extrapolating from laboratory animals to humans in the derivation of the RfD. For the RfC, a
factor of 3 was adopted by convention where an adjustment from an animal-specific NOAELadj
to a NOAELhec has been incorporated. Application of an UF of 10 would depend on two areas
of uncertainty (i.e., toxicokinetic and toxicodynamic uncertainties). In this assessment, the
toxicokinetic component is mostly addressed by the determination of a HEC as described in the
RfC methodology (U.S. EPA, 1994b). The toxicodynamic uncertainty is also accounted for to a
101
DRAFT - DO NOT CITE OR QUOTE
-------
certain degree by the use of the applied dosimetry method and an UF of 3 is retained to account
for this component.
Heterogeneity among humans is another uncertainty associated with extrapolating doses
from animals to humans. Uncertainty related to human variation needs consideration in
extrapolating dose from a subset or smaller sized population, say of one sex or a narrow range of
life stages typical of occupational epidemiologic studies, to a larger, more diverse population. In
the absence of HCE-specific data on human variation, a factor of 10 was used to account for
uncertainty associated with human variation in the derivation of both the RfD and RfC. Human
variation may be larger or smaller; however, HCE-specific data to examine the potential
magnitude of over- or under-estimation are unavailable.
Uncertainties associated with data gaps in the HCE database have been identified. Data
more fully characterizing potential multigenerational reproductive effects associated with both
oral and inhalation HCE exposure are lacking. The oral database includes studies in laboratory
animals, including chronic and subchronic dietary exposure studies and two oral developmental
toxicity studies. The developmental studies show effects at doses higher than those observed to
induce renal toxicity in the subchronic and chronic toxicity studies. Therefore, in consideration
of the entire oral database for HCE, a database UF of 3 was considered appropriate to account for
the lack of a two-generational reproductive study. There are no available human occupational or
epidemiological studies of inhalation exposure to HCE. There are no standard chronic toxicity
or multigeneration reproductive toxicity animal studies available for inhalation exposure to HCE.
The toxicity data on inhalation exposure to HCE is limited and largely restricted to one
subchronic (6-week) inhalation study (Weeks et al., 1979) in rats, male dogs, male rabbits, and
quail. The same investigators performed a developmental study and an acute study in rats. The
developmental study in rats did not provide any evidence of teratogenic effects. However, these
data do not abrogate concern given the paucity of the inhalation database for HCE. In addition,
the inhalation database lacks studies of developmental neurotoxicity, endpoints of concern based
on the available inhalation data (critical effect for the RfC). Therefore, in consideration of the
inhalation database for HCE, a database UF of 10 is was applied.
5.4. CANCER ASSESSMENT
There are no available studies on cancer in humans associated with exposure to HCE.
NTP (1989) provided evidence of renal adenomas and carcinomas and pheochromocytomas and
malignant pheochromocytomas in male F344/N rats in a 2-year cancer bioassay. NCI (1978)
provided evidence of hepatocellular carcinomas in male and female B6C3Fi mice in a 91-week
cancer bioassay. Additionally, HCE was shown to be a promoter, but not an initiator, in an
Osborne-Mendel rat liver foci assay (Milman et al., 1988; Story et al., 1986). Binding of
radiolabeled carbon to DNA, RNA, and protein was observed following [14C]-HCE
administration in both in vitro and in vivo assays in mice and rats (Lattanzi et al., 1988).
102
DRAFT - DO NOT CITE OR QUOTE
-------
Under the Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005a), HCE is "likely
to be carcinogenic to humans" based on dose-dependent, statistically significant increases in the
incidence of renal adenoma or carcinoma combined in male F344/N rats, statistically significant
increases in the incidence of pheochromocytomas/malignant pheochromocytomas combined in
male F344/N rats (NTP, 1989), and statistically significant increases in the incidence of
hepatocellular carcinomas in male and female B6C3Fi mice (NCI, 1978).
5.4.1. Choice of Study/Data—with Rationale and Justification
Two animal studies were selected for BMD analysis and subsequent quantitative cancer
assessment. In the first study, NTP (1989) reported statistically significantly elevated incidences
of renal adenomas and carcinomas combined and pheochromocytomas, malignant
pheochromocytomas, and complex pheochromocytomas combined in male F344 rats
administered HCE via gavage for 2 years. Female rats in this study did not exhibit any
HCE-related tumors. In the second study, NCI (1978) reported statistically significantly elevated
incidences of hepatocellular carcinomas in both sexes of B6C3Fi mice administered HCE via
gavage for 78 weeks. However, male mice in this study demonstrated a dose-response
relationship, while female mice did not.
5.4.2. Dose-response Data
NTP (1989) administered, via gavage, TWA doses of 7 or 14 mg/kg-day HCE to male
F344/N rats and TWA doses of 57 or 114 mg/kg-day HCE to female F344/N rats for 103 weeks.
No HCE-related tumors were observed in female rats. Renal adenomas and carcinomas
combined were observed in 2, 4, and 14% (statistically significant) of male rats administered 0
(controls), 7, and 14 mg/kg-day HCE, respectively. Male rats also exhibited increased
incidences of pheochromocytomas and malignant pheochromocytomas combined; 28, 58
(statistically significant), and 39% in the control, 7 and 14 mg/kg-day dose groups, respectively
(NTP, 1989). The NCI (1978) gavage study administered TWA doses of 0, 360, and 722 mg/kg-
day HCE to male and female B6C3Fi mice for 91 weeks. Statistically significant increases in
the incidence of hepatocellular carcinomas were observed in 15, 30, and 63% of males and 10,
40, and 31% of females in the control, 360, and 722 mg/kg-day dose groups, respectively.
Both NTP (1989) and NCI (1978) are well-designed studies, conducted in both sexes of
two species with 50 animals/sex/dose. Each study utilized two dose groups of HCE and an
untreated control group, with examination of a wide range of toxicological endpoints in both
sexes of the rodents. Tumor incidences were elevated over controls at two sites in rats (NTP,
1989) and at one site in mice (NCI, 1978). Some limitations associated with the NCI (1978)
study in mice include changes to the dosing regimen 9 weeks into the study, cyclical dosing
periods, and decreased survival in all study groups for the male mice. Individual animal data
103
DRAFT - DO NOT CITE OR QUOTE
-------
were unavailable to perform time-to-tumor modeling or adjust the tumor incidences for survival
before BMD modeling. The cancer incidence data are summarized in Table 5-5.
Table 5-5. Summary of incidence data in rodents orally exposed to HCE for
use in cancer dose-response assessment
Study
Sex/strain/species
Endpoint
HCE dose
(mg/kg-day)
Incidence
NTP (1989)
Male F344 rats
Kidney adenoma or
carcinoma
0
1/50 (2%)
7.1
2/50 (4%)
14.3
7/50 (14%)a
NTP (1989)
Male F344 rats
Pheochromocytomas/
malignant
pheochromocytomas
0
14/50 (28%)
7.1
26/45 (58%)a
14.3
19/49 (39%)
NCI (1978)
Male B6C3Fi mice
Hepatocellular
carcinoma
0
3/20 (15%)
360
15/50 (30%)a
722
31/49 (63%)a
NCI (1978)
Female B6C3Fi mice
Hepatocellular
carcinoma
0
2/20 (10%)
360
20/50 (40%)a
722
15/49 (31%)a
"Denotes statistical significance.
5.4.3. Dose Adjustments and Extrapolation Methods
The HCE doses administered to laboratory animals were scaled to human equivalent
doses (HEDs) according to EPA guidance (U.S. EPA, 2005a, 1992). More specifically, animal
doses were converted to HEDs by assuming that doses in animals and humans are toxicologically
equivalent when scaled by body weight raised to the 3/4 power, as follows:
Dose(mg / day) \animal~\ Dose(mg / day) \human\
BW 3/4 [animal] BW ^[human]
The body weights for the laboratory animals used in the scaled human dose conversions
are the mean body weights reported in the studies for each dose group. The following formula
was used for the conversion of oral animal doses to oral HEDs:
Scaled human dose (HED) = animal dose x (animal body weight/human body weight)'4
Therefore, the HCE doses of 7 and 14 mg/kg-day employed by NTP (1989) in rats were
converted to HEDs, as follows:
104
DRAFT - DO NOT CITE OR QUOTE
-------
Scaled human dose (HED) = 7 mg/kg-day x (0.483 kg/70 kg) "
= 2.05 mg/kg-day
Scaled human dose (HED) = 14 mg/kg-day x (0.471 kg/70 kg) "
= 4.10 mg/kg-day
Similarly, the HCE doses of 360 and 722 mg/kg-day employed by NCI (1978) in mice
were converted to HEDs, as follows:
Scaled human dose (HED) = 360 mg/kg-day x (0.033 kg/70 kg) "
= 53.05 mg/kg-day
Scaled human dose (HED) = 722 mg/kg-day x (0.030 kg/70 kg) "
= 103.88 mg/kg-day
These scaled human doses are used in the dose-response modeling described below.
The multistage model was the primary model considered for fitting the dose-response
data and is given by:
P(d) = 1 - exp[-(q0 + qid + q2d2 + ... + qkdk)],
where:
P(d) = lifetime risk (probability) of cancer at dose d
q, = parameters estimated in fitting the model, i = 1, ..., k
And extra risk is defined as (P(d) -P(0))/(1-P(0)).
The multistage model in BMDS (version 2.0) (U.S. EPA, 2008) was fit to the incidence
data summarized in Table 5-5 using the calculated HEDs in order to derive an oral slope factor
for HCE. The BMR selected was the default value of 10% extra risk recommended for
dichotomous models (U.S. EPA, 2000b). No data were excluded from the BMD multistage
modeling.
As stated above, the multistage model was fit to the incidences of renal adenomas or
carcinomas combined in male rats and hepatocellular carcinomas in male and female mice. In all
cases, the 2° multistage model provided the best fit. The multistage model was also fit to the
incidence of pheochromocytomas or malignant pheochromocytomas in male rats. The model
exhibited a significant lack of fit for the pheochromocytomas (according to the %2 statistic with
p < 0.01, see Appendix B for modeling output). Thus, this dataset was not useful for
dose-response assessment because the tumor incidence is not a monotonic increasing function of
dose, as demonstrated by the Cochran-Armitage Trend Test. Therefore, the BMD modeling
105
DRAFT - DO NOT CITE OR QUOTE
-------
results for the kidney and liver tumors in rats and mice, respectively, are summarized in Table
5-6, with more detailed results contained in Appendix B.
Table 5-6. Summary of BMD modeling results for oral cancer
assessment of HCE
Study
Sex/strain/
species
Endpoint
"Best-fit"
model
BMR
BMD10
BMDL10
or POD
Oral slope
factor
(mg/kg-d)1
NTP
(1989)
Male F344
rats
Renal
adenomas/carcinomas
combined
2° Multistage
0.1
3.73
2.44
0.040984
NCI
(1978)
Male
B6C3Fi
mice
Hepatocellular
carcinomas
2° Multistage
0.1
37.03
14.44
0.006925
NCI
(1978)
Female
B6C3Fi
mice
Hepatocellular
carcinomas
2° Multistage
0.1
286.24
136.88
0.000730
The U.S. EPA Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005a)
recommend that the method used to characterize and quantify cancer risk from a chemical is
determined by what is known about the mode of action of the carcinogen and the shape of the
cancer dose-response curve. The linear approach is used as a default option if the mode of action
of carcinogenicity is not understood (U.S. EPA, 2005a). In the case of HCE, the mode of
carcinogenic action of HCE in the kidneys and livers of rats and mice, respectively, is unknown.
There are some data in experimental animals evaluating a2U-globulin accumulation and toxicity
in the kidney. As described in Section 4.7.3.1, two principal factors contribute to the conclusion
that the available data do not support an a2U-globulin mode of action for the development of renal
tumors: (1) the lack of information identifying the a2U-globulin protein in HCE-treated rats, and
(2) evidence of nephropathy in female rats as well as male and female mice (because the
a2u-globulin-related mode of action is specific for male rats). Therefore, a linear low-dose
extrapolation approach was used to estimate human carcinogenic risk associated with HCE
exposure.
5.4.4. Oral Slope Factor and Inhalation Unit Risk
The candidate oral slope factors were derived by linear extrapolation to the origin from
the POD by dividing the BMR by the BMDLio (the lower bound on the exposure associated with
a 10% extra cancer risk). The oral slope factor represents an upper bound estimate on cancer risk
associated with a continuous lifetime exposure to HCE. In accordance with the U.S. EPA
guidelines (2005a), an oral slope factor for renal tumors in male rats of 0.04 (mg/kg-day)"1 was
calculated by dividing the BMR of 0.1 by the human equivalent BMDLio of 2.44 mg/kg-day
(Appendix B). An oral slope factor for hepatocellular tumors in male mice of 0.007 (mg/kg-
106
DRAFT - DO NOT CITE OR QUOTE
-------
day)"1 was calculated by dividing the BMR of 0.1 by the human equivalent BMDLio of 14.44
mg/kg-day (Appendix B). An oral slope factor for hepatocellular tumors in female mice of
0.0007 (mg/kg-day)"1 was calculated by dividing the BMR of 0.1 by the human equivalent
BMDLio of 136.88 mg/kg-day (Appendix B). The rats exhibited greater sensitivity to
HCE-induced carcinogenicity than the mice. Thus, the risk estimate associated with the male
rats that developed renal adenomas or carcinomas was selected as the oral slope factor of 0.04
(mg/kg-day)"1 for HCE. The slope of the linear extrapolation from the central estimate (i.e.,
BMD) is 0.1/37.03 mg/kg-day or 3 x 10"3 (mg/kg-day)"1.
In the absence of data on the carcinogenicity of HCE via the inhalation route, an
inhalation unit risk has not been derived.
5.4.5. Uncertainties in Cancer Risk Values
Extrapolation of data from animals to estimate potential cancer risks to human
populations from exposure to HCE yields uncertainty. Several types of uncertainty may be
considered quantitatively, whereas others can only be addressed qualitatively. Thus, an overall
integrated quantitative uncertainty analysis cannot be developed. Major sources of uncertainty in
the cancer assessment for HCE are summarized in Section 5.4.5.1 and in Table 5-7.
107
DRAFT - DO NOT CITE OR QUOTE
-------
Table 5-7. Summary of uncertainties in the HCE cancer risk assessment
Consideration/
approach
Impact on oral slope
factor
Decision
Justification
Human relevance
of rodent tumor
data
Human risk could j or f,
depending on relative
sensitivity; if rodent
tumors proved not to be
relevant to humans, oral
cancer risk estimate
would not apply (i.e.,
human risk would j)
Kidney and adrenal
gland tumors in male
rats and liver tumors in
male and female mice
are relevant to human
exposure
It was assumed that rodent tumors are relevant
to humans; true correspondence is unknown.
The carcinogenic response occurs across
species. HCE is a multi-site carcinogen,
although direct site concordance is generally
not assumed (U.S. EPA, 2005a); consistent
with this view, some human tumor types are
not found in rodents.
Bioassay
Alternatives could t or j
oral slope factor by an
unknown extent
NTP study
Alternative bioassays in rats were unavailable.
A NCI (1978) bioassay in mice was available,
although mice were less sensitive than rats to
HCE carcinogenicity and were not utilized in
estimating carcinogenic risk to humans.
Species/gender
choice
Human risk could t or I,
depending on relative
sensitivity
Incidence of renal
adenoma/carcinoma in
male rats
It was assumed that humans are as sensitive as
the most sensitive rodent gender/species
tested; true correspondence is unknown.
Increased tumor incidence in mice resulted in
a lower risk estimate than rats. No increase of
kidney tumors was observed in female rats.
Dose metric
Alternatives could t or j
oral slope factor by an
unknown extent
Used administered
exposure
Experimental evidence supports a role for
metabolism in toxicity, but actual responsible
metabolites are not identified. If the
responsible metabolites are generated in
proportion to administered dose, the estimated
slope factor is an unbiased estimate.
Low-dose
extrapolation
procedure
Alternatives could t or j
oral slope factor by an
unknown extent
Multistage model to
determine POD, linear
low-dose extrapolation
from POD (default
approach)
Available mode-of-action data do not inform
selection of dose-response model; linear
approach employed in absence of support for
an alternative approach.
Cross-species
scaling
Alternatives could j or f
the oral slope factor
(e.g., 3.5-fold i [scaling
by body weight] or f
2-fold [scaling by
BW2/3])
BW3/4 (default
approach)
There are no data to support alternatives.
Because the dose metric was not an area under
the curve, BW3 4 scaling was used to calculate
equivalent cumulative exposures for
estimating equivalent human risks.
Statistical
uncertainty at
POD
i oral slope factor
1.5-fold ifBMD used as
the POD rather than
lower bound on POD
BMDL (preferred
approach for
calculating reasonable
upper bound slope
factor)
Limited size of bioassay results in sampling
variability; lower bound is 95% confidence
interval on administered exposure.
Human
population
variability in
metabolism and
response/sensitive
subpopulations
Low-dose risk f or J, to
an unknown extent
Considered
qualitatively
No data to support range of human
variability/sensitivity, including whether
children are more sensitive.
t = increase; j = decrease
108 DRAFT - DO NOT CITE OR QUOTE
-------
5.4.5.1. Sources of Uncertainty
Relevance to humans. The modes of action for the kidney (adenomas/carcinomas) and
adrenal gland tumors (pheochromocytomas) in male rats and liver tumors (hepatocellular
carcinomas) in male and female mice are unknown. There are some data in experimental
animals evaluating a2U-globulin accumulation and toxicity in the kidney. As described in
Section 4.7.3, two principal factors contribute to the conclusion that the available data do not
support an a2U-globulin mode of action for the development of renal tumors. First, the presence
of kidney effects in HCE-exposed male and female mice, which generally do not accumulate the
a2u-globulin protein, suggests a mode of action other than a2U-globulin nephropathy. Second,
none of the HCE studies performed the necessary immunohistochemical assays to confirm the
presence of a2U-globulin protein within the hyaline droplets observed following administration of
HCE (NTP, 1996, 1989). This represents a data gap, as the presence of a2U-globulin is necessary
to support an a2U-globulin mode of action.
The relevance of the mode of action of liver tumor induction to humans was considered
in Section 4.7.2. There is no available information regarding hepatic cancer associated with
HCE exposure in humans. The experimental animal literature, however, shows that oral
exposure to HCE induces liver tumors in male and female mice. It is possible that the HCE-
induced hepatocellular carcinomas in mice occur as a result of the binding of HCE metabolites to
liver macromolecules and the generation of free radicals during HCE metabolism, causing key
events in the carcinogenic process such as cytotoxicity, inflammation, and regenerative cell
proliferation. Limited information exists to distinguish the similarities and differences between
experimental animals and humans in terms of HCE metabolism or toxicity. However, these
potential key events have not been evaluated for HCE.
Pheochromocytomas are catecholamine-producing neuroendocrine tumors. The
relevance of rodent pheochromocytomas as a model for human cancer risk has been the subject
of discussion in the scientific literature (e.g., Greim et al., 2009; Powers et al., 2008). In humans,
pheochromocytomas are rare and usually benign, but may also present as or develop into a
malignancy (Eisenhofer et al., 2004; Lehnert et al., 2004; Elder et al., 2003; Goldstein et al.,
1999). Hereditary factors in humans have been identified as important in the development of
pheochromocytomas (Eisenhofer et al., 2004). Pheochromocytomas are more common in
laboratory rats, though evidence suggests that certain rat pheochromocytomas may have
similarity to human pheochromocytomas (Powers et al., 2009). Furthermore, mechanisms of
action inducing pheochromocytomas in rats are expected to occur in humans as well (Greim et
al., 2009). Therefore, in the absence of information indicating otherwise, the kidney and adrenal
gland tumors in male rats and liver tumors in male and female mice are considered relevant to
humans.
109
DRAFT - DO NOT CITE OR QUOTE
-------
Bioassay selection. The study by NTP (1989) was used for the development of an oral
slope factor. This study was conducted in both sexes of F344/N rats and used 50 male and
50 female rats per dose group. Test animals were allocated among two dose levels of HCE and
an untreated control group. Animals were observed twice daily and examined weekly (for
14 weeks) then monthly for body weight and monthly for feed consumption. Animals were
necropsied and all organs and tissues were examined grossly and microscopically for
histopathological lesions for a comprehensive set of toxicological endpoints in both sexes.
Choice of species/gender. The oral slope factor for HCE was quantified using the tumor
incidence data for male rats, which were found to be more sensitive than male or female mice to
the carcinogenicity of HCE. The oral slope factor calculated from male rats was higher than the
slope factors calculated from male and female mice. As there is no information to inform which
species or gender of animals would be most applicable to humans, the most sensitive group was
selected for the basis of the oral slope factor. Though the mode of action for observed kidney
tumors in rodents is unknown, the evidence suggesting the kidney as a target organ of HCE
toxicity in both species lends strength to the concern for human carcinogenic potential.
Dose metric. HCE is likely metabolized to PERC and pentachloroethane; however, it is
unknown whether a metabolite or some combination of parent compound and metabolites is
responsible for the observed toxicity and carcinogenicity of HCE. If the actual carcinogenic
moiety(ies) is(are) proportional to administered exposure, then use of administered exposure as
the dose metric provides an unbiased estimate of carcinogenicity. On the other hand, if this is
not the most relevant dose metric, then the impact on the human equivalent slope factor is
unknown; the low-dose cancer risk value may be higher or lower than that estimated, by an
unknown amount.
Choice of low-dose extrapolation approach. The mode of action is a key consideration in
clarifying how risks should be estimated for low-dose exposure. A linear-low-dose extrapolation
approach was used to estimate human carcinogenic risk associated with HCE exposure, in the
absence of information to inform the dose-response at low doses. The extent to which the
overall uncertainty in low-dose risk estimation could be reduced if the mode of action for HCE
were known is of interest, but data on the mode of action of HCE are limited and the mode of
action is not known. If an a2u-globulin-associated mode of action is, in fact, responsible for male
rat tumor formation, then these tumors would not have been utilized for quantitation of cancer
risk as they would have been characterized as not relevant to humans.
Etiologically different tumor types were not combined across sites prior to modeling, in
order to allow for the possibility that different tumor types can have different dose-response
relationships because of varying time courses or other underlying mechanisms or factors. The
human equivalent oral slope factors estimated from the tumor sites with statistically significant
increases ranged from 0.007 to 0.04 per mg/kg-day, a range less than one order of magnitude,
with greater risk coming from the male rat kidney data.
110
DRAFT - DO NOT CITE OR QUOTE
-------
Choice of model. All risk assessments involve uncertainty, as study data are extrapolated
to make inferences about potential effects in humans from environmental exposure. The largest
sources of uncertainty in the HCE cancer risk estimates are interspecies extrapolation and low-
dose extrapolation. There are no human data from which to estimate human cancer risk;
therefore, the risk estimate must rely on data from studies of rodents exposed to levels greater
than would occur from environmental exposures.
Without human cancer data or better mechanistic data, the relevance of the rodent cancer
results to humans is uncertain. The occurrence of increased incidences of kidney and adrenal
gland tumors in male rats, and liver tumors in male and female mice exposed to HCE from the
oral route of exposure suggests that HCE is potentially carcinogenic to humans as well.
However, the lack of concordance in tumor sites between the two rodent species makes it more
difficult to quantitatively estimate human cancer risk.
Regarding low-dose extrapolation, in the absence of mechanistic data for biologically
based low-dose modeling or mechanistic evidence supporting a nonlinear approach (see the
discussion at the beginning of Section 5.4.3), a linear low-dose extrapolation was carried out
from the BMDLio. It is expected that this approach provides an upper bound on low-dose cancer
risk for humans. The true low-dose risks cannot be known without additional data.
With respect to uncertainties in the dose-response modeling, the two-step approach of
modeling only in the observable range (U.S. EPA, 2005a) and extrapolating from a POD in the
observable range is designed in part to minimize model dependence. Measures of statistical
uncertainty require assuming that the underlying model and associated assumptions are valid for
the data under consideration. The multistage model used provided an adequate fit to all the
datasets for kidney and liver tumors. For the multistage model applied to the incidence of
tumors, the BMDLs should generally be within a factor of 3 of the BMDs. This indicates that
there is a reasonably typical degree of uncertainty at the 10% extra risk level. A large difference
between the BMD and BMDL raises concern that the algorithm for the calculation of the BMDL
is not accurate (U.S. EPA, 2000b). The ratios of the BMDio values to the BMDLio values did
not exceed a value of 2.6, indicating that the estimated risk is not influenced by any unusual
variability in the model and associated assumptions.
Cross-species scaling. An adjustment for cross-species scaling (BW3 4) was applied to
address toxicological equivalence of internal doses between rats and humans, consistent with the
Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005a). It is assumed that equal risks
result from equivalent constant lifetime exposures.
Human population variability. The extent of inter-individual variability or sensitivity to
the potential carcinogenicity of HCE is unknown. There are no data exploring whether there is
differential sensitivity to HCE carcinogenicity across life stages. In addition, neither the extent
of interindividual variability in HCE metabolism nor human variability in response to HCE has
been characterized. Factors that could contribute to a range of human responses to HCE include
111
DRAFT - DO NOT CITE OR QUOTE
-------
variations in CYP450 levels because of age-related differences or other factors (e.g., exposure to
other chemicals that induce or inhibit microsomal enzymes), nutritional status, alcohol
consumption, or the presence of underlying disease that could alter metabolism of HCE or
antioxidant protection systems. This lack of understanding about potential susceptibility
differences across exposed human populations thus represents a source of uncertainty. Humans
are expected to be more genetically heterogenous than inbred strains of laboratory animals
(Calderon, 2000), and this variability is likely to be influenced by ongoing or background
exposures, diseases, and biological processes.
5.4.6. Previous Cancer Assessment
The previous HCE cancer assessment was based on the incidence of hepatocellular
carcinomas in male mice in the NCI (1978) study. The current risk value is derived from the
incidence of renal adenomas or carcinomas in male rats (NTP, 1989), resulting in an oral slope
factor approximately 2.8-fold higher than the one derived in the previous assessment.
In addition, the scaled human doses were calculated using a slightly different formula
than is current practice:
1/3
Scaled human dose = animal dose x (animal weight/human body weight) x (546/637)
The difference in the animal-to-human dose scaling procedure is due to the fact that
current practice bases dose equivalence on the 3/4 power of body weight instead of the previous
% power of body weight.
112
DRAFT - DO NOT CITE OR QUOTE
-------
6. MAJOR CONCLUSIONS IN Till CHARACTERIZATION OF HAZARD AND DOSE
RESPONSE
6.1. HUMAN HAZARD POTENTIAL
HCE is a halogenated hydrocarbon consisting of six chlorines attached to an ethane
backbone. HCE was produced in the United States from 1921 to 1967, but is currently not
commercially distributed. HCE is primarily used in the military for smoke pots, smoke
grenades, and pyrotechnic devices. In the past, HCE was used as antihelminthic for the
treatment of sheep flukes, but is no longer used for this purpose since the FDA withdrew
approval for this use in 1971. HCE has also been used as a polymer additive, a moth repellant, a
plasticizer for cellulose esters, and an insecticide solvent, and in metallurgy for refining
aluminum alloys.
There is limited information on the toxicity of HCE in humans. Current understanding of
HCE toxicology is based on the limited database of animal studies. After absorption by oral
exposure, HCE is primarily distributed to fat tissue. Toxicokinetic studies in animals indicated
that HCE is also localized and metabolized in the liver and kidney. Kidney concentrations of
HCE were higher in male rats than female rats (Gorzinski et al., 1985; Nolan and Karbowski,
1978). Studies of HCE metabolism indicated that the major CYP450 enzymes involved are
phenobarbital-inducible, which include the 2A, 2B, and 3A subfamilies (Salmon et al., 1985,
1981; Town and Leibman, 1984; Nastainczyk et al., 1982, 1981). HCE is putatively metabolized
via a pentachloroethyl free radical to PERC and pentachloroethane. Pentachloroethane is then
metabolized to TCE. TCE and PERC are further metabolized by hepatic oxidation to several
urinary metabolites including TCA, trichloroethanol, oxalic acid, dichloroethanol, dichloroacetic
acid, and monochloroacetic acid (Mitoma et al., 1985; Nastainczyk et al., 1982, 1981; Bonse and
Henschler, 1976; Fowler, 1969; Jondorf et al., 1957). Metabolism is minimal based on the few
studies that provided quantitative data on metabolites. However, several of these metabolites
have demonstrated liver and kidney toxicities similar to HCE.
The kidney has consistently been shown as the target for toxicity in acute, subchronic,
and chronic toxicity bioassays in animals (NTP, 1996, 1989; Gorzinski et al., 1985; NCI, 1978).
Noncancer effects include kidney degeneration (tubular nephropathy, necrosis of renal tubular
epithelium, hyaline droplet formation, tubular regeneration, and tubular casts) and hepatocellular
necrosis. Hepatotoxicity was noted in animals exposed to HCE, although endpoints of this
nature have not been evaluated in laboratory animals as fully as the renal effects. Hepatocellular
necrosis was reported in female rats (NTP, 1989), but was not evaluated in a chronic exposure
study of mice (NCI, 1978). The mouse study (NCI, 1978) focused on tumorigenic endpoints
rather than noncancer effects.
113
DRAFT - DO NOT CITE OR QUOTE
-------
There is no information available describing the metabolism of HCE following exposure
via inhalation. The inhalation database for HCE contains one acute (Weeks and Thomasino,
1978) and one subchronic (Weeks et al., 1979) study. Neurological effects, such as tremors and
ataxia, were observed in male Beagle dogs, male and female rats, and pregnant rats. Other
effects included reduced body weight gain and increased relative liver weight in rats and guinea
pigs exposed to HCE via inhalation. Male rats also displayed increased relative spleen and testes
weights.
Cancer effects observed in animal studies include hepatocellular carcinomas in mice and
renal adenomas or carcinomas and pheochromocytomas in rats. Under EPA's Guidelines for
Carcinogen Risk Assessment (U.S. EPA, 2005a), HCE is "likely to be carcinogenic to humans"
because HCE induced kidney and adrenal gland tumors in male rats and liver tumors in male and
female mice. Studies evaluating the carcinogenicity in humans exposed to HCE are unavailable.
The carcinogenicity incidence data in male rats (NTP, 1989) were used to develop a quantitative
cancer risk assessment for HCE. The consistency of the kidney and liver as target organs in
different species for HCE distribution and metabolism, and both noncancer and cancer endpoints,
provides support for the evaluation of these endpoints as relevant to humans.
6.2. DOSE RESPONSE
6.2.1. Oral Noncancer
Subchronic and chronic bioassays in rats and mice have identified the following
endpoints after exposure to HCE: tubular nephropathy, atrophy and degeneration of renal
tubules, and hepatocellular necrosis. In female rats, tubular nephropathy, atrophy and
degeneration of the renal tubules, and hepatocellular necrosis were observed in a statistically
significant dose-response manner (NTP, 1989; Gorzinski et al., 1985; NCI, 1978). Tubular
nephropathy, severity of nephropathy, and atrophy and degeneration of the renal tubules in male
rats demonstrated a statistically significant dose response. Although mice were evaluated in a
chronic exposure study (NCI, 1978), noncancer effects were not reported because this study was
focused on tumorigenic endpoints.
The most sensitive endpoint identified for HCE by oral exposure relates to kidney
toxicity in the 16-week feeding study by Gorzinski et al. (1985) in male rats. Gorzinski et al.
(1985) was selected as the principal study and atrophy and degeneration of renal tubules in male
rats were chosen as the critical effect for the derivation of the oral RfD. This study included both
sexes of F344 rats, 10 animals/sex/dose, and three dose groups plus controls (0, 1, 15, and 62
mg/kg-day). Dose-response analyses of the noncancer endpoint, atrophy and degeneration of
renal tubules (Gorzinski et al., 1985), using EPA's BMDS, resulted in a POD of
0.728 mg/kg-day. A composite UF of 3,000 was applied to the POD to derive an oral RfD of
2 x 10"4 mg/kg-day.
114
DRAFT - DO NOT CITE OR QUOTE
-------
Confidence in the principal study, Gorzinski et al. (1985), is high. The 16-week study is
a well-conducted study that used three dose groups plus a control. NTP (1989) also conducted
16-day, 13-week, and 103-week studies that supported the results observed in the 16-week study.
Application of BMD modeling provided a POD upon which to base the derivation of the RfD.
The critical effect on which the RfD is based is well-supported by other oral short-term,
subchronic, and chronic studies. Confidence in the database is low to medium because the
database includes acute, short-term, subchronic, and chronic toxicity studies and developmental
toxicity studies in rats and chronic carcinogenicity bioassays in rats and mice. The database
lacks a multigenerational reproductive study and studies in other species. Overall confidence in
the RfD is low to medium.
6.2.2. Inhalation Noncancer
The inhalation toxicity database is limited to a single 6-week repeat-exposure study by
Weeks et al. (1979). This study reported a NOAEL of 465 mg/m3 and a LOAEL of 2,517 mg/m3
in several species including Sprague-Dawley rats, male Beagle dogs, and male Hartley guinea
pigs. The effects described in this report include neurotoxicity, reduced body weight gain, and
increased relative liver, spleen, and testes weights. Based on neurological effects in Sprague-
Dawley rats, the NOAEL of 465 mg/m3 was selected to serve as the POD. Adjustments for
continuous exposure and for the HEC, resulted in the POD[Hec] of 83 mg/m3. An UF of 3,000
was applied to derive an inhalation RfC of 3 x 10"2 mg/m3. Confidence in the principal study,
Weeks et al. (1979), is low. The 6-week study was conducted in several species (including male
dogs, male and female rats, male guinea pigs, and quail). The study used three exposure groups
(145, 465, and 2,517 mg/m3) plus a control. The study is limited by the relatively short exposure
duration (6 weeks) and minimal reporting of effects, especially quantitative changes.
Application of BMD modeling was precluded based on a 100% response in animals for the
neurological effects and the lack of quantitative information. Therefore, a NOAEL served as the
POD. The critical effect on which the RfD is based is supported by the oral short-term study
conducted by the same investigators and two oral subchronic studies. Confidence in the database
is low because the database includes one acute and one subchronic toxicity study in multiple
species and one developmental toxicity study in rats. The database lacks studies by another
laboratory and a multigenerational reproductive study. Overall confidence in the RfC is low.
6.2.3. Cancer
Under EPA's Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005a), HCE is
"likely to be carcinogenic to humans" by all routes of exposure. This descriptor is based on
evidence of carcinogenicity from animal studies. HCE induced statistically significant increases
in the incidence of kidney and adrenal gland tumors in male rats and liver tumors in male and
female mice. The NTP (1989) rat study was selected for dose-response assessment based on
115
DRAFT - DO NOT CITE OR QUOTE
-------
statistically significant increased incidences of renal adenomas and carcinomas and adrenal
pheochromocytomas and malignant pheochromocytomas in male rats. This study was used for
development of an oral slope factor. This was a well-designed study, conducted in both sexes of
F344 rats with 50 rats/sex/dose, typical of carcinogenicity bioassays. Test animals were
allocated among two dose levels (7 and 14 mg/kg-day) and an untreated control group. Animals
were observed twice daily and examined weekly (for 14 weeks) and then monthly for body
weight and monthly for feed consumption. Animals were necropsied and all organs and tissues
were examined grossly and microscopically for histopathological lesions for a comprehensive set
of toxicological endpoints in both sexes.
Renal adenomas and carcinomas and pheochromocytomas and malignant
pheochromocytomas observed in male rats (NTP, 1989) were not seen in female rats or other
species orally-exposed to HCE. Hepatocellular carcinomas were observed in male and female
mice, but not in the rats. The male B6C3Fi mice tumor incidence data (NCI, 1978)
demonstrated evidence of carcinogenicity and a low-dose quantitative risk estimate was derived.
The cancer risk associated with mice exposed to HCE was less sensitive than that of rats. Thus,
the oral slope factor derived for HCE is based on the increased incidence of kidney tumors in
male rats.
A linear approach was applied in the dose-response assessment for HCE, in which the
mode of action is unknown, consistent with U.S. EPA's (2005a) Guidelines for Carcinogen Risk
Assessment. The guidelines recommend the use of a linear extrapolation as a default approach
when the available data are insufficient to establish a mode of action for a tumor site. As
discussed in Section 4.7, the mechanism leading to the formation of the kidney and adrenal
tumors in rats and the liver tumors in mice following oral exposure to HCE is unknown. The
database for HCE lacks information on the mode of action and the shape of the curve in the
region below the POD; therefore, a linear extrapolation was performed in determining the oral
slope factor in the derivation of a quantitative estimate of cancer risk for ingested HCE.
Increased incidence of renal adenomas and carcinomas in a 2-year rat bioassay (NTP,
1989) served as the basis for the oral cancer dose-response analysis. A multistage model using
linear extrapolation from the POD was performed to derive an oral slope factor of
4 x 10"2(mg/kg-day)~' for HCE. Extrapolation of the experimental data to estimate potential
cancer risk in human populations introduces uncertainty in the risk estimation for HCE.
Uncertainty can be considered quantitatively; however, some uncertainty can only be addressed
qualitatively. For this reason, an overall integrated quantitative uncertainty analysis cannot be
developed. However, EPA's development of the cancer quantitative assessment for HCE
included consideration of potential areas of uncertainty.
A biologically-based model was not supported by the available data; therefore, a
multistage model was the preferred model. The multistage model can accommodate a wide
116
DRAFT - DO NOT CITE OR QUOTE
-------
variety of dose-response shapes and provides consistency with previous quantitative dose-
response assessments for cancer. Linear low-dose extrapolation from a POD determined by an
empirical fit of tumor data has been judged to lead to plausible upper bound risk estimates at low
doses for several reasons. However, it is unknown how well this model or the linear low-dose
extrapolation predicts low dose risks for HCE. An adjustment for cross-species scaling (BW3 4)
was applied to address toxicological equivalence of internal doses between rats and humans
based on the assumption that equal risks result from equivalent constant lifetime exposures.
An inhalation unit risk was not derived in this assessment. Data on the carcinogenicity of
the compound via the inhalation route are unavailable, and route-to-route extrapolation was not
possible due to the lack of a PBPK model. However, it is proposed that HCE is likely to be
carcinogenic to humans by the inhalation route since the compound is absorbed and, in oral
studies, induces tumors at sites other than the portal of entry.
117
DRAFT - DO NOT CITE OR QUOTE
-------
7. REFERENCES
ACGIH (American Conference of Governmental Industrial Hygienists). (1991) Documentation of the threshold limit
values and biological exposure indices. 6th edition. Cincinnati, OH: American Conference of Governmental
Industrial Hygienists.
ACGIH (2001) Hexachloroethane. In: TLV chemical substances 7th edition. Cincinnati, OH: American Conference
of Governmental Industrial Hygienists.
Allen, MB; Crisp, A; Snook, N; et al. (1992) Smoke-bomb pneumonitis. RespirMed 86:165-166.
Ashby, J; Tennant, RW. (1988) Chemical structure, salmonella mutagenicity and extent of carcinogenicity as
indicators of genotoxic carcinogenesis among 222 chemicals tested in rodents by the U.S. NCI/NTP. Mutat Res
204:17-115.
ATSDR (Agency for Toxic Substances and Disease Registry). (1997c) Toxicological profile for hexachloroethane.
Atlanta, GA: U.S. Department of Health and Humans Services. Available from
http ://www.atsdr. cdc.gov/toxprofiles/tp97 .pdf.
ATSDR (1997b) Toxicological profile for tetrachloroethylene. Atlanta, GA: U.S. Department of Health and
Humans Services. Available online at http://www.atsdr.cdc.gov/.
ATSDR (1997a) Toxicological profile for trichloroethylene. Atlanta, GA: U.S. Department of Health and Humans
Services. Available online at http://www.atsdr.cdc.gov/.
ATSDR (2008) Toxicological profile for 1,1,2,2-tetrachloroethane. Atlanta, GA: U.S. Department of Health and
Humans Services. Available online at http://www.atsdr.cdc.gov/.
Axelson, O. (1985) Halogenated alkanes and alkenes and cancer: epidemiological aspects. In: Fishbein, L; O'Neill,
IK, eds. Environmental carcinogens: selected methods of analysis. Vol. 7. Lyon, France: International Agency for
Research on Cancer, pp. 5-20.
Beurskens, JE; Stams, AJ; Zehnder, AJ; et al. (1991) Relative biochemical reactivity of three
hexachlorocyclohexane isomers. Ecotoxicol Environ Saf 21:128-136.
Blanco, JG; Harrison, PL; Evans, W; et al. (2000) Human cytochrome P450 maximal activities in pediatric versus
adult liver. Drug Metab Disp 28(4):379-382.
Bonse, G; Henschler, D. (1976) Chemical reactivity, biotransformation, and toxicity of polychlorinated aliphatic
compounds. Crit Rev Toxicol 4:395-409.
Bronzetti, G; Morichetti, E; Del Carratore, R; et al. (1989) Tetrachloroethane, pentachloroethane, and
hexachloroethane: genetic and biochemical studies. Teratog Carcinog Mutagen 9:349-357.
Bronzetti, G; Morichetti, E; Vellosi, R; et al. (1990) Genotoxicity and effects on microsomal enzymes of three
chlorinated ethanes. Mutat Res 234:429-430.
Budavari, S; O'Neil, MJ; Smith, A; et al., eds. (1989) The Merck index: an encyclopedia of chemicals, drugs, and
biologicals. 11th edition. Rahway, NJ: Merck & Co., Inc., p. 740.
Bull, RJ; Sanchez, IM; Nelson, MA; et al. (1990) Liver tumor induction in B6C3F1 mice by dichloroacetate and
trichloroacetate. Toxicology 63:34lB359.
Calderon, RL. (2000) Measuring Risks in Humans: the Promise and Practice of Epidemiology. Food Chem Toxicol
38: S59-S63.
118
DRAFT - DO NOT CITE OR QUOTE
-------
Cal EPA (California Environmental Protection Agency). (2001) Public health goal for tetrachloroethylene in
drinking water. Office of Environmental Health Hazard Assessment. Available online at
http ://oehha. ca. gov/water/phg/pdf/PCE Aug2001 .pdf.
Cazeneuve, C; Pons, G; Rey, E; et al. (1994) Biotransformation of caffeine in human liver microsomes from
foetuses, neonates, infants and adults. Br J Clin Pharmacol 37:405-412.
ChemlDplus Advanced. (2005) Hexachloroethane. ChemFinder.com database & internet searching. Available
online at http://chem.sis.nlm.nih.gov/chemidplus/.
Crebelli, R; Benigni, R; Franekic, J; et al. (1988) Induction of chromosome malsegregation by halogenated organic
solvents inAspergillue nidulans: unspecific or specific mechanism? Mutat Res 201:401-411.
Crebelli, R; Andreoli, C; Carere, A; et al. (1992) The induction of mitotic chromosome malsegregation in
Aspergillus nidulans. Quantitative structure activity relationship (QSAR) analysis with chlorinated aliphatic
hydrocarbons. Mutat Res 266:117-134.
Crebelli, R; Andreoli, C; Carere, A; et al. (1995) Toxicology of halogenated aliphatic hydrocarbons: structural and
molecular determinants for the disturbance of chromosome segregation and the induction of lipid peroxidation.
Chem Biol Interact 98(2): 113-129.
Crebelli, R; Carere, A; Leopardi, P; et al. (1999) Evaluation of 10 aliphatic halogenated hydrocarbons in the mouse
bone marrow micronucleus test. Mutagenesis 14(2):207-215.
Doherty, AT; Ellard, S; Parry, EM; et al. (1996) An investigation into the activation and deactivation of chlorinated
hydrocarbons to genotoxins in metabolically competent human cells. Mutagenesis 11(3):247-274.
Doi, AM; Hill, G; Seely, J; et al. (2007) a2u-Globulin nephropathy and renal tumors in national toxicology program
studies. Toxicol Pathol 35(4): 533-540.
Dome, JLCM. (2004) Impact of inter-individual difference in drug metabolism and pharmacokinetics on safety
evaluation. Fundam Clin Pharmacol 18:609-620.
Elder, EE; Xu, D; Hoog, A; et al. (2003) KI-67 AND hTERT expression can aid in the distinction between
malignant and benign pheochromocytoma and paraganglioma. Mod Pathol 16(3):246-255.
Eisenhofer, G; Huynh, TT; Pacak, K; et al. (2004) Distinct gene expression profiles in norepinephrine- and
epinephrine-producing hereditary and sporadic pheochromocytomas: activation of hypoxia-driven angiogenic
pathways in von Hippel-Lindau syndrome. EndocrRelat Cancer 11(4):897-911.
Evans, WE; Relling, MF; de Graaf, S; et al. (1989) Hepatic drug clearance in children: studies with indocyanine
green as a model substrate. J Pharmacol Exp Ther 78:452-456.
Fiserova-Bergerova, V; Pierce, JT; Droz, PO. (1990) Dermal absorption potential of industrial chemicals: criteria for
skin notation. Am J Ind Med 17:617-635.
Fishbein, L. (1979) Potential halogenated industrial carcinogenic and mutagenic chemicals. II. Halogenated
saturated hydrocarbons. Sci Total Environ 11:163-195.
Fowler, JS. (1969) Some hepatotoxic action of hexachloroethane and its metabolites in sheep. Br J Pharmacol
35:530-542.
Fox, JG; Cohen, BJ; Loew, FM; eds. (1984) Laboratory animal medicine. New York, NY: Academic Press..
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.
Gargas, ML; Andersen, ME. (1989) Determining kinetic constants of chlorinated ethane metabolism in the rat from
rates of exhalation. Toxicol Appl Pharmacol 99:344-353.
119
DRAFT - DO NOT CITE OR QUOTE
-------
Gargas, ML; Seybold, PG; Andersen, ME. (1988) Modeling the tissue solubilities and metabolic rate constant Vmax
of halogenated methanes, ethanes and ethylenes: symposium on quantitative toxicology held at the 17th conference
on toxicology; November 3-5; Dayton, Ohio, USA. Toxicol Lett 43:235-256.
Goldstein, RE; O'Neill, JA, Jr; Holcomb, GW, III; et al. (1999) Clinical experience over 48 years with
pheochromocytoma. Ann Surg 229(6):755-764.
Goodman, DG.; Ward, JM; Squire, RA; et al. (1980) Neoplastic and non-neoplastic lesions in aging Osborne-
Mendel rats. Toxicol Appl Pharmacol 55:433-447.
Gorzinski, SJ; Nolan, RJ; McCollister, SB; et al. (1985) Subchronic oral toxicity, tissue distribution and clearance of
hexachloroethane in the rat. Drug Chem Toxicol 8:155-169.
Gorzinski, SJ; Wade, CE; McCollister, SB; et al. (1980) Hexachloroethane: results of a 16 week toxicity study in the
diet of CDF Fischer 344 rats. Midland, MI: Dow Chemical Company.
Greim, H: Hartwig, A; Reuter, U; et al. (2009) Chemically induced pheochromocytomas in rats: mechanisms and
relevance for human risk assessment. Critical Rev Toxicol 39(8):695-718.
Hard, GS; Rodgers, IS; Baetcke, KP; et al. (1993) Hazard evaluation of chemicals that cause accumulation of
-globulin. hyaline droplet nephropathy, and tubule neoplasia in the kidneys of male rats. Environ Health Perspect
99:313-349.
Haworth, S; Lawlor, T; Mortelmans, K; et al. (1983) Salmonella mutagenicity test results for 250 chemicals.
Environ Mutagen 5(Suppl 1):3—142.
Hodge, HC; Sterner, JH. (1949) Tabulation of toxicity classes. Am Ind Hyg Assoc Q 10:93-96.
Holmes, DD. (1984) Clinical laboratory animal medicine. Ames, IA: Iowa State University Press.
Howard, PH; ed. (1989) Handbook of environmental fate and exposure data for organic chemicals. Vol. I. Large
production and priority pollutants. Chelsea, MI: Lewis Publishers.
I ARC (International Agency for Research on Cancer). (1979) I ARC monographs on the evaluation of the
carcinogenic risk of chemicals to humans. Vol. 20. Some halogenated hydrocarbons. Lyon, France: International
Agency for Research on Cancer, p. 467.
IARC. (1999) IARC scientific publications no. 147. Species differences in thyroid, kidney and urinary bladder
carcinogenesis. Lyon, France.
Jackson, MA; Stack, HF; Waters, MD. (1993) The genetic toxicology of putative nongenotoxic carcinogens. Mutat
Res 296:241-277.
Jondorf, WR; Parke, DV; Williams, RT. (1957) The metabolism of [14C]hexachloroethane. Biochem J 65:14-15.
JISA (Japan Industrial Safety Association). (1993) Carcinogenicity study of tetrachloroethylene by inhalation in rats
and mice. Data No. 3-1. Kanagawa, Japan. Available from: IRIS Information Desk, U.S. Environmental Protection
Agency, Washington, DC.
Kinkead, ER; Wolfe, RE. (1992) Single oral toxicity of various organic compounds. J Am Coll Toxicol 11(6):713.
Kulig, B; Alleva, E; Bignami, G; et al. (1996) Animal behavioral methods in neurotoxicity assessment: SGOMSEC
joint report. Environ Health Perspect 104(Suppl 2): 193-204.
Lacroix, D; Sonnier, M; Moncion, A; et al. (1997) Expression of CYP3 A in the human liver-evidence that the shift
between CYP3A7 and CYP3A4 occurs immediately after birth. Eur J Biochem 247:625-634.
120
DRAFT - DO NOT CITE OR QUOTE
-------
Lattanzi, G; Colacci, A; Grilli, S; et al. (1988) Binding of hexachloroethane to biological macromolecules from rat
and mouse organs. J Toxicol Environ Health 24:403-411.
Legator, MS; Harper, BL. (1988) Mutagenicity screening/in vitro testing—the end of an era; animal and human
studies—the direction for the future. Ann NY Acad Sci 534:833-844.
Lehnert, H; Mundschenk, J; Hahn, K. (2004) Malignant pheochromocytoma. Front Horm Res 31:155-162.
Loh, CH; Chang, YW; Liou, SH; et al. (2006) Case report: hexachloroethane smoke inhalation: a rare cause of
severe hepatic injuries. Environ Health Perspect 114(5):763-765.
Loh, CH; Liou, SH; Chang, YW; et al. (2008) Hepatic injuries of hexachloroethane smoke inhalation: the first
analytical epidemiological study. Toxicology 247(2-3): 119-122.
Lohman, PHM; Lohman, WJA. (2000) Genetic activity profiles 2000 (program version 1.3.0), data record for
hexachloroethane. Data base and software are a joint effort of the U.S. Environmental Protection Agency and the
International Agency for Research on Cancer.
Lutz, WK. (1979) In vivo covalent binding of organic chemicals to DNA as a quantitative indicator in the process of
chemical carcinogenesis. Mutat Res 65:289-356.
Lutz, WK. (1986) Quantitative evaluation of DNA binding data for risk estimation and for classification of direct
and indirect carcinogens. J Cancer Res Clin Oncol 112:85-91.
Mather, GG; Exon, JH; Koller, LD. (1990) Subchronic 90-day toxicity of dichloroacetic and trichloroacetic acid in
rats. Toxicology 64:71-80.
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.
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.
Nakamura, S; Oda, Y; Shimada, T; et al. (1987) SOS-inducing activity of chemical carcinogens and mutagens in
Salmonella typhimurium TA1535/pSK1002: examination with 151 chemicals. Mutat Res 192:239-246.
Nastainczyk, W; Ahr, H; Uhich, V; et al. (1981) The mechanism of the reductive dehalogenation of polyhalogenated
compounds by microsomal cytochrome P450. Adv Exp Med Biol 136(A):799-808.
Nastainczyk, W; Ahr, HJ; Ullrich, V. (1982) The reductive metabolism of halogenated alkanes by liver microsomal
cytochrome P450. Biochem Pharmacol 131:391-396.
NCI (National Cancer Institute). (1976) Carcinogenesis bioassay of trichloroethylene (CAS No. 79-01-6). Public
Health Service, U.S. Department of Health and Human Services; NTP TR-2. Available from: National Institute of
Environmental Health Sciences, Research Triangle Park, NC. Available online at
http://ntp.mehs.nih.gov/index.cfm?objectid=07028C7F-AB6E-6D29-3FClCC9D48574701.
NCI. (1977) Bioassay of tetrachloroethylene for possible carcinogenicity. Public Health Service, U.S. Department
of Health, Education, and Welfare; NTP TR-13. Available from: National Cancer Institute, Bethesda, MD.
Available online at http://ntp.mehs.nih.gov/index.cfm?objectid=0702B823-CEA9-1089-DFBDC6F9207C56F2.
NCI. (1978) Bioassay of hexachloroethane for possible carcinogenicity. Public Health Service, U.S. Department of
Health, Education, and Welfare; NTP TR-68. Available from: National Cancer Institute, Bethesda, MD. Available
online at http://ntp.niehs.nih.gov/ntp/htdocs/LT_rpts/tr068.pdf.
121
DRAFT - DO NOT CITE OR QUOTE
-------
Nolan, RJ; Karbowski, RJ. (1978) Hexachloroethane: tissue clearance and distribution in Fischer 344 rats. Midland,
MI: Dow Chemical Company.
NRC (National Research Council) (1983) Risk assessment in the federal government: managing the process.
Washington, DC: National Academy Press.
NTP (National Toxicology Program). (1983) Carcinogenesis studies of pentachloroethane (CAS No. 76-01-7) in
F344/N rats and B6C3F1 mice (gavage study). Public Health Service, U.S. Department of Health and Human
Services; NTP TR-232. Available from National Institute of Environmental Health Sciences, Research Triangle
Park, NC. Available online at http://ntp.niehs.nih.gov/ntp/htdocs/LT_rpts/tr232.pdf.
NTP. (1986) Toxicology and carcinogenesis studies of tetrachloroethylene (perchloroethylene) (CAS No. 127-18-4)
in F344/N rats and B6C3Fi mice (inhalation studies). Public Health Service, U.S. Department of Health and Human
Services; NTP TR-311. Available from National Institute of Environmental Health Sciences, Research Triangle
Park, NC. Available online at http://ntp.niehs.nih.gov/ntp/htdocs/LT_rpts/tr311.pdf.
NTP. (1988) Toxicology and carcinogenesis studies of trichloroethylene (CAS No. 79-01-6) in four strains of rats
(ACI, August, Marshall, Osborne-Mendel)(gavage studies). Public Health Service, U.S. Department of Health and
Human Services; NTP TR-273. Available from National Institute of Environmental Health Sciences, Research
Triangle Park, NC. Available online at http://ntp.niehs.nih.gov/ntp/htdocs/LT_rpts/tr273.pdf.
NTP. (1989) Toxicology and carcinogenesis studies of hexachloroethane (CAS No. 67-72-1) inF344/N rats (gavage
studies). Public Health Service, U.S. Department of Health and Human Services; NTP TR-361. Available from
National Institute of Environmental Health Sciences, Research Triangle Park, NC. Available online at
http://ntp.niehs.nih.gov/ntp/htdocs/LT_rpts/tr361 .pdf.
NTP. (1990) Carcinogenesis studies of trichloroethylene (without epichlorohydrin) (CAS No. 79-01-6) in F344/N
rats and B6C3F1 mice (gavage study). Public Health Service, U.S. Department of Health and Human Services; NTP
TR-243. Available from National Institute of Environmental Health Sciences, Research Triangle Park, NC.
Available online at http://ntp.mehs.nih.gov/index.cfm?objectid=07067B36-09A5-8398-7E70FB2C35377215.
NTP. (1996) NTP technical report on renal toxicity studies of selected halogenated ethanes administered by gavage
to F344/N rats. Public Health Service, U.S. Department of Health and Human Services; NTP TOX-45. Available
from National Institute of Environmental Health Sciences, Research Triangle Park, NC. Available online at
http://ntp.niehs.nih.gov/ntpweb/index.cfm?objectid=D 1512B41-F1F6-975E-7FBA3D4A2132F1C1.
NTP. (2005) 11th report on carcinogens. Public Health Service, U.S. Department of Health and Human Services,
Research Triangle Park, NC. Available online at http://ntp-server.niehs.nih.gov.
Odabasi, M. (2008) Halogenated volatile organic compounds from the use of chlorine-bleach-containing household
products. Environ Sci Technol 42(5): 1445-1451.
Omiecinski, CJ; Remmel, RP; Hosagrahara, VP. (1999) Concise review of the cytochrome P450s and their roles in
toxicology. Toxicol Sci 48:151-156.
Onfelt, A. (1987) Spindle disturbances in mammalian cells. III. Toxicity, c-mitosis and aneuploidy with 22 different
compounds. Specific and unspecific mechanisms. MutatRes 182(3): 135—154.
Powers, JF; Picard, KL; Nyska, A; et al. (2008) Adrenergic Differentiation and Ret Expression in Rat
Pheochromocytomas. Endocr Pathol 19:9-16.
Ramsey, JC; Andersen, ME. (1984) A physiologically based description of the inhalation pharmacokinetics of
styrene monomer in rats and humans. Toxicol Appl Pharmacol 73:159-175.
Reynolds, ES. (1972) Comparison of early injury to liver endoplasmic reticulum by halomethanes,
hexachloroethane, benzene, toluene, bromobenzene, ethionine, thioacetamide and dimethylnitrosamine. Biochem
Pharmacol 21:2555-2561.
122
DRAFT - DO NOT CITE OR QUOTE
-------
Roldan-Aijona, T; Garcia-Pedrajas, MD; Luque-Romero, FL; et al. (1991) An association between mutagenicity in
rodents for 16 halogenated aliphatic hydrocarbons. Mutagenesis 6:199-205.
Salmon, AG; Jones, RB; Mackrodt, WC. (1981) Microsomal dechlorination of chloroethanes: structure reactivity
relationships. Xenobiotica 11:723-734.
Salmon, AG; Nash, JA; Walkin, CM; et al. (1985) Dechlorination of halocarbons by microsomes and vesicular
reconstituted cytochrome P-450 systems under reductive conditions. Br J Ind Med 42:305-311.
Sandstedt, K; Berglof, A; Feinstein, R; et al. (1997) Differential susceptibility to Mycoplasma pulmonis intranasal
infection in X-linked immunodeficient (xid), severe combined immunodeficient (scid), and immunocompetent mice.
Clin Exp Immunol 108:490-496.
Selden, A; Jacobson, G; Berg, P; et al. (1989) Hepatocellular carcinoma and exposure to hexachlorobenzene: a case
report. Br J Ind Med 46:138-140.
Selden, A; Nygren, M; Kvamlof, A; et al. (1993) Biological monitoring of hexachloroethane. Int Ach Occup
Environ Health 65(Suppl 1):S111—S114.
Selden, A; Kvarnlof, A; Bodin, L; et al. (1994) Health effects of low level occupational exposure to
hexachloroethane. J Occup Med Toxicol 3(10):73-79.
Selden, Al; Nygren, Y; Westberg, HB; et al. (1997) Hexachlorobenzene and octachlorostyrene in plasma of
aluminium foundry workers using hexachloroethane for degassing. Occup Environ Med 54(8):613-618.
Selden, Al; Floderus, Y; Bodin, LS; et al. (1999) Porphyrin status in aluminum foundry workers exposed to
hexachlorobenzene and octachlorostyrene. Arch Environ Health 54(4):248-253.
Shimizu, M; Noda, T; Yamano, T; et al. (1992) Safety evaluation of chemicals for use in household products
(XVII). A teratological study on hexachloroethane in rats. Osaka City Institute of Public Health and Environmental
Sciences 54:70-75. (Japanese)
Simmon, VF; Kauhanen, K. (1978) In vitro microbiological mutagenicity assays of hexachloroethane. SRI
International, Menlo Park, CA. Prepared for U.S. Environmental Protection Agency, National Environmental
Research Center, Water Supply Research Laboratory, Cincinnati, OH.
Southcott, WH. (1951) The toxicity and antihelmintic efficiency of hexachloroethane in sheep. Aust Vet J 27:18-
21.
Spanggord, RJ; Chou, TW; Mill, T; et al. (1985) Environmental fate of nitroguanidine, diethyleneglycol dinitrate,
and hexachloroethane smoke. SRI International, Menlo Park, CA. Prepared for U.S. Army Medical Research and
Development Command.
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.
Tafazoli, M; Baeten, A; Geerlings, P; et al. (1998) In vitro mutagenicity and genotoxicity study of a number of
short-chain chlorinated hydrocarbons using the micronucleus test and the alkaline single cell gel electrophoresis
technique (Comet assay) in human lymphocytes: a structure-activity relationship (QSAR) analysis of the genotoxic
and cytotoxic potential. Mutagenesis 13(2): 115—126.
Town, C; Leibman, KC. (1984) The in vitro dechlorination of some poly chlorinated ethanes. Drug Metab Disp
12:4-8.
Treluyer, JM; Jacqz-Aigrain, E; Alvarez, F; et al. (1991) Expression of CYP2D6 in developing human liver. Eur J
Biochem 202:583-588.
Tu, AS; Murray, TA; Hatch, KM; et al. (1985) In vitro transformation of BALB/c-3T3 cells by chlorinated ethanes
and ethylenes. Cancer Lett 28:85-92.
123
DRAFT - DO NOT CITE OR QUOTE
-------
U.S. EPA (Environmental Protection Agency). (1979) Water-related environmental fate of 129 priority pollutants.
Vol. II. Monitoring and Data Support Division, Washington, DC; EPA440/4-79-029b. PB80-204381.
U.S. EPA. (1982) Aquatic fate process data for organic priority pollutants. Monitoring and Data Support Division,
Washington, DC; EPA 440/4-81-014.
U.S. EPA. (1986a) Guidelines for the health risk assessment of chemical mixtures. Federal Register
51(185):34014-34025. Available online at http://www.epa.gov/iris/backgrd.html.
U.S. EPA. (1986b) Guidelines for mutagenicity risk assessment. Federal Register 51(185):34006-34012. Available
online at http://www.epa.gov/iris/backgrd.html.
U.S. EPA. (1988) Recommendations for and documentation of biological values for use in risk assessment.
Prepared by the Environmental Criteria and Assessment Office, Office of Health and Environmental Assessment,
Cincinnati, OH for the Office of Solid Waste and Emergency Response, Washington, DC; EPA 600/6-87/008.
Available online at http://www.epa.gov/iris/backgrd.html.
U.S. EPA. (1991a) Guidelines for developmental toxicity risk assessment. Federal Register 56(234):63798-63826.
Available online at http://www.epa.gov/iris/backgrd.html.
U.S. EPA. (1991b) Health advisory for hexachloroethane. Office of Water, Washington, DC; EPA/625/3-91/019F;
PB91-159657.
U.S. EPA. (1991c) Alpha2u-globulin: association with chemically induced renal toxicity and neoplasia in the male
rat. Risk Assessment Forum, Washington, DC; EPA/625/3-91/019F; PB92-143668.
U.S. EPA. (1992) Draft report: a cross-species scaling factor for carcinogen risk assessment based on equivalence of
mg/kg3/4/day. Federal Register 57(109):24152-24173.
U.S. EPA. (1994a) Interim policy for particle size and limit concentration issues in inhalation toxicity studies.
Federal Register 59(206):53799. Available online at http://www.epa.gov/iris/backgrd.html.
U.S. EPA. (1994b) Methods for derivation of inhalation reference concentrations and application of inhalation
dosimetry. Office of Research and Development, Washington, DC; EPA/600/8-90/066F. Available online at
http://www.epa.gov/iris/backgrd.html.
U.S. EPA. (1995) Use of the benchmark dose approach in health risk assessment. Risk Assessment Forum,
Washington, DC; EPA/630/R-94/007. Available online at
http://cfpub.epa.gov/ncea/raf/recordisplay .cfm?deid=42601.
U.S. EPA. (1996) Guidelines for reproductive toxicity risk assessment. Federal Register 61(212):56274-56322.
Available online at http://www.epa.gov/iris/backgrd.html.
U.S. EPA. (1998) Guidelines for neurotoxicity risk assessment. Federal Register 63(93):26926-26954.
U.S. EPA. (2000a) Science policy council handbook: risk characterization. Office of Science Policy, Office of
Research and Development, Washington, DC; EPA 100-B-00-002. Available online at
http://www.epa.gov/iris/backgrd.html.
U.S. EPA. (2000b) Benchmark dose technical guidance. External review draft. Risk Assessment Forum,
Washington, DC; EPA/630/R-00/001. Available online at http://www.epa.gov/iris/backgrd.html.
U.S. EPA. (2000c) Supplementary guidance for conducting health risk assessment of chemical mixtures. Risk
Assessment Forum, Washington, DC; EPA/630/R-00/002. Available online at
http://www.epa.gov/iris/backgrd.html.
U.S. EPA. (2002) A review of the reference dose and reference concentration processes. Risk Assessment Forum,
Washington, DC; EPA/630/P-02/0002F. Available online at http://www.epa.gov/iris/backgrd.html.
124
DRAFT - DO NOT CITE OR QUOTE
-------
U.S. EPA. (2005a) Guidelines for carcinogen risk assessment. Risk Assessment Forum, Washington, DC;
EPA/630/P-03/001F. Available online at http://www.epa.gov/iris/backgrd.html.
U.S. EPA. (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/backgrd.html.
U.S. EPA. (2006a) Science policy council handbook: peer review. Third edition. Office of Science Policy, Office of
Research and Development, Washington, DC; EPA/100/B-06/002. Available online at
http://www.epa.gov/iris/backgrd.html.
U.S. EPA. (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= 15 8363.
U.S. EPA. (2008) Benchmark dose software (BMDS) version 2.0. Available online at
http://www.epa.gov/ncea/bmds.html.
Van Dyke, RA. (1977) Dechlorination mechanisms of chlorinated olefins. Environ Health Perspect 21:121-124.
Van Dyke, RA; Wineman, CG. (1971) Enzymatic dechlorination of chloroethanes and propanes in vitro. Biochem
Pharmacol 20:463-470.
Verschueren, K. (1983) Handbook of environmental data on organic chemicals. 2nd ed. New York, NY: Van
Nostrand Reinhold Company.
Vieira, I; Sonnier, M; Cresteil, T. (1996) Developmental expression of CYP2E1 in the human liver:
hypermethylation control of gene expression during the neonatal period. Eur J Biochem 238:476-483.
Vogel, EW; Nivard, MJ. (1993) Performance of 181 chemicals in a Drosophila assay predominantly monitoring
interchromosomal mitotic recombination. Mutagenesis 8:57-81.
Weast, RC, ed. (1986) CRC handbook of chemistry and physics. 67th ed. Boca Raton, FL: CRC Press.
Webb, DR; Ridder, GM; Alden, CL. (1989) Acute and subchronic nephrotoxicity of d-limonene in dogs. Food
Chem Toxicol 28:669-675.
Weeks, MH; Thomasino, JA. (1978) Assessment of acute toxicity of hexachloroethane in laboratory animals. U.S.
Army Environmental Hygiene Agency, Aberdeen Proving Ground, MD; Report No. 51-0075-78.
Weeks, MH; Angerhofer, RA; Bishop, R; et al. (1979) The toxicity of hexachloroethane in laboratory animals. Am
Ind Assoc J 40:187-199.
Weisburger, EK. (1977) Carcinogenicity studies on halogenated hydrocarbons. Environ Health Perspect 21:7-16.
Xu, X; Zhang, D; Lyubynska, N; et al. (2006) Mast cells protect mice from mycoplasma pneumonia. Am J Respir
Crit Care Med 173:219-225.
Yamakage, A; Ishikawa, H. (1982) Generalized morphea-like scleroderma occurring in people exposed to organic
solvents. Dermatologica 165:186-193.
Yanagita, K; Sagami, I; Shimizu, T. (1997) Distal site and surface mutations of cytochrome P450 1A2 markedly
enhance dehalogenation of chlorinated hydrocarbons. Arch Biochem Biophys 346(2):269-276.
Yanagita, K; Sagami, I; Daff, S; et al. (1998) Marked enhancement in the reductive dehalogenation of
hexachloroethane by a Thr319Ala mutation of cytochrome P450 1A2. Biochem Biophys Res Commun 249(3):678-
682.
125
DRAFT - DO NOT CITE OR QUOTE
-------
Yoshikawa, K. (1996) Anomalous nonidentity between Salmonella genotoxicants and rodent carcinogens and
genotoxic noncarcinogens. Environ Health Perspect 104:40-46.
Younglai, EV; Foster, WG; Hughes, EG; et al. (2002) Levels of environmental contaminants in human follicular
fluid, serum, and seminal plasma of couples undergoing in vitro fertilization. Arch Environ Contam Toxicol
43(1): 121—126.
126
DRAFT - DO NOT CITE OR QUOTE
-------
APPENDIX A: SUMMARY OF EXTERNAL PEER REVIEW AND PUBLIC
COMMENTS AND DISPOSITION
PAGE INTENTIONALLY LEFT BLANK
A-l
DRAFT - DO NOT CITE OR QUOTE
-------
APPENDIX B: BENCHMARK DOSE MODELING OUTPUT
Table B-l. Dose-response modeling results using BMDS (version 2.0) based
on non-cancerous kidney and liver effects in rats following oral exposure to
HCE
Study
Endpoint
Sex/species
Fitted model"
/7-Value
AIC
BMD10
(mg/kg-d)
BMDL10
(mg/kg-d)
Kidney effects
NCI
(1978)
Tubular nephropathy
Male rat
Gamma
0.93
133.68
21.22
16.99
Multistage 1°
0.93
133.66
21.25
17.01
Weibull
0.93
133.68
21.22
16.99
Female rat
Gamma
1.00
117.47
87.24
50.63
Multistage 2°
0.94
116.09
80.63
41.89
Logistic
0.42
118.61
95.19
73.25
Probit
0.53
118.14
91.25
69.20
Weibull
1.00
117.47
84.22
48.62
NTP
(1989)
Moderate to marked
Tubular nephropathy
Male rat
Logistic
0.99
205.88
3.84
2.62
Multistage 1°
0.87
205.90
3.20
1.88
Probit
0.99
205.88
3.81
2.60
Quantal-linear
0.87
205.90
3.20
1.88
Mild to moderate
Tubular nephropathy
Female rat
Gamma
0.86
191.90
15.17
10.72
Logistic
0.46
192.42
23.06
18.33
Multistage 1°
0.78
192.96
15.91
11.14
Probit
0.47
192.40
22.55
18.04
Quantal-
linear
0.86
191.90
15.17
10.72
Weibull
0.86
191.90
15.17
10.72
NTP
(1989)
Linear mineralization
Male rat
Logistic
0.36
148.11
4.30
3.45
Multistage 1°
0.20
148.90
1.75
1.40
Probit
0.51
147.66
3.98
3.22
NTP
(1989)
Hyperplasia of the
pelvic transitional
epithelium
Male rat
Gamma
0.42
84.64
7.33
4.87
LogLogistic
0.48
84.42
7.05
4.48
Multistage 2°
0.42
84.64
7.33
4.87
Weibull
0.42
84.64
7.33
4.87
Quantal-linear
0.42
84.64
7.33
4.87
Gorzinski
et al.
(1985)
Atrophy and
degeneration of renal
tubules
Male rat
Gamma
0.70
32.94
1.34
0.728
Multistage 1°
0.93
32.94
1.34
0.728
Logistic
0.89
32.97
3.30
1.98
Probit
0.89
32.95
3.08
1.95
Quantal-
linear
0.93
32.94
1.34
0.728
Weibull
0.69
34.92
1.72
0.729
Female rat
Gamma
0.99
42.47
13.80
4.56
Multistage 1°
0.93
40.61
8.54
4.49
B-l DRAFT - DO NOT CITE OR QUOTE
-------
Table B-l. Dose-response modeling results using BMDS (version 2.0) based
on non-cancerous kidney and liver effects in rats following oral exposure to
HCE
Study
Endpoint
Sex/species
Fitted model3
/7-Value
AIC
BMD10
(mg/kg-d)
BMDL10
(mg/kg-d)
Logistic
0.98
40.51
17.40
11.07
Probit
0.99
40.49
16.10
10.51
Quantal-linear
0.93
40.61
8.54
4.49
Weibull
0.98
42.47
13.71
4.56
Gorzinski
et al.
(1985)
Slight hypertrophy
and/or dilation of
proximal convoluted
tubules
Male rat
Gamma
0.99
20.88
1.22
0.710
Logistic
0.66
23.91
4.85
2.71
LogLogistic
0.68
23.89
1.23
0.308
LogProbit
0.54
24.26
2.11
1.01
Multistage 2°
0.94
22.84
1.33
0.713
Probit
0.67
23.85
4.28
2.54
Weibull
0.99
20.88
1.22
0.710
Quantal-
linear
0.99
20.88
1.22
0.710
Liver effects
NTP
(1989)
Hepatocellular
necrosis
Female rat
Gamma
0.93
38.62
118.04
60.18
Multistage 1°
0.68
40.56
53.82
35.19
Logistic
0.55
41.58
156.22
107.49
Probit
0.61
40.95
148.49
102.71
Weibull
0.91
38.91
114.68
56.75
Tor all models, a BMR of 0.1 was employed in deriving the estimates of the benchmark dose (BMD10) and its
95% lower CL (BMDL10). Modeling output is provided for models that represent the POD for each of the kidney
endpoints; these models are highlighted in bold font.
Table B-l presents the dose-response modeling results using BMDS (version 2.0) based
on non-cancerous kidney and liver effects in rats following oral exposure to HCE. Based on the
incidence of tubular nephropathy in male rats (NCI, 1978), the logistic and probit models
exhibited significant lack-of-fit (p < 0.1), while the gamma, multistage (1°) and Weibull models
had /(-values >0.1. All three of these models that showed adequate fit yielded the same AIC
values, as well as nearly equivalent BMDio and BMDLio estimates of 21.22 and 16.99 mg/kg-
day, respectively. Therefore, the candidate POD selected for this dataset is 16.99 mg/kg-day.
Based on the incidence of tubular nephropathy in female rats (NCI, 1978), only the
1° multistage model exhibited significant lack-of-fit. Of the models that did not show significant
lack-of-fit (i.e., gamma, multistage 2°, logistic, probit, and Weibull models), the BMDLio
estimates were within a factor of three of each other, suggesting no appreciable model
dependence. As the BMDLio values did not show large variation, the model with the lowest AIC
value was selected. Therefore, the multistage 2° model BMDLio of 41.89 mg/kg-day was
selected as the candidate POD for this dataset.
B-2
DRAFT - DO NOT CITE OR QUOTE
-------
In fitting the available dichotomous dose-response models to the incidence of moderate to
marked tubular nephropathy in male rats (NTP, 1989), the gamma and Weibull models exhibited
significant lack-of-fit (p < 0.1). The models that did not show significant lack-of-fit (i.e.,
logistic, multistage 1°, probit, and quantal-linear) yielded BMDLio estimates that were within a
factor of three of each other, suggesting no appreciable model dependence. As the BMDLio
values did not show large variation, the model with the lowest AIC value was selected. The AIC
values for the logistic and probit models were the lowest (and identical); therefore, the probit
model with the lowest BMDLio, of 2.60 mg/kg-day was selected as the candidate POD for this
dataset.
Based on the incidence of mild to moderate tubular nephropathy in female rats (NTP,
1989), none of the models exhibited significant lack-of-fit. These models (i.e., gamma, logistic,
multistage 1°, probit, quantal-linear, and Weibull models) yielded BMDLio estimates that were
within a factor of three of each other, suggesting no appreciable model dependence. As the
BMDLio values did not show large variation, the model with the lowest AIC value was selected.
The gamma, quantal-linear, and Weibull models had identical AIC values; therefore, the model
with the lowest BMDLio was selected. The BMDLio values for these models were identical;
therefore, the BMDLio of 10.72 mg/kg-day was selected as the candidate POD for this dataset.
In fitting the available dichotomous dose-response models to the incidence of linear
mineralization in male rats (NTP, 1989), the gamma and the Weibull models exhibited
significant lack-of-fit (p< 0.1). Of the models that did not show significant lack-of-fit (i.e.,
logistic, multistage 1°, and probit), the resulting BMDLio estimates were within a factor of three
of each other, suggesting no appreciable model dependence. As the BMDLio values did not
show large variation, the model with the lowest AIC value was selected. Therefore, the probit
model BMDLio of 3.22 mg/kg-day was selected as the candidate POD for this dataset.
In fitting the available dichotomous dose-response models to the incidence of hyperplasia
of the pelvic transitional epithelium in male rats (NTP, 1989), the logistic, logprobit, and probit
models exhibited significant lack-of-fit (p< 0.1). Of the models that did not show significant
lack-of-fit (i.e., gamma, loglogistic, multistage 2°, Weibull, and quantal-linear), the resulting
BMDLio estimates were within a factor of three of each other, suggesting no appreciable model
dependence. As the BMDLio values did not show large variation, the model with the lowest AIC
value was selected. Therefore, the loglogistic model BMDLio of 4.48 mg/kg-day was selected as
the candidate POD for this dataset.
In fitting the available dichotomous dose-response models to the incidence of atrophy and
degeneration of renal tubules in male and female rats (Gorzinski et al., 1985), none of the models
exhibited a significant lack-of-fit in either sex. For male rats, these models (i.e., gamma,
multistage 1°, logistic, probit, quantal-linear, and Weibull) yielded BMDLio estimates that were
within a factor of three of each other, suggesting no appreciable model dependence. As the
BMDLio values did not show large variation, the model with the lowest AIC value was selected.
B-3 DRAFT - DO NOT CITE OR QUOTE
-------
The AIC values for the gamma, multistage 1°, and quantal-linear were identical; therefore, the
model with the lowest BMDLio was selected. All of the BMDLio values were identical for these
models; therefore, the BMDLio of 0.728 mg/kg-day was selected as the candidate POD for this
dataset.
For female rats, these models (i.e., gamma, multistage 1°, logistic, probit, quantal-linear,
and Weibull) yielded BMDLio estimates that were within a factor of three of each other,
suggesting no appreciable model dependence. As the BMDLio values did not show large
variation, the model with the lowest AIC value was selected. The probit BMDLio of 10.51
mg/kg-day was selected as the candidate POD for this dataset.
In fitting the available dichotomous dose-response models to the incidence of slight
hypertrophy and/or dilation of proximal convoluted tubules in male rats (Gorzinski et al., 1985),
none of the models exhibited a significant lack-of-fit. For male rats, these models (i.e., gamma,
logistic, loglogistic, logprobit, multistage 2°, probit, Weibull, and quantal-linear) yielded
BMDLio estimates that were within a factor of three of each other, suggesting no appreciable
model dependence. As the BMDLio values did not show large variation, the model with the
lowest AIC value was selected. The gamma, Weibull, and quantal-linear models yielded the
lowest ( and identical) AICs. All of the BMDLio values were identical for these models;
therefore, the BMDLio of 0.710 mg/kg-day was selected as the candidate POD for this dataset.
Based on the incidence of hepatocellular necrosis in female rats (NTP, 1989), none of the
dichotomous dose-response models exhibited a significant lack-of-fit. All of these models (i.e.,
gamma, multistage 1°, logistic, probit, and Weibull) yielded BMDLio estimates that were within
a factor of three of each other, suggesting no appreciable model dependence. As the BMDLio
values did not show large variation, the model with the lowest AIC value was selected.
Therefore, the gamma model BMDLio of 60.18 mg/kg-day was selected as the candidate POD
for this dataset.
For comparison purposes, BMD modeling for the above endpoints was also conducted
using BMRs of 5 and 1%. The modeling results are included in Table B-2.
B-4
DRAFT - DO NOT CITE OR QUOTE
-------
Table B-2. Dose-response modeling results using BMDS (version 2.0) for
BMRs of 1, 5, and 10% based on noncancerous kidney and liver effects in
rats following oral exposure to HCE
Study
Endpoint
Sex/
species
Fitted
model3
BMD10
(mg/
kg-d)
BMDL10
(mg/
kg-d)
BMDos
(mg/
kg-d)
BMDLos
(mg/
kg-d)
BMD01
(mg/
kg-d)
BMDL01
(mg/
kg-d)
Kidney effects
NCI
(1978)
Tubular
nephropathy
Male
rat
Gamma and
Weibull
21.22
16.99
10.33
8.27
2.02
1.62
Multistage 1°
21.25
17.01
10.35
8.28
2.03
1.62
Female
rat
Multistage 2°
80.63
41.89
56.26
21.18
24.90
4.28
NTP
(1989)
Moderate to
marked
tubular
nephropathy
Male
rat
Probit
3.81
2.60
1.93
1.32
0.39
0.27
Mild to
moderate
tubular
nephropathy
Female
rat
Gamma,
Quantal-
linear, and
Weibull
15.17
10.72
7.39
5.22
1.45
1.02
NTP
(1989)
Linear
mineralization
Male
rat
Probit
3.98
3.22
2.36
1.80
0.58
0.40
NTP
(1989)
Hyperplasia of
the pelvic
transitional
epithelium
Male
rat
LogLogistic
7.05
4.48
3.34
2.12
0.64
0.41
Gorzinski
et al.
(1985)
Atrophy and
degeneration
of renal
tubules
Male
rat
Gamma,
Multistage
1°, and
Quantal-
linear
1.34
0.73
0.66
0.35
0.13
0.07
Female
rat
Probit
16.10
10.51
8.89
5.60
1.97
1.18
Gorzinski
et al.
(1985)
Slight
hypertrophy
and/or dilation
of proximal
convoluted
tubules
Male
rat
Gamma,
Weibull, and
Quantal-
linear
1.22
0.71
0.60
0.35
0.12
0.07
Liver effects
NTP
(1989)
Hepatocellular
necrosis
Female
rat
Gamma
118.03
60.18
84.66
33.34
41.75
8.60
B-5 DRAFT - DO NOT CITE OR QUOTE
-------
Modeling for Noncancer Assessment
NCI (1978) Tubular Nephropathy in Male Rats
Gamma Model
Gamma Model. (Version: 2.13; Date: 05/16/2008)
Input Data File: C:\USEPA\BMDS2\Temp\tmpCDF.(d)
Gnuplot Plotting File: C:\USEPA\BMDS2\Temp\tmpCDF.plt
Thu Apr 09 14:55:06 2009
BMDS Model Run NCI 1978 Tubular Nephropathy Male Rat - Gamma Model
The form of the probability function is:
P[response]= background+(1-background)*CumGamma[siope*dose,power],
where CumGamma(.) is the cummulative Gamma distribution function
Dependent variable = PercentPositiveModerateMarkedTubularNephropathy
Independent variable = ularNephropathy
Power parameter is restricted as power >=1
Total number of observations = 3
Total number of records with missing values = 0
Maximum number of iterations = 25 0
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial (and Specified) Parameter Values
Background = 0.0238095
Slope = 0.00474439
Power = 1.01848
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -Background -Power
have been estimated at a boundary point, or have been specified by
the user,
and do not appear in the correlation matrix )
Slope
Slope 1
Parameter Estimates
Interval
Variable
Limit
Background
Slope
0.00632309
Power
Estimate
0
0.00496352
1
Std. Err.
NA
0. 000693669
NA
95.0% Wald Confidence
Lower Conf. Limit Upper Conf.
0.00360396
B-6
DRAFT - DO NOT CITE OR QUOTE
-------
NA - Indicates that this parameter has hit a bound
implied by some inequality constraint and thus
has no standard error.
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced model
AIC:
Log(likelihood)
-65.7706
-65.8419
-82.1514
133.684
# Param's Deviance Test d.f. P-value
3
1 0.142715 2 0.9311
1 32.7616 2 <.0001
Dose
Goodness of Fit
Est._Prob. Expected Observed Size
Scaled
Residual
0.0000
113.0000
227.0000
Chi^2 = 0.14
0.0000
0.4293
0.6759
d.f. = 2
0.000 0.000 20
21.035 22.050 49
33.795 33.000 50
P-value = 0.9308
0. 000
0.293
-0.240
Benchmark Dose Computation
Specified effect
Risk Type
Confidence level
BMD
BMDL
0.1
Extra risk
0. 95
21.227
16.9904
B-7
DRAFT - DO NOT CITE OR QUOTE
-------
Gamma Multi-Hit Model with 0.95 Confidence Level
~o
CD
o
CD
£=
o
O
TO
Gamma Mu ti-Hit
BMDU BMD
dose
14:55 04/09 2009
B-8
DRAFT - DO NOT CITE OR QUOTE
-------
Multistage 1°
Multistage Model. $Revision: 2.1 $ $Date: 2000/08/21 03:38:21 $
Input Data File: C:\BMDS\UNSAVED1.(d)
Gnuplot Plotting File: C:\BMDS\UNSAVEDl.plt
Thu Sep 14 09:09:29 2006
BMDS Model Run NCI 1978 Tubular Nephropathy Male Rat - Multistage 1 degree Model
The form of the probability function is:
P[response] = background + (1-background)*[1-EXP(
-betal*doseAl) ]
The parameter betas are restricted to be positive
Dependent variable = PercentPositiveModerateMarkedTubularNephropathy
Independent variable = ularNephropathy
Total number of observations = 3
Total number of records with missing values = 0
Total number of parameters in model = 2
Total number of specified parameters = 0
Degree of polynomial = 1
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial Parameter Values
Background = 0.0201528
Beta(1) = 0.00475168
the user,
Beta(1)
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -Background
have been estimated at a boundary point, or have been specified by
and do not appear in the correlation matrix )
Beta(1)
1
Interval
Variable
Limit
Background
Beta(1)
Parameter Estimates
Estimate
0
Std. Err.
0.00495719 *
* - Indicates that this value is not calculated.
95.0% Wald Confidence
Lower Conf. Limit Upper Conf.
B-9
DRAFT - DO NOT CITE OR QUOTE
-------
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced model
AIC:
Log(likelihood)
-65.7706
-65.8277
-82.1514
133.655
# Param's Deviance Test d.f. P-value
3
1 0.114158 2 0.9445
1 32.7616 2 <.0001
Dose
Goodness of Fit
Est._Prob. Expected Observed Size
Scaled
Residual
0.0000
113.0000
227.0000
Chi^2 = 0.14
0.0000
0.4289
0.6754
d.f. = 2
0.000 0.000 20
21.015 22.050 49
33.772 33.000 50
P-value = 0.9307
0. 000
0.299
-0.233
Benchmark Dose Computation
Specified effect
Risk Type
Confidence level
BMD
BMDL
BMDU
0.1
Extra risk
0. 95
21.2541
17.0107
26.9612
Taken together, (17.0107, 26.9612) is a 90
interval for the BMD
two-sided confidence
Multistage Model with 0.95 Confidence Level
Multistage
< 0.4
BMDL BMP
0 50
100
150
dose
200
250
300
09:09 09/14 2006
B-10
DRAFT - DO NOT CITE OR QUOTE
-------
Weibull
Weibull Model $Revision: 2.2 $ $Date: 2000/03/17 22:27:16 $
Input Data File: C:\BMDS\UNSAVED1.(d)
Gnuplot Plotting File: C:\BMDS\UNSAVEDl.plt
Thu Sep 14 09:13:24 2006
BMDS Model Run NCI 1978 Tubular Nephropathy Male Rat - Weibull Model
The form of the probability function is:
P[response] = background + (1-background)*[1-EXP(-slope*dose/spower)]
Dependent variable = PercentPositiveModerateMarkedTubularNephropathy
Independent variable = ularNephropathy
Power parameter is restricted as power >=1
Total number of observations = 3
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial (and Specified) Parameter Values
Background = 0.0238095
Slope = 0.00453277
Power = 1.00295
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -Background -Power
have been estimated at a boundary point, or have been specified by
the user,
and do not appear in the correlation matrix )
Slope
Slope 1
Parameter Estimates
Interval
Variable
Limit
Background
Slope
0.00632309
Power
Estimate
0
0.00496352
1
Std. Err.
NA
0.000693669
NA
95.0% Wald Confidence
Lower Conf. Limit Upper Conf.
0.00360396
NA - Indicates that this parameter has hit a bound
implied by some ineguality constraint and thus
has no standard error.
Analysis of Deviance Table
Model Log(likelihood) # Param's Deviance Test d.f. P-value
B-ll
DRAFT - DO NOT CITE OR QUOTE
-------
Full model -65.7706 3
Fitted model -65.8419 1 0.142715 2 0.9311
Reduced model -82.1514 1 32.7616 2 <.0001
AIC: 133.684
Dose
Goodness of Fit
Est._Prob. Expected Observed Size
Scaled
Residual
0.0000
113.0000
227.0000
0.0000
0.4293
0.6759
0.000
21.035
33.795
0.000
22.050
33.000
20
49
50
0. 000
0.293
-0.240
Chi^2 = 0.14
d.f. = 2
P-value = 0.9308
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Extra risk
Confidence level = 0.95
BMD = 21.227
BMDL = 16.9904
B-12
DRAFT - DO NOT CITE OR QUOTE
-------
Wei bull Model with 0.95 Confidence Level
Wei bull
0.8
0.7
0.6
0.5
< 0.4
c
o
0.3
Ll_
0.2
BMD
50
100
150
200
250
300
dose
09:13 09/14 2006
B-13 DRAFT - DO NOT CITE OR QUOTE
-------
NCI (1978) Tubular Nephropathy in Female Rats
Multistage 2°
Multistage Model. $Revision: 2.1 $ $Date: 2000/08/21 03:38:21 $
Input Data File: C:\BMDS\UNSAVED1.(d)
Gnuplot Plotting File: C:\BMDS\UNSAVEDl.plt
Thu Apr 09 16:18:29 2009
BMDS Model Run - NCI 1978 Tubular Nephropathy Female Rat - Multistage 2 degree Model
The form of the probability function is:
P[response] = background + (1-background)*[1-EXP(
-betal*dose/sl-beta2*dose/s2) ]
The parameter betas are restricted to be positive
Dependent variable = PercentPositiveModerateMarkedTubularNephropathy
Independent variable = ularNephropathy
Total number of observations = 3
Total number of records with missing values = 0
Total number of parameters in model = 3
Total number of specified parameters = 0
Degree of polynomial = 2
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial Parameter Values
Background = 0
Beta(l) = 0
Beta(2) = 1.74381e-005
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -Background -Beta(l)
have been estimated at a boundary point, or have been specified by
the user,
and do not appear in the correlation matrix )
Beta(2)
Beta (2) 1
Parameter Estimates
Interval
Variable
Limit
Background
Beta(1)
Beta(2)
Estimate
0
0
Std. Err.
95.0% Wald Confidence
Lower Conf. Limit Upper Conf.
1.62048e-005 *
* - Indicates that this value is not calculated.
B-14
DRAFT - DO NOT CITE OR QUOTE
-------
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced model
AIC:
Log(likelihood)
-56.7357
-57. 0429
-74.4688
116.086
# Param's
3
1
1
Deviance Test d.f.
0.614339
35.466
P-value
0.7355
<.0001
Dose
Goodness of Fit
Est._Prob. Expected Observed Size
Scaled
Residual
0.0000
113.0000
227.0000
Chi^2 =0.13
0.0000
0.1869
0.5661
d.f. = 2
0.000 0.000 20
9.346 9.000 50
27.741 28.910 49
P-value = 0.9374
0. 000
-0.125
0.337
Benchmark Dose Computation
Specified effect
Risk Type
Confidence level
BMD
BMDL
BMDU
0.1
Extra risk
0. 95
80.6338
41.8864
93.2552
Taken together, (41.8864, 93.2552) is a 90
interval for the BMD
two-sided confidence
B-15
DRAFT - DO NOT CITE OR QUOTE
-------
Multistage Model with 0.95 Confidence Level
Multistage
0.7
0.6
0.5
0.4
0.3
0.2
0
BMDL
BMD
0
50
100
150
200
250
300
dose
09:21 09/14 2006
B-16 DRAFT - DO NOT CITE OR QUOTE
-------
NTP (1989) Male Rat Nephropathy
Probit Model
Probit Model. (Version: 3.1; Date: 05/16/2008)
Input Data File: C:\USEPA\BMDS2\Temp\tmpA0E.(d)
Gnuplot Plotting File: C:\USEPA\BMDS2\Temp\tmpA0E.plt
Wed Apr 08 13:27:38 2009
BMDS Model Run NTP 1989 Tubular Nephropathy Male Rat - Probit Model
The form of the probability function is:
P[response] = CumNorm(Intercept+Slope*Dose) ,
where CumNorm(.) is the cumulative normal distribution function
Dependent variable = PercentPositiveModerateMarkedTubularNephropathy
Independent variable = ularNephropathy
Slope parameter is not restricted
Total number of observations = 3
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial (and Specified) Parameter Values
background = 0 Specified
intercept = -0.354714
slope = 0.0433259
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -background
have been estimated at a boundary point, or have been specified by
the user,
and do not appear in the correlation matrix )
intercept slope
intercept 1 -0.78
slope -0.78 1
Parameter Estimates
95.0% Wald Confidence
Interval
Variable Estimate Std. Err. Lower Conf. Limit Upper Conf.
Limit
intercept -0.35763 0.165052 -0.681127
0.0341335
slope 0.0436991 0.0182219 0.00798493
0.0794134
B-17
DRAFT - DO NOT CITE OR QUOTE
-------
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced model
AIC:
Log(likelihood)
-100.939
-100.939
-103.852
205.878
# Param's Deviance Test d.f.
3
2 0.000120944 1
1 5.82641 2
P-value
0.9912
0.0543
Dose
Goodness of Fit
Est._Prob. Expected Observed Size
Scaled
Residual
0.0000
0.3603
18.016
18.000
50
-0.005
7.0000
0.4794
23.968
24.000
50
0. 009
14.0000
0.6003
30.016
30.000
50
-0.005
Chi^2 = 0.00
d.f.
= 1 P-
-value = 0.9912
Benchmark Dose Computation
Specified effect
Risk Type
Confidence level
BMD
BMDL
0.1
Extra risk
0. 95
3.81407
2 .59812
Probit Model with 0.95 Confidence Level
0.7
0.6
"O
"G
% 0.5
£=
o
U—'
o
to
0.4
0.3
0.2
13:27 04/08
B-18 DRAFT - DO NOT CITE OR QUOTE
Probit
BMDL
dose
2009
-------
NTP (1989) Female Rat Nephropathy
Gamma Model
Gamma Model. (Version: 2.13; Date: 05/16/2008)
Input Data File: C:\USEPA\BMDS2\Temp\tmpD9.(d)
Gnuplot Plotting File: C:\USEPA\BMDS2\Temp\tmpD9.plt
Fri Apr 10 10:19:37 2009
BMDS Model Run NTP 198 9 Tubular Nephropathy Female Rat - Gamma Model
The form of the probability function is:
P[response]= background+(1-background)*CumGamma[siope*dose,power] ,
where CumGamma(.) is the cummulative Gamma distribution function
Dependent variable = PercentPositiveModerateMarkedTubularNephropathy
Independent variable = ularNephropathy
Power parameter is restricted as power >=1
Total number of observations = 3
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial (and Specified) Parameter Values
Background = 0.2 45 098
Slope = 0.0111213
Power = 1.3
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -Power
have been estimated at a boundary point, or have been specified by
the user,
and do not appear in the correlation matrix )
Background Slope
Background 1 -0.55
Slope -0.55 1
Parameter Estimates
Interval
Variable
Limit
Background
0.358621
Slope
0.0102497
Power
Estimate
0.242452
0.00694477
1
Std. Err.
0.0592711
0.0016862
NA
95.0% Wald Confidence
Lower Conf. Limit Upper Conf.
0.126283
0.00363988
NA - Indicates that this parameter has hit a bound
B-19
DRAFT - DO NOT CITE OR QUOTE
-------
implied by some inequality constraint and thus
has no standard error.
Analysis of Deviance Table
Model Log(likelihood)
Full model -93.93 62
# Param's Deviance Test d.f.
3
P-value
Fitted model
¦93.9519
2
0.0312372
1
0.8597
Reduced model
-102.85
1
17.
8276
2
0.0001345
AIC:
191.904
Goodness
of Fit
Scaled
Dose Est. Prob.
Expected
Observed
Size
Residual
0.0000 0.2425
12.123
12 .
000
50
-0.040
57.0000 0.4901
24.504
25 .
000
50
0.140
114.0000 0.6568
32.182
31.
850
49
-0.100
Chi^2 = 0.03 d.f.
= 1 P-
-value
= 0.8596
Benchmark Dose Computation
Specified effect
Risk Type
Confidence level
BMD
BMDL
0.1
Extra risk
0. 95
15.1712
10.7248
B-20
DRAFT - DO NOT CITE OR QUOTE
-------
Quantal Linear Model with 0.95 Confidence Level
Quantal Linear
BMD Lower Bound
BMDL
0 20 40 60 80 100
dose
12:21 12/16 2008
B-21
DRAFT - DO NOT CITE OR QUOTE
-------
Quantal-linear Model
Quantal Linear Model using Weibull Model (Version: 2.12; Date: 05/16/2008)
Input Data File: C:\USEPA\BMDS2\Temp\tmpE4.(d)
Gnuplot Plotting File: C:\USEPA\BMDS2\Temp\tmpE4.plt
Fri Apr 10 10:36:29 2009
BMDS Model Run NTP 1989 Tubular Nephropathy Female Rat - Quantal-linear Model
The form of the probability function is:
P[response] = background + (1-background)*[1-EXP(-slope*dose)]
Dependent variable = PercentPositiveModerateMarkedTubularNephropathy
Independent variable = ularNephropathy
Total number of observations = 3
Total number of records with missing values = 0
Maximum number of iterations = 25 0
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial (and Specified) Parameter Values
Background = 0.245098
Slope = 0.00666772
Power = 1 Specified
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -Power
have been estimated at a boundary point, or have been specified by
the user,
and do not appear in the correlation matrix )
Background Slope
Background 1 -0.55
Slope -0.55 1
Parameter Estimates
95.0% Wald Confidence
Interval
Variable Estimate Std. Err. Lower Conf. Limit Upper Conf.
Limit
Background 0.242451 0.0592711 0.126282
0.358621
Slope 0.00694478 0.0016862 0.00363989
0.0102497
Analysis of Deviance Table
Model Log(likelihood) # Param's Deviance Test d.f. P-value
Full model -93.9362 3
B-22
DRAFT - DO NOT CITE OR QUOTE
-------
Fitted model -93.9519 2 0.0312372 1 0.8597
Reduced model -102.85 1 17.8276 2 0.0001345
AIC: 191.904
Goodness of Fit
Scaled
Dose Est._Prob. Expected Observed Size Residual
0.0000
0.2425
12.123
12.000
50
-0.040
57.0000
0.4901
24.504
25.000
50
0.140
114.0000
0.6568
32.182
31.850
49
-0.100
Chi^2 = 0.03 d.f. = 1 P-value = 0.8596
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Extra risk
Confidence level = 0.95
BMD = 15.1712
BMDL = 10.7248
B-23
DRAFT - DO NOT CITE OR QUOTE
-------
Quantal Linear Model with 0.95 Confidence Level
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
10:36 04/10
Quantal Linear
BMDL
B-24
DRAFT - DO NOT CITE OR QUOTE
-------
Weibull Model
Weibull Model using Weibull Model (Version: 2.12; Date: 05/16/2008)
Input Data File: C:\USEPA\BMDS2\Temp\tmpE3.(d)
Gnuplot Plotting File: C:\USEPA\BMDS2\Temp\tmpE3.plt
Fri Apr 10 10:34:27 2009
BMDS Model Run NTP 198 9 Tubular Nephropathy Female Rat - Weibull Model
The form of the probability function is:
P[response] = background + (1-background)*[1-EXP(-slope*dose/spower)]
Dependent variable = PercentPositiveModerateMarkedTubularNephropathy
Independent variable = ularNephropathy
Power parameter is restricted as power >=1
Total number of observations = 3
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial (and Specified) Parameter Values
Background = 0.2 45 098
Slope = 0.00666772
Power = 1
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -Power
have been estimated at a boundary point, or have been specified by
the user,
and do not appear in the correlation matrix )
Background Slope
Background 1 -0.55
Slope -0.55 1
Parameter Estimates
Interval
Variable
Limit
Background
0.358621
Slope
0.0102497
Power
Estimate
0.242451
0.00694478
1
Std. Err.
0.0592711
0.0016862
NA
NA - Indicates that this parameter has hit a bound
implied by some ineguality constraint and thus
95.0% Wald Confidence
Lower Conf. Limit Upper Conf.
0.126282
0.00363989
B-25
DRAFT - DO NOT CITE OR QUOTE
-------
has no standard error.
Model
Full model
Fitted model
Reduced model
AIC:
Analysis of Deviance Table
Deviance Test d.f.
Log(likelihood)
-93.9362
-93.9519
-102.85
191.904
# Param's
3
2
1
0.0312372
17. 8276
P-value
0.8597
0.0001345
Goodness of Fit
Scaled
Dose
Est. Prob.
Expected
Observed
Size
Residual
0.0000
0.2425
12.123
12.000
50
-0.040
57.0000
0.4901
24.504
25.000
50
0.140
114.0000
0.6568
32.182
31.850
49
-0.100
Chi^2 =0.03
d.f. = 1
P-value = 0.8596
Benchmark Dose Computation
Specified effect
Risk Type
Confidence level
BMD
BMDL
0.1
Extra risk
0. 95
15.1712
10.7248
B-26
DRAFT - DO NOT CITE OR QUOTE
-------
Weibull Model with 0.95 Confidence Level
0.7
0.6
0.5
0.4
0.3
0.2
0.1
10:34 04/10
B-27 DRAFT - DO NOT CITE OR QUOTE
Weibull
BMDL
60
dose
2009
-------
NTP (1989) Linear Mineralization in Male Rats
Probit Model
Probit Model. (Version: 3.1; Date: 05/16/2008)
Input Data File: C:\USEPA\BMDS2\Temp\tmpA33.(d)
Gnuplot Plotting File: C:\USEPA\BMDS2\Temp\tmpA33.plt
Wed Apr 08 14:24:02 2009
BMDS Model Run NTP 1989 Linear Mineralization Male Rat - Probit Model
The form of the probability function is:
P[response] = CumNorm(Intercept+Slope*Dose) ,
where CumNormf .) is the cumulative normal distribution function
Dependent variable = PercentPositiveLinearMineralization
Independent variable = ion
Slope parameter is not restricted
Total number of observations = 3
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial (and Specified) Parameter Values
background = 0 Specified
intercept = -1.67551
slope = 0.149038
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -background
have been estimated at a boundary point, or have been specified by
the user,
and do not appear in the correlation matrix )
intercept slope
intercept 1 -0.87
slope -0.87 1
Interval
Variable
Limit
intercept
1.14919
slope
0.191579
Estimate
-1.62793
0.144885
Parameter Estimates
Std. Err.
0.244257
0.0238239
95.0% Wald Confidence
Lower Conf. Limit Upper Conf.
-2.10666
0.0981906
B-28
DRAFT - DO NOT CITE OR QUOTE
-------
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced model
AIC:
Log(likelihood)
-71.6113
-71. 8283
-94.7689
147.657
# Param's
3
2
1
Deviance Test d.f.
P-value
0. 433989
46.3152
0.51
<.0001
Dose
Goodness of Fit
Est._Prob. Expected Observed Size
Scaled
Residual
0.0000
7.0000
14.0000
Chi^2 = 0.43
0.0518
0.2697
0.6556
d.f. = 1
2.589 2.000 50
13.485 15.000 50
32.780 32.000 50
P-value = 0.5129
-0.376
0.483
-0.232
Benchmark Dose Computation
Specified effect
Risk Type
Confidence level
BMD
BMDL
0.1
Extra risk
0. 95
3.98089
3.21773
B-29
DRAFT - DO NOT CITE OR QUOTE
-------
Probit Model with 0.95 Confidence Level
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
14:24 04/08
B-30 DRAFT - DO NOT CITE OR QUOTE
Probit
BMDL
dose
2009
-------
NTP (1989) Male Rat Hyperplasia of Pelvic Transitional
Epithelium
LogLogistic Model
Logistic Model. (Version: 2.12; Date: 05/16/2008)
Input Data File: C:\USEPA\BMDS2\Temp\tmp4D5.(d)
Gnuplot Plotting File: C:\USEPA\BMDS2\Temp\tmp4D5.plt
Wed Aug 12 14:26:53 2009
BMDS Model Run - NTP 989 - Male Rat - Hyperplasia - LogLogistic Model
The form of the probability function is:
P[response] = background+(1-background)/[1+EXP(-intercept-siope*Log(dose)) ]
Dependent variable = Effect
Independent variable = DOSE
Slope parameter is restricted as slope >= 1
Total number of observations = 3
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
User has chosen the log transformed model
Default Initial Parameter Values
background = 0
intercept = -3.7 612
slope = 1
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -background -slope
have been estimated at a boundary point, or have been specified by
the user,
and do not appear in the correlation matrix )
intercept
intercept 1
Parameter Estimates
Interval
Variable
Limit
background
intercept
slope
Estimate
-4.15077
1
Std. Err.
95.0% Wald Confidence
Lower Conf. Limit Upper Conf.
Indicates that this value is not calculated.
B-31
DRAFT - DO NOT CITE OR QUOTE
-------
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced model
AIC:
Log(likelihood)
-40.4963
-41.2103
-46.5274
84.4207
# Param's
3
1
1
Deviance Test d.f.
1.42796
12.0622
P-value
0.4897
0.002403
Dose
Goodness of Fit
Est._Prob. Expected Observed Size
Scaled
Residual
0.0000
7.0000
14.0000
Chi^2 = 1.48
0.0000
0.0993
0.1807
d.f. = 2
0.000 0.000 50
4.966 7.000 50
9.034 7.000 50
P-value = 0.47 61
0. 000
0. 962
-0.748
Benchmark Dose Computation
Specified effect
Risk Type
Confidence level
BMD
BMDL
0.1
Extra risk
0. 95
7.05365
4.48322
B-32
DRAFT - DO NOT CITE OR QUOTE
-------
Log-Logistic Model with 0.95 Confidence Level
0.3
Log-Logistic
0.25
0.2
15
1
0.05
0
BMDL
BMD
0
2
4
6
8
10
12
14
dose
14:26 08/12 2009
B-33
DRAFT - DO NOT CITE OR QUOTE
-------
Gorzinski (1985) Atrophy and Degeneration of renal tubules in
Male Rats
Gamma Model
Gamma Model. (Version: 2.13; Date: 05/16/2008)
Input Data File: C:\USEPA\BMDS2\Temp\tmpF14.(d)
Gnuplot Plotting File: C:\USEPA\BMDS2\Temp\tmpF14.plt
Thu Oct 08 08:59:00 2009
ndDegenRenalTubulesDataNoSeverityMaleRat.dax
The form of the probability function is:
P[response]= background+(1-background)*CumGamma[siope*dose,power],
where CumGamma(.) is the cummulative Gamma distribution function
Dependent variable = Effect
Independent variable = DOSE
Power parameter is restricted as power >=1
Total number of observations = 4
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial (and Specified) Parameter Values
Background = 0.136364
Slope = 0.0871864
Power = 1.3
Asymptotic Correlation Matrix of Parameter Estimates
Background Slope Power
Background 1 0.52 0.64
Slope 0.52 1 0.93
Power 0.64 0.93 1
Parameter Estimates
Interval
Variable
Limit
Background
0.320747
Slope
0.244756
Power
3.0996
Estimate
0.110626
0.0787607
1.00164
95.0% Wald Confidence
Std. Err. Lower Conf. Limit Upper Conf.
0.107207 -0.0994949
0.0846932 -0.0872348
1.07041 -1.09632
B-34
DRAFT - DO NOT CITE OR QUOTE
-------
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced model
AIC:
Log(likelihood)
-14.3635
-14.4712
-27.7259
34.9424
# Param's
4
3
1
Deviance Test d.f.
P-value
0.215359
26.7248
0.6426
<.0001
Dose
Goodness of Fit
Est. Prob.
Expected
Observed
Size
Scaled
Residual
0.0000
1.0000
15.0000
62.0000
Chi^2 =0.15
0.1106
0.1777
0.7265
0.9932
d.f. = 1
1.106 1.000 10
1.777 2.000 10
7.265 7.000 10
9.932 10.000 10
P-value = 0.6994
-0.107
0.185
-0.188
0.261
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Extra risk
Confidence level = 0.95
BMD = 1.34399
BMDL = 0.727509
"O
0
-l—1
o
CD
o
CO
Gamma Multi-Hit Model with 0.95 Confidence Level
Gamma Multi-Hit
EMDLBMD
08:59 10/08 2009
B-35
DRAFT - DO NOT CITE OR QUOTE
-------
Multistage 1 degree Model
Multistage Model. (Version: 3.0; Date: 05/16/2008)
Input Data File: C:\USEPA\BMDS2\Temp\tmpF17.(d)
Gnuplot Plotting File: C:\USEPA\BMDS2\Temp\tmpF17.plt
Thu Oct 08 09:00:57 2009
ndDegenRenalTubulesDataNoSeverityMaleRat.dax
The form of the probability function is:
P[response] = background + (1-background)*[1-EXP(
-betal*doseAl) ]
The parameter betas are restricted to be positive
Dependent variable = Effect
Independent variable = DOSE
Total number of observations = 4
Total number of records with missing values = 0
Total number of parameters in model = 2
Total number of specified parameters = 0
Degree of polynomial = 1
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial Parameter Values
Background = 0
Beta(1) = 1.66732e+018
Asymptotic Correlation Matrix of Parameter Estimates
Background Beta(l)
Background 1 -0.4
Beta (1) -0.4 1
Parameter Estimates
95.0% Wald Confidence
Interval
Variable Estimate Std. Err. Lower Conf. Limit Upper Conf.
Limit
Background 0.11052 * * *
Beta(1) 0.0786399 * * *
* - Indicates that this value is not calculated.
Analysis of Deviance Table
Model Log(likelihood) # Param's Deviance Test d.f. P-value
B-36
DRAFT - DO NOT CITE OR QUOTE
-------
Full model -14.3635 4
Fitted model -14.4712 2 0.215361 2 0.8979
Reduced model -27.7259 1 26.7248 3 <.0001
AIC: 32.9424
Dose
Est. Prob.
Goodness of Fit
Expected
Observed
Size
Scaled
Residual
0.0000
1.0000
15.0000
62.0000
0.1105
0.1778
0.7266
0.9932
1.105
1.778
7.266
9.932
1.000
2.000
7.000
10.000
10
10
10
10
-0.106
0.184
-0.189
0.261
Chi^2 =0.15
d.f. = 2
P-value = 0.92 83
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Extra risk
Confidence level = 0.95
BMD = 1.33978
BMDL = 0.727509
BMDU = 2.66189
Taken together, (0.727509, 2.66189) is a 90 % two-sided confidence
interval for the BMD
B-37
DRAFT - DO NOT CITE OR QUOTE
-------
Multistage Model with 0.95 Confidence Level
~o
CD
o
(D
-------
Quantal-linear Model
Quantal Linear Model using Weibull Model (Version: 2.12; Date: 05/16/2008)
Input Data File: C:\USEPA\BMDS2\Temp\tmpF18.(d)
Gnuplot Plotting File: C:\USEPA\BMDS2\Temp\tmpF18.plt
Thu Oct 08 09:02:11 2009
ndDegenRenalTubulesDataNoSeverityMaleRat.dax
The form of the probability function is:
P[response] = background + (1-background)*[1-EXP(-slope*dose)]
Dependent variable = Effect
Independent variable = DOSE
Total number of observations = 4
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial (and Specified) Parameter Values
Background = 0.136364
Slope = 0.047491
Power = 1 Specified
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -Power
have been estimated at a boundary point, or have been specified by
the user,
and do not appear in the correlation matrix )
Background Slope
Background 1 -0.2 9
Slope -0.29 1
Parameter Estimates
95.0% Wald Confidence
Interval
Variable Estimate Std. Err. Lower Conf. Limit Upper Conf.
Limit
Background 0.11052 0.0819804 -0.0501583
0.271199
Slope 0.0786399 0.0310542 0.0177749
0.139505
Analysis of Deviance Table
Model Log(likelihood) # Param's Deviance Test d.f. P-value
Full model -14.3635 4
Fitted model -14.4712 2 0.215361 2 0.8979
B-39
DRAFT - DO NOT CITE OR QUOTE
-------
Reduced model
AIC:
-27.7259
32.9424
26.7248
<.0001
Dose
Goodness of Fit
Est._Prob. Expected Observed Size
Scaled
Residual
0.0000
1.0000
15.0000
62.0000
Chi^2 =0.15
0.1105
0.1778
0.7266
0.9932
d.f. = 2
1.105 1.000 10
1.778 2.000 10
7.266 7.000 10
9.932 10.000 10
P-value = 0.92 83
-0.106
0.184
-0.189
0.261
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Extra risk
Confidence level = 0.95
BMD = 1.33978
BMDL = 0.727509
Quantal Linear Model with 0.95 Confidence Level
Quantal Linear
0.6
0.6
0.4
0.2
BflVIDL^BMD
0
09:02 10/08 2009
10
20
30
dose
40
50
60
B-40
DRAFT - DO NOT CITE OR QUOTE
-------
Gorzinski (1985) Atrophy and Degeneration of renal tubules in
Female Rats
Probit Model
Probit Model. (Version: 3.1; Date: 05/16/2008)
Input Data File: C:\USEPA\BMDS2\Temp\tmpF0E.(d)
Gnuplot Plotting File: C:\USEPA\BMDS2\Temp\tmpF0E.plt
Thu May 06 10:06:12 2010
ubulesDataNoSeverityFemaleRat.dax
The form of the probability function is:
P[response] = CumNorm(Intercept+Slope*Dose) ,
where CumNormf .) is the cumulative normal distribution function
Dependent variable = Effect
Independent variable = DOSE
Slope parameter is not restricted
Total number of observations = 4
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial (and Specified) Parameter Values
background = 0 Specified
intercept = -1.21184
slope = 0.0236401
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -background
have been estimated at a boundary point, or have been specified by
the user,
and do not appear in the correlation matrix )
intercept slope
intercept 1 -0.69
slope -0.69 1
Interval
Variable
Limit
intercept
0.628881
slope
0.0417261
Estimate
-1.26508
0.0246481
Parameter Estimates
Std. Err.
0.324595
0. 00871343
95.0% Wald Confidence
Lower Conf. Limit Upper Conf.
-1.90127
0.00757005
B-41
DRAFT - DO NOT CITE OR QUOTE
-------
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced model
Log(likelihood)
-18.2358
-18.2465
-22.4934
# Param's
4
2
1
Deviance Test d.f.
0.0214055
8.51521
P-value
0.9894
0.03648
AIC:
40.493
Dose
Goodness of Fit
Est. Prob.
Expected
Observed
Size
Scaled
Residual
0.0000
1.0000
15.0000
62.0000
Chi^2 =0.02
0.1029
0.1074
0.1853
0.6038
d.f. = 2
1.029 1.000 10
1.074 1.000 10
1.853 2.000 10
6.038 6.000 10
P-value = 0.98 93
-0.030
-0.076
0.120
-0.024
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Extra risk
Confidence level = 0.95
BMD = 16.0998
BMDL = 10.5128
Probit Model with 0.95 Confidence Level
Probit
8
0.6
0.4
0.2
0
BMDL
BMD
0
10
20
30
40
50
60
dose
10:06 05/06 2010
B-42
DRAFT - DO NOT CITE OR QUOTE
-------
Gorzinski et al. (1985) Male Rat Hypertrophy and/or Dilation of
Proximal Tubules
Gamma Model
Gamma Model. (Version: 2.13; Date: 05/16/2008)
Input Data File: C:\USEPA\BMDS2\Temp\tmp4D6.(d)
Gnuplot Plotting File: C:\USEPA\BMDS2\Temp\tmp4D6.plt
Wed Aug 12 14:31:38 2009
BMDS Model Run - Gorzinski et al (1985) - Male Rat - Hypertrophy/Dilation of Proximal
Tubules - Gamma Model
The form of the probability function is:
P[response]= background+(1-background)*CumGamma[siope*dose,power],
where CumGamma(.) is the cummulative Gamma distribution function
Dependent variable = Effect
Independent variable = DOSE
Power parameter is restricted as power >=1
Total number of observations = 4
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial (and Specified) Parameter Values
Background = 0.0454545
Slope = 0.0907614
Power = 1.3
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -Background -Power
have been estimated at a boundary point, or have been specified by
the user,
and do not appear in the correlation matrix )
Slope
Slope 1
Parameter Estimates
Interval
Variable
Limit
Background
Slope
0.143889
Power
Estimate
0
0.0860249
1
Std. Err.
NA
0.029523
NA
NA - Indicates that this parameter has hit a bound
implied by some ineguality constraint and thus
95.0% Wald Confidence
Lower Conf. Limit Upper Conf.
0.0281609
B-43
DRAFT - DO NOT CITE OR QUOTE
-------
has no standard error.
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced model
AIC:
Log(likelihood)
-9.35947
-9.44226
-27.5256
20. 8845
# Param's Deviance Test d.f. P-value
4
1 0.165576 3 0.9829
1 36.3322 3 <.0001
Dose
Goodness of Fit
Est. Prob.
Expected
Observed
Size
Scaled
Residual
0.0000
1.0000
15.0000
62.0000
Chi^2 =0.12
0.0000
0.0824
0.7248
0.9952
d.f. = 3
0.000 0.000 10
0.824 1.000 10
7.248 7.000 10
9.952 10.000 10
P-value = 0.98 93
0. 000
0.202
-0.176
0.220
Benchmark Dose Computation
Specified effect
Risk Type
Confidence level
BMD
BMDL
0.1
Extra risk
0. 95
1.22477
0.710032
B-44
DRAFT - DO NOT CITE OR QUOTE
-------
Gamma Multi-Hit Model with 0.95 Confidence Level
~o
CD
o
(D
-------
Weibull Model
Weibull Model using Weibull Model (Version: 2.12; Date: 05/16/2008)
Input Data File: C:\USEPA\BMDS2\Temp\tmp4D9.(d)
Gnuplot Plotting File: C:\USEPA\BMDS2\Temp\tmp4D9.plt
Wed Aug 12 14:35:51 2009
BMDS Model Run - Gorzinski et al (1985) - Male rats - Hypertrophy/Dilation of
Proximal Tubules - Weibull Model using Weibull Model (Version: 2.12; Date:
05/16/2008)
The form of the probability function is:
P[response] = background + (1-background)*[1-EXP(-slope*dose/spower)]
Dependent variable = Effect
Independent variable = DOSE
Power parameter is restricted as power >=1
Total number of observations = 4
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial (and Specified) Parameter Values
Background = 0.0454545
Slope = 0.0491052
Power = 1
the user,
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -Background -Power
have been estimated at a boundary point, or have been specified by
and do not appear in the correlation matrix )
Slope
Slope 1
Parameter Estimates
Interval
Variable
Limit
Background
Slope
0.143889
Power
Estimate
0
0.086025
1
Std. Err.
NA
0.0295231
NA
NA - Indicates that this parameter has hit a bound
implied by some ineguality constraint and thus
has no standard error.
95.0% Wald Confidence
Lower Conf. Limit Upper Conf.
0.0281608
B-46
DRAFT - DO NOT CITE OR QUOTE
-------
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced model
AIC:
Log(likelihood)
-9.35947
-9.44226
-27.5256
20. 8845
# Param's
4
1
1
Deviance Test d.f.
P-value
0.165576
36.3322
0.9829
<.0001
Dose
Goodness of Fit
Est. Prob.
Expected
Observed
Size
Scaled
Residual
0.0000
1.0000
15.0000
62.0000
Chi^2 =0.12
0.0000
0.0824
0.7248
0.9952
d.f. = 3
0.000 0.000 10
0.824 1.000 10
7.248 7.000 10
9.952 10.000 10
P-value = 0.98 93
0. 000
0.202
-0.176
0.220
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Extra risk
Confidence level = 0.95
BMD = 1.22477
BMDL = 0.710032
"O
0
-l—1
o
CD
o
CO
0.6
0.6
0.4
0.2
B
Weibull Model with 0.95 Confidence Level
Weibull
MDLBMD
0
14:35 08/12 2009
10
20
30
dose
40
50
60
B-47
DRAFT - DO NOT CITE OR QUOTE
-------
Quantal-linear Model
antal-linear Model using Weibull Model (Version: 2.12; Date: 05/16/2008)
Input Data File: C:\USEPA\BMDS2\Temp\tmp4DA.(d)
Gnuplot Plotting File: C:\USEPA\BMDS2\Temp\tmp4DA.plt
Wed Aug 12 14:37:26 2009
BMDS Model Run - Gorzinski et al (1985) - Male rats - Hypertrophy/Dilation Proximal
Tubules - Quantal-linear Model using Weibull Model (Version: 2.12; Date: 05/16/2008)
The form of the probability function is:
P[response] = background + (1-background)*[1-EXP(-slope*dose/spower)]
Dependent variable = Effect
Independent variable = DOSE
Power parameter is set to 1
Total number of observations = 4
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial (and Specified) Parameter Values
Background = 0.0454545
Slope = 0.0491052
Power = 1 Specified
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -Background -Power
have been estimated at a boundary point, or have been specified by
the user,
and do not appear in the correlation matrix )
Slope
Slope 1
Parameter Estimates
95.0% Wald Confidence
Interval
Variable Estimate Std. Err. Lower Conf. Limit Upper Conf.
Limit
Background 0 NA
Slope 0.0860249 0.029523 0.0281608
0.143889
NA - Indicates that this parameter has hit a bound
implied by some ineguality constraint and thus
has no standard error.
Analysis of Deviance Table
B-48
DRAFT - DO NOT CITE OR QUOTE
-------
Model
Full model
Fitted model
Reduced model
Log(likelihood)
-9.35947
-9.44226
-27.5256
# Param's
4
1
1
Deviance Test d.f.
0.165576
36.3322
P-value
0.9829
<.0001
AIC:
20.8845
Dose
Goodness of Fit
Est._Prob. Expected Observed Size
Scaled
Residual
0.0000
1.0000
15.0000
62.0000
Chi^2 =0.12
0.0000
0.0824
0.7248
0.9952
d.f. = 3
0.000 0.000 10
0.824 1.000 10
7.248 7.000 10
9.952 10.000 10
P-value = 0.98 93
0. 000
0.202
-0.176
0.220
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Extra risk
Confidence level = 0.95
BMD = 1.22477
BMDL = 0.710032
Weibull Model with 0.95 Confidence Level
"O
0
-i—1
O
(D
o
CO
14:37
Weibull
0 10 20 30 40 50 60
dose
08/12 2009
B-49
DRAFT - DO NOT CITE OR QUOTE
-------
NTP (1989) Female Rat Hepatocellular Necrosis
Gamma Model
Gamma Model. (Version: 2.13; Date: 05/16/2008)
Input Data File: C:\USEPA\BMDS2\Temp\tmpB62.(d)
Gnuplot Plotting File: C:\USEPA\BMDS2\Temp\tmpB62.plt
Thu Apr 09 09:14:08 2009
BMDS Model Run NTP 1989 Hepatocellular Necrosis Female Rat - Gamma Model
The form of the probability function is:
P[response]= background+(1-background)*CumGamma[siope*dose,power],
where CumGamma(.) is the cummulative Gamma distribution function
Dependent variable = PercentPositiveHepatocellularNecrosis
Independent variable = rosis
Power parameter is restricted as power >=1
Total number of observations = 6
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial (and Specified) Parameter Values
Background = 0.0454545
Slope = 0.00743289
Power = 2.82109
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -Background
have been estimated at a boundary point, or have been specified by
the user,
and do not appear in the correlation matrix )
Slope Power
Slope 1 0.95
Power 0.95 1
Parameter Estimates
Interval
Variable
Limit
Background
Slope
0.0150393
Power
4.823
Estimate
0
0.00723384
2.58447
Std. Err.
NA
0. 00398244
1.14213
NA - Indicates that this parameter has hit a bound
implied by some ineguality constraint and thus
95.0% Wald Confidence
Lower Conf. Limit Upper Conf.
-0.000571608
0.345944
B-50
DRAFT - DO NOT CITE OR QUOTE
-------
has no standard error.
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced model
Log(likelihood) # Param'
-16.7382 6
-17.3091 2
-32.5964 1
Deviance Test d.f.
1.14186
31.7164
P-value
0.8876
<.0001
AIC:
38.6182
Goodness of Fit
Scaled
Dose Est._Prob. Expected Observed Size Residual
0.0000
0.0000
0.000
0.000
10
0. 000
33.5000
0. 0059
0.059
0.000
10
-0.244
67.1000
0.0300
0.300
0.000
10
-0.556
134.3000
0.1289
1.289
2.000
10
0. 671
267.8000
0.4095
4.095
4.000
10
-0.061
535 .7000
0. 8159
8.159
8.000
10
-0.130
Chi^2 = 0.84 d.f. = 4 P-value = 0.9331
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Extra risk
Confidence level = 0.95
BMD = 118.037
BMDL = 60.1812
B-51
DRAFT - DO NOT CITE OR QUOTE
-------
Gamma Multi-Hit Model with 0.95 Confidence Level
0.8
0.6
0.4
0.2
0
09:14 04/09
B-52 DRAFT - DO NOT CITE OR QUOTE
Gamma Multi-Hit
BMDL
BMD
dose
2009
-------
Modeling for Cancer Assessment
NTP (1989) BMP Modeling of Adenoma/Carcinoma in Male Rats
Multistage 2°Model
Multistage Cancer Model. (Version: 1.7; Date: 05/16/2008)
Input Data File: C:\USEPA\BMDS2\Temp\tmp6E8.(d)
Gnuplot Plotting File: C:\USEPA\BMDS2\Temp\tmp6E8.plt
Mon Apr 13 14:38:06 2009
BMDS Model Run NTP 1989 Kidney Adenoma-Carcinoma Male Rat - Multistage Cancer 2
degree Model
The form of the probability function is:
P[response] = background + (1-background)*[1-EXP(
-betal*dose/sl-beta2*dose/s2) ]
The parameter betas are restricted to be positive
Dependent variable = PercentAdenomaCarcinoma
Independent variable = DOSE
Total number of observations = 3
Total number of records with missing values = 0
Total number of parameters in model = 3
Total number of specified parameters = 0
Degree of polynomial = 2
Maximum number of iterations = 250
Relative Function Convergence has been set to: 2.22045e-016
Parameter Convergence has been set to: 1.49012e-008
**** We are sorry but Relative Function and Parameter Convergence ****
**** are currently unavailable in this model. Please keep checking ****
**** the web sight for model updates which will eventually ****
**** incorporate these convergence criterion. Default values used. ****
Default Initial Parameter Values
Background = 0.014541
Beta(l) = 0
Beta(2) = 0.00799069
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -Beta(l)
have been estimated at a boundary point, or have been specified by
the user,
and do not appear in the correlation matrix )
Background Beta (2)
Background 1 -0.67
Beta (2) -0.67 1
B-53
DRAFT - DO NOT CITE OR QUOTE
-------
Parameter Estimates
Interval
Variable
Limit
Background
Beta(1)
Beta(2)
Estimate
0.0177261
0
0.00751246
Std. Err.
95.0% Wald Confidence
Lower Conf. Limit Upper Conf.
Indicates that this value is not calculated.
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced model
AIC:
Log(likelihood) # Param's Deviance Test d.f. P-value
-33.5473 3
-33.6008 2 0.106829 1 0.7438
-36.7395 1 6.38433 2 0.04108
71.2015
Dose
Est. Prob.
Goodness of Fit
Expected Observed Size
Scaled
Residual
i: 1
0.0000
i: 2
2 . 0400
i: 3
4.0900
Chi-square =
0. 0177
0.0481
0.1343
0. 10
0. 887
2 .407
6. 717
DF= 1
1
2
7
P-value =
50
50
50
0.7510
0. 129
-0. 178
0. 049
Benchmark Dose Computation
Specified effect
Risk Type
Confidence level
BMD
BMDL
BMDU
0.1
Extra risk
0. 95
3.74496
2.45283
9.24921
Taken together, (2.45283, 9.24921) is a 90
interval for the BMD
two-sided confidence
Multistage Cancer Slope Factor =
0.0407692
B-54
DRAFT - DO NOT CITE OR QUOTE
-------
Multistage Cancer Model with 0.95 Confidence Level
0.3
Multistage Cancer
Linear extrapolation
0.25
0.2
0.15
0.05
BMDL
BMD
0
0.5
1
1.5
2
2.5
3
3.5
4
dose
14:38 04/13 2009
B-55
DRAFT - DO NOT CITE OR QUOTE
-------
NCI (1978) BMP Modeling of Hepatocellular Carcinoma in Male Mice
Multistage 2°
Multistage Cancer Model. (Version: 1.7; Date: 05/16/2008)
Input Data File: C:\USEPA\BMDS2\Temp\tmp7B8.(d)
Gnuplot Plotting File: C:\USEPA\BMDS2\Temp\tmp7B8.plt
Tue Apr 14 08:30:03 2009
BMDS Model Run NCI 1978 Hepatocellular Carcinoma Male Mice - Multistage Cancer 2
degree Model
The form of the probability function is:
P[response] = background + (1-background)*[1-EXP(
-betal*dose/sl-beta2*dose/s2) ]
The parameter betas are restricted to be positive
Dependent variable = PercentHepatocellularCarcinoma
Independent variable = DOSE
Total number of observations = 3
Total number of records with missing values = 0
Total number of parameters in model = 3
Total number of specified parameters = 0
Degree of polynomial = 2
Maximum number of iterations = 250
Relative Function Convergence has been set to: 2.22045e-016
Parameter Convergence has been set to: 1.49012e-008
**** We are sorry but Relative Function and Parameter Convergence ****
**** are currently unavailable in this model. Please keep checking ****
**** the web sight for model updates which will eventually ****
**** incorporate these convergence criterion. Default values used. ****
Default Initial Parameter Values
Background = 0.141096
Beta(l) = 0
Beta(2) = 7.77012e-005
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -Beta(l)
have been estimated at a boundary point, or have been specified by
the user,
and do not appear in the correlation matrix )
Background Beta (2)
Background 1 -0.73
Beta (2) -0.73 1
Parameter Estimates
B-56
DRAFT - DO NOT CITE OR QUOTE
-------
95.0% Wald Confidence
Interval
Variable Estimate Std. Err. Lower Conf. Limit Upper Conf.
Limit
Background 0.146344 * * *
Beta(1) 0 * * *
Beta(2) 7.26074e-005 * * *
* - Indicates that this value is not calculated.
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced model
Log(likelihood)
-71.2862
-71.7199
-80.5752
# Param's
3
2
1
Deviance Test d.f.
0.867331
18.5779
P-value
0.3517
<.0001
AIC:
147.44
Dose
Goodness of Fit
Est._Prob. Expected Observed Size
Scaled
Residual
0.0000
53.0500
103.8800
0.1463
0.3041
0.6101
2.927
15 .206
29.892
3.000
15.000
30.870
20
50
49
0. 046
-0.063
0.286
Chi^2 = 0.09
d.f. = 1
P-value = 0.7666
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Extra risk
Confidence level = 0.95
BMD = 38.0933
BMDL = 13.8018
BMDU = 4 9.5091
Taken together, (13.8018, 49.5091) is a 90 % two-sided confidence
interval for the BMD
Multistage Cancer Slope Factor = 0.00724545
B-57
DRAFT - DO NOT CITE OR QUOTE
-------
Multistage Cancer Model with 0.95 Confidence Level
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
08:30 04/14
Multistage Cancer
Linear extrapolation
BMDL
B-58
DRAFT - DO NOT CITE OR QUOTE
-------
NTP (1989) BMP Modeling of Pheochromocvtoma/Malignant
Pheochromocvtomas in Male Rats
Multistage 2°
Multistage Cancer Model. (Version: 1.7; Date: 05/16/2008)
Input Data File: C:\USEPA\BMDS2\Temp\tmp70C.(d)
Gnuplot Plotting File: C:\USEPA\BMDS2\Temp\tmp70C.plt
Mon Apr 13 15:55:38 2009
tomaMaleRat.dax
The form of the probability function is:
P[response] = background + (1-background)*[1-EXP(
-betal*dose/sl-beta2*dose/s2) ]
The parameter betas are restricted to be positive
Dependent variable = PercentPheochromocytomaMalignantPheochromocytoma
Independent variable = Pheochromocytoma
Total number of observations = 3
Total number of records with missing values = 0
Total number of parameters in model = 3
Total number of specified parameters = 0
Degree of polynomial = 2
Maximum number of iterations = 250
Relative Function Convergence has been set to: 2.22045e-016
Parameter Convergence has been set to: 1.49012e-008
**** We are sorry but Relative Function and Parameter Convergence ****
**** are currently unavailable in this model. Please keep checking ****
**** the web sight for model updates which will eventually ****
**** incorporate these convergence criterion. Default values used. ****
Default Initial Parameter Values
Background = 0.381549
Beta(1) = 0.0404371
Beta(2) = 0
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -Beta(2)
have been estimated at a boundary point, or have been specified by
the user,
and do not appear in the correlation matrix )
Background Beta(l)
Background 1 -0.78
Beta (1) -0.78 1
Parameter Estimates
B-59
DRAFT - DO NOT CITE OR QUOTE
-------
Interval
Variable
Limit
Background
Beta(1)
Beta(2)
Estimate
0.341708
0.055345
0
Std. Err.
95.0% Wald Confidence
Lower Conf. Limit Upper Conf.
Indicates that this value is not calculated.
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced model
AIC:
Log(likelihood)
-93.0295
-96.701
-97.5291
197.402
# Param's Deviance Test d.f. P-value
3
2 7.34302 1 0.006732
1 8.99926 2 0.01111
Dose
Goodness of Fit
Est._Prob. Expected Observed Size
Scaled
Residual
0.0000
2.0500
4.1000
Chi^2 = 7.50
0.3417
0.4123
0.4753
d.f. = 1
17.085 14.000 50
18.554 26.100 45
23.292 19.110 49
P-value = 0.00 62
-0.920
2.285
-1.196
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Extra risk
Confidence level = 0.95
BMD = 1.9037
BMDL = 0.811704
BMDU did not converge for BMR = 0.100000
BMDU calculation failed
BMDU = Inf
B-60
DRAFT - DO NOT CITE OR QUOTE
-------
Multistage Cancer Model with 0.95 Confidence Level
15:55 04/13 2009
Multistage Cancer
Linear extrapolation
BMDL
0 0.5 1
BMD
1.5 2 2.5 3
dose
3.5
B-61
DRAFT - DO NOT CITE OR QUOTE
------- |