EPA/63 5/R-01/001
?/EPA
TOXICOLOGICAL REVIEW
OF
CHLOROFORM
(CAS No. 67-66-3)
In Support of Summary Information on the
Integrated Risk Information System (IRIS)
October 2001
U.S. Environmental Protection Agency
Washington, DC
-------
DISCLAIMER
This document has been reviewed in accordance with U.S. Environmental Protection Agency
policy. Mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
Note: This document may undergo revisions in the future. The most up-to-date version will be
made available electronically via the IRIS Home Page at http://www.epa.gov/iris
-------
CONTENTS - TOXICOLOGICAL REVIEW OF CHLOROFORM
(CAS No. 67-66-3)
LIST OF TABLES v
LIST OF FIGURES v
ACRONYM LIST vi
FOREWORD vii
AUTHORS, CONTRIBUTORS, AND REVIEWERS viii
SUMMARY OF SCIENCE ADVISORY BOARD RECOMMENDATIONS AND EPA
RESPONSES xi
1. INTRODUCTION 1
2. CHEMICAL AND PHYSICAL INFORMATION 2
3. TOXICOKINETICS 2
3.1. ABSORPTION 2
3.2. DISTRIBUTION 3
3.3. METABOLISM 3
3.3.1. Oxidative and Reductive Pathways 3
3.3.2. Fate of Reactive Metabolites 4
3.3.3. Relative Importance of Oxidative and Reductive Pathways 6
3.4. EXCRETION 6
3.5. PHYSIOLOGICALLY BASED PHARMACOKINETIC (PBPK) MODELS ... 6
4. HAZARD IDENTIFICATION 9
4.1. STUDIES IN HUMANS 9
4.1.1. Inhalation Studies in the Workplace 9
4.1.2. Exposure to Chloroform in Drinking Water 10
4.2. PRECHRONIC AND CHRONIC STUDIES AND CANCER BIOASSAYS
IN ANIMALS 11
4.2.1. Oral Studies 11
4.2.2. Inhalation Studies 18
4.3. REPRODUCTIVE/DEVELOPMENTAL STUDIES 21
4.3.1. Oral Studies 21
4.3.2. Inhalation Studies 24
4.4. OTHER STUDIES 26
4.4.1. Other Effects 26
4.4.2. Mutagenicity 27
4.4.3. Studies Related to Mode of Action 31
in
-------
CONTENTS (continued)
4.4.4. Studies of Interactions With Other Chemicals 34
4.5. SYNTHESIS AND EVALUATION OF MAJOR NONCANCER
EFFECTS AND MODE OF ACTION 35
4.6. WEIGHT-OF-EVIDENCE EVALUATION AND CANCER
CHARACTERIZATION 37
4.6.1. Mode of Action 37
4.6.2. Weight of Evidence 42
4.7. SUSCEPTIBLE POPULATIONS 43
4.7.1. Possible Childhood Susceptibility 43
4.7.2 Possible Gender Differences 47
4.7.3 Other Factors that May Increase Susceptibility 48
5. DOSE-RESPONSE ASSESSMENTS 49
5.1. ORAL REFERENCE DOSE 49
5.1.1. NOAEL-LOAEL Approach 49
5.1.2. Benchmark Dose Approach 51
5.1.3. Summary of Oral RfD Derivation 55
5.2. INHALATION REFERENCE CONCENTRATION 55
5.3. ORAL CANCER ASSESSMENT 56
5.3.1. Choice of Approach 56
5.3.2. Quantification of Cancer Risk 56
5.4. INHALATION CANCER ASSESSMENT 62
6. MAJOR CONCLUSIONS IN THE CHARACTERIZATION OF HAZARD
AND DOSE-RESPONSE 62
6.1. HUMANHAZARD POTENTIAL 62
6.1.1. Exposure Pathways 62
6.1.2. Toxicokinetics 63
6.1.3. Characterization of Noncancer Effects 63
6.1.4. Reproductive Effects and Risks to Children 64
6.1.5. Mode of Toxicity 64
6.1.6. Characterization of Human Carcinogenic Potential 64
6.2. DOSE RESPONSE 65
6.2.1. Oral RfD 65
6.2.2. Inhalation RfC 66
6.2.3. Oral Cancer Risk 66
6.2.4. Inhalation Cancer Risk 66
7. REFERENCES 67
APPENDIX A. EXTERNAL PEER REVIEW—SUMMARY OF COMMENTS AND
DISPOSITION A-l
APPENDIX B. QUANTITATIVE DOSE-RESPONSE MODELING B-l
IV
-------
LIST OF TABLES
Table 1. Summary of PBPK parameters 8
Table 2. Summary of chloroform-induced cytotoxicity and cell proliferation via inhalation ... 20
Table 3. Correlation of carcinogen!city and regenerative cell hyperplasia 39
Table 4. Summary of oral noncancer studies in animals 50
Table 5. Dose-response data sets used for BMD modeling 53
Table 6. Summary of noncancer BMD modeling results 54
Table 7. Summary of inhalation noncancer studies in humans and animals 57
Table 8. Summary of oral cancer studies in animals 60
Table 9. Dose-response modeling of male rat kidney tumor data 61
LIST OF FIGURES
Figure 1. Metabolic Pathways of Chloroform Biotransformation 5
Figure 2. SGPT Levels in Dogs Exposed to Chloroform for 7 Years 14
-------
ACRONYM LIST
AIC
ATP
BDCM
BMD
BMDL
BMDS
BMR
BrdU
CHO
CYP2E1
DEN
DNA
EPA
GOT
GOT
ICPEMC
ILSI
IRIS
LD50
LDH
LI
LOAEL
NCI
NOAEL
PBPK
ppm
RBC
RfD
SAP
SGPT
THM
TTHM
U.S. EPA
Akaike information criterion
Adenosine tri-phosphate
Bromodichloromethane
Benchmark dose
A lower one-sided confidence limit on the BMD
Benchmark dose software
Benchmark response
Bromodeoxyuridine
Chinese hamster ovary
Cytochrome P-450-2E1
Diethylnitrosamine
Deoxyribonucleic acid
Environmental Protection Agency
Gamma glutamyl transferase
Glutamate oxaloacetate transaminase (aspartate aminotransferase)
International Commission for Protection against Environmental Mutagens
International Life Sciences Institute
Integrated Risk Information System
Lethal Dose 50 (dose causing death in 50% of the exposed animals)
Lactate dehydrogenase
Labeling index
Lowest-observed-adverse-effect-level
National Cancer Institute
No-observed-adverse-effect-level
Physiologically based pharmacokinetic models
Parts per million
Red blood cell
Reference dose
Serum alkaline phosphatase
Serum glutamate pyruvate transaminase (alanine aminotransferase)
Trihalomethane
Total trihalomethanes
United States Environmental Protection Agency
VI
-------
FOREWORD
The purpose of this Toxicological Review is to provide scientific support and rationale
for the hazard and dose-response assessments in IRIS pertaining to chronic exposure to
chloroform. It is not intended to be a comprehensive treatise on the chemical or toxicological
nature of chloroform.
In Section 6, EPA has characterized its overall confidence in the quantitative and
qualitative aspects of hazard and dose response. Matters considered in this characterization
include knowledge gaps, uncertainties, quality of data, and scientific controversies. This
characterization is presented in an effort to make apparent 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.
vn
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Chemical Manager
Julie T. Du, Ph.D.
Office of Science and Technology
Office of Water
Washington, DC
Reviews
This Toxicological Review of Chloroform was based in part on the Health Risk
Assessment/Characterization of the Drinking Water Disinfectant Byproduct Chloroform and the
Draft Chloroform Risk Assessment (mode-of-action analysis for the carcinogen!city of
chloroform). Both documents have been peer-reviewed. The mode-of-action analysis was
reviewed by the Agency's Science Advisory Board (SAB) in October 1999. The SAB reviewers
and consultants are listed below, and the SAB report can be found on the web at
http://www.epa.gov/sab/fiscalOO.htm. The Agency response to SAB comments is shown
following the names of SAB reviewers. The Health Risk Assessment/Characterization of the
Drinking Water Disinfectant Byproduct Chloroform is peer-reviewed both by EPA scientists (see
Internal EPA Reviewers) and by independent scientists external to EPA (see External Peer
Reviewers). Summaries of the external peer reviewers' comments and the disposition of their
recommendations are in Appendix A. Subsequent to the external review and incorporation of
comments, this Toxicological Review of Chloroform and IRIS Summaries have been written and
undergone an Agencywide review process whereby the IRIS program manager has achieved a
consensus approval among the Office of Research and Development; Office of Air and
Radiation; Office of Prevention, Pesticides, and Toxic Substances; Office of Solid Waste and
Emergency Response; Office of Water; Office of Policy; Office of Children's Health Protection;
and the Regional Offices.
Before the reviews mentioned above, International Life Sciences Institute (ILSI) provided
a formal review of chloroform mode of action as part of a cooperative agreement with EPA. A
panel often scientific experts reviewed the literature and issued a report on the carcinogen risk
assessment of chloroform in November 1997. Similar to the SAB report, the ILSI report
supported a nonlinear approach for risk estimation.
As recommended by the SAB, a systematic analysis of the genotoxicity of chloroform,
including the most recent in vivo and in vitro studies, is included in this document and in the
IRIS summaries. A brief discussion of the epidemiological studies of chlorinated drinking water
(a mixture of disinfection byproducts including chloroform) is also included in this document.
On the noncancer endpoint, a more complete RfD analysis is performed including the traditional
NOAEL/LOAEL and the benchmark dose approaches. The final value is coincidentally the same
as the one previously on IRIS.
Vlll
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS (continued)
Internal EPA Reviewers
Penelope Fenner-Crisp, Ph.D.
Office of Pesticide Programs
Vicki Dellarco, Ph.D.
Health Effects Division
Office of Pesticide Programs
Steve Nesnow, Ph.D.
National Health and Environmental Effects Research Laboratory
Jennifer Seed, Ph.D.
Risk Assessment Division
Office of Pollution Prevention and Toxics
Vanessa Vu, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
External Peer Reviewers and Affiliation
External peer reviewers who provided comments on EPA's evaluation of chloroform are listed
below:
James A. Swenberg, D.V.M., Ph.D., University of North Carolina
Lorenz Rhomberg, Ph.D., Gradient Corporation
R. Julian Preston, Ph.D., Chemical Industry Institute of Toxicology
Summaries of the external peer reviewers' comments and the disposition of their
recommendations are presented in Appendix A.
SAB Review of the Mode of Action of Chloroform
Co-chairs, members, and consultants of the SAB who provided review comments on EPA's
evaluation of chloroform are listed below:
Dr. Richard J. Bull, Battelle Pacific Northwest National Laboratory (Co-chair)
Dr. Mark J. Utell, University of Rochester Medical Center (Co-chair)
IX
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS (continued)
Dr. Mary Davis, West Virginia University (member)
Dr. George Lambert, Robertwood Johnson University (member)
Dr. Lauren Zeise, California Environmental Protection Agency (member)
Dr. James E. Klaunig, Indiana University School of Medicine (consultant)
Dr. Richard Okita, Washington State University (consultant)
Dr. David Savitz, University of North Carolina, School of Public Health (consultant)
Dr. Verne Ray, Toxicologist (consultant)
Dr. Robert Maronpot, NIEHS (Federal Expert)
A summary of the comments provided by the SAB and EPA's response to those comments is
presented in the following two pages.
x
-------
SUMMARY OF SCIENCE ADVISORY BOARD RECOMMENDATIONS
AND EPA RESPONSES
In October 1999 the Chloroform Risk Assessment Review Subcommittee of the Science
Advisory Board met to consider the Office of Science and Technology health assessment of
chloroform. Summaries of the major parts of the subcommittee's advice and our responses
follow. The documents reviewed were a final hazard and dose-response characterization
document and a draft mode-of-action framework analysis.
1. The subcommittee agreed with EPA that sustained or repeated cytotoxicity with
secondary regenerative hyperplasia in the liver and/or the kidney of rats and mice
precedes, and is probably a causal factor for, hepatic and renal neoplasia. Some members
of the subcommittee were concerned about possible mutagenic activity, and the
subcommittee recommended that the risk assessment further address the possible role of
mutagenicity as a mode of action.
EPA Response: The Office of Water has included a more complete analysis of mutagenic
potential in the final Toxicological Review of Chloroform.
2. The Subcommittee concluded that a nonlinear margin-of-exposure approach is
scientifically reasonable for the liver tumor response because of the strong role
cytotoxicity appears to play. In contrast, the application of the standard linear approach to
the liver tumor data is likely to substantially overstate the low-dose risk. In addition,
there is considerable question about this response because it is not produced when
chloroform is administered to mice in drinking water.
For the kidney response, because sustained cytotoxicity plays a clear role in the rat, a
margin of exposure (MOE) is a scientifically reasonable approach. Most members of the
subcommittee thought that genotoxicity might possibly contribute to low-dose response
in this organ, while some thought it unlikely.
EPA Response: The Office of Water has utilized the MOE approach recommended by
SAB, but has also noted the reservations of some committee members regarding a
potential role for genotoxicity.
3. The subcommittee concluded that the extensive epidemiologic evidence relating drinking
water disinfection (specifically chlorination) with cancer has little bearing on the
determination of whether chloroform is a carcinogen. It added recommendations for
discussion of endpoints and the potential meaning of these data to the assessment of
chloroform.
EPA Response: The hazard and dose-response assessment document reviewed by SAB
did not contain the complete analysis of epidemiologic studies and the population-
attributable risk analysis. The latter were separately provided to the subcommittee. The
xi
-------
Toxicological Review for Chloroform does provide a summary of these studies along with
a discussion of their limitations in evaluating cancer risk from chloroform in humans.
4. The subcommittee found that the draft mode-of-action document addressed children's
risks quite adequately, based on the scientific information currently available. The major
conclusions were believed correct, the role of CYP2E1 should be expressed as important,
and its definitive role in the developing human or (other) mammal has yet to be
confirmed. Nevertheless, the subcommittee report discussed knowledge of children's
potential risk in several areas, such as exposure latency and transplacental and
transmamillary exposure, that can be improved.
EPA Response: The Office of Water has revised the Toxicological Review in accord with
the SAB recommendations. As the advice on some issues appears to be applicable
beyond the chloroform assessment and to carry implications for Agency guidance
documents, the advice will be discussed with the EPA Risk Assessment Forum.
xn
-------
1. INTRODUCTION
This document presents background and justification for the hazard and dose-response
assessment summaries in EPA's Integrated Risk Information System (IRIS). IRIS summaries
may include an oral reference dose (RfD), inhalation reference concentration (RfC), and a
carcinogen!city assessment.
The RfD and RfC provide quantitative information for noncancer dose-response
assessments. The RfD is based on the assumption that thresholds exist for certain toxic effects
such as cellular necrosis but may not exist for other toxic effects such as some carcinogenic
responses. It is expressed in units of mg/kg/day. In general, the RfD is 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 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). It is generally expressed in units of mg/m3.
The carcinogenicity assessment provides information on the carcinogenic hazard potential
of the substance in question and quantitative estimates of risk from oral exposure and inhalation
exposure. 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 are presented in three ways. The slope factor is the result
of application of a low-dose extrapolation procedure and is presented as the risk per mg/kg/day.
The unit risk is the quantitative estimate in terms of either risk per |ig/L drinking water or risk
per |ig/m3 air breathed. Another form in which risk is presented is as a drinking water or air
concentration providing cancer risks of 1 in 10,000; 1 in 100,000; or 1 in 1,000,000.
Development of these hazard identification and dose-response assessments for
chloroform has followed the general guidelines for risk assessment as set forth by the National
Research Council (1983). EPA guidelines that were used in the development of this assessment
may include the following: the Guidelines for Carcinogen Risk Assessment (U.S. EPA, 1986a),
Guidelines for the Health Risk Assessment of Chemical Mixtures (U.S. EPA, 1986b), Guidelines
for Mutagenicity Risk Assessment (U.S. EPA, 1986c), Guidelines for Developmental Toxicity
Risk Assessment (U.S. EPA, 1991), Proposed Guidelines for Neurotoxicity Risk Assessment (U.S.
EPA, 1998a), Proposed Guidelines for Carcinogen Risk Assessment (U.S. EPA, 1996a), Draft
Revisions of the Guidelines for Carcinogen Risk Assessment (U.S. EPA, 1999), Reproductive
Toxicity Risk Assessment Guidelines (U.S. EPA, 1996b); Recommendations for and
Documentation of Biological Values for Use in Risk Assessment (U.S. EPA, 1988); (proposed)
Interim Policy for Particle Size and Limit Concentration Issues in Inhalation Toxicity (U.S. EPA,
1994a); Methods for Derivation of Inhalation Reference Concentrations and Application of
Inhalation Dosimetry (U.S. EPA, 1994b); Peer Review and Peer Involvement at the U.S.
Environmental Protection Agency (U.S. EPA, 1994c); Use of the Benchmark Dose Approach in
Health Risk Assessment (U.S. EPA, 1995a); Science Policy Council Handbook: Peer Review
(U.S. EPA, 1998b); and memorandum from EPA Administrator, Carol Browner, dated March 21,
1995, Subject: Guidance on Risk Characterization.
-------
The literature search strategies employed for this compound were based on the CASRN
and at least one common name. At a minimum, the following databases were searched: RTECS,
HSDB, TSCATS, CCRIS, GENETOX, EMIC, EMICBACK, DART, ETICBACK, TOXLINE,
CANCERLINE, MEDLINE, and MEDLINE backfiles. Any pertinent scientific information
submitted by the public to the IRIS Submission Desk was also considered in the development of
this document.
2. CHEMICAL AND PHYSICAL INFORMATION
Chloroform (trichloromethane) is a colorless, volatile liquid with a distinct odor.
Chloroform is nonflammable. It is slightly soluble in water and is readily miscible with most
organic solvents (Lewis 1993). Selected chemical and physical properties of chloroform are
listed below (Howard and Meylan 1997):
CASRN: 67-66-3
Empirical formula: CHC13
Molecular weight: 119.38
Density: 1.483 g/mL
Vapor pressure: 197 mm Hg at25°C
Henry's Law Constant: 3E-03 atm-m3/mole (0.12 mg/L in air per mg/L in water)
Water solubility: 7.95 g/L at 25°C
LogKow: 1.97
Conversion factor (air): 1 ppm = 4.88 mg/m3
1 mg/m3 = 0.205 ppm
Because chloroform is relatively volatile, it tends to escape from contaminated environmental
media (e.g., water or soil) into air, and may also be released in vapor form from some types of
industrial or chemical operations. Therefore, humans may be exposed to chloroform not only by
ingestion of chloroform in drinking water, food, or soil, but also by dermal contact with
contaminated media (especially water) and by inhalation of vapor (especially in indoor air).
3. TOXICOKINETICS
3.1. ABSORPTION
Studies in animals indicate that gastrointestinal absorption of chloroform is rapid (peak
blood levels at about 1 hour) and extensive (64% to 98%) (U.S. EPA, 1997; ILSI, 1997; U.S.
EPA, 1998c). Limited data indicate that gastrointestinal absorption of chloroform is also rapid
and extensive in humans, with more than 90% of an oral dose recovered in expired air (either as
unchanged chloroform or carbon dioxide) within 8 hours (Fry et al., 1972).
-------
Most studies of chloroform absorption following oral exposure have used oil-based
vehicles and gavage dosing (U.S. EPA, 1994d, 1998c). This is of potential significance because
most humans are exposed to chloroform by ingestion in drinking water. Withey et al. (1983)
compared the rate and extent of gastrointestinal absorption of chloroform following gavage
administration in either aqueous or corn oil vehicles. Twelve male Wistar rats were administered
single oral doses of 75 mg chloroform/kg via gavage. The time-to-peak blood concentration of
chloroform was similar for both vehicles; however, the concentration of chloroform in the blood
was lower at all time points for the animals administered chloroform in the oil vehicle compared
with animals administered the water vehicle. The authors interpreted this to indicate that the rate
of chloroform absorption was higher from water than from oil, although differences in the rate of
first-pass metabolism in the liver might contribute to the observed difference (U.S. EPA, 1994d,
1998c).
Dermal and inhalation absorption of chloroform by humans during showering was
investigated by Jo et al. (1990). Chloroform concentrations in exhaled breath were measured in
six human subjects before and after a normal shower, and following inhalation-only shower
exposure. Breath levels measured at 5 minutes following either exposure correlated with tap
water levels of chloroform. Breath levels following inhalation exposure only were about half
those following a normal shower (both inhalation and dermal contact). These data indicate that
humans absorb chloroform by both the dermal and inhalation routes (U.S. EPA, 1994d).
3.2. DISTRIBUTION
Absorbed chloroform appears to distribute widely throughout the body (U.S. EPA, 1994d,
1998c). In postmortem samples from eight humans, the highest levels of chloroform were
detected in the body fat (5-68 • g/kg), with lower levels (1-10 • g/kg) detected in the kidney,
liver, and brain (McConnell et al., 1975). Studies in animals indicate rapid uptake of chloroform
by the liver and kidney (U.S. EPA, 1997). In mice receiving chloroform via gavage in either
corn oil or water, maximal uptake of chloroform was achieved within 10 minutes in the liver and
within 1 hour in the kidney (Pereira, 1994). Following intraperitoneal injection of 150 mg/kg
14C-chloroform, peak radioactivity levels were achieved in the liver, kidney, and blood of male
mice within 10 minutes of dosing, and had returned to background levels within 3 hours (Gemma
etal., 1996).
3.3. METABOLISM
3.3.1. Oxidative and Reductive Pathways
Chloroform is metabolized in humans and animals by cytochrome P450-dependent
pathways. In the presence of oxygen (oxidative metabolism), the chief product is
trichloromethanol, which rapidly and spontaneously dehydrochlorinates to form phosgene
(CC12O):
2 CHC13 + NADPH + FT + O2 2 CC13OH + NADP+
CCUOH CCLO + HC1
-------
In the absence of oxygen (reductive metabolism), the chief metabolite is dichloromethyl free
radical (CHC12) (U.S. EPA, 1997; ILSI, 1997).
Nearly all tissues of the body are capable of metabolizing chloroform, but the rate of
metabolism is greatest in liver, kidney cortex, and nasal mucosa (ILSI, 1997). These tissues are
also the principal sites of chloroform toxicity, indicating the importance of metabolism in the
mode of action of chloroform toxicity.
At low chloroform concentrations, metabolism occurs primarily via cytochrome P450-
2E1 (CYP2E1) (Constan et al., 1999). The level of this isozyme (and hence the rate of
chloroform metabolism) is induced by a variety of alcohols (including ethanol) and ketones, and
may be inhibited by phenobarbital. At high chloroform concentrations, metabolism is also
catalyzed by cytochrome P450-2B1/2 (CYP2B1/2) (ILSI, 1997; U.S. EPA, 1997, 1998c).
Because chloroform metabolism is enzyme-dependent, the rate of metabolism displays saturation
kinetics. Under low dose-rate conditions, nearly all of a dose is metabolized. However, as the
dose or the dose rate increases, metabolic capacity may become saturated and increasing fractions
of the dose are excreted as the unmetabolized parent (Fry et al., 1972).
3.3.2. Fate of Reactive Metabolites
The products of oxidative metabolism (phosgene) and reductive metabolism
(dichloromethyl free radical) are both highly reactive. Phosgene is electrophilic and undergoes
attack by a variety of nucleophiles. The predominant reaction is hydrolysis by water, yielding
carbon dioxide and hydrochloric acid:
CC12O+ H2O CO2 + 2HC1
The rate of phosgene hydrolysis is very rapid, with a half-time of less than 1 second (De Bruyn et
al., 1995). Phosgene also reacts with a wide variety of other nucleophiles, including primary and
secondary amines, hydroxy groups, and thiols (Schneider and Diller, 1991). For example,
phosgene reacts with the thiol group of glutathione (GSH), yielding S-chloro-carbonyl
glutathione, which in turn can either interact further with glutathione to form diglutathionyl
dithiocarbonate, or form glutathione disulfide and carbon monoxide (ILSI, 1997):
CC12O + GSH GSCOC1 + HC1
GSCOC1 + GSH GS-CO-SG + HC1
GSCOC1 + GSH GSSG+CO + HC1
Phosgene also undergoes attack by nucleophilic groups (-SH, -OH, -NH2) in cellular
macromolecules such as enzymes, proteins, or the polar heads of phospholipids, resulting in
formation of covalent adducts (Pohl et al., 1977, 1980, 1981; Pereira and Chang, 1981; Pereira et
al., 1984; Noort et al., 2000). Formation of these molecular adducts can interfere with molecular
function (e.g., loss of enzyme activity), which in turn may lead to loss of cellular function and
subsequent cell death (ILSI, 1997; WHO, 1998).
-------
o
CO
V)
O
6
o
I
o
£M
o
o
a
M
O O CO
I ,
o=c.
/ 1
(
* 1
CM M
W rn CO O fll ^
O ° CD 0 c O _
!->
•*! 0 '
^^ CM
o W 4- O
-*¥ to f»I
o3 ,^J ^
O -*n
¥
1
I
H
Q.
UJ
P
^
O
X
o
1— »• _
c
c
« ^ ,
3
I
,1
o
10
t
Q_
ON
V ^
^— O
0)
D)
"
Q.
c
1
c
1
i
UJ
>
1-
o
^^
Q
UJ
o
a.
i
c
i
c
o
$
_D
O)
N
T3
O
CO
CO
o
of
o
!c
"S
•D
05
o
D
T3
(U
CO
o _
38
o -^
Ifr
0)
= °
8 "
u P
ao
[a>
O)
co
TJ
ro
CO
O)
St ™
;,Q- o>
o E
omasi
(D
T3
ro
w
I
5
CO
o:
o
u.
w
I
O
ffl
tt
o
X
u
LL
O
I
O
J
o
m
UJ
2
£
O
0) 7? TJ
o S £
3 CD Q.
c •;= co
I
-------
Free radicals that are formed under conditions of low oxygen are also extremely reactive,
forming covalent adducts with microsomal enzymes and the fatty acid tails of phospholipids,
probably quite close to the site of free radical formation (cytochrome P450 in microsomal
membranes). This results in a general loss of microsomal enzyme activity, and can also result in
lipid peroxidation (ILSI, 1997; U.S. EPA, 1998c).
3.3.3. Relative Importance of Oxidative and Reductive Pathways
A priori, it might be expected that the oxidative pathway of chloroform metabolism
would predominate in vivo, because tissues of healthy animals are oxygenated. However, some
investigators have noted that the centrilobular region of the liver, where chloroform
hepatotoxicity is largely localized, is physiologically hypoxic, with oxygen partial pressures from
0.1% to 8% (U.S. EPA, 1998c; ILSI, 1997).
Nevertheless, two lines of evidence suggest that metabolism occurs mainly via the
oxidative pathway. First, reductive metabolism of chloroform is observed only in phenobarbital-
induced animals or in tissues prepared from them, with negligible reducing activity observed in
uninduced animals (ILSI, 1997). Second, in vitro studies using liver and kidney microsomes
from mice indicate that, even under relatively low (2.6%) oxygen partial pressure (approximately
average for the liver), more than 75% of the phospholipid binding was to the fatty acid heads.
This pattern of adduct formation on phospholipids is consistent with phosgene, not free radicals,
as the main reactive species, indicating metabolism was chiefly by the oxidative pathway (U.S.
EPA, 1998c; ILSI, 1997). Addition of glutathione to the incubation system completely negated
binding to liver microsomes, with only residual binding remaining in kidney microsomes (ILSI,
1997). This quenching by glutathione is expected for the products of oxidative but not reductive
metabolism. Taken together, these observations strongly support the conclusion that chloroform
metabolism in vivo occurs primarily via the oxidative pathway, except under special conditions
of high chloroform doses in preinduced animals (ILSI 1997, U.S. EPA 1998c).
3.4. EXCRETION
Excretion of chloroform occurs primarily via the lungs (U.S. EPA, 1998c). Results from
studies in humans indicate that approximately 90% of an oral dose of chloroform was exhaled
(either as chloroform or as carbon dioxide), with less than 0.01% of the dose excreted in the
urine (U.S. EPA, 1994d). In mice and rats, 45%-88% of an oral dose of chloroform was
excreted from the lungs either as chloroform or carbon dioxide, with l%-5% excreted in the
urine (U.S. EPA, 1998c).
No data are available regarding the bioaccumulation or retention of chloroform following
repeated exposure. However, because of the rapid excretion and metabolism of chloroform,
combined with low levels of chloroform detected in human postmortem tissue samples, marked
accumulation and retention of chloroform is not expected (U.S. EPA, 1994d).
3.5. PHYSIOLOGICALLY BASED PHARMACOKINETIC (PBPK) MODELS
The concentration of a chemical that reaches a target tissue following some external
exposure depends not only on the external dose administered to the organism (human or animal),
-------
but also on a number of physiological parameters that may differ significantly from organism to
organism. Likewise, the rate and extent of metabolism of the chemical to less toxic or more
toxic intermediates may also vary from tissue to tissue and from organism to organism.
Therefore, extrapolation of toxicological observations from dose to dose, from route to route, and
from organism to organism are all quite uncertain unless a detailed understanding exists
regarding the absorption, distribution, metabolism, and clearance of the chemical. Mathematical
models that describe the rate and extent of absorption, distribution, metabolism, and clearance as
a function of dose, time, route, and organism-specific physiological parameters are referred to as
physiologically based pharmacokinetic (PBPK) models.
Corley et al. (1990) developed a PBPK model for chloroform. In brief, the model
consists of a series of differential equations that describe the rate of chloroform entry into and
exiting from each of a series of body compartments, including (1) gastrointestinal tract, (2) lungs,
(3) arterial blood, (4) venous blood, (5) liver, (6) kidney, (7) other rapidly perfused tissues,
(8) slowly perfused tissues, and (9) fat. In general, the rate of input to each compartment is
described by the product of (a) the rate of blood flow to the compartment, (b) the concentration
of chloroform in arterial blood, and (c) the partition coefficient between blood and tissue.
Absorption of chloroform into the blood from the lungs or stomach is modeled by assuming first-
order absorption kinetics. Material absorbed from the stomach is assumed to flow via the portal
system directly to the liver (the "first-pass effect"), while material absorbed from the lungs enters
the arterial blood. Each tissue compartment is assumed to be well mixed, with venous blood
leaving the tissue being in equilibrium with the tissue. Metabolism of chloroform is assumed to
occur in both the liver and the kidney. The rate of metabolism is assumed to be saturable and is
described by Michaelis-Menten type equations. Chloroform metabolism is assumed to lead to
binding of a fraction of the total metabolites to cellular macromolecules, and the amount bound is
one indicator of the delivered dose. Binding of reactive metabolites to cell macromolecules is
also assumed to cause a loss of some of the metabolic capacity of the cell. This metabolic
capacity (enzyme level) is then resynthesized at a rate proportional to the amount of decrease
from the normal level. Based on a review of published physiological and biochemical data, as
well as several studies specifically designed to obtain model parameter estimates, Corley et al.
(1990) provided recommended values for each of the model inputs for three organisms (mouse,
rat, and human). These values are shown in Table 1. On the basis of these inputs, the model
predicted that the amount of chloroform metabolized per unit dose per kg of tissue (liver or
kidney) would be highest in the mouse, intermediate in the rat, and lowest in the human. This
difference between species is due to the lower rates of metabolism, ventilation, and cardiac
output in larger species compared to smaller species. If equal amounts of metabolite binding to
cellular molecules were assumed to be equitoxic to tissues, then the relative potency of
chloroform would be mice > rats > humans.
The model was extended by Reitz et al. (1990), who added equations describing the effect
of chloroform metabolism on cell killing in the liver. It was assumed that cells were subject to
risk of death when the rate of metabolism exceeded the ability of the cell to detoxify the
metabolic products, with the probability of any particular cell dying being characterized by a
normal distribution function. In addition, it was assumed that cell death did not occur instantly,
but depended on both the rate of metabolism and the time of exposure. Results from this model
-------
Table 1. Summary of PBPK parameters
Parameter
Body weight (kg)
Percentage of body
weight
Flows (L/hr)
Tissue blood flow (%
cardiac output)
Partition coefficients
Metabolic constants
Gastric absorption rate
constants
Tissue/compartment
—
Liver
Kidney
Fat
Rapidly perfused
Slowly perfused
Alveolar ventilation
Cardiac output
Liver
Kidney
Fat
Rapidly perfused
Slowly perfused
Blood/air
Liver/air
Kidney/air
Fat/air
Rapidly perfused/air
Slowly perfused/air
VmaxC (mg/kg/hr)
Km (mg/L)
kioss (L/mg)
kresvsn (1/hr)
A (kidney/liver)
fMMB in liver (1/hr)
fMMB in kidney (1/hr)
kas from corn oil (1/hr)
kas from water (1/hr)
Mouse
0.0285
5.86
1.70
6.00
3.30
74.1
2.01
2.01
25.0
25.0
2.0
29.0
19.0
21.3
19.1
11.0
242
19.1
13.0
22.8
0.352
5.72E-4
0.125
0.153
0.003
0.010
0.6
5.0
Rat
0.230
2.53
0.71
6.30
4.39
77.1
5.06
5.06
25.0
25.0
5.0
26.0
19.0
20.8
21.1
11.0
203
21.1
13.9
6.8
0.543
0
0
0.052
0.00104
0.0086
0.6
5.0
Human
70.0
3.14
0.44
23.1
3.27
61.1
347.9
347.9
25.0
25.0
5.0
26.0
19.0
7.43
17.0
11.0
280
17.0
12.0
15.7
0.448
0
0
0.033
0.00202
0.00931
0.6
5.0
All values are derived from Corley et al., 1990.
-------
predicted that the number of cells killed depended on the dose route, with higher toxicity via
gavage exposure than drinking water exposure. This supports the view that the hepatotoxicity of
chloroform (and hence the potential for carcinogenicity) is strongly dependent on rate of
metabolism, which in turn is dependent on dose rate.
The Corley model was adapted by Blancato and Chiu (1994) to include dermal exposure
from water while bathing or swimming. The EPA model was validated by comparing results
with those obtained by Corley et al. for identical input assumptions, and by comparing results for
the same model established in a separate simulation environment (SEVIUSOLVE). In both cases,
model results were nearly identical for all cases compared, indicating that the model is
mathematically valid. Blancato and Chiu (1994) applied the model to several human exposure
scenarios where data were available on the amount of chloroform in exhaled air. The model
predictions fit the data well, supporting the accuracy of the underlying model and the
pharmacokinetic input values.
Smith et al. (1995) also adapted the basic Corley et al. (1990) model to evaluate the
relative merits of various estimates of internal dose as predictors of rodent tumor bioassay data.
These workers found that dose-rate-dependent measures (maximal rate of metabolism and
percentage of hepatocytes killed per day) correlated well for the rodent liver bioassay data. In
contrast, none of the model dose parameters predicted the kidney bioassay data as well as dose
scaled to body surface area.
4. HAZARD IDENTIFICATION
4.1. STUDIES IN HUMANS
4.1.1. Inhalation Studies in the Workplace
A number of epidemiological studies have been performed to investigate the occurrence
of adverse effects in populations of workers exposed to chloroform vapors in the workplace. In
general, these studies must be interpreted cautiously, because data on actual chloroform exposure
are generally lacking, and most workplace studies involved exposures to other chemicals besides
chloroform.
Based on the limited data available, and subject to the cautions mentioned above, it
appears that long term exposure to concentrations of 100-1,000 mg/m3 (20-200 ppm) of
chloroform produce mainly neurological effects, with increased incidence of symptoms such as
fatigue, nausea, vomiting, lassitude, dry mouth, and anorexia (Phoon et al., 1983; Challen et al.,
1958; Li et al., 1993; Bomski et al., 1967). Some studies also observed effects on the liver,
including jaundice, increased serum enzyme levels, and increased liver size (Phoon et al., 1983;
Bomski et al., 1967). Available data are not adequate to define with confidence the inhalation
dose-response curve in humans for either neurological or hepatic effects, but data from Li et al.
(1993) suggest hepatic effects are not likely at exposure concentrations of 30 mg/m3 (6 ppm) or
lower, and essentially no effects were detected at concentrations of about 13 mg/m3 (2.6 ppm).
An association between chloroform exposure and increased risk of spontaneous abortion was
reported for workers in biomedical research laboratories (Wennborg et al., 2000), but no data on
-------
actual exposure levels were presented, and the workers were also known to be exposed to
numerous other laboratory solvents. No data were located on cancer incidence in workers
exposed to chloroform vapors.
4.1.2. Exposure to Chloroform in Drinking Water
There have been no studies of toxicity or cancer incidence in humans chronically exposed
to chloroform (alone) via drinking water. However, there have been a number of
epidemiological studies on cancer risk in humans exposed to chlorinated drinking water (e.g.,
Cantor et al., 1985; McGeehin et al., 1993; King and Marrett, 1996; Doyle et al., 1997; Freedman
et al., 1997; Cantor et al., 1998; Hildesheim et al., 1998). Chlorinated drinking water typically
contains chloroform, along with other trihalomethanes and a wide variety of other disinfection
byproducts (U.S. EPA, 1994d). It should be noted that humans exposed to chloroform in
drinking water are likely to be exposed both by direct ingestion and by inhalation of chloroform
gas released from water into indoor air.
Some of these epidemiological studies have detected a weak association between
exposure to chlorinated water and cancer (mainly bladder cancer). Based on the studies of
Cantor et al. (1985), McGeehin et al. (1993); King and Marrett (1996); Freedman et al. (1997),
and Cantor et al. (1998), EPA calculated that the population-attributable risk (the fraction of a
disease that could be eliminated if the exposure of concern were eliminated) for bladder cancer
ranged from 2% to 17% (U.S. EPA, 1998g). However, these calculations are based on a number
of assumptions, including the assumption that there is a cause-effect relationship between
exposure to chlorinated drinking water and increased risk of bladder cancer. This assumption is
subject to considerable uncertainty, especially because findings are not consistent within or
between studies. Evaluation of these studies by application of standard criteria for establishing
causality from epidemiological observations (strength of association, consistency of findings,
specificity of association, temporal sequence, dose-response relation, biological plausibility) has
led EPA to conclude that the current data are insufficient to establish a causal relationship
between exposure to chloroform in drinking water and increased risk of cancer (SAB, 2000; U.S.
EPA, 1998c; ATSDR, 1997; IPCS, 2000). Moreover, even if, in the future, the weight of
evidence does reach a point where a causal link is established between exposure to chlorinated
water and increased risk of bladder or other types of cancer, it could not be concluded from
epidemiological studies of this type that chloroform per se is carcinogenic in humans, because
chlorinated water contains numerous disinfection byproducts besides chloroform that are
potentially carcinogenic (U.S. EPA, 1994d, 1998c).
There have also been a number of epidemiological studies that have investigated the
association between human exposure to chloroform and other disinfection byproducts in
chlorinated water and the occurrence of adverse reproductive outcomes. Several such studies are
summarized below:
10
-------
Study
Kramer et al., 1992
Boveetal., 1995
Gallagher et al., 1998
Waller etal., 1998
Study type
Case-control
Cross-sectional
Retrospective
Prospective
Index of exposure
TTHM (chloroform)
TTHM
TTHM
TTHM (BDCM)
Associated effects
Intrauterine growth
retardation
Low birth weight
Small for gestational age
CNS defects
Oral cleft defects
Cardiac defects
Retarded fetal growth
Spontaneous abortion
As seen, statistically significant correlations between exposure to total trihalomethanes and one
or more adverse reproductive outcomes have been detected in several different types of
epidemiological study design. In one case (Kramer et al., 1992), there was a significant
relationship between chloroform levels and decreased intrauterine growth. In another case
(Waller et al., 1998), an association was noted between increased risk of spontaneous abortion
and bromodichloromethane (but not chloroform) levels. As noted earlier, although
epidemiological studies of this type are useful in evaluating whether chlorinated drinking water
can increase the risk of adverse reproductive effects in exposed populations, the studies are not
adequate to establish a causal link between ingestion of chloroform and the occurrence of adverse
reproductive effects in humans, because chlorinated drinking water contains many different
potentially toxic disinfection byproducts.
4.2. PRECHRONIC AND CHRONIC STUDIES AND CANCER BIOASSAYS IN
ANIMALS
A number of studies in animals have investigated the chronic toxicity and carcinogenic
potential of chloroform. This includes studies both by oral exposure and by inhalation exposure.
Presented below are summaries of the most important of these investigations.
4.2.1. Oral Studies
4.2.1.1. Eschenbrenner, AB; Miller, E. (1945) Induction ofhepatomas in mice by repeated
oral administration of chloroform, with observations on sex differences. J Natl
Cancer Inst 5:251-255.
Eschenbrenner and Miller (1945) exposed Strain A mice (five/sex/group) to chloroform
at dose levels of 0, 150, 300, 600, 1,200, or 2,400 mg/kg in olive oil by gavage,. The animals
were dosed every 4 days over a period of 120 days (a total of 30 doses) and were examined for
hepatomas 30 days after the last dose. No males administered doses of at least 600 mg/kg and
no females in the high-dose group survived the study. All deaths occurred 24 to 48 hours after
the first or second chloroform dose. All surviving females dosed with chloroform at 600 or
1,200 mg/kg developed hepatomas. Liver necrosis was observed in both sexes in the three
highest dose groups. Necrosis of hepatoma cells was not observed. The hepatomas did not
11
-------
appear invasive and no metastasis was found. Males in all treatment groups developed kidney
necrosis, whereas kidney necrosis was not apparent in any females. The severity of renal necrosis
was dose related.
4.2.1.2. National Cancer Institute (NCI). (1976) Report on carcinogenesis bioassay of
chloroform. Springfield, VA: NTIS PB-264018.
The carcinogenic potential of chloroform was evaluated by NCI (1976) in Osborne-
Mendel rats. Male rats were administered concentrations of 90 or 180 mg chloroform/kg/day in
corn oil, via oral gavage, 5 days/week for 78 weeks. Female rats were administered
concentrations of 125 or 250 mg/kg/day for 22 weeks, after which the doses were reduced to 90
or 180 mg/kg/day, with the average dose over the course of the study being 100 or 200
mg/kg/day. Three additional groups of animals served as matched, colony, and positive
controls. At week 111, all rats were sacrificed.
Survival rates and weight gains were decreased for rats in all chloroform treatment
groups. A statistically significant increase (24%) in the incidence of kidney epithelial tumors
was observed in male rats (12/50) in the high-dose group when compared with males in the
control group (0/98). A statistically significant increase in the incidence of thyroid tumors was
also observed in female rats, but this finding was not considered biologically significant (U.S.
EPA, 1994d).
NCI (1976) also evaluated the carcinogenic potential of chloroform using B6C3F1 mice.
The average dose levels for the study were 138 or 277 and 238 or 477 mg/kg/day for males and
females, respectively. All mice were sacrificed at weeks 92 or 93. Three additional groups of
animals served as matched (20/sex/group), colony (99 males and 98 females), and positive
(100/sex/group) controls.
Comparable survival rates and weight gains were observed between the treated and
control groups, except for the high-dose females. The incidence of hepatocellular carcinomas
was significantly increased in males and females in both the low- and high-dose groups when
compared to controls. Many of the male mice in the low-dose group that did not develop
hepatocellular carcinoma had nodular hyperplasia of the liver. The incidence of kidney epithelial
tumors was comparable between treatment and control groups.
4.2.1.3. Roe, FJC; Palmer, AK; Warden, AN; et al. (1979) Safety evaluation of toothpaste
containing chloroform: I. Long-term studies in mice. J Environ Pathol Toxicol
2:799-819.
Roe et al. (1979) reported three experiments in mice to evaluate the potential
carcinogenicity of chloroform. In three different studies, 10-week-old mice were administered
chloroform by gavage 6 days per week for 80 weeks, followed by a 13- to 24-week observation
period. The design of each study is summarized below:
12
-------
Study
I
II
III
Strain (gender)
ICI (male, female)
ICI (male)
C57BL, CBA, CF/1, ICI (male)
N
52/sex
52
52 per strain
Doses
17,60
60
60
There were no statistically significant differences in survival, body weight, or food
consumption between chloroform-treated and control groups in any of the experiments. In
experiment I, a slight increase in moderate to severe fatty degeneration of the liver was seen in
ICI mice given 60 mg, but not 17 mg, chloroform/kg/day. Kidney tumors were statistically
higher in high-dose male mice than in controls, while all other tumor incidences were
comparable to control. In experiment II, a decrease in liver and kidney weights was observed in
chloroform-treated male mice, and the incidence of kidney tumors was increased. In experiment
III, treatment with chloroform was associated with increased incidence of moderate to severe
kidney lesions in CBA and CF/1 mice. No increases in liver or kidney tumors were observed
except in ICI male mice.
4.2.1.4. Palmer, AK; Street, AE; Roe, FJC; et al (1979) Safety evaluation of toothpaste
containing chloroform: II. Long-term studies in rats. J Environ Pathol Toxicol
2:821-833.
Sprague-Dawley rats (50/sex/group) were administered concentrations of 0 or 60 mg
chloroform/kg/day in toothpaste by gavage, 6 days/week for 80 weeks. No significant
differences in mortality were observed between treated and control animals. A marginal
decrease in body weight gain (about 10%) was observed in both treated males and females when
compared to controls. A statistically significant decrease in relative liver weight was observed in
treated females. Histologic examination of the liver revealed only minor changes, with no severe
fatty infiltration, fibrosis, or marked bile duct abnormalities reported. The incidence of
moderate to severe glomerulonephritis was reported to be slightly increased in treated males.
4.2.1.5. Heywood, R; Sortwell, RJ; Noel, PRB; et al. (1979) Safety evaluation of toothpaste
containing chloroform: III. Long-term study in beagle dogs. J Environ Pathol
Toxicol 2:835-851.
Heywood et al. (1979) exposed groups of eight male and eight female beagle dogs to
doses of 15 or 30 mg chloroform/kg/day. The chemical was given orally in a toothpaste base in
gelatin capsules, 6 days/week for 7.5 years. This was followed by a 20- to 24-week recovery
period. A group of 16 male and 16 female dogs received toothpaste base without chloroform and
served as the vehicle control group. Eight dogs of each sex served as an untreated group and a
final group of 16 dogs (8/sex) received an alternative nonchloroform toothpaste. Four male dogs
(one each from the low- and high-dose chloroform groups, the vehicle control group, and the
untreated control group) and seven female dogs (four from the vehicle control group and three
from the untreated control group) died during the study. Results for serum glutamate pyruvate
transaminase (SGPT, now known as alanine aminotransferase or ALT) levels are shown in
Figure 2. Although there is substantial variability in individual measurements, SGPT levels
13
-------
160
140 -
120 -
100 -
cl 80
O Control
X30 mg/kg/day
415 mg/kg/day
X
X
o o
0
0 50 100 150 200 250
Weeks on Treatment
300
350
400
Data are from Heywood et al, 1979. SGPT = serum glutamate pyruvate transaminase.
Figure 2. SGPT levels in dogs exposed to chloroform for 7 years.
tended to be about 30%-50% higher in the low-dose group (15 mg/kg/day) than in control
animals. These increases were statistically significant for weeks 130-364. For the high-dose
group (30 mg/kg/day), the typical increase in SGPT was about twofold, and the differences were
statistically significant for the entire exposure duration (weeks 6-372). After 14 weeks of
recovery, SGPT levels remained significantly increased in the high-dose group but not in the
low-dose group, when compared with the controls. After 19 weeks of recovery, SGPT levels
were not significantly increased in either treated group when compared with the controls. The
authors concluded that the increases in SGPT levels were likely the result of minimal liver
damage. Serum alkaline phosphatase (SAP) and SGPT levels were also moderately increased
(not statistically significant) in the treated dogs at the end of the treatment period when compared
with the controls. Microscopic examinations were conducted on the major organs. The most
prominent microscopic effect observed in the liver was the presence of "fatty cysts," which were
described as aggregations of vacuolated histiocytes. The fatty cysts were observed in the control
and treated dogs, but were larger and more numerous (i.e., higher incidence of cysts rated as
14
-------
"moderate or marked," as opposed to "occasional or minimal") in the treated dogs at both doses
than in the control dogs. The prevalence of moderated or marked fatty cysts was 1/27 in control
animals, 9/15 in low-dose animals, and 13/15 in high-dose animals. Nodules of altered
hepatocytes were observed in both treated and control animals, and therefore were not considered
related to treatment. No other treatment-related nonneoplastic or neoplastic lesions were
reported for the liver, gall bladder, cardiovascular system, reproductive system, or urinary
system. A NOAEL was not identified in this study. However, a LOAEL of 15 mg/kg/day was
identified, based on elevated SGPT levels and increased incidence and severity of fatty cysts
(U.S. EPA, 1998c).
4.2.1.6. Jorgenson, TA; Rushbrook, CJ. (1980) Effects of chloroform in the drinking
water of rats and mice: ninety-day subacute toxicity study. United States
Environmental Protection Agency Publication No. EPA-600/1-80-030.
Seven groups of 6-week-old female B6C3F1 mice (30 mice/group) were given water
containing either 0, 200, 400, 600, 900, 1,800, or 2,700 ppm chloroform for 30-90 days.
Calculated dose levels were 0, 32, 64, 97, 145, 290, or 436 mg/kg/day based on reported water
intakes. At week 1, a significant decrease in body weight was observed in the 900, 1,800, and
2,700 ppm chloroform treatment groups; however, all body weights of the treated animals were
comparable to controls after week 1. On days 30, 60, and 90, ten animals from each treatment
group were sacrificed for gross and microscopic pathologic examination, as well as for
measurement of organ fatorgan weight ratios. A 160%-250% increase in liver fat was observed
in the high-dose group. Histological examination of the liver revealed mild centrilobular fatty
changes in the 1,800 and 2,700 ppm groups. On day 30, reversible fatty changes in the liver were
observed at doses as low as 400 ppm chloroform. Treatment-related atrophy of the spleen was
observed at the high dose. Based on the observation of mild effects of chloroform exposure via
the drinking water on liver and other tissues, the LOAEL in this study was 290 mg/kg/day, while
the NOAEL was 145 mg/kg/day (U.S. EPA 1994d).
4.2.1.7. Jorgenson, TA; Rushbrook, CJ; Jones, DCL. (1982) Dose-response study of
chloroform carcinogenesis in the mouse and rat: status report. Environ Health
Perspect 46:141-149.
This study was an interim report of a 2-year bioassay conducted by Jorgenson et al.
(1985) (see below). Male Osborne-Mendel rats and female B6C3F1 mice were exposed to
chloroform in drinking water (0, 200, 400, 900, or 1,800 mg/L) for 1-6 months. The time-
weighted average doses, based on measured water intake and body weights, were 0, 19, 38, 81, or
160 mg/kg in rats and 0, 34, 65, 130, or 263 mg/kg in mice. An additional group of matched
controls received the same water volume as the high-dose groups.
In male rats, some changes were observed in body weight and in some hematological and
serum biochemical parameters, but the authors judged these changes to be a secondary effect of
reduced water intake. Gross and microscopic pathology findings in the rats generally were slight
or mild in severity, were not dose related, and either appeared adaptive (occurred in rats
sacrificed after 30 or 60 days, but not in those sacrificed after 90 days) or were sporadic (by
15
-------
nature and/or incidence) and not considered treatment-related. This study identifies a NOAEL of
160 mg/kg/day in the male rat.
In mice, mortality within the first 3 weeks was significantly increased in the two highest
dose groups (130 and 263 mg/kg/day), but was comparable to controls after that time. Early
mortality and behavioral effects (e.g., lassitude, lack of vigor) were apparently related to reduced
water consumption. A significant increase in liver fat in mice was noted at doses of 65
mg/kg/day and higher at 3 months, and at doses of 130 and 263 mg/kg/day at 6 months. This
study identifies a NOAEL of 34 mg/kg/day and a LOAEL of 65-130 mg/kg/day in mice, based on
increased liver fat.
4.2.1.8. Jorgenson, TA; Meierhenry, EJ; Rushbrook, CJ; et al. (1985) Carcinogenicity of
chloroform in drinking water to male Osborne-Mendel rats and female B6C3F1
mice. Fundam Appl Toxicol 5:760- 769.
Jorgenson et al. (1985) exposed male Osborne-Mendel rats and female B6C3F1 mice to
chloroform in drinking water (0, 200, 400, 900, or 1,800 mg/L) for 104 weeks. Time-weighted
average doses, based on measured water intake and body weights, were 0, 19, 38, 81, or 160
mg/kg/day for rats and 0, 34, 65, 130, or 263 mg/kg/day for mice. An additional group of
animals that served as controls was limited to the same water intake as the high-dose groups.
The number of animals in the dose groups (from low to high) was 330, 150, 50, and 50 for rats
and 430, 150, 50, and 50 for mice.
In male rats, survival at 104 weeks was greater in exposed groups than in controls. In
female mice, survival was similar to controls following an initial decline in survival of mice that
refused to drink for the first week of the study.
A statistically significant dose-related increase in the incidence of kidney tumors (tubular
cell adenomas and adenocarcinomas) was observed in male rats in the high-dose group (160
mg/kg). A statistically significant increase in the incidence of lymphomas and leukemias and a
statistically significant decrease in the incidence of adrenal cortical adenomas, adrenal
pheochromocytomas, and thyroid c-cell adenomas was observed in male rats in the high-dose
group when compared with controls. However, study authors suggested that the incidence of
renal tumors was the only endpoint that was biologically significant with respect to chloroform
treatment (U.S. EPA, 1994d).
Chloroform in the drinking water did not increase the incidence of hepatocellular
carcinomas in female B6C3F1 mice. The combined incidence of hepatocellular adenomas and
carcinomas was 2% in the high-dose group compared with 6% in the control groups. The authors
speculated that the differences observed between this study and the NCI (1976) bioassay may be
related to differences in the mode of administration (in drinking water versus in corn oil by
gavage).
In reports from the original study (Jorgenson et al., 1982, 1985), histological findings
indicative of renal cytotoxicity were not reported. Recently, histological slides of rat kidney
from this study have been re-examined to assess whether evidence of renal cytotoxicity could be
16
-------
detected (ILSI, 1997; Hard and Wolf, 1999; Hard et al., 2000). Based on this reexamination, it
was found that animals exposed to average doses of 81 or 160 mg/kg/day of chloroform
displayed low-grade renal tubular injury with regeneration, mainly in the mid to deep cortex.
The changes included faint basophilia, cytoplasmic vacuolation, and simple hyperplasia in
proximal convoluted tubules. In some animals, single-cell necrosis, mitotic figures, and
karyomegaly were also observed. Hyperplasia was visualized as an increased number of nuclei
crowded together in tubule cross-sections. These changes were observable in the 160 mg/kg/day
dose group at 12, 18, and 24 months, and in the 81 mg/kg/day dose group at 18 and 24 months.
Cytotoxic changes were not seen in either of the lower dose groups (19 or 38 mg/kg/day). Based
on histological evidence of renal cytotoxicity in rats, this study identifies a LOAEL of 81
mg/kg/day.
4.2.1.9. Bull, RJ; Brown, JM; Mderhenry, EA; et al. (1986) Enhancement of the
hepatotoxicity of chloroform in B6C3F1 mice by corn oil: implications for
chloroform carcinogenesis. Environ Health Perspect 69:49-58.
The effect of the vehicle on the hepatotoxicity of chloroform was evaluated using male
and female B6C3F1 mice. Doses of 0, 60, 130, or 270 mg/kg/day in corn oil or in 2% emulphor
were administered via gavage for 90 days. Based on measurements of serum enzyme levels,
serum and tissue triglyceride levels, and histological examination of the livers, the authors
concluded that hepatotoxic effects were enhanced by the administration of chloroform via corn
oil versus chloroform administered in an aqueous suspension. The authors suggested that the
cause may be absorption kinetics or interaction between chloroform and the corn oil vehicle
(U.S. EPA, 1994d). A LOAEL of 270 mg/kg/day was identified for chloroform when
administered in corn oil, but 270 mg/kg/day was considered a NOAEL for chloroform when
administered in aqueous vehicle (U.S. EPA, 1994d).
4.2.1.10. Tumasonis, CF; McMartin, DN; Bush, B. (1987) Toxicity of chloroform and
bromodichloromethane when administered over a lifetime in rats. J Environ
Pathol Toxicol Oncol 7:55-64.
Male and female Wistar rats were administered chloroform in drinking water at
concentrations of 0 or 2,900 mg/L for 72 weeks. Concentrations of chloroform were then
reduced to 1,450 mg/L for an additional 113 weeks until all animals had died (approximately 185
weeks). The average dose for males and females was approximately 200 or 150 mg/kg/day,
respectively (U.S. EPA, 1994d). Exposed animals had a decrease in body-weight gain compared
to controls. Treated females (but not males) showed a statistically significant increase in the
incidence of hepatic neoplastic nodules, and both males and females had a statistically significant
increase in the incidence of hepatic adenofibrosis. It is unclear if the nodules and adenofibroses
were considered to be tumors (U.S. EPA, 1994d).
4.2.1.11. Voronin, VM; Litvirov, NN; Kazachkov, VI. (1987) Carcinogenicity of'chloroform
in the mouse. Vopr Onkol 33(8):81-85.
The potential carcinogenicity of chloroform was evaluated in mice following oral
administration via oil or water. When administered in oil, 250 mg chloroform/kg/day produced
17
-------
an increased incidence in tumors (tissue not specified), whereas there were no increases in the
incidence of tumors observed in mice treated with 15 mg/kg/day. No increases in tumor
incidence were observed in mice treated with up to 42 mg/kg/day via drinking water (U.S. EPA,
1994d).
4.2.1.12. DeAngelo, A. (1995) Evaluation of the ability of chloroform administered in the
drinking water to enhance renal carcinogenesis in male F344 rats (letter summary
from A. DeAngelo to N. Chiu, October 1995).
DeAngelo (1995) exposed male F-344 rats to chloroform in drinking water for 100
weeks. Exposure levels were 0, 900, or 1,800 ppm. Assuming ingestion of about 0.05 L/day of
water per kg body weight, this corresponds to doses of approximately 45 and 90 mg/kg/day.
Exposure began when the animals were 8-10 weeks of age. Interim sacrifices of groups of 6
animals were performed at 26, 52, and 78 weeks, and the final sacrifice at 100 weeks included 50
animals per group. At each time point, liver and kidney were examined for gross and
microscopic lesions.
In the liver, there were borderline significant (p = 0.05-0.10) increases in the prevalence
of hepatocellular proliferative lesions at 100 weeks. In addition, there was a statistically
significant increase (p < 0.05) in the multiplicity of adenomas and carcinomas in the group
exposed to 1,800 ppm, and a significant dose trend (p < 0.05) for hyperplastic nodules, neoplasia,
and total proliferative lesions.
Chloroform cone.
in water (ppm)
0
900
1,800
Hepatocellular proliferative lesions
Prevalence
5.6%
2.3%
20.5%
Multiplicity
0.06
0.02
0.28
With the exception of midzonal vacuolization (probably due to fat accumulation), there were no
hepatic histopathological lesions observed at any of the sacrifice periods other than those
normally associated with aging rats. In kidney, a wide variety of chronic nephropathies were
observed in both control and exposed animals. The incidence of these nephropathies was not
considered to be different than spontaneous background rates. No renal neoplasms were
observed in any of the chloroform-exposed groups.
4.2.2. Inhalation Studies
4.2.2.1. Mery, S; Larson, JL; Butterworth, BE; et al. (1994). Nasal toxicity of chloroform
in male F-344 rats and female B6C3F1 mice following a 1-week inhalation
exposure. Toxicol Appl Pharmacol 125:214-22 7.
Mery et al. (1994) exposed rats and mice to chloroform for 6 hours/day for 7 consecutive
days. Exposure concentrations ranged from 1 to 300 ppm. Examination of the nasal passages
18
-------
revealed that chloroform caused a complex set of responses in the ethmoid turbinates,
predominantly in rats. These lesions were most severe peripherally and generally spared the
tissue adjacent to the medial airways. The changes were characterized by atrophy of Bowman's
glands, new bone formation, and increased labeling index in periosteal cells. The only change
noted in the mouse was increased cell proliferation without osseous hyperplasia. The NOAEL
values for these responses ranged from 3-100 ppm, with histological and induced cell
proliferation being the most sensitive indices of effect.
4.2.2.2. Larson, JL; Templin, MV; Wolf, DC; et al (1996) A 90-day chloroform inhalation
study in female and male B6C3F1 mice: implications for cancer risk assessment.
FundamAppl Toxicol 30:118-137.
Templin, MV; Larson, JL; Butterworth, BE; et al. (1996a) A 90-day chloroform
inhalation study in F-344 rats: profile oftoxicity and relevance to cancer studies.
Fundam Appl Toxicol 32:109-125.
Templin, MV; Constan, AA; Wolf, DC; et al. (1998) Patterns of chloroform-
induced regenerative cell proliferation in BDF1 mice correlate with organ
specificity and dose-response of tumor formation. Carcinogenesis 19:187-193.
Larson et al. (1996) and Templin et al. (1996a, 1998) performed a series of prechronic
studies on the toxicity of inhaled chloroform in B6C3F1 mice and F344 rats. Animals were
exposed to concentrations of chloroform ranging from 2-300 ppm (10-1460 mg/m3) for 6 hours
per day, either 5 or 7 days per week, for up to 13 weeks (90 days). All animals were examined
for histological lesions of liver, kidney, and nasal epithelium. Some animals were administered
bromodeoxyuridine (BrdU) via osmotic pump prior to sacrifice in order to measure the labeling
index (LI).
The results of these studies are summarized in Table 2. Exposure to chloroform caused
histopathological lesions in liver, kidney, and nasal epithelium of both rats and mice. Lesions in
liver were characterized by scattered vacuolated hepatocytes and necrotic foci, sometimes with
inflammation, mainly in the centrilobular and midzonal regions. Renal lesions occurred
primarily in the epithelial cells of the proximal convoluted tubules in the cortex. Changes
included vacuolation, a basophilic appearance, tubule cell necrosis, and enlarged cell nuclei.
Nasal lesions were characterized as atrophy of olfactory epithelium, mainly in the ethmoid
portion of the nasal passage. In most cases, histological effects in liver and kidney were not
observed until exposure levels were about 30 ppm or higher. However, atrophy of the nasal
epithelium was observed in rats at the lowest exposure level tested (2 ppm). Histological
changes were generally accompanied by statistically significant increases in Labeling Index,
although not always at exactly the same exposure level. These increases in Labeling Index are
interpreted as evidence that the cytotoxic responses in these tissues triggers a regenerative
hyperplasia. Increased cell proliferation was not found in either sex of rats exposed to
chloroform for 6 weeks and held (without exposure) until week 13, suggesting that cell
proliferation is dependent on the presence of chloroform and represents a regenerative response
to cytotoxicity.
19
-------
Table 2. Summary of chloroform-induced cytotoxicity and cell proliferation via inhalation
Reference
Larson et al.,
1996
Templin et al.,
1996a
Templin et al.,
1998
Species
Mouse
B6C3F1
Rat
F344
Mouse
BDF1
Sex
Male
Female
Male
Female
Male
Female
Exposure
duration
(days)
90a
90b
90a
90b
90a
90b
90a
90b
90b
90b
Liver
Histopath.
NOAEL,
ppm
10
10
10
10
30
30
30
90
5
30
LOAEL,
pprn
30
90
30
90
90
90
90
300
30
90
LI
NOAEL,
pprn
30
10
30
10
90
90
90
90
30
30
LOAEL,
ppm
90
90
90
90
300
300
300
300
90
90
Kidney
Histo
NOAEL,
ppm
10
10
90
90
30
90
90
90
5
90
path.
LOAEL,
ppm
30
90
—
90
300
300
300
30
-
LI (Cortex)
NOAEL,
ppm
10
~
90
90
10
30
10
30
5
90
LOAEL,
ppm
30
10
—
30
90
30
90
30
-
Nasal epithelium
Histopath.
NOAEL,
ppm
90
90
90
90
—
~
~
~
NA
NA
LOAEL,
ppm
~
—
2
30
2
30
NA
NA
LI
NOAEL,
ppm
90
90
90
90
2
~
2
~
NA
NA
LOAEL
ppm
~
—
10
30
10
30
NA
NA
to
o
Exposure was 7 days/week.
bExposure was 5 days/week.
NOAEL = no-observed-adverse-effect level.
LOAEL = lowest-observed-adverse-effect level.
LI = labeling index.
-------
4.2.2.3. Nagano, K; Nishizawa, T; Yamamoto, S; et al. (1998) Inhalation carcinogenesis
studies of six halogenated hydrocarbons in rats and mice. In: Advances in the
prevention of occupational respiratory diseases. Chiyotani, K; Hosoda, Y; Aizawa,
Y; eds. Elsevier Science B. V.
Nagano et al. (1998) evaluated the chronic hepatotoxicity of chloroform in F344 rats and
BDF1 mice. This study has also been summarized in abstract form by Yamamoto et al. (1994).
Groups of male and female rats and mice were exposed to target chloroform vapor
concentrations of 0, 10, 30, or 90 ppm or 0, 5, 30, or 90 ppm, respectively, 6 hours/day,
5 days/week for 104 weeks. To avoid lethality in the high-dose groups, mice in the 30-ppm and
90-ppm groups were exposed to chloroform concentrations of 5 and 10 ppm for 2 weeks each
and then 30 ppm for 100 weeks or 5, 10, and 30 ppm for 2 weeks each and then 90 ppm for 98
weeks, respectively. The time-weighted average for the 30-ppm group was 29.1 ppm and for the
90-ppm group 85.7 ppm.
The authors reported that both male and female rats and mice showed necrosis and
metaplasia of the olfactory epithelium and goblet cell hyperplasia of the respiratory epithelium.
Ossification was observed in the nasal turbinate and in the nasal septum of rats and mice,
respectively, at the lowest exposure levels. Statistically significant increases in the incidence of
overall renal cell adenoma and renal cell carcinoma were observed in male mice in the 30 (7/50)
and 90 (12/48) ppm groups when compared with controls (0/50). The overall incidence rates of
renal cell carcinoma were statistically significantly increased in males in the 90-ppm group
(11/48) when compared with controls (0/50). There were no statistically significant changes in
tumor incidence for female mice or for male or female rats in any exposure group.
Templin et al. (1998) duplicated the exposure regimen in mice (including the
acclimatization period) in order to study whether the treatment caused cytotoxicity and
regenerative hyperplasia. These authors observed cytotoxicity and hyperplasia in the kidneys of
male mice exposed to 30 or 90 ppm throughout a 90-day exposure period. No renal lesions or
hyperplasia were observed in female mice. These observations are consistent with the hypothesis
that cytotoxicity and regenerative hyperplasia are key events in the neoplastic response to
chloroform.
4.3. REPRODUCTIVE/DEVELOPMENTAL STUDIES
4.3.1. Oral Studies
4.3.1.1. Thompson, DJ; Warner, SD; Robinson, VB. (1974) Teratology studies on orally
administered chloroform in the rat and rabbit. Toxicol Appl Pharmacol 29:
348-357.
A preliminary study was conducted to evaluate embryonic and fetal development of
Sprague-Dawley rats administered chloroform in corn oil at doses of 79, 126, 300, 316, or 516
mg chloroform/kg/day via oral gavage on days 6-15 of gestation. Alopecia, rough hair, and
eczema were observed in the dams in all dose groups. Significantly decreased food consumption
and body weights were noted in dams administered 126 mg/kg/day or greater. Fetotoxicity, acute
21
-------
toxic nephrosis, hepatitis, and maternal death occurred in animals administered 316 mg
chloroform/kg/day and higher.
In the main study, groups of pregnant rats (25/group) were administered 0, 20, 50, or 126
mg chloroform/kg/day via intubation on days 6-15 of gestation. Caesarean section was
performed 1 or 2 days prior to expected parturition and fetuses were removed and examined.
Maternal toxicity, including decreased weight gain and mild fatty changes in the liver, occurred
in dams administered 50 or 126 mg chloroform. A statistically significant increase in the
frequency of bilateral extra lumbar ribs and a statistically significant decrease in fetal weight
were observed in fetuses from the 126 mg/kg/day dose groups when compared with controls. For
the dams, a NOAEL of 20 mg/kg/day and a LOAEL of 50 mg/kg/day were identified in this
study. For the fetuses, a NOAEL for this study was 50 mg/kg/day and a LOAEL was 126
mg/kg/day.
Pregnant Dutch-belted rabbits were administered 0, 25, 63, 100, 159, 251, or 398 mg
chloroform/kg/day in corn oil on days 6-18 of gestation in a preliminary range-finding study.
Results showed decreased maternal survival (60%-100%) in dams administered 100 mg/kg/day
or greater. Anorexia, weight loss, diarrhea, abortion, and one maternal death were observed in
females administered 63 mg/kg/day. Dams administered 25 mg/kg/day showed signs of mild
diarrhea and intermittent anorexia (U.S. EPA, 1994d).
A main study was conducted in which 0, 20, 35, or 50 mg chloroform/kg/day via oral
intubation was administered to pregnant rabbits (15/group) on days 6-18 of gestation. Decreased
weight gain was reported in dams in the high-dose group. Hepatotoxicity was the cause of four
maternal deaths in the high-dose group. No microscopic treatment-related effects were reported
in the liver, kidney, or breast of the high-dose dams. A statistically significant decrease in body
weight was observed in fetuses from the 20 and 50 mg/kg/day groups when compared with
controls. Fetuses from the 20 and 35 mg/kg/day groups had a statistically significant increase in
the frequency of incompletely ossified skull bones when compared with controls. However, this
effect was not statistically significantly increased when the litter was used as the statistical unit of
comparison and in the absence of a dose-response (this effect was not observed in the high-dose
group). These findings were not considered evidence of teratogenicity or fetotoxicity by the
study authors. Therefore, a NOAEL of 35 mg/kg/day and a LOAEL of 50 mg/kg/day were
identified for maternal effects based on the fact that maternal toxicity was observed at doses
lower than the doses of chloroform that induced fetotoxicity (U.S. EPA, 1994d).
4.3.1.2. NTP. (1988) Chloroform reproduction and fertility assessment in CD-I mice when
administered by gavage. Report by Environmental Health Research and Testing,
Inc., Lexington, KY, to National Toxicology Program, NTP-89-018. NTIS PB89-
148639.
The reproduction and fertility of CD-I (ICR) BR outbred albino mice (20/sex/group)
administered chloroform in corn oil via gavage at concentrations of 6.6, 16, or 41 mg/kg/day, 7
days/week for 18 weeks was investigated. An additional group of animals (40/sex/group) served
as controls. The basis of dose selection was the death of one male animal administered 100
mg/kg/day for 13 days in a range-finding study. Additionally, Fx mice (20/sex/group) from the
22
-------
control and high-dose groups were administered the same concentrations of chloroform as their
parents from postnatal day 22 until they were sacrificed after the birth of the F2 generation.
Mating of the Ft generation occurred at 64-84 days of age.
No significant differences in reproductive parameters, such as fertility index, number of
litters per pair, litter size, proportion of live pups, proportion of male pups, or pup weight at
birth, occurred between treated and control groups. The Ft generation also had no adverse effects
on fertility or reproduction. However, all females in the Fx generation exposed to 41 mg/kg/day
showed increased liver weight and liver lesions characterized by degeneration of centrilobular
hepatocytes. Treated males of the Fx generation had statistically significantly increased
epididymal weights, when compared to controls. Sperm motility, sperm density, and percent
abnormal sperm were not altered by chloroform treatment in the Fx generation. However,
vacuolar degeneration of ductal epithelium in the cauda epididymis was observed in 8/20 treated
and 3/20 control Fx males. The F2 generation was not examined microscopically. Study authors
concluded that mild to moderate liver histopathology was observed at 41 mg chloroform/kg/day
in Fj females but not males, and that minimal epididymal histopathology was observed in Fx
males. A NOAEL could not be defined in this study because histopathology was not performed
on animals in the low- and mid-dose levels.
4.3.1.3. Ruddick, JA; Villeneuve, DC; Chu, I. (1983) A teratological assessment of four
trihalomethanes in the rat. J Environ Sci Health 18(3):333-349.
A study was conducted to determine the potential developmental toxicity of chloroform
following administration via oral gavage in rats. On gestational days 6 through 15, pregnant
dams (8 to 14 animals/dose group) were administered 0, 100, 200, or 400 mg chloroform/kg in
corn oil. On day 22 of gestation, dams were anesthetized with ether and their viscera, including
the uteri, were examined. The fetuses were removed, weighed, and examined for viability and
external malformations. Histological examination was performed on two fetuses from each dam.
Maternal endpoints evaluated included hematology (hemoglobin concentration, erythrocyte and
leucocyte counts, hematocrit, mean corpuscular hemoglobin concentration, and mean corpuscular
hemoglobin), clinical chemistry (alkaline phosphatase, sodium, total bilirubin, cholesterol,
glucose, potassium, inorganic phosphorus, calcium, uric acid, LDH, GOT, and total protein), and
gross examination of the organs.
A significant decrease in weight gain, hemoglobin levels, and hematocrit levels as well as
enlargement of the liver in dams occurred at all dose levels. A significant increase in serum
inorganic phosphorus, cholesterol levels, and kidney weights and a decrease in RBC count were
observed in dams in the high-dose group. Also in the high-dose group, a statistically significant
(19%) decrease in fetal body weight was observed when compared with the controls. There were
no fetal malformations upon gross examination; however, a dose-dependent increase in the
incidence of sternebra aberrations was observed in the 200 and 400 mg chloroform/kg/day
exposure groups. Deviations were also observed at the high dose. However, statistical analyses
were not performed on the observed variations.
23
-------
4.3.2. Inhalation Studies
4.3.2.1. Baeder, C; Hoffman, T. (1988) Initial submission: inhalation embryotoxicity study
of chloroform in Wistar rats (final report) with attachment and cover letter dated
02/21/92. Pharma Res Toxicol Pathol. Conducted for Occidental Chem Corp. U.S.
EPA/OTS Public Files, Document Number: 88-920001208.
The potential developmental toxicity of chloroform vapor was evaluated following
inhalation exposure in rats. On gestational days 7 to 16, groups of 20 pregnant Wistar rats were
exposed to 0, 30, 100, or 300 ppm (0, 146, 488, 1,464 mg/m3) chloroform via inhalation for 7
hours/day. On gestational day 21, dams were sacrificed and fetuses were removed by Caesarian
section, weighed, sexed, and measured for crown-rump length. Half of the fetuses were
examined for skeletal anomalies, while the other half were examined for organ anomalies.
Maternal endpoints evaluated included food consumption, body weight, clinical signs of toxicity,
selected organ weights (heart, liver, kidneys, and spleen), and reproductive viability (number of
live and dead fetuses, number of corpora lutea, embryonic resorption sites, and placentas).
A dose-related decrease in maternal food consumption with increasing chloroform
concentrations occurred throughout the gestational period. On gestational days 14, 17, and 21,
maternal body weight and body weight gain values (18%, 24%, and 29% at 30, 100, and 300
ppm, respectively) were also decreased in a concentration-related manner when compared to
controls. A significant decrease (6%) in mean fetal weights was observed for the high-
concentration group. At all exposure concentrations, an increase in the number of dead fetuses
(there were no live fetuses in 2 dams at 30 ppm, 3 dams at 100 ppm, and 8 dams at 300 ppm) and
a significant decrease in fetal crown-rump length was observed. Fetal skeletal development for
all treatment groups was comparable to controls. Based on maternal toxicity and fetal lethality,
the study authors identified a LOAEL of 30 ppm (146 mg/m3). This corresponds to a time-
weighted average concentration of 43 mg/m3. A NOAEL was not identified for this study.
4.3.2.2. Baeder, C; Hoffman, T. (1991) Initial submission—chloroform: supplementary
inhalation embryotoxicity study in Wistar rats (final report) with attachments and
cover letter dated 12/24/91. NTIS/OTS0535017. EPA/OTS Doc#8-920000566.
Study title: Chloroform: supplementary inhalation embryotoxicity study in Wistar
rats. September 12, 1991. Performed by Hoechst Aktiengesellschaff, Germany,
Sponsored by Hoechst AG and Dow Europe SA. Report No. 91.0902.
Baeder and Hoffmann (1991) exposed groups of 20 pregnant Wistar rats to 0, 3, 10, or 30
ppm chloroform via inhalation for 7 hours/day on gestational days 7 to 16. The actual delivered
concentrations of chloroform were 0, 3.1, 10.7, or 30.2 ppm (0, 15, 52.2, or 147 mg/m3). On
gestational day 21, dams were sacrificed and fetuses were removed by Caesarian section,
weighed, sexed, and measured for crown-rump length. Half of the fetuses were examined for
skeletal anomalies while the other half were examined for internal anomalies. Maternal
endpoints examined included food consumption, body weight, clinical signs of toxicity, selected
organ weights (heart, liver, kidneys, and spleen), and reproductive viability (number of live and
dead fetuses, resorptions, corpora lutea, and placentas). Maternal food consumption was
significantly decreased in all exposure groups, and maternal body weight was significantly
24
-------
decreased in the 10-and 30-ppm treated groups. A concentration-related decrease in overall body
weight gains for dams for all exposure groups was reported. At 30-ppm, significant increases in
maternal kidney weights and significant decreases in fetal body weights and crown-rump lengths
were observed. One dam exposed to 30-ppm chloroform via inhalation exhibited only empty
implantation sites (i.e., no fetuses were present). A statistically significant increase in the
incidence of fetuses with body weights <3 grams and in the incidence of fetuses with slight or no
ossification of individual skull bones was observed in the 30-ppm exposed group when compared
with controls. The incidence of fetuses with body weights <3 grams was increased in a dose-
related fashion (3.2%, 14.2%, 24%, and 26.9% at 0, 3, 10, and 30-ppm, respectively); this trend
did not appear to be due to variations in litter size. However, when the litter was used as the
statistical unit of comparison, only litters from the high-concentration group had a significant
number of fetuses weighing 3 grams or less. A significant increase in the incidence of fetuses
with ossification of less than two caudal vertebral centers was observed at all concentrations. A
dose response was observed for the incidence of litters with this effect; however, the effect was
not statistically significant. Finally, all exposure groups exhibited a significant increase in the
incidence of both litters and fetuses with nonossified or weakly ossified sternebrae; however,
there was no statistically significant concentration-related trend for this effect. Based on
decreased body weight gain in dams, and slight retardation in growth of fetuses, a NOAEL of 3
ppm and a LOAEL of 10 ppm were identified. Even though there were increases in low-weight
fetuses at the two lowest concentrations, this effect was not considered adverse. Therefore, the
NOAEL for developmental effects in this study was 10.7 ppm (52.2 mg/m3) based on the weight
of evidence of the data, including comparison to historical controls and the higher concentration
study.
4.3.2.3. Schwetz, BA; Leong, BJK; Gehring, PJ. (1974) Embryo- andfetotoxicity of
inhaled chloroform in rats. Toxicol Appl Pharmacol 28:442-451.
Four groups of 68, 22, 23, and 3 pregnant Sprague-Dawley rats were exposed to either 0,
30, 100, or 300 ppm chloroform, respectively, via inhalation for 7 hours/day from gestational
days 6 to 15. Because in an earlier experiment marked anorexia was observed in dams exposed
to 300 ppm chloroform, an additional control group (starved) that was allowed only 3.7 grams of
food per day was also used. Actual delivered concentrations of chloroform were 0, 30, 95, or
291 ppm (0, 146, 464, or 1,420 mg/m3). The low percent pregnancy observed at the high-
concentration group was not considered to be treatment-related because of the timing of
exposure; however, the use of such a small number of animals in the 300-ppm group decreased
the statistical sensitivity of any adverse effects observed in this group. On gestational day 21,
dams were sacrificed and fetuses were removed by Caesarian section, weighed, measured, and
sexed. Half of the fetuses were examined for skeletal anomalies while the other half were
examined for organ anomalies. A concentration-related decrease in body weight gain and food
consumption was observed in dams of all exposure groups. A significant increase in relative
liver weights in dams exposed to 100- and 300-ppm chloroform was observed at study
termination, with a significant decrease in absolute liver weight reported in dams exposed to 300
ppm chloroform. In the high-concentration group, 61% of the implantations were resorbed
(statistically significant). This high resorption rate was not observed in the "starved" control
group; therefore, weight loss cannot account for the observed effect. Fetal body weights (40%)
and fetal crown-rump lengths were significantly decreased at 300 ppm. Fetal crown-rump
25
-------
lengths were significantly decreased in the 30- and 300-ppm groups by 2% and 15%,
respectively. At 100 ppm chloroform, the frequencies of litters with acaudia or imperforate anus
were significantly increased when compared with the controls. All exposure groups exhibited an
increase in the frequency of litters with delayed ossification. Also, there were statistically
significant increases in wavy ribs at 30 ppm and in missing ribs and subcutaneous edema at 100
ppm. Fetal malformations were not observed at the high-dose group; however, there were only
three litters at this concentration. The study authors concluded that exposure to 100 and 300 ppm
chloroform via inhalation was embryotoxic and fetotoxic, with embryo death a significant effect
at 300 ppm.
4.3.2.4. Murray, FJ; Schwetz, BA; McBride, JG; et al (1979) Toxicity of inhaled
chloroform in pregnant mice and their offspring. Toxicol Appl Pharmacol
50:515-522.
The potential developmental toxicity of chloroform vapor was investigated following
inhalation exposure in mice. Groups of 34 to 40 pregnant CF-1 mice were exposed to either 0 or
100 ppm chloroform (0 or 490 mg/m3) 7 hours/day on gestational days 1-7, 6-15, or 8-15, and
sacrificed on gestational day 18. Chloroform exposure at 100 ppm was teratogenic in mice
exposed on gestational days 8-15, and fetotoxic in mice exposed on gestational days 1-7 or
6-15. A significant decrease in the ability of mice to maintain pregnancy was observed in the
group exposed on gestational days 1-7 or 6-15, and there was a slight, but not statistically
significant, decrease in pregnancies in the group exposed on gestational days 8-15. Fetal weight
and length were significantly decreased in the groups exposed on gestational days 1-7 and 8-15,
but not 6-15. A significant increase in the incidence of litters with cleft palate and delayed
ossification of sternebrae was observed in the gestational day 8-15 exposed group or all exposure
groups, respectively. Cleft palate was not observed in the gestational day 1-7 or the 6-15
exposed groups and was mostly in those fetuses with retarded growth. Delayed ossification of
skull bones was significantly increased in all exposure groups. A significant increase in the
incidence of delayed ossification of the sternebrae was reported in the group exposed on
gestational days 1-7 and 8-15, but not days 6-15. No other malformations were significantly
increased in any chloroform treatment group. The study authors suggested that the lack of
malformations in the gestational day 6-15 exposed group may have been due to the lethality of
chloroform on the early embryo. Liver weights and SGPT activity were increased in dams
exposed to chloroform.
4.4. OTHER STUDIES
4.4.1. Other Effects
As summarized above, the available data indicate that the characteristic effects of
chloroform exposure include cytotoxicity in liver, kidney, and nasal epithelium, with
neurological effects following relatively high-dose inhalation exposures. No studies were
located that identified toxic effects on other tissues such as the immune system.
26
-------
4.4.2. Mutagenicity
4.4.2.1. Overview
A number of studies have been performed to evaluate the mutagenicity of chloroform. In
reviewing and evaluating these studies, it is important to recognize the following potential
concerns regarding study design: (1) because chloroform is relatively volatile, test systems not
designed to prevent chloroform escape to the air may yield unreliable results; (2) because it is the
metabolites of chloroform (e.g., phosgene, dichloromethyl free radical) rather than the parent
compound that are most likely to react with DNA, studies in which appropriate P450-based
metabolic activation systems are absent are also likely to be unreliable; (3) because of the
relatively high reactivity of the metabolites, tests using exogenous activation systems (i.e., the
metabolites are formed outside of the test organism) are likely to be less relevant than tests using
endogenous activation systems (i.e., metabolites are formed inside the test organism); (4) studies
(especially older studies) that used ethanol as a solvent or preservative for chloroform may be
confounded by formation of ethyl or diethyl carbonate, which are potent alkylating agents; and
(5) tests performed in vitro (e.g., clastogenicity tests) or in vivo under highly toxic doses can
produce positive results as a secondary consequence of severe cytotoxicity, resulting from
lysosomal or other releases (Brusick, 1986). Also, chloroform-induced cycles of cytotoxicity and
cell proliferation in vivo or in vitro could cause the expression of preexisting genetic damage in
cells that, under normal conditions, have low mitotic indices. Therefore, one should exercise
caution in interpreting the mutagenicity test results.
4.4.2.2. Evaluation of Available Data
4.4.2.2.1. In Vitro Studies. Data on the genotoxic potential of chloroform in subcellular systems
are limited, but two investigators reported DNA binding in studies with calf thymus DNA in the
presence of exogenous activation (DiRenzo et al., 1982, Colacci et al., 1991). The study by
DiRenzo et al. (1982) utilized ethanol as a solvent, suggesting that ethyl carbonate formation
might be a problem. In the study by Colacci et al. (1991) addition of SKF-525A inhibited DNA
binding, suggesting that binding was mediated by a cytochrome P-450 pathway, as would be
expected for chloroform. In interpreting these studies, it is important to remember that cell-free
systems may not always be a good model for intact cellular processes.
Gene mutation studies in Salmonella typhimurium and E. coli (Ames assay), including
tests done under conditions designed to reduce evaporation, are mostly negative, with or without
activation with microsomes from liver or kidney of rats or mice (Rapson et al., 1980; San
Agustin and Lim-Sylianco, 1978; Van Abbe et al., 1982; Uehleke et al., 1977; Gocke et al., 1981;
Roland-Arjona et al., 1991; Le Curieux et al., 1995; Kirkland et al., 1981; Simmon et al., 1977).
However, four studies have showed positive results in bacteria. Varma et al. (1988) reported that
chloroform caused mutagenicity in five strains of S. typhimurium, but the response was noted
only at the lowest dose tested, and all higher doses were no different from control. This unusual
pattern casts some doubt on these results. San Agustin and Lim-Sylianco (1997) reported that
chloroform caused DNA damage in Bacillus subtilis, and Wecher and Scher (1982) reported that
chloroform caused mutations in Photobacteriumphosphoreum. However, neither study reported
the exposure concentrations that caused these effects, so the relevance of these reports is
27
-------
uncertain. In addition, the studies by Varma et al. (1988) and Wecher and Scher (1982) each
used ethanol as a diluent, raising the possibility that the positive effect might be related to ethyl
carbonate formation rather than to chloroform. The majority of results reported for S.
typhimurium and E. coli exposed to the vapor phase were also negative (Van Abbe et al., 1982;
Pegram et al., 1997; Simmon, 1977; Sasaki et al., 1998). Pegram et al. (1997) reported that
chloroform was weakly positive at vapor concentrations greater than 19,200 ppm (about 770
mg/L in the aqueous phase). Employing physiologically based pharmacokinetic models, the
authors estimated the oral doses needed to produce the effect would exceed 2,000 mg/kg
(approximately twice the LD50).
Tests of genotoxicity are also mainly negative in fungi (Gualandi, 1984; Mehta and von
Bortsel, 1981; Kassinova et al., 1981; Jagannath et al., 1981). However, chloroform was shown
to induce intrachromosomal recombination in Saccharomyces cerevisiae at concentrations of
6,400 mg/L (Callen et al., 1980) or 750 mg/L (Brennan and Schiestl, 1998). In the Brennan and
Schiestl study, addition of 7V-acetylcysteine reduced chloroform-induced toxicity and
recombination, suggesting a free radical may have been involved. Chromosome malsegregation
was also reported in Aspergillus nidulcms (Crebelli et al., 1988), but only at concentrations above
1,600 mg/L. In all three of these positive studies, doses that caused positive results also caused
cell death, indicating that exposures were directly toxic to the test cells.
Studies in intact mammalian cells are mainly negative (Larson et al., 1994a; Perocco and
Prodi, 1981; Butterworth et al., 1989; Kirkland et al., 1981; White et al., 1979; Sturrock, 1977),
although positive results have been reported in a few systems. Increased sister chromatid
exchange was reported in human lymphocytes at a concentration of about 1,200 mg/L without
exogenous activation (Morimoto and Koizumi, 1983), and at a lower concentration (12 mg/L)
with exogenous activation (Sobti, 1984). In the study by Sobti, the increase was quite small (less
than 50%), and there was an increase in the number of cells that did not exclude dye. This
suggests that the exposure levels causing the mutagenic effect may have been directly toxic to the
cells. In addition, ethanol was used as a dose vehicle. Mitchell et al. (1988) did not detect an
increase in mutation in mouse lymphoma cells at an exposure level of 2,100 mg/L in the absence
of exogenous activation, but did detect an effect at a concentration of 59 mg/L with exogenous
activation.
4.4.2.2.2. In vivo studies. A number of different endpoints of chloroform genotoxicity have
been measured in intact animals exposed to chloroform either orally or by inhalation. In studies
of DNA binding in liver and kidney of mice and rats, negative results have been reported at doses
of 742 mg/kg, 119 mg/kg, and 48 mg/kg (Diaz Gomez and Castro, 1980; Reitz et al., 1982;
Pereira et al., 1982). However, positive results have been reported at doses as low as 2.9 mg/kg
(Colacci et al., 1991). In the study by Colacci et al (1991), no significant difference in binding
was noted between multiple tissue (liver, kidney, lung, and stomach), and there was no increase
in binding with phenobarital pretreatment. This suggests the binding may not have been related
to chloroform metabolism.
Studies based on signs of DNA damage or repair have been uniformly negative (Larson et
al., 1994a; Potter et al., 1996; Reitz et al., 1982; Mirsalis et al., 1982). However, studies based
on various signs of chromosomal abnormalities have been mixed, with some studies reporting
28
-------
negative findings at doses of 371 mg/kg and 800 mg/kg (Shelby and Witt, 1995; Topham, 1980),
while other studies report positive results at doses as low as 1.2 mg/kg (Fujie et al., 1990).
However, the positive result at low dose in the study by Fujie et al. (1990) was observed
following intraperitoneal exposure, and positive results following oral exposure were not
observed until dose levels of 119 mg/kg. Morimoto and Koizumi (1983) observed an increase in
the frequency of sister chromatid exchange in bone marrow cells at a dose of 50 mg/kg/day, but
at 200 mg/kg/day, all of the mice died. As discussed before, mutagenicity results observed
following highly toxic doses may have been confounded by cytotoxic responses and should be
viewed as being of uncertain relevance.
Several studies have reported negative findings for the micronucleus test in rats and mice
(Gocke et al., 1981; Salamone et al., 1981; Le Curieux, 1995), but several other studies have
detected a positive result, mainly at exposure levels of 400-600 mg/kg (San Agustin and Lim-
Sylianco, 1982; Robbiano etal., 1998; Sasaki etal., 1998; Shelby and Witt, 1995). This suggests
that chloroform may be clastogenic, but it is important to note that these doses are well above the
level that causes cytotoxicity in liver and kidney in most oral exposure studies in rodents.
Butterworth et al. (1998) did not detect an increase in mutation frequency in male mice
exposed by inhalation at an exposure level of 90 ppm, even though this exposure did cause an
increase in tumors in the study by Nagano et al. (1998). Increased incidence of spermhead
abnormalities was reported in mice exposed at 400 ppm (Land et al. 1981), but were not
observed in mice exposed to 371 mg/kg intraperitoneally (Topham 1980).
In Drosophila melanogaster larvae exposed to chloroform vapor, gene mutation (Gocke
et al. 1981) and mitotic recombination tests (Vogel and Nivard 1993) were both negative.
Grasshopper embryos (Melanoplus sanguinipes) did not display mitotic arrest at vapor
concentrations of 30,000 ppm, but an effect was seen at 150,000 ppm (Liang et al. 1983). San
Agustin and Lim-Syllianco (1981) reported a single positive and negative result for host-
mediated mutagenicity in Salmonella typhimurium, but exposure levels were not reported in
either case.
4.4.2.3. Reviews by Other Groups
Data on the mutagenicity of chloroform have recently been reviewed and evaluated by
several groups, including the International Commission for Protection against Environmental
Mutagens and Carcinogens (ICPEMC), ILSI (1997), and WHO (1998). The findings of these
review efforts are summarized in the following paragraphs.
4.4.2.3.1. ICPEMC. The ICPEMC has developed a comprehensive, quantitative weight-of-
evidence approach for assessing genotoxic potential and has used this approach to evaluate more
than 100 chemicals with large genetic toxicity databases (Lohman et al., 1992). In this approach,
scores are developed for relative DNA reactivity. For a particular chemical, the maximum
possible score is 100 and the minimum possible score is -100. The highest actual score obtained
using this approach was 49.7 (triazaquone) and the lowest score was -27.7 (ethanol). When this
approach was applied to chloroform, the score based on the results of more than 40 studies was
29
-------
-14.33. Thus, ICPEMC concluded that the weight of evidence indicates that chloroform should
be classified as nongenotoxic (Brusick et al., 1992; Lohman et al., 1992).
4.4.2.3.2. ILSI. ILSI (1997) performed a review of the available data on the mutagenicity of
chloroform. The committee noted that phosgene is highly reactive and might be expected to
have the capacity to interact directly with DNA, but that phosgene has not been tested in any
standard mutagenicity test system. The committee also noted that, because of its high reactivity,
phosgene formed in the cytosol following chloroform metabolism would likely react with cellular
components prior to reaching the cell nucleus, and concluded that direct effects on DNA would
be unlikely. Based on their review of the available data, the ILSI committee (ILSI, 1997)
concluded that no subset of observations points unequivocally to a specific genotoxic mode of
action associated with chloroform, and that the preponderance of the evidence indicates that
chloroform is not strongly mutagenic. Based on this, the committee concluded that chloroform
would not be expected to produce rodent tumors via a genotoxic mechanism.
4.4.2.3.3. WHO. WHO (1998) noted that studies on the mutagenicity of chloroform must be
considered in light of the fact that (1) chloroform is volatile, so tests that do not prevent
volatilization are unreliable, and (2) most chloroform contains ethanol, which may react with
phosgene generated from chloroform metabolism to yield ethyl or diethyl carbamates (potentially
causing false positive results). The WHO committee noted that largely negative results have
been obtained in Salmonella typhimurium and Escherichia coli (with and without activation), in
gene mutation tests in CHO cells and human lymphocytes, in mouse micronucleus tests, and in
tests of unscheduled DNA synthesis both in vitro and in vivo. Given the large number of
sensitive assays that have been used to investigate the genotoxicity of chloroform, the committee
considered it noteworthy that the positive responses were so few, and that the positive results
were randomly distributed among the various assays. Taken together, WHO (1998) concluded
that the weight of evidence indicates that neither chloroform nor its metabolites appear to interact
directly with DNA or possess genotoxic activity.
4.4.2.4. Overall Weight-of-Evidence Conclusion on Mutagenicity
In summary, the results of the mutagenicity assays that have been conducted with
chloroform are mixed. By number, the majority of tests are negative, and many of the positive
studies have been conducted under high exposure conditions that resulted in severe cytotoxicity.
As expressed in SAB (2000):
Genotoxicity endpoints have to be interpreted cautiously when used as evidence for
potential carcinogenicity. In vitro clastogenicity can be a product of severe cytotoxicity
resulting from lysosomal or other releases (Brusick, 1986). This may be important with
substances such as chloroform, where there is evidence of cytotoxicity and cell
proliferation in target tissues. Also, cycles of cytotoxicity and cell proliferation could
cause the expression of preexisting genetic damage in target tissues which, under normal
conditions, have low mitotic indices.
Consequently, the relevance of many of the positive studies is questionable. Therefore, based on
the preponderance of negative findings and the uncertain relevance of the positive findings, EPA
30
-------
concludes that the weight of evidence indicates that even though a role for mutagenicity cannot
be excluded with certainty, chloroform is not a strong mutagen and that neither chloroform nor
its metabolites readily bind to DNA. Based on these results and the results of studies that
evaluated other endpoints of DNA reactivity, it seems likely that chloroform does not produce
carcinogenic effects primarily by a specific mutagenic mode of action.
4.4.3. Studies Related to Mode of Action
The precise mode of action by which chloroform produces toxic effects is not yet certain,
but it is evident that metabolism of chloroform to toxic metabolites plays a critical role (U.S.
EPA, 1994d). Representative studies that provide information on the role of metabolism in
chloroform-induced toxicity and on the mechanism of metabolism-induced toxicity are
summarized below.
4.4.3.1. Studies That Demonstrate That Metabolism is Required for Toxicity
A large number of studies support the conclusion that metabolism of chloroform is
required for toxicity. Brown et al. (1974) reported that pretreatment of rats with phenobarbital (a
cytochrome P-450 inducer) resulted in increased hepatic toxicity following chloroform exposure.
Similarly, Gopinath and Ford (1975) indicated that chloroform hepatotoxicity in rats was
increased by phenobarbitone, phenylbutazone, and chlorpromazine, all inducers of microsomal
enzymes. Conversely, inhibitors of microsomal enzymes, such as SKF-525A, sodium diethyl-
dithiocarbamate, and carbon disulfide, decreased the hepatic toxicity of chloroform. Constan et
al. (1999) showed that 1-aminobenztriazole, which is a general cytochrome P450 inhibitor,
prevented chloroform-induced toxicity in liver and kidney of mice following inhalation exposure.
ILSI (1997) summarized several studies to correlate the degree of hepatic metabolism
with toxicity. Pohl and Krishna (1978) indicate that metabolism is essential for chloroform to
induce liver toxicity in rats and mice. Smith and Hook (1983) evaluated the nephrotoxicity of
chloroform in kidney slices. At equimolar concentrations, ^-chloroform, which is readily
metabolized, induced nephrotoxicity more readily than 2H-chloroform, which is not as readily
metabolized as ^-chloroform.
Further evidence of the role of metabolism is derived from the finding that variations in
toxicity between tissues, genders, and species generally correlate with differences in metabolic
rate. For example, male mice are more sensitive to chloroform-induced renal toxicity than
female mice, and this difference in toxicity is paralleled in a difference in metabolism in
proximal tubular cells (Ilett et al., 1973). Renal cytochrome levels in mice are increased by
testosterone (Mohla et al., 1988, Henderson et al., 1989; Hong et al., 1989), and male mice are
more sensitive to chloroform-induced renal toxicity than are females. Female mice treated with
testosterone have increased renal toxicity along with increased covalent binding of chloroform
metabolites (Taylor et al., 1974; Smith et al., 1979; Pohle et al., 1984). Conversely, male mice
that were castrated had lower levels of chloroform-derived radioactivity accumulated in the
kidneys (Eschenbrenner and Miller, 1945; Culliford and Hewitt, 1957; Taylor et al., 1974; Smith
etal., 1984).
31
-------
4.4.3.2. Identification of Specific Enzymes Responsible for Metabolism
Nakajima et al. (1995a,b) evaluated the magnitude and localization of liver tissue injury
in mice pretreated with chemicals known to induce specific P450 enzymes. These chemicals
were n-hexane, an inducer of CYP2E1, phenobarbital, an inducer of CYP2B1/2, and 2-hexanone,
an inducer of both enzymes. Hepatocyte necrosis was associated more with CYP2B1/1 induction
whereas ballooning of cells was observed more frequently with CYP2E1 induction. The results
of the histologic examinations indicated that liver damage was associated more with CYP2E1
induction, with the damage localized primarily to the centrilobular regions, than with CYP2B1/2
induction, where damage was more generalized. At low doses, chloroform is metabolized more
extensively by CYP2E1 and activity of this enzyme correlates with tissue damage (TLSI, 1997).
Most recently, Constan et al. (1999) compared the toxicity of chloroform in three strains
of mice: B6C3F1, Sv/129 wild type, and Sv/129 CYP2E1 knockout mice. Exposure to 90 ppm
chloroform for 6 hrs/day for 4 days produced clear hepatotoxicity and renal toxicity
(histopathology, increased labeling index) in the B6C3F1 mice and the Sv/129 wild type, but not
in the Sv/129 CYP2E1 knockout mice. The authors concluded that metabolism of chloroform by
CYP2E1 was obligatory for toxicity, at least at the dose tested.
4.4.3.3. Role of Covalent Binding
The precise mode of action by which chloroform metabolism leads to cell toxicity is not
known with certainty, but covalent binding of phosgene with key cellular molecules is considered
to be a likely pathway. Pohl et al. (1980) reported that the level of covalent binding correlated
directly with injury to the liver tissue and concluded that phosgene was the metabolite
responsible for the covalent binding to liver macromolecules. Brown et al. (1974) reported that
pretreatment of rats with phenobarbital (a cytochrome P-450 inducer) resulted in increased
formation of covalent adducts and increased hepatic toxicity following chloroform exposure.
Studies by Ilett et al. (1973) and Tyson et al. (1983) also show that covalent binding to proteins
in rats and mice is more prevalent in areas of necrosis than in areas where tissue damage is not
severe. DBA mice have a metabolism rate twice as fast as that of C57BL mice (Pohl et al., 1984)
and also much greater covalent binding to renal microsomes (Clemens et al., 1979). The results
of in vitro studies also indicate that metabolism is necessary for covalent binding to
macromolecules (Cresteil et al., 1979).
4.4.3.4. Role of Glutathione
Reaction of chloroform metabolites (phosgene) with glutathione is a probable
detoxification pathway. Acute chloroform toxicity is associated with glutathione depletion
(Brown et al., 1974; Stevens and Anders, 1981), and it has been reported that glutathione levels
decrease in a dose-dependent manner prior to microscopic evidence of liver pathology (Brown et
al., 1974; Docks and Krishna, 1976). Glutathione depletion was reported in chloroform-exposed
mice pretreated with phenobarbital; however, this effect was not observed in mice that received
chloroform alone (Brown et al., 1974). The results of in vitro studies indicate that glutathione
inhibits covalent binding in liver cells (Cresteil et al., 1979; Sipes et al., 1977; Smith and Hook,
1984). After glutathione depletion, continued chloroform exposures resulted in increased
32
-------
covalent binding and lipid peroxidation (Brown et al., 1974). Glutathione depletion has been
observed primarily following acute high exposures or in phenobarbital-treated rats (ILSI, 1997),
and not in animals exposed to lower doses over a longer period of time (Munson et al., 1982).
4.4.3.5. Role of Dose Vehicle
The results of some animal studies have suggested that the vehicle used to administer
chloroform may affect the toxicity (Bull et al., 1986; Jorgenson et al., 1985; Lilly, 1992; Larson
et al., 1994b, 1995a). Lilly (1992) reported that hepatic and renal toxicity was greater in rats that
received chloroform in a corn oil vehicle when compared with the toxicity observed in rats that
received chloroform in an aqueous vehicle. Similar results were reported by Bull et al. (1986), in
that hepatotoxicity was greater in rats that received chloroform via corn oil for 90 days, when
compared with rats that received chloroform in water. Larson et al. (1994b, 1995a) indicated that
following the administration of chloroform via corn oil gavage, there were dose-related increases
in centrilobular necrosis and hepatic cell proliferation in female B6C3F1 mice and male F344
rats. These effects were not observed in rats that received chloroform via drinking water. The
authors suggested that these differences may have been due to the delivery of a higher dose to the
liver following a single gavage dose, when compared to the delivery of smaller doses over a
prolonged period as would be the case with administration via drinking water.
4.4.3.6. Studies on Initiation-Promotion of Cancer Effects
ILSI (1997) reviewed studies that have been conducted to evaluate the potential for
chloroform to promote tumor formation when administered in initiation-promotion protocols
(Pereira et al., 1982; Klaunig et al., 1986; Herren-Freund and Pereira, 1986; Reddy et al., 1992;
Oesterle and Demi, 1985). The results of the studies by Klaunig et al. (1986), Herren-Freund and
Pereira (1986), and Reddy et al. (1992) indicated that chloroform, when administered in the
drinking water, did not promote the development of liver tumors in rats or in two strains of mice,
that chloroform in some cases inhibited the development of hepatic lesions, and that chloroform
did not act as initiator or cocarcinogen. Pereira et al. (1982) reported that chloroform, when
administered as a single dose of 180 mg/kg in tricaprylin, did not demonstrate initiating activity.
However, a significant increase in GGT-positive hepatic foci was reported in rats initiated with
diethylnitrosamine (DEN) and treated with 180 mg chloroform/kg twice weekly for 2 months.
Oesterle and Demi (1985) reported that chloroform had initiating activity, as indicated by
increased incidences of GGT-positive and ATPase-deficient lesions in the livers of female rats
initiated with DEN and treated with chloroform in olive oil.
4.4.3.7. Role of Altered Gene Expression in Carcinogenicity
Several studies have investigated the potential for chloroform to alter gene expression as
a carcinogenic mode of action. These studies have been summarized by ILSI (1997) and U.S.
EPA (1998c). Fox et al. (1990) evaluated the mutation frequency of the H-ras oncogene in liver
tumors in male B6C3F1 mice, which have a high spontaneous incidence of liver tumors
(2%-30%). In the spontaneous tumors, mutations activating the H-ras oncogene were present in
about 64% of the tumors. However, in mice treated with 200 mg chloroform/kg twice weekly via
corn oil gavage for 1 year, only about 21% of the liver tumors had mutations that activated the H-
33
-------
ras oncogene. Based on these results, the authors concluded that mutations activating the H-ras
oncogene may be a mechanism for the formation of spontaneous tumors, but that chloroform-
induced liver tumors occurred by a different mechanism.
The expression of various oncogenes in the liver was evaluated by Sprankle et al. (1996).
Oncogene expression was evaluated in the livers of B6C3F1 mice that had received 350 mg
chloroform/kg and rats that received 180 mg/kg via a single oral corn oil gavage. In the female
mouse liver, transient increases in mRNA for the myc andfos genes were reported; however,
mRNA levels for the Ha-ras and met genes and for hepatocyte growth factor were similar to the
levels observed in the controls. The authors noted similar gene responses were reported for other
carcinogens that are cytotoxic and concluded that the changes in expression of the myc andfos
gene may be a mechanism by which chloroform induces regenerative cell proliferation.
The methylation state of genes in chloroform-induced liver tumors has been evaluated by
Vorce and Goodman (1991) and Dees and Travis (1994). The methylation status of ras
oncogenes was evaluated in liver tumors in male B6C3F1 mice that had received chloroform
(200 mg/kg) via corn oil gavage twice weekly for 1 year (Vorce and Goodman, 1991). In all
liver tumors examined in the treated and control groups, the Ha-ras was hypomethylated and
occasional hypomethylation of the Ki-ras gene was also observed. The methylation status of the
myc gene was not altered.
Exposures to chloroform at concentrations of 0.5%-2% (v/v) resulted in
hypermethylation of the p53 protein in rat liver epithelial cells and in Saos-2 human sarcoma
cells transfected with the gene for p53 (Dees and Travis, 1994). The authors noted that
concurrent administration of other chemicals (phorbol myristate acetate, toluene, and benzene)
resulted in greater hypermethylation and that each of these chemicals, including chloroform, has
the capacity to stimulate protein kinase C. The authors proposed that the hypermethylation of
p53 may have been due to the stimulation of protein kinase C, and that this may represent an
alternative mode of action to chloroform carcinogen!city.
4.4.4. Studies of Interactions With Other Chemicals
Because of the importance of CYP2E1 in the metabolism and toxicity of chloroform, any
chemical that induces the level of CYP2E1 activity is likely to also increase the toxicity of
chloroform. A number of such chemicals are known, including many alcohols (including
ethanol), aldehydes, aromatics, ethers, halogenated solvents, and heterocyclics (Ronis et al.,
1996). The mechanisms by which these agents induce CYP2E1 appear to be complex and
varied, including transcriptional, translational, and posttranslational mechanisms.
Other chemicals, specifically certain ketones, may potentiate the toxic effects of
chloroform by mechanisms other than enzyme induction. For example, hepatic microsomal
enzymes were induced to a greater extent with the insecticide mirex, when compared with its
ketone analogue, chlordecone (Cianflone et al., 1980). However, the binding of chloroform to
hepatic constituents was greater following pretreatment with chlordecone than with pretreatment
with mirex. Hewitt et al. (1979) reported that pretreatment with chlordecone followed by
exposure to chloroform resulted in an altered pattern of histological hepatic lesions, when
34
-------
compared with the pattern of histological lesions observed following administration of the
appropriate dose of chloroform that resulted in a comparable number of abnormal hepatocytes.
Alcohols, which are metabolized to ketones, and other ketones have also been reported to
potentiate the toxic effects of chloroform. Alloxan-induced diabetic rats, which were in a state of
metabolic ketosis, were reported to be more sensitive to the effects of chloroform (Hanasono
et al., 1975). The precise mode of action by which ketones potentiate chloroform toxicity is
unclear; however, possible mechanisms include alterations in calcium pump activity (Moore and
Ray, 1983) or an increased sensitivity to chloroform following exposure to ketones (Hewitt et al.,
1990).
Other chemicals that may potentiate chloroform toxicity include dichloroacetic acid
(Davis, 1992) and carbon tetrachloride (Borzelleca et al., 1990). In the study by Davis (1992),
the administration of nontoxic doses of dichloroacetic acid to female Sprague-Dawley rats
resulted in increased toxicity in the liver, based on increased plasma alanine aminotransferase
levels, and in the kidney, based on increased blood urea nitrogen levels. Histological
examinations of the liver and kidney were not conducted. The administration of trichloroacetic
acid resulted in the potentiation of chloroform-induced nephrotoxicity. Borzelleca et al. (1990)
reported greater evidence of hepatotoxicity, measured by increases in several serum enzymes, in
rats that received concurrent oral administrations of carbon tetrachloride and chloroform, when
compared with rats that received carbon tetrachloride or chloroform alone.
The toxicity of chloroform may be decreased by some chemicals as well. For example,
hepatic and renal toxicity was decreased in rats that received concurrent oral administrations of
chloroform and trichloroethylene, when compared with the toxicity observed in rats that received
chloroform alone (Lilly, 1992). This effect was reported to be independent of the dosing vehicle.
4.5. SYNTHESIS AND EVALUATION OF MAJOR NONCANCER EFFECTS AND
MODE OF ACTION
The noncancer effects of chloroform exposure have been well characterized in numerous
studies in animals. These studies reveal that oral or inhalation exposure to chloroform results in
toxicity to the liver, kidney, and nasal epithelium. Data from studies of exposed humans are
more limited, but support the conclusion that hepatotoxicity is the noncancer effect of chief
concern following chronic exposure.
In laboratory animals, evidence of hepatic and renal damage is usually based on
histological detection of fatty infiltration and degeneration, cellular necrosis, and cellular
vacuolization, along with changes in serum enzyme levels, altered liver and kidney weight,
and/or altered organ function. In some cases, evidence of cellular regeneration (presumably in
response to antecedent cellular necrosis) can be detected using the labeling index even when
frank cytotoxicity is not readily apparent.
Exposure to chloroform during pregnancy can also result in reproductive or
developmental toxicity (U.S. EPA, 1998c). However, available studies indicate that these effects
occur at the same or higher doses as those that cause effects on the dam (Thompson et al., 1974,
1988; Ruddick et al., 1983; Baeder and Hoffman, 1988, 1991; Schwetz et al., 1974), suggesting
35
-------
that most of the effects are secondary to maternal toxicity. No studies were located that
demonstrate that the fetus is more sensitive to chloroform toxicity than the mother.
There is strong evidence that the toxicity of chloroform is a result of the metabolism of
the parent to toxic intermediates. This conclusion is based mainly on the observation that
toxicity of chloroform is increased by chemicals that enhance metabolism and is reduced by
chemicals that inhibit metabolism. Further, variations in chloroform toxicity between tissues and
between species and genders tend to correlate with the level of metabolic enzymes and the
metabolic capacity of the tissues and species.
Metabolism of chloroform may occur through one or both of two pathways: oxidative and
reductive metabolism. Both pathways result in the formation of highly reactive metabolites:
phosgene (oxidative) and dichloromethyl free radical (reductive) (U.S. EPA, 1994d). These
reactive intermediates are capable of forming covalent adducts with cellular molecules (Pohl et
al., 1980; Brown et al., 1974; Tyson et al., 1983), presumably resulting in impairment of cellular
function and contributing to cell injury and death. In general, covalent binding of reactive
metabolites to cellular molecules is highest in areas of the liver and kidney where cytotoxicity is
greatest (e.g., Ilett et al., 1973). Free radicals produced via the reductive pathway may also
induce lipid peroxidation; however, there are only limited data available to support this
conclusion (U.S. EPA, 1994d).
The relative importance of the oxidative and reductive pathways has been investigated by
several researchers, and the results suggest that metabolism occurs mainly via the oxidative
pathway. First, reductive metabolism of chloroform is observed only in phenobarbital-induced
animals or in tissues prepared from them, with negligible reducing activity observed in
microsomes from uninduced animals (ILSI, 1997). Second, the reactive intermediates formed
by oxidative metabolism bind to the polar heads of phospholipids, whereas the reductive
metabolites bind to the fatty acid tails (ILSI, 1997). Thus, the pattern of lipid adducts formed can
be used to distinguish which chloroform metabolic pathways are occurring under a specified test
condition. Using this approach, Ade et al. (1994) showed that even under relatively low (2.6%)
oxygen partial pressure (approximately average for the liver), more than 75% of the phospholipid
binding was to the fatty acid heads, indicating metabolism was chiefly by the oxidative pathway
(U.S. EPA, 1998c; ILSI, 1997). Third, addition of glutathione to the incubation system
completely negated binding to liver microsomes, with only residual binding remaining in kidney
microsomes (ILSI, 1997). This quenching by glutathione is expected for the products of
oxidative but not reductive metabolism. Finally, the gender-specific pattern of renal toxicity
observed in mice in vivo correlates with the level of adduct formation in proteins and lipids in
vitro only under aerobic conditions. Taken together, these observations strongly support the
conclusion that chloroform metabolism in vivo occurs primarily via the oxidative pathway except
under special conditions of high chloroform doses in preinduced animals (ILSI, 1997; U.S. EPA,
1998c).
36
-------
4.6. WEIGHT-OF-EVIDENCE EVALUATION AND CANCER
CHARACTERIZATION
4.6.1. Mode of Action
4.6.1.1. Summary of Postulated Mode of Action
Studies in humans are inadequate to determine if chloroform is carcinogenic. Studies in
animals reveal that chloroform can cause an increased incidence of kidney tumors in male rats
and an increased incidence of liver tumors in male and female mice. Tumors are produced only
at dose levels that result in cytotoxicity. These induced tumor responses are postulated to be
secondary to sustained or repeated cytotoxicity and secondary regenerative hyperplasia.
Chloroform's carcinogenic effects in rodent liver and kidney are attributed to oxidative
metabolism-mediated cytotoxicity in the target organs. Although chloroform undergoes both
oxidative and reductive cytochrome P450-mediated metabolism, it is the oxidative (CYP2E1)
metabolic pathway that predominates at low chloroform exposures. This oxidative pathway
produces highly tissue-reactive metabolites (in particular phosgene) that lead to tissue injury and
cell death. It is likely that the electrophilic metabolite phosgene causes cellular toxicity by
reaction with tissue proteins and cellular macromolecules as well as phospholipids, glutathione,
free cysteine, histidine, methionine, and tyrosine. The liver and kidney tumors induced by
chloroform depend on persistent cytotoxic and regenerative cell proliferation responses. The
persistent cell proliferation presumably would lead to higher probabilities of spontaneous cell
mutation and subsequent cancer. The weight of the evidence indicates that a mutagenic mode of
action via DNA reactivity is not a significant component of the chloroform carcinogenic process.
4.6.1.2. Identification of Key Events
There are essentially three key steps in the sequence of events that lead to chloroform-
induced tumorigenesis in the liver and kidneys of rodents. The first step is oxidative metabolism
of chloroform in the target organs, kidney and liver. Numerous binding and metabolism studies
(as described in ILSI, 1997, and U.S. EPA, 1998c) support the conclusion that chloroform is
metabolized by the oxidative cytochrome P450 (CYP2E1) pathway. This conclusion is
supported by the study of Constan et al. (1999) in Sv/129 wild type, Sv/129 CYP2E1 null, and
B6C3F1 mice. In the wild type of each strain, exposure to 90 ppm chloroform for 6 hours per
day for 4 consecutive days resulted in severe hepatic and renal lesions along with increased cell
proliferation. With the same exposure, neither the cytotoxicity nor cell proliferation occurred in
the CYP2E1 null mouse or in the wild type of either strains treated with the P450 inhibitor 1-
aminobenztriazole.
Available evidence indicates that metabolism by CYP2E1 predominates at low exposures
and is rate-limiting to chloroform's carcinogenic potential. Reductive metabolism, if it occurs,
can lead to free radicals and tissue damage, but this pathway is absent or minor under normal
physiological conditions. The next key step is the resultant cytotoxicity and cell death caused by
the oxidative metabolites (with phosgene as the significant toxic intermediate). Regenerative cell
proliferation follows the hepatotoxicity and renal toxicity as measured by labeling index in
mouse kidney and liver and rat kidney from chloroform-treated animals.
37
-------
This increase in cell division can lead to an increased probability of cancer by one or both
of two alternative modes of action. First, cells that are undergoing cell division are inherently
more susceptible to initiation than are slowly growing or nondividing cells. This is because
DNA undergoing replication is more exposed to nucleophilic attack than DNA that is covered
with histones and arranged in nucleosomes (Ames and Gold, 1991a,b). Also, any gene damage
that occurs in a cell undergoing division has less time to be repaired before mitosis than in a
slowly growing cell, so a larger fraction of DNA alterations could be converted into mutations.
Second, chemicals that promote cell division may convey a selective growth advantage to
preexisting initiated cells in comparison with normal cells, thereby facilitating clonal expansion
of initiated cells. This could occur because initiated cells are more responsive than normal cells
to growth stimuli, because they are less susceptible to the toxicity of the chemical, or because
they are less susceptible to endogenous regulatory signals that trigger programmed cell death
(apoptosis). In any case, the ratio of cell birth to cell death of initiated cells increases compared
with normal cells, leading to increased likelihood that a clone of initiated cells will form and
survive. A key characteristic of this mode of action is that the effect is reversible: the clones of
induced cells will tend to regress if the promoter (mitogen, cytotoxicant) is withdrawn (Pitot et
al., 1987; Schulte-Hermann et al., 1993).
4.6.1.3. Strength, Consistency, Specificity of Association
Table 3 summarizes information on the correlation between the occurrence of statistically
significant increases in cancer prevalence and evidence of cytotoxicity and regenerative
hyperplasia (mainly in the form of increased labeling index) in animals exposed to chloroform.
Inspection of Table 3 reveals two main points:
• There are numerous cases where exposure to chloroform causes an increase in the LI
without any observable increase in cancer incidence. These data indicate that chloroform
exposures that are adequate to cause cytotoxicity and regenerative cell proliferation do
not always lead to cancer.
• There are no cases in which a tumorigenic response has been observed where evidence of
cell regeneration is not also observed at the same or lower dose as that which caused an
increase in tumors. This consistency of evidence (i.e., cell regeneration is detected in all
cases of tumorigenicity) is strong evidence supporting the conclusion that cell
regeneration is a mandatory precursor for tumorigenicity.
Evidence for a link between sustained cytotoxicity/regenerative hyperplasia and cancer is
strongest in the kidney. In male Osborne Mendel rats exposed to chloroform in water for 2 years
(Jorgenson et al., 1985), a statistically significant increase in renal tumors was observed at a
concentration of 1,800 ppm (160 mg/kg/day). A reanalysis of the histopathological slides from
this study (Hard et al., 2000) revealed evidence for sustained cytotoxicity and cell proliferation in
the kidney at exposures of 900 ppm (81 mg/kg/day) or higher. Likewise, in BDFj mice exposed
38
-------
a
a>
=
a>
0/j
a>
•a
=
w
'S
oil
o
o
=
o
U
of cell regeneration
u
S
a
g>
X
O
j=
u
§
O
i
I
'-
-*i
9
1
e
Cj
OJ
Dose level
1
1
Cj
=
OJ
(2
1
s
o
CJ
3
5
L =
3 2
•e
e
0
, u
j '3
- &
1/3
i 3
i-a
Os
"^
Larson et
34 (4 days)
d,
+
E
on
o
^
m
15
^
} r.
O
OO
1—1
+
B6C3F1
a
O
00
c\
2
i
1
H
90 ppm (7 wks)
d,
+
Q
m
00
Os
^
-H
tj
1
M
1
O
OS
Q
m
o
•a
3
.o
S
OS
•ar
Larson et
238 (4 days)
d,
+
E
on
o
^
m
15
Os
} ^
O
00
on
B6C3F1
$"
O
1
o
OS
•ar
Larson et
CN
d,
1
PH"
f>
O
^D
ffl
OO
2
r1
rt
_
i
a
o
o-,
CN
1
B6C3F1
M
•g
Q
OJ
1
CM
00
2
i
1
H
I
O
d,
+
Q
m
00
Os
^
-H
tj
1
M
1
O
OS
Q
m
o
•a
3
j3
OS
•ar
c
1
H
t
d,
i
0
r?
o>
t •>
o
0
00
O
s1
O
b
J
Os
•ar
Larson et
0
d,
i
PH
OO
2
r1
c3
_
i
a
o
0
•^
O
M
•g
Q
jj
S
51
Cfs
•ar
1
H
m
i
OH
OH
0
o
dx
+
PH
OO
ON
1—1
•
"S
§
3i
c3
^
1
O
0>
PH
O
•a
3
*j
»H
^
~3
Larson et
1
O
0
d,
+
m
PH
(5
0>
t •-
O
0
o
^
-H
"S
1
3)
1
O
0
2
o
•a
3
OJ
1
fe
Os
t^
"3r
Larson et
34 (4 days)
d,
+
E
on
o
^
m
15
Os
} ^
O
t~
fN
B6C3F1
M
O
00
c\
2
i
1
H
1
O
d,
Q
ffl
OO
0>
^
-H
"S
§
M
1
O
+
Q
m
o
•a
3
(D
S
OS
•ar
Larson et
477 (4 days)
d,
E
on
o
v>
m
15
Os
} r.
O
t~
*
B6C3F1
M
O
1
o
t^
•ar
Larson et
1
d.
E
m
O
^D
ffl
OO
2
r1
rt
_
i
wj
o
o-,
CN
B6C3F1
M
•g
Q
OJ
1
fe
00
t^
2
i
1
H
I
O
d,
Q
ffl
OO
0>
^
-H
"S
§
M
1
O
OS
Q
m
o
•a
3
j3
OS
•ar
c
1
H
0
d.
0
(5
0>
t •>
o
0
00
O
s1
O
&
a
S8
•— 1 "-^
OO
^^
&§,
+ +
2°
OO
r1
rt
_
1
a
o
0
•^
O
M
•g
Q
jj
S
a
Os
•ar
c
1
H
30 ppm (3 wks)
d,
+
m
PH
OO
0>
^
-H
"S
§
M
1
O
0>
PH
O
•a
3
*j
»H
<^
*a
Larson et
1
o
0
d-
+
m
PH
(5
0>
t •-
O
0
o
(N
o
a
O
a
Os
•ar
c
1
(D
30 ppm (13 wks)
d-
m
PH
OO
0>
^
-H
"S
1
3)
1
O
PH
O
•a
3
-------
to chloroform by inhalation at 5, 30, or 90 ppm for 6 h/day, 5 days/week (Nagano et al., 1998),
increased incidence of renal tumors was observed in male mice at the two higher doses while
females showed no significant tumor response. Templin et al. (1998) duplicated this exposure
regimen in order to study whether the treatment caused cytotoxicity and regenerative hyperplasia.
These authors observed cytotoxicity and hyperplasia in the kidneys of male mice exposed to 30
or 90 ppm throughout a 90-day exposure period, but not in females. This observation is
consistent with the hypothesis that sustained cytotoxicity and regenerative hyperplasia are key
events in the neoplastic response of the kidney to chloroform.
Available data also indicate that cytotoxicity and regenerative hyperplasia are required for
liver cancer, although the strength of this conclusion is somewhat limited because most of the
observations are based on short-term rather than long-term histological or labeling index
measurements. For example, in the B6C3F1 mouse, corn oil gavage (bolus dosing) at the same
doses that resulted in liver tumors in the study by NCI (1976) also caused hepatic cytolethality
and a cell proliferative response at 4 days and 3 weeks (Larson et al. 1994b,c). Similarly,
exposure of female B6C3F1 mice to chloroform in drinking water at levels that did not induce
liver tumors (Jorgenson et al., 1985) also did not induce hepatic cytolethality or cell proliferation
at 4 days or 3 weeks (Larson et al., 1994b). This consistency of the data (i.e., evidence of
cytolethality and/or regenerative hyperplasia is always observed in cases of increased liver
tumors) supports the conclusion that this liver cancer also occurs via a mode of action involving
regenerative hyperplasia.
4.6.1.4. Dose-Response Relationship
Chloroform-induced liver tumors in mice are only seen after bolus corn oil dosing.
Mouse liver tumors are not found following administration by other routes (drinking water and
inhalation). Rat liver tumors are not induced by chloroform following either drinking water or
corn oil gavage administration. Kidney tumors are found in mice exposed to chloroform via
inhalation or in toothpaste preparations, and in rats when exposed via drinking water or corn oil
gavage. Kidney and liver tumors develop only at doses that cause persistent cytotoxicity and
regenerative proliferation, regardless of route of exposure or dosing regime. The dose-response
curves for the cytotoxicity and cell proliferation responses are nonlinear. All key events and
tumor effects depend on the dose rate, as shown by the difference in oil gavage versus drinking
water administration (ILSI, 1997; U.S. EPA, 1998c).
4.6.1.5. Temporal Relationship
As noted above, there is very strong evidence from short-term and long-term histological
and labeling index studies in mice and rats that cytotoxicity and cell proliferation always precede
increased kidney or liver tumor effects in long-term bioassays. For example, a reevaluation of
serial sacrifice data from the chloroform 2-year drinking water bioassay in Osborne-Mendel rats
revealed a linkage between toxicity in the renal tubules and tumor development and showed that
renal toxicity preceded tumor development (Hard and Wolf, 1999; Hard et al., 2000).
40
-------
4.6.1.6. Biological Plausibility and Coherence
The theory that sustained cell proliferation to replace cells killed by toxicity, viral, or
other insult such as physical abrasion of tissues can be a significant risk factor for cancer is
plausible and generally accepted (Correa, 1996). It is logical to deduce that sustained
cytotoxicity and regenerative cell proliferation may result in a greater likelihood of spontaneous
mutations being perpetuated, with the possibility of one or more of these resulting in
uncontrolled growth. It may also be that continuous stimulus of proliferation by growth factors
involved in inflammatory responses increases the probability that damaged cells may slip through
cell cycle checkpoints carrying DNA alterations that would otherwise be repaired. Current views
of cancer processes support both possibilities. There are no data on chloroform that allow the
events that occur during cell proliferation to be directly observed. A high proliferation rate alone
is not assumed to cause cancer; tissues with naturally high rates of turnover do not necessarily
have high rates of cancer, and tissue toxicity in animal studies does not invariably lead to cancer.
Nevertheless, regenerative proliferation associated with persistent cytotoxicity appears to be a
risk factor of consequence.
4.6.1.7. Role of Mutagenicity
The question whether chloroform or a metabolite is mutagenic has been tested
extensively across different phylogenetic orders (i.e., bacterial, eukaryotic, and mammalian
systems). Predominately negative results are reported in all test systems, with no pattern of
mutagenicity seen in any one system considered to be a competent predictor. Positive results
appear sporadically in the database, but are outnumbered by negative results in other tests in the
same system. ILSI (1997) considered results from 40 tests by the quantitative weight-of-
evidence method for heterogeneous genetic toxicology databases from the International
Commission for Protection against Environmental Mutagens and Carcinogens (ICEMC)
(Lohman et al., 1992). This method scores relative DNA reactivity with a maximum positive
score being +100, and maximum negative of-100. The maximum positive score obtained
among 100 chemical databases has been +49.7 (triazaquone) and the maximum negative has
been -27.7. The score for chloroform was -14.3.
Testing of chloroform in the p53 heterozygous knockout mouse shows no tumor effect
(Gollapudi et al., 1999). Heterozygous p53 males were dosed up to 140 mg/kg and females up to
240 mg/kg via corn oil gavage for 13 weeks. This model is known to respond to most mutagenic
carcinogens.
Products of oxidative and reductive metabolism of chloroform are highly reactive. Such
species are unstable and will likely react with cytoplasmic molecules before reaching nuclear
DNA. Such reactive species (e.g., phosgene) have not been evaluated separately for genetic
toxicity, and because of reactivity would not be amenable to study and would not likely be able
to transport from the cellular site of production to the nucleus.
Comparative examination of both oxidative and reductive metabolism for structural
analogues and chloroform has revealed that carbon tetrachloride, which is largely metabolized to
a free radical via the reductive pathway, results in cell toxicity, not mutagenicity. Moreover,
chloroform and carbon tetrachloride show very different patterns of liver toxicity (i.e., carbon
41
-------
tetrachloride's toxicity is more consistent with free radical production and chloroform's is not).
For methylene chloride, glutathione conjugation results in mutagenic metabolites. When rat
glutathione transferase gene copies are introduced into Salmonella, bromodichloromethane
produces mutagenic metabolites; the fact that chloroform in this system did so only marginally
and only at high toxic doses (Pegram et al., 1997) provides support for a conclusion that the
reductive pathway does not contribute to chloroform's toxicity and carcinogen!city.
In initiation-promotion studies, chloroform at the highest test dose of the drinking water
bioassay, does not promote development of hepatic lesions in rats or two strains of mice, nor
does it initiate or act as a co-carcinogen. Administered in oil, chloroform was a promoter in the
rat liver in initiation-promotion protocols. These results are more consistent with the postulated
mode of action than with any mutagenic potential.
4.6.1.8. Conclusion Regarding Cancer Mode of Action
The weight of the evidence supports the conclusion that chloroform-induced tumors in
liver and kidney are only produced at dose levels that result in repeated or sustained cytotoxicity
and regenerative cell proliferation. A wide range of evidence across different species, sexes, and
routes of exposure implicates oxidative CYP2E1 metabolism leading to persistent cytotoxicity
and regenerative cell proliferation as events that precede and are associated with tumor
formation. The cytochrome P450 oxidative metabolism that leads to oxidative damage and
ensuing cell growth, involving basic tissue responses to cellular toxicity and death, is common to
humans and rodents. No data exist that indicate the mode of action observed in rodents is not
also likely to apply to humans.
Available data on the mutagenic and genotoxic potential of chloroform are mixed, but the
majority of tests are negative, and some of the positive results are observed only at extreme
exposure conditions. Thus, the weight of the evidence indicates that chloroform is not a strong
mutagen and that neither chloroform nor its metabolites readily bind to DNA. Based on these
results and the results of studies that evaluated other endpoints of genotoxicity, it seems likely
that even though a role for mutagenicity cannot be excluded with certainty, chloroform does not
produce carcinogenic effects primarily by a specific genotoxic mechanism.
4.6.2. Weight of Evidence
Under the 1986 U.S. EPA Guidelines for Carcinogen Risk Assessment, chloroform has
been classified as Group B2, probable human carcinogen, based on sufficient evidence of
carcinogenicity in animals (U.S. EPA, 1998d).
Under the Proposed Guidelines for Carcinogen Risk Assessment (U.S. EPA, 1996a,
Federal Register 61[79]:17960-18011; U.S. EPA, 1999), chloroform is likely to be carcinogenic
to humans by all routes of exposure under dose conditions that lead to cytotoxicity and
regenerative hyperplasia in susceptible tissues (U.S. EPA, 1998c,d). Chloroform is not likely to
be carcinogenic to humans by all routes of exposure at dose levels that do not cause cytotoxicity
and cell regeneration. This weight-of-evidence conclusion is based on: (1) Observations in
animals exposed by both oral and inhalation pathways indicate that sustained or repeated
cytotoxicity with secondary regenerative hyperplasia precedes, and is probably required for,
42
-------
hepatic and renal neoplasia; (2) there are no epidemiological data specific to chloroform and, at
most, equivocal epidemiological data related to drinking water exposures that cannot necessarily
be attributed to chloroform amongst multiple other disinfection byproducts; and (3) the weight of
evidence of the genotoxicity data on chloroform supports a conclusion that chloroform is not
strongly mutagenic, and that genotoxicity is not likely to be the predominant mode of action
underlying the carcinogenic potential of chloroform. Although no cancer data exist for exposures
via the dermal pathway, the weight-of-evidence conclusion is considered to be applicable to this
pathway as well, because chloroform absorbed through the skin and into the blood is expected to
be metabolized and to cause toxicity in much the same way as chloroform absorbed by other
exposure routes.
4.7. SUSCEPTIBLE POPULATIONS
A susceptible population is any group of people who may be at increased risk of
experiencing an adverse effect from an environmental chemical compared with other members of
the population. In general, two factors may contribute to increased susceptibility: higher than
average risk of exposure, and higher than average adverse response per unit exposure.
Individuals may have higher exposure to a chemical because they reside or work in an area with
elevated concentrations in the environment, or because they have higher than average intake of
contaminated environmental media. For example, children often have higher intakes per unit
body weight of water, air, and food than do adults, and this may contribute to an increased
exposure rate during childhood compared to adulthood. Individuals may have a higher than
average adverse response per unit exposure for a number of reasons, including increased
absorption, decreased excretion, higher or lower metabolism (depending on whether metabolism
increases or decreases toxicity), decreased cellular defense and repair mechanisms, etc. The
following section discusses available data on whether there are any subpopulations that may be
especially susceptible to the adverse effects of chloroform.
4.7.1. Possible Childhood Susceptibility
The central questions asked in a mode of action analysis are, 1) whether the standard
assumption that a mode of action observed in animals is relevant to humans holds true in a
particular case, and 2) what the nature of the mode of action implies about the shape of the dose
response relationship. In the case of chloroform the conclusions have been that the rodent mode
of action can be assumed to be relevant to humans and that a nonlinear approach is most
appropriate. The next question is whether the data lead one to anticipate similarities or
differences in response by sex or age.
Ideally, one would have adequate data to compare each of the key events of chloroform
toxicity and subsequent carcinogenicity in tissues of adults with those of the developing fetus and
young. This kind of information is currently not to be found. In the absence of data on the fetus
and young specific to chloroform, an evaluation is made as to whether a cogent biological
rationale exists for determining that the postulated mode of action is applicable to children (EPA,
1999). There is no suggestion from available studies of chloroform to indicate that children or
fetuses would be qualitatively more sensitive to its effects than adults. The developing organism
would not be expected to be particularly sensitive to cytotoxic agents at minimally toxic levels
43
-------
because cell division is proceeding rapidly and repair capacity at the molecular and cellular level
is high. This is reflected by the relatively low incidence of spontaneous tumors in developing
and young organisms. Moreover, the reproductive and developmental studies available, while
they have limitations, show that fetal effects are seen only at doses at which maternal toxicity is
evident. Research would be needed to further explore whether there are circumstances in which
this relationship does not hold. Research would also be needed to discover whether there is
some other mode of action, not seen in rodents, that might be possible. Presently, there are no
clues from in vivo or in vitro studies as to what alternative mode of action might be considered.
In keeping with traditional toxicologic evaluations, chloroform has been tested in lifetime studies
with high level doses to provide maximal opportunities for toxicologic effects to manifest
themselves in multiple tissues and organs through multiple mechanisms. In the absence of data
to the contrary, this approach is considered to provide evidence for lack of potential for
significant response, other than those noted, even for sensitive individuals and life stages.
The mode of action analyzed as well as all other potential modes of action identified
required that chloroform be metabolized by cytochrome P450 (CYP2E1) (SAB (2000), p.2).
When this is considered along with the comparison of this enzyme activity between adults and
the young there is confidence in assuming similarity in response among life stages. Further
research on the processes of cell injury, death and regeneration would increase this confidence by
addressing any uncertainty about potential quantitative similarity. The literature does not reveal
any such quantitative data at present.
Given the above, it is reasonable to assume that: 1) The reactive metabolite inside the
cell should have similar effects by reacting with and disrupting macromolecules in the cells of
fetuses, children and adults, 2) Cell necrosis and reparative replication are not likely to be
qualitatively different in various stages in life, 3) Cancer risk to the fetus or children would be a
function of cytotoxic injury, like in adults, and protecting these life stages from sufficient
cytotoxicity to elicit this response should protect against cancer risks. Further research would be
needed to assess whether there are significant quantitative differences between life stages which
have not yet been elucidated.
It can be noted that if data indicated that it were appropriate to apply a linear approach to
part of a lifetime, such as the first 3 years of life, the resulting risk would be represented by a
small increment of the total dose per body weight over a lifetime since most of a 70 year life is
at an adult body weight. When this total is divided by 70 years to derive the lifetime average
daily dose, the small increment of early dose does not significantly increase risk. The RfD (in
mg/kg/dy) estimated for chloroform's cancer effect would fall at about 4.2 x 10"5 on a line of
linear extrapolation if dose were at the RfD level for a lifetime. If one assumed a higher dose (in
mg/kg/dy) for the first three years of life, e.g., three times, the added risk would be 4.21 x 10"5 for
a lifetime.
4.7.1.1. U.S. Incidence of Kidney and Liver Cancer in Children
The National Cancer Institute's SEER Pediatric Monograph on the cancer incidence and
survival among children and adolescents reports that Wilms' tumor (a tumor due to a germ cell
mutation) accounts for 95% of the renal cancers in children and those younger than 20 (Reis et
al., 1999). Renal cancer accounts for about 6.3% of cancer diagnoses in children younger than 15
44
-------
an 4.4% of diagnoses in those younger than 20. As to risk factors the report says in part: "A small
proportion of Wilms' tumor cases appear to be heritable including those patients with bilateral
tumors, those occurring in association with aniridia and other congenital disorders, and those few
cases arising in the small number of families with one or more additional cases of Wilms' tumor
in close family members....Most of the analytical and epidemiologic investigations of childhood
renal cancer have focused on Wilms' tumor, and very little is known about risk factors for
childhood renal carcinoma or the other rarer childhood renal cancer subtypes. Several
epidemiological studies have investigated occupational, environmental, and lifestyle
characteristics as potential risk factors for Wilms' tumor. A number of parental and childhood
exposures have been found to be associated with an increased risk of Wilms' tumor. Most of
these associations have not been replicated in multiple high quality studies. However, some
warrant further evaluation including paternal occupational exposures, pesticide exposure, and
certain maternal exposures during pregnancy." Liver cancer is rare in children; 100-150 children
younger than 20 are diagnosed with liver cancer yearly, about 1% of childhood cancers.
Hepatoblastoma, a congenital cancer, is the most frequent liver neoplasm in infancy to 4 years,
and thus is not related to adult onset of liver cancer. Less frequent is hepatocarcinoma, which
increases in proportion with age and is the prevalent adult tumor type. Recent studies have
suggested an association of hepatoblastoma with prematurity and its treatment. For both liver
and kidney cancer, there are studies reporting evidence of association with specific parental
occupational exposure to metals and organic chemicals and with maternal medication or other
exposures; as yet, no identification of a causal agent or agents has been made.
A potential limitation to the use of cancer incidence data to assess the relative sensitivity
of children versus adults is that exposures during childhood may not result in a neoplastic
response until later in life. Thus, the low incidence of renal and hepatic tumors in children
should be interpreted as evidence that is consistent with (but not proof of) the idea that children
are not more susceptible than adults. Indeed, there are no direct data that indicate whether
exposures to chloroform during childhood are or are not associated with increased risk of cancer
later in life.
4.7.1.2. Liver Toxicity in Younger Versus Older Rodents
In a two-generation study (NTP, 1988), CD-I (ICR)BR mice were exposed to chloroform
in utero, during lactation, and then by gavage as young mice through "young" adulthood. The
only liver effect observed was mild to moderate liver histopathology (degeneration of
centrilobular hepatocytes, accompanied by occasional single-cell necrosis) in females at 41
mg/kg/day, the only dose at which systemic effects were evaluated. Thus, the only dose tested in
this study, 41 mg/kg/day, was a LOAEL for liver histopathology (U.S. EPA, 1998c). No effects
of chloroform on reproductive function were identified (NTP, 1988). Oral developmental
toxicity studies have found decreased fetal weight (Thompson et al., 1974), and inhalation
developmental studies have found an increased incidence of delayed ossification in Wistar rats
(Baeder and Hofmann, 1991), but these effects occurred at doses above those causing
hepatotoxicity.
4.7.1.3. Metabolism of Chloroform in Fetuses, Infants, and Children Compared with
Adults—Implications for Quantitative Dose-Response Relationship
45
-------
Metabolism of chloroform is essential to its toxicity (U.S. EPA, 1998c). Moreover,
metabolism by cytochrome P450 CYP2E1 is required for toxicity to both liver and kidney of
B6C3F1 and Sv/129 male mice (Constan et al., 1999). Because of the role of CYP2E1 in
chloroform's mode of carcinogenic action, it is important to evaluate CYP2E1 activity in tissues
of the young compared with adults to determine whether the young might respond at a lower
dose than adults.
Most studies on CYP2E1 levels in humans indicate that this enzyme is expressed in
human adults but not in human fetuses, even when measured using sensitive assays (reviewed in
Hakkola et al., 1998). In these studies, levels of both CYP2E1 protein and of the associated
enzyme activity were undetectable before birth, but rose rapidly shortly after birth. For example,
Vieira et al. (1996) found that CYP2E1 protein could not be detected immunochemically in fetal
human liver, and there was only minimal evidence of CYP2E1 mRNA or CYP2E1 activity in
fetal liver microsomes. (The difference in assay results may be due to differences in sensitivity,
or to cross-reaction of CYP1 Al activity, but the authors noted that "it is generally assumed that
CYP1 Al is not expressed in appreciable amounts in human livers.") The authors found,
however, that CYP2E1 protein levels rise rapidly in the first few hours after birth, with a slow
increase in protein levels and in CYP2E1 RNA levels during childhood. However, a few studies
indicate CYP2E1 is expressed in fetal liver or cephalic tissue (Boutelet-Bochan et al., 1997;
Carpenter et al., 1996). Boutelet-Bochan et al. (1997) detected low levels of CYP2E1 mRNA
transcription in human fetal brains (gestation days 52-117, or 7-17 weeks), and levels tended to
increase with gestational age. Transcription was detected with a very sensitive assay (reverse
transcriptase-polymerase chain reaction, RT-PCR) or the moderately sensitive RNase protection
assay. Transcription in fetal liver was much lower and was detectable in only two of six samples.
Also using the RNase technique, Carpenter et al. (1996) found transcription of CYP2E1 mRNA
in the liver of human fetuses at 19-24 weeks gestation, but not at 10 weeks gestation. Fetal liver
microsomes could metabolize the CYP2E1 substrate ethanol, but at a rate only 12%-27% of
adult liver microsomes. Most of the observed activity was specific to CYP2E1, as it was
inhibited by an anti-CYP2El antibody. Like adult hepatocytes, fetal hepatocytes exposed to
ethanol had induced levels of CYP2E1. Thus, maternal exposures to ethanol and other inducers
of CYP2E1 might increase fetal levels of the enzyme and hence might increase sensitivity to
chloroform exposures that occurred during or after birth.
Studies in humans indicate CYP2E1 enzymic activity in human fetuses is either absent or
low compared with that in adult tissues. However, the enzyme is rapidly induced upon birth,
although the amount of CYP2E1 at birth may be less than that present in the adult. Given that
metabolism of chloroform is necessary to its carcinogenicity, the data on CYP2E1 and
considering that the amount of this enzyme's activity appears to be less than adult amounts in the
fetus and less or equal in children depending on age, the data provide no evidence to suggest that
fetuses or children are more susceptible than adults due to this metabolic activity.
Studies in animals on the developmental regulation of CYP2E1 provide uniform evidence
of the rapid induction of this gene soon after birth (Song et al., 1986; Umeno et al., 1988;
Schenkman et al., 1989; Ueno and Gonzalez, 1990). The idea suggested by some scientists that
the enzyme activity peaks before weaning with a gradual decrease to adult levels has not been
consistently reported in the three studies that compared expression over this period of time.
46
-------
For example, Schenkman et al. (1989) indicate that CYP2E1 protein is present in low
levels in neonates, rises to a peak level at age 2 weeks, and subsequently decreases to adult levels
by puberty. Analysis of protein levels quantified from western blots showed a maximum at 2
weeks with decreasing levels at 4 and 12 weeks. The protein level at 12 weeks was
approximately 50% of the level at 2 weeks. The authors did not provide a statistical analysis of
this result, but it appears from the error bars that the 2-week and 12-week levels (but not 4-week
levels) were significantly different.
Song et al. (1986) conducted a similar analysis and reported a rapid transcriptional
induction of CYP2E1 within 1 week following birth that remained elevated throughout 12 weeks.
The authors did not quantify the western blots, but visual inspection indicates a small decline in
protein levels by 12 weeks. However, in this same study, enzyme activity gradually increased
over time, reaching a maximum at adulthood.
Ueno and Gonzalez (1990) showed that extracts from 3-day-old and 12-week-old rat
liver, but not fetal or newborn rat liver, were able to generate significant CYP2E1 transcription in
vitro. The ability of the extract to drive transcription of CYP2E1 was slightly greater at 12 weeks.
Taken together, these animal studies do not provide conclusive evidence of an early
period of increased enzymatic activity in young animals when compared with adults. While the
animal data remain unclear regarding the potential for a period of increased CYP2E1 activity
above that in the adult, for humans, a gradual increase of CYP2E1 activity throughout childhood
with a maximum level at adulthood, as described by Hakkola et al. (1998), appears to be the most
likely situation.
4.7.1.4. Conclusion Regarding Risks to Children
The evidence provides no basis to conclude that the mode of action of chloroform
(CYP2E1-dependent generation of phosgene leading to cytotoxicity in liver and kidney) would
differ between children and adults. Neither the fetus, nor the child appears to more sensitive
based on level of CYP2E1 activity. Studies in humans indicates CYP2E1 enzymic activity in
human fetuses is either absent or low compared with that in adult tissues. However, the enzyme
is rapidly induced upon birth, although the amount of CYP2E1 at birth may be less than that
present in the adult.
4.7.2. Possible Gender Differences
Cancer statistics for the United States (Wingo et al., 1995) indicate that liver cancer
incidence is similar in males and females, with an annual incidence rate in 1995 of about
0.007%. Alcohol consumption and viral infection are the two risk factors most often cited.
Kidney cancer occurs with an annual incidence of about 0.01% and is about 50% more common
in men than in women. Most kidney tumors are renal adenocarcinomas. The risk factors cited
have been smoking, radiation, obesity, and pharmaceuticals, and kidney cancer does not appear
to be associated with occupational exposures.
47
-------
Studies in animals reveal gender-specific differences in the renal toxicity or
carcinogenicity of chloroform. These differences are generally thought to be a result of gender-
specific differences in the level of cytochrome P450 enzymes responsible for the metabolism of
chloroform. For example, male mice are more sensitive to the nephrotoxic effects of chloroform
than are female mice (Culliford and Hewitt, 1957; Eschenbrenner and Miller, 1945), and the
metabolism of chloroform in the kidneys of male mice is greater than that in female mice (Taylor
et al., 1974). The concentration of CYP2E1 (and hence the toxicity of chloroform) is induced by
testosterone (ILSI, 1997), and as a consequence renal toxicity following chloroform
administration is low in female mice and marked in male mice. This is supported by the finding
that testosterone treatment of female mice increases renal toxicity following administration of
chloroform (ILSI, 1997).
Gender-specific differences in hepatic toxicity also have been reported in rats and mice.
In female rats, evidence of liver toxicity or carcinogenicity has been observed following
administration of chloroform, whereas similar responses were not observed in males
administered similar concentrations of chloroform. In the study conducted by Tumasonis et al.
(1987), an increase in the incidence of hepatic adenomas was observed in female rats (0/18 in
controls versus 10/40 in chloroform-treated animals) administered chloroform in drinking water.
No increase in the incidence of hepatic lesions was observed in male rats, even though available
data indicate that the CYP2E1 isozyme is expressed in higher amounts in the male rat than in the
female rat (Ronis et al., 1996). In mice, a similar increase in the incidence of liver lesions
(degeneration of centralobular hepatocytes, accompanied by occasional single-cell necrosis) was
observed in females administered chloroform by gavage in corn oil, with no increase in the
incidence of this lesion observed in male mice. The reason for the gender-specific difference in
hepatic toxicity is not clear, but might be gender-specific differences in hepatocellular protective
mechanisms (e.g., glutathione levels).
These studies provide evidence of gender-specific susceptibilities in experimental
animals. Whether gender-specific differences in chloroform toxicity also occur in human
populations is currently unknown.
4.7.3. Other Factors That May Increase Susceptibility
As noted earlier, CYP2E1 is induced by a wide variety of alcohols, ketones, and other
chemicals, so people exposed to these substances may have higher enzyme levels (and hence a
greater capacity to metabolize chloroform) than the average individual. This includes people
with excess intake of ethanol.
Dietary status may also influence chloroform metabolism. Wang et al. (1995) reported
that overnight food deprivation resulted in a threefold increase in the metabolism and toxicity of
chloroform compared to fed animals. Cytochrome P450 levels were not significantly increased
by fasting, suggesting that the effect may not have been mediated by enzyme induction.
48
-------
5. DOSE-RESPONSE ASSESSMENTS
5.1. ORAL REFERENCE DOSE
Data on the noncancer effects of chloroform were used to estimate RfD values using two
different approaches: the traditional NOAEL-LOAEL approach and the benchmark dose (BMD)
modeling approach (U.S. EPA, 1995).
5.1.1. NOAEL-LOAEL Approach
5.1.1.1. Data Summary and Choice of Principal Study
Table 4 summarizes oral exposure studies that were considered as candidates for the
derivation of the chronic oral RfD for chloroform. No reliable long-term oral exposure studies in
humans were located, so only studies in animals were evaluated. In order to ensure that the most
sensitive endpoint was selected, the list of candidate studies includes not only longer term studies
of liver and kidney toxicity, but also shorter term studies of reproductive and developmental
effects, as well as several short-term studies based on labeling index in liver or kidney.
Many studies indicate that liver toxicity is the most sensitive non cancer endpoint
following chronic oral exposure of animals to chloroform. Effects on the liver can be detected in
a number of ways, including increased liver fat and/or histological evidence of hepatic
cytotoxicity. The principal study selected to derive the RfD was the report by Heywood et al.
(1979), in which there was an increase in the incidence of moderate to marked hepatic fatty cysts
in dogs. This study was selected because it identifies the lowest LOAEL, and because it is also
the longest duration study (7.5 years). The lesions observed in this study were characterized by
aggregations of vacuolated histiocytes. Although fatty cysts were observed in the control group
as well as all treated groups, both the size and severity of these lesions were significantly
increased in treated animals. Although fatty cysts in liver are not a common endpoint, fat
accumulation in the liver is a common and characteristic effect of chloroform exposure (see Roe
et al., 1979; Jorgenson and Rushbrook, 1980; Jorgenson et al., 1982; DeAngelo et al., 1995;
Thompson et al., 1974), supporting the view that fatty cysts are an authentic and lexicologically
relevant endpoint. This is supported by the observation that chloroform exposure caused a
sustained and dose-responsive increase in SGPT levels in the exposed animals, indicative of low-
level hepatocytotoxicity.
5.1.1.2. No-Observed-Adverse-Effect Level and Lowest-Observed-Adverse Effect Level
The study by Heywood et al. (1979) identified a LOAEL of 15 mg/kg/day, based on the
increase in the number and severity of hepatic fatty cysts in dogs. Because this was the lowest
dose tested, a NOAEL was not identified.
5.1.1.3. Derivation of the Oral Reference Dose
The LOAEL of 15 mg/kg/day identified by Heywood et al. (1979) is used to derive a
chronic oral RfD for chloroform as follows:
49
-------
Table 4. Summary of oral noncancer studies in animals
Endpoint
Systemic
(Body weight,
liver and/or renal
effects)
Reproductive or
developmental
effects
Reference
Roeetal., 1979
Palmer etal., 1979
Hey wood et al., 1979
Jorgenson et al., 1980
Jorgenson et al., 1982
Jorgenson et al., 1982
Jorgenson et al., 1985
Hard etal., 2000
Larson etal., 1994b
Larson etal., 1995b
Bull etal., 1986
Bull etal., 1986
Thompson et al., 1974
Thompson et al., 1974
NTP, 1988
Ruddick et al., 1983
Species/strain
Mouse: ICI,
C57BL, CBA, CFI
Rat: Sprague-
Dawley
Dog: Beagle
Mouse: B6C3F1
Mouse: B6C3F1
Rat: Osborne-
Mendel
Rat: Osborne-
Mendel
Rat: Osborne-
Mendel
Mouse: B6C3F1
Rat: F344
Mouse: B6C3F1
Mouse: B6C3F1
Rat: Sprague-
Dawley
Rabbit: Dutch-
belted
Mouse: CD-I
(ICR) BR
Rat: Sprague-
Dawley
Gender
M,F
M,F
M,F
F
F
M
M
M
F
F
M,F
M,F
F
F
M,F
F
Vehicle
Toothpaste
Toothpaste
Toothpaste
Drinking
water
Drinking
water
Drinking
water
Drinking
water
Drinking
water
DW
Corn oil
Corn oil
Emulphor
Corn oil
Corn oil
Corn oil
Corn oil
Duration
80 wks
SOwks
7.5 yr
30-90 d
3-6 mo
1-6 mo
104 wks
6-24 mo
3 wks
3 wks
90 d
90 d
6-15 of gest.
6-18ofgest.
Fl generation
6-15 of gest.
NOAEL
17
—
—
145
34
160
38
38
16
34
—
270
20
50
35
—
—
LOAEL
60
60
15
290
65-130
—
81
81
43
100
270
—
50
126
50
41
100
Basis
Moderate/severe fatty liver
Decreased weight gain in males and females
(10%), increased liver weight in females
Increased incidence and severity of fatty cysts in
liver, increased SGPT
Increased liver fat
Increased liver fat
No increase in liver fat
Histological evidence of renal tubular
cytotoxicity
Histological evidence of renal tubular
cytotoxicity
Small increase in LI in kidney (no effect in
liver)
Increased LI in liver, kidney, nasal turbinates
Diffuse hepatic degeneration, mild cirrhosis
No hepatic toxicity observed
Dams (decreased wt, mild fatty changes in liver)
Fetus (decreased weight)
Maternal toxicity (no fetotoxicity)
Liver histopathology in females
Decreased weight in dams and fetuses
-------
RfD=15mg/kg/day.(6days/7days)=1E_()2
1,000 & & y
where:
15 mg/kg/day = LOAEL identified by Hey wood et al (1979)
6 days/7 days = Adjustment to account for exposure 6 days/week
1,000= Uncertainty factor. This uncertainty factor includes a factor of 10
to extrapolate from a LOAEL to a NOAEL, a factor of 10 to
extrapolate from an animal species (dog) to humans, and a factor of
10 to account for potential sensitive human subpopulations.
5.1.2. Benchmark Dose Approach
The benchmark dose (BMD) approach utilizes mathematical models to characterize the
dose-response curve for a given endpoint. Given a specified benchmark response (BMR) (e.g., a
10% increase in extra risk), the dose-response equation is used to calculate the BMD (the dose
that yields the BMR), as well as the lower confidence limit of the BMD (referred to as the
BMDL). The BMDL and/or the BMD are then used to derive the RfD (U.S. EPA, 1995).
5.1.2.1. Selection of Data Sets for Modeling
In accord with U.S. EPA guidance (U.S. EPA, 1995), several data sets in addition to the
data set with the lowest LOAEL (Heywood et al., 1979) were selected for BMD modeling. This
is because the study that identifies the lowest LOAEL may not always be suitable for modeling,
or might not always yield the lowest BMD. The data sets selected for modeling included the
following:
l)Incidence of fatty cysts in liver of dogs (Heywood et al., 1979)
2)Histological evidence of renal cytotoxicity in male rats exposed via drinking water (Hard
et al., 2000)
3)Increased labeling index in kidney of female mice exposed via drinking water (Larson et
al., 1994b)
4) Increased labeling index in liver of female rats exposed via gavage in corn oil (Larson et
al., 1995b)
These studies were chosen because they all provide quantitative dose-response data for sensitive
indicators of chloroform toxicity.
5.1.2.2. BMD Modeling of Selected Data Sets
The software employed for benchmark dose modeling was BMDS Version 1.2,
downloaded from EPA's NCEA Web site.
Table 5 summarizes the five data sets that were used for BMD modeling. The data for
dichotomous endpoints were fit to each of the dichotomous models provided in the software,
51
-------
including gamma, logistic, multistage, probit, quantal-linear, quantal-quadratic, and Weibull.
The BMR for dichotomous endpoints was a 10% increase in extra risk (U.S. EPA, 1995). Data
for continuous endpoints were fit to each of the continuous models offered in the BMDS
software (linear, polynomial, power, Hill), using a BMR of one standard deviation.
The detailed results of the BMD model fitting are presented in Appendix B. Within a
data set, the preferred model was selected based on the quality of the model fit to the data.
Models that yielded p values lower than 0.100 were judged to be inadequate and were not
considered further. Models that yielded/* values above 0.100 were assessed using Akaike's
Information Criterion (AIC), along with a visual inspection of the quality of the fit (especially at
low doses). The results are summarized in Table 6.
As seen, the kidney LI data set from Larson et al. (1994b) could not be adequately
described by any of the continuous models. This is because even though the response was
statistically significant, the magnitude of the response was small in comparison to normal
variability, and the data did not form a smooth dose-response relationship (tending to first
increase and then decrease as dose increased). The liver and kidney LI data sets from Larson et
al. (1995b) were reasonably well fit by the Hill equation, with BMD values of 64-75 mg/kg/day.
However, the software was not able to estimate a BMDL value in either case. The data sets from
the studies by Hard et al. (2000) and by Hey wood et al. (1979) were adequately fit by one or
more of the dichotomous models, with the best fit being given by the log-logistic and the quantal-
linear models, respectively. The preferred BMD of 70 mg/kg/day based on the renal cytotoxicity
data of Hard et al. (2000) is similar to the BMD values derived for the LI data from Larson et al.
(1995b), but is significantly higher than the preferred BMD based on the incidence of fatty cysts
in dogs (1.7 mg/kg/day) reported by Heywood et al. (1979). The basis for this marked difference
in BMD between studies is not known, but the data suggest that liver toxicity in the dog is a more
sensitive endpoint of chloroform toxicity than renal or liver cytotoxicity in rodents.
5.1.2.3. Calculation of the BMD-Based RfD
Based on the calculations above, the BMDL value of 1.2 mg/kg/day derived from the
study by Heywood et al. (1979) is selected as the most appropriate basis for the derivation of the
RfD. Because this value is based on exposures that occurred 6 days per week, the value is
adjusted as follows:
BMDL = (1.2 mg/kg/day) * (6/7) = 1.0 mg/kg/day
The RfD is derived from the BMDL by application of appropriate uncertainty factors
(U.S. EPA, 1995). In this case, an uncertainty factor of 10 is used to account for interspecies
extrapolation and a factor of 10 is used to protect potentially sensitive human subpopulations.
Additional uncertainty factors are not required because the database for chloroform is complete.
Bioassays are available in the dog (Heywood et al., 1979), and the rat and mouse (NCI, 1976;
Jorgenson et al., 1982, 1985). Developmental toxicity studies are available in rats and rabbits
exposed via the oral route (Thompson et al., 1974), and in rats (Baeder and Hoffmann, 1988,
52
-------
Table 5. Dose-response data sets used for BMP modeling
Study
Larson etal, 1994b
Study
Larson etal., 1995b
Study
Larson etal., 1995b
Study
Hard et al., 2000
Study
Heywood et al., 1979
Endpoint
LI in kidney (medulla) of female
mice exposed by drinking water for 3
weeks
Endpoint
LI in kidney of female F344 rats
exposed by gavage (3 wks)
Endpoint
LI in liver of female F344 rats
exposed by gavage (3 wks)
Endpoint
Renal proliferation in male OM rats
(18 months)
Endpoint
Incidence of moderate to marked
fatty cysts in liver of beagle dogs
Dose
0
15.7
42.7
82.5
184
329
Dose
0
34
100
200
Dose
0
34
100
200
400
Dose
0
19
38
81
160
Dose
0
15
30
N
14
14
14
14
14
14
N
10
10
10
10
N
10
10
10
10
10
N
18
16
19
19
17
N
27
15
15
Mean
2.3
2.2
3.8
3.5
2.7
3.2
Mean
1.3
1.5
22.4
33.8
Mean
0.6
0.8
2.7
14.0
11.8
% Positive
0
0
0
58%
100%
Incidence
1
9
13
Stdev
0.7
0.8
0.2
0.5
1.5
0.6
Stdev
1.0
0.3
20.9
20.9
Stdev
0.5
0.4
1.5
9.0
15.9
Notes
Values are estimated from
graph in Figure 6B of the
report
Notes
High-dose group excluded
from BMD modeling
Notes
High-dose group excluded
from BMD modeling
Notes
Based on study by Jorgenson
etal.
Notes
Control group is based on
vehicle control and does not
include untreated animals
-------
Table 6. Summary of noncancer BMD modeling results
Study
Larson et al,
1994b
Larson et al.,
1995b
Larson et al.,
1995b
Hard et al.,
2000
Hey wood et al.,
1979
Endpoint
LI in kidney (medulla) of
female mice exposed by
drinking water for 3 weeks
LI in kidney of female F344
rats exposed by gavage (3 wks)
LI in liver of female F344 rats
exposed by gavage (3 wks)
Renal proliferation in male
OM rats (18 months)
Incidence of moderate to
marked fatty cysts in liver of
beagle dogs
Range across models
P-Value
0.000-0.000
0.000-0.000
0.000-0.007
0.003-1.000
0.168-0.810
BMD
25.6-734
7.6-74.7
15.7-63.6
11.7-75.3
1.7-6.2
Preferred model
Type3
None
Hill
Hill
LL
QL
P-Value
—
NA(c)
NA(c)
1.000
0.810
AIC
—
69.33
158.55
27.86
44.58
BMD"
—
75
64
70
1.7
BMDL"
—
Failed
Failed
46
1.2
aLL = log-logistic, QL = quantal linear.
bAll BMD (benchmark dose) and BMDL (lower confidence limit on the BMD) values are reported to two significant figures.
°P value not calculated because degrees of freedom 0.
-------
1991; Schwetz et al., 1974; Stanford Research Institute, 1978) and mice (Murray et al., 1979)
exposed by the inhalation route. These studies indicate that effects on the fetus do not occur
except at doses that cause maternal toxicity. A two-generation reproduction study (NTP, 1988)
found no effects on fertility or reproduction at doses that resulted in liver histopathology.
Finally, chloroform is rapidly metabolized and excreted and thus is not expected to
bioaccumulate.
Based on all of these considerations, a total uncertainty factor of 100 is applied and the
resulting RfD is 1E-02 mg/kg/day:
RfD = 1.0 mg/kg/day / 100 = 1E-02 mg/kg/day.
5.1.3. Summary of Oral RfD Derivation
In the derivation of the RfD for chloroform, both the traditional NOAEL/LOAEL and the
benchmark dose approaches were used. In general, the NOAEL/LOAEL approach for derivation
of an RfD is subject to a number of limitations, including the following (U.S. EPA, 1995):
• Identification of the NOAEL is often judgmental and sometimes controversial
• Experiments with low power tend to yield higher NOAELs
• The scope of the dose response curve plays little role in determining the NOAEL
• The NOAEL is restricted to doses tested
Most of these limitations are addressed by use of the BMD approach (U.S. EPA, 1995). Thus,
the RfD based on the benchmark dose approach is generally preferred.
In the BMD analysis, we modeled several studies. The study by Heywood et al. (1979)
was selected because it identified the lowest LOAEL and because it yielded the lowest BMD
(more protective).
In this case, the dose-response data set from the critical study (Heywood et al., 1979) is
composed of only two doses plus a control group. This is not ideal, because the shape of the
dose-response curve is difficult to define with only three values, especially when the lowest dose
yields a response that is well above the BMR. Nevertheless, the data do yield curve fits of
adequate quality (see Appendix B), so the results of the BMD approach are considered to be
preferable to the NOAEL/LOAEL approach.
Note that in this particular case, the two approaches (NOAEL/LOAEL, benchmark) yield
equal RfD values. This is consistent, albeit coincidental, with the result from the default
LOAEL/NOAEL approach.
5.2. INHALATION REFERENCE CONCENTRATION
Not available. The Agency is currently reviewing the literature and will develop a RfC at a later
date.
55
-------
5.3. ORAL CANCER ASSESSMENT
5.3.1. Choice of Approach
In accord with proposed EPA guidelines for cancer risk assessment (U.S. EPA, 1996a),
the method used to characterize and quantify cancer risk from a chemical depends on what is
known about the mode of action of carcinogenicity and the shape of the cancer dose-response
curve for that chemical. A default assumption of linearity is appropriate when evidence supports
a mode of action of gene mutation due to DNA reactivity or supports another mode of action that
is anticipated to be linear. The linear approach is used as a matter of policy if the mode of action
of carcinogenicity is not understood. A default assumption of nonlinearity is appropriate when
there is no evidence for linearity and sufficient evidence to support an assumption of
nonlinearity. Alternatively, the mode of action may theoretically have a threshold, e.g., the
carcinogenicity may be a secondary effect of toxicity that is itself a threshold phenomenon (U.S.
EPA, 1996a).
In the case of chloroform, the mode of action of carcinogenicity is reasonably well
understood. Available data indicate that chloroform is not strongly mutagenic and that
chloroform is not expected to produce rodent tumors via a genotoxic mode of action (ILSI,
1997). Rather, there is good evidence that carcinogenic responses observed in animals are
associated with regenerative hyperplasia that occurs in response to cytolethality (ILSI, 1997; U.S.
EPA, 1998c; U.S. EPA, 1998d). Because cytolethality occurs only at exposure levels above
some critical dose, a nonlinear approach is considered to be the most appropriate method for
characterizing the cancer risk from chloroform.
5.3.2. Quantification of Cancer Risk
The Proposed Guidelines for Carcinogenic Risk Assessment (U.S. EPA, 1996a) describe
two alternative methods for quantifying the cancer risk from a chemical that acts via a nonlinear
mode of action: a margin of exposure (MOE) evaluation based on quantitative modeling of
tumor dose-response data, or use of an RfD approach. Each of these two alternatives were
assessed as described below.
5.3.2.1. Margin Of Exposure Analysis
5.3.2.1.1. Choice of principal study. Table 7 summarizes studies that were considered
candidates for the point-of-departure/margin-of-exposure quantification of the cancer risk from
ingestion of chloroform. Chloroform has been reported to be carcinogenic in several chronic
animal bioassays, with significant increases in the incidence of liver tumors in male and female
mice and significant increases in the incidence of kidney tumors in male rats and mice (U.S.
EPA, 1994d, 1998c). NAS (1987) reviewed all the available data on the carcinogenicity of
chloroform and concluded that the quantitative risk estimate for oral exposure to chloroform
should be based on the incidence of kidney tumors in male rats reported in the study by
Jorgenson et al. (1985). This study is more appropriate than other candidate data sets in which
exposure occurred via corn oil gavage, because exposure was via drinking water, the most
applicable route of exposure for humans (U.S. EPA, 1998c). Use of data from drinking water
56
-------
Table 7. Summary of inhalation noncancer studies in humans and animals
Effect category
Systemic
effects
Reproductive/
developmental
effects
Reference
Phoonet al,
1983
Lietal., 1993
Challen et al.,
1958
Bomski et al.,
1967
Templin et al.,
1996a
Templin et al.,
1998
Larson et al.,
1996
Nagano et al.,
1998
Baeder and
Hoffman, 1988
Baeder and
Hoffman, 1991
Schwetz et al.,
1974
Murray et al.,
1979
Duration
13wks
(6 hr/d,
7d/wk)
13wks
(5 d/wk, 6 hr/day)
13wks
(6 hr/d, 7 d/wk)
2yr
(6 hr/d, 7 d/wk)
d. 7-16 of gest
(7 hr/d)
d. 7-16 of gest
(7 hr/d)
d. 6-15 of gest.
(7 hr/d)
d. 6-15 or 8-15 of
gest
(7 hr/d)
Species/
strain
1-6 mo.
1-15 yrs
1-10 yrs
1-4 yrs
Rat: F344
Mouse:
B6C3F1
Mouse:
B6C3F1
Rat: F344
Mouse:
BDF1
Rat:
Wistar
Rat:
Wistar
Rat:
Sprague-
Dawley
Mouse:
CF1
Gender
Humans
Humans
Humans
Humans
M,F
M,F
M,F
M,F
M,F
F
F
F
F
NOAEL
ppm
(raw)
5
10
—
~
3.1
30
mg/m3
(raw)
~
24
49
—
~
15
146
mg/m3
(TWA)
~
4.3
12
—
~
4.4
43
LOAEL
ppm
(raw
~
2.0
23
30
8.7
5
30
10.7
95
100
mg/m3
(raw)
9.7
112
144
42
24
146
52
464
488
mg/m3
(TWA)
70-2,000
13-30
110-
1,200
10-1,000
2.4
20
36
7.6
4.4
43
15
135
142
Basis
Jaundice
Neurobehavioral effects; no effects on
liver
Subjective CNS effects; no clinical
effects on liver
Enlarged liver, increased serum
enzymes
Olfactory epithelial atrophy; effects on
liver and kidney not apparent until 88
ppm (430 mg/m3)
Kidney histopathology in males (nasal
effects not studied)
Incidence of liver effects in
females and kidney effects in
males
Necrosis and metaplasia of
olfactory epithelium, ossification
of turbinates
Maternal wt loss, embryolethality
Decreased weight in dams; fetal
toxicity at 30 ppm
Embryotoxicity, fetotoxicity
Fetotoxicity, teratogenicity
TWA: time-weighted average.
-------
studies eliminates a number of uncertainties regarding the potential impact of a corn oil vehicle,
as well as reducing potential issues due to bolus vs. continuous low-dose exposure rate. The
Jorgenson et al. (1985) data set is also preferred because of the multiple doses of chloroform used
in the study (i.e., five doses including a control), and the large number of animals in the low-dose
groups provides greater power in resolving the low end of the dose-response curve (U.S. EPA,
1998d).
5.3.2.1.2. Dose-response data. Jorgenson et al. (1985) administered chloroform in drinking
water at concentrations of 0, 200, 400, 900, or 1,800 mg/L to male Osborne-Mendel rats for 104
weeks. Based on measured water intake and body weights, time-weighted average doses were 0,
19, 38, 81, or 160 mg/kg/day. The incidence of all kidney tumors combined (metastatic
carcinomas, transitional cell carcinomas, tubular cell adenomas, and adenocarcinomas and
nephroblastomas) in male rats increased in a dose-related manner. The incidences of all kidney
tumors combined were 5/301 (2%), 1/50 (2%), 6/313 (2%), 7/148 (5%), 3/48 (6%), and 7/50
(14%) for the control, matched control, 200, 400, 900, and 1,800 mg/L groups, respectively. For
the quantitative modeling performed here, the incidence of all kidney tumors in the matched
control and the four treated groups was used.
5.3.2.1.3. Dose conversion. Because the exposure to rats in the Jorgenson et al. (1985) study
was continuous in drinking water for the lifespan of the animals, no adjustment is needed to
account for duration of exposure or duration of study. Doses in rats are converted to human
equivalent doses by assuming that intakes (mg/day) are equivalent when scaled by body weight
raised to the 3/4 power (U.S. EPA, 1992). When doses are expressed as mg/kg/day, this yields
the following:
( BW V/4
Human dose = animal dose • T-^—
Based on an average body weight of rats in this study of 0.375 kg and an assumed human body
weight of 70 kg, the human equivalent doses are 5.1, 10.3, 21.9, and 43.2 mg/kg/day.
5.3.2.1.4. Point of departure and margin of exposure. In accord with the proposed guidelines
(U.S. EPA, 1996a), evaluation of the carcinogenic risk from a chemical that is believed to act via
a nonlinear mechanism is done in two steps, as follows:
Step 1: Calculate the point of departure
The first step involves dose-response modeling of the cancer incidence data in the
observable range to derive a point of departure (Pdp). The ED10 is defined as the dose that causes
a 10% increase in the incidence of tumors, and the LED10 is defined as the lower 95% confidence
limit on the dose associated with an estimated 10% increase in tumors. In general, the LED10 is
taken as the point of departure.
Using the methods described in U.S. EPA (1995) and U.S. EPA (1996a), and employing
EPA's BMDS Version 1.2 software, the dose-response data for male rat kidney were fit to a
58
-------
variety of alternative mathematical models that are suitable for dichotomous responses. The
results are presented in Appendix B and are summarized in Table 9. All of the models yielded an
adequate fit with the data (all of the p values were greater than 0.50), and all yielded BMD values
in the range of 34 to 38 mg/kg/day. This indicates that the estimates of the ED10 and the LED10
from this data set are not strongly model-dependent. In accord with EPA guidance, the preferred
model was selected on the basis of the lowest AIC. This is the quantal-linear model, which yields
an ED10 of 36 mg/kg/day and an LED10 of 23 mg/kg/day.
Step 2: Calculate a MOE
The MOE is calculated as the ratio of the Pdp divided by the measured or predicted
environmental concentration. When evaluating the extent of the MOE from a risk management
perspective, the assessor/manager can take into account guidance factors such as the following in
determining whether the exposure is sufficiently protective of public health (U.S. EPA,
1998c,e):
1) A factor to account for potential intrahuman variability in susceptibility
2) A factor to account for the 10% effect level modeled by the BMDL10 software for a
cancer effect
3) A factor to account for remaining sources of uncertainty, including residual
pharmacodynamic differences in sensitivity between the rat and human that may not be
accounted for in the body-weight scaling; the relatively shallow slope of the dose-
response curve; and using data from tumor rather than a precursor effect.
Other factors that could be considered in determining the MOE include the size of the
population exposed, the duration and magnitude of human exposure, persistence in the
environment and in the body, and the relative susceptibility of children and other potentially
susceptible subpopulations. Humans are exposed to relatively low levels of chloroform (average
24 ng/L), below the level that is likely to induce a cytotoxic response. Further, there are no data
to indicate that children are more susceptible than adults.
Comparing the point of departure (LED10) of 23 mg/kg/day to the RfD of 0.01 mg/kg/day
leads to a MOE of 2,000. The Agency considers this large MOE to indicate that the RfD is
adequately protective of public health for cancer effects, based on the nonlinear dose response for
chloroform and the mode of action for both cancer and noncancer effects having a common link
through cytotoxicity.
59
-------
Table 8. Summary of oral cancer studies in animals
Reference
JNC1, 19/6
Roeetal., 1979
Jorgenson et al.,
1985
Tumasonis et al.,
1987
DeAngelo et al.,
1995
Species
Rat:
Osborne-
Mouse:
B6C3F1
IVToii *sp'
ICI,
CBA, CF/1,
Rat:
Osborne-
Mouse:
B6C3F1
Rat: Wistar
Rat: F344
Sex
M,F
M,F
M,F
M,F
M, F
M,F
M
Vehicle
Corn oil
Corn oil
Toothpaste
or oil
Water
Water
Water
Water
Duration
78 wks
93 wks
80 wks
104 wks
104 wks
185 wks
100 wks
Doses
(mg/kg/day)
90, 180
138, 237 (males)
238, 477
17,60
19,38, 81, 160
34, 65, 130, 263
150-200
45,90
N/grp
50
50
52
53-330
50-430
32-45
50
Observation
Increased incidence of kidney
epithelial tumors in males
Increased incidence of
hepatocellular carcinoma in males
Increased incidence of kidney
tumors in ICI males
Increased incidence of kidney
tumors in male rats
No increase in tumors in female
mice
Increased hepatic neoplastic
nodules (females) and
adenofibromas (males and
Increased prevalence or
multiplicity of hepatic hyperplastic
or neoplastic lesions
-------
Table 9. Dose-response modeling of male rat kidney tumor data
Ref:
Data Set:
Data Values:
Jorgenson et al. 1985
Male Kidney Tumors
Rat
Dose
0
19
38
81
160
HED
0
5.1
10.3
21.9
43.2
N
50
313
148
48
50
Incidence
1
6
7
3
7
Model
Gamma
Logistic
Log-Logistic
Multi-Stage
Probit
Log-Probit
Quantal-Linear
Quantal-Quadratic
Weibull
BMR
10% ER
10% ER
10% ER
10% ER
10% ER
10% ER
10% ER
10% ER
10% ER
AIC
195.33
193.94
195.33
195.44
193.78
194.41
193.64
194.18
195.34
P value
0.637
0.673
0.639
0.601
0.711
0.546
0.744
0.601
0.633
BMD
34.5
37.9
34.4
35.1
37.1
36.1
36.2
37.0
34.6
BMDL
22.8
29.9
22.5
23.3
28.4
25.4
23.0
27.6
22.9
SETTINGS
Restrict
Power > 1
yes
yes
Restrict
Slope > 1
yes
yes
Restrict
Betas > 0
yes
N
2
Range Across Models
p-value 0.546 0.744
BMD 34.4 37.9
Preferred Model
AIC 193.64
Type Quantal-Linear
p-Value 0.744
BMD 36.2
BMDL 23.0
61
-------
5.3.2.2. RfD Approach
The Proposed Guidelines for Carcinogenic Risk Assessment (U.S. EPA, 1996a) state that
when the mode-of-action analysis based on available data indicates that "the carcinogenic
response is secondary to another toxicity that has a threshold, the margin-of-exposure analysis
performed for toxicity is the same as is done for a noncancer endpoint, and an RfD for that
toxicity may be considered in the cancer assessment."
This is the case for chloroform. That is, available evidence indicates that chloroform-
induced carcinogenicity is secondary to cytotoxicity and regenerative hyperplasia, and that doses
below the RfD do not result in cytolethality (and hence do not result in increased risk of cancer).
Accordingly, the RfD developed above 1E-02 mg/kg/day for protection against noncancer effects
(including cytolethality and regenerative hyperplasia) is also judged to be protective against
increased risk of cancer.
5.3.2.3. Summary of Oral Cancer Risk Evaluation
As presented above, based on the mode of carcinogenic action for chloroform, a nonlinear
approach is considered appropriate; two methods (MOE approach and the RfD method) were
discussed for quantification of cancer risk from ingestion of chloroform. The RfD approach
identified a dose of 0.01 mg/kg/day is not expected to produce either cancer or noncancer effects.
Thus, this RfD is selected as protective against both noncancer and cancer effects of chloroform
by the oral route.
5.4. INHALATION CANCER ASSESSMENT
[RESERVED]
6. MAJOR CONCLUSIONS IN THE CHARACTERIZATION OF HAZARD AND
DOSE-RESPONSE
6.1. HUMAN HAZARD POTENTIAL
6.1.1. Exposure Pathways
Chloroform is a volatile liquid that is sparingly soluble in water. It occurs in many
public drinking water supplies as a byproduct of chlorination. Because chloroform is volatile,
humans may be exposed to it in water not only by ingestion of drinking water, but also by
inhalation of chloroform vapors released from water into indoor air. Dermal exposure may also
occur during showering or bathing. Certain foods may contain chloroform. Chloroform may
also be released to the environment from some chemical or industrial operations.
62
-------
6.1.2. Toxicokinetics
Chloroform is readily absorbed following oral, dermal, and inhalation exposure. Once
absorbed, chloroform is widely distributed throughout the body. Although chloroform may
accumulate to some degree in body fat, most of an ingested or inhaled dose is readily excreted
via the lungs, and there is little tendency for long-term bioaccumulation in the body.
Chloroform is metabolized mainly in the liver, but metabolism also occurs in other
tissues such as the kidney. Metabolism may occur via two pathways, oxidative and reductive.
Both pathways occur in the cytoplasm and are mediated by cytochrome P450. Oxidative
metabolism results in the conversion of chloroform to phosgene and hydrochloric acid, both of
which are toxic to cells. Reductive metabolism results in formation of dichloromethyl free
radical, which is highly reactive and can cause lipid peroxidation. Under normal conditions,
oxidative metabolism is thought to be the primary metabolic pathway, and reductive metabolism
is believed to be relatively minor.
6.1.3. Characterization of Noncancer Effects
Data on the noncancer effects of chloroform in humans are limited. Central nervous
system effects have been reported in workers exposed to high levels of chloroform in workplace
air as well as in individuals ingesting high doses. However, central nervous system effects are not
likely to be of concern following low level inhalation or oral exposures. Humans exposed to
chloroform by either inhalation or oral routes often display signs of liver damage (Phoon et al.,
1983; Bomski et al., 1967; Schroeder, 1965), supporting the conclusion that the adverse effects in
humans are similar to those observed in animals.
Chloroform toxicity has been extensively studied in animals. In general, similar patterns
of adverse effects are observed following both oral and inhalation exposure. Effects occur
mainly in liver, kidney (epithelial cells of the proximal tubule), and the olfactory nasal
epithelium. It should be noted that the nasal effects are a result of absorbed chloroform rather
than an effect of direct contact, as nasal lesions occur following both oral and inhalation
exposure (Larson et al., 1995b; Templin et al., 1996a), and the spatial pattern of lesions in the
nasal passage does not correlate with the airflow-driven pattern of chloroform contact with
epithelial surfaces (Mery et al., 1994).
Effects on liver, kidney, and nasal epithelium are generally characterized by cellular
degeneration, vacuolization, and necrosis, often accompanied by changes in organ weights and/or
organ function. In some cases, evidence of cellular regeneration (presumably in response to
antecedent cellular necrosis) can be detected even when frank cytotoxicity is not readily apparent.
Effects on liver are largely reversible, whereas effects in the kidney and the nasal epithelium are
poorly reversible.
The severity of liver injury following oral exposure depends in part on the dosing vehicle.
For example, effects that occur following exposure in water are generally less than following an
equal dose administered in corn oil. The basis for this vehicle effect is not certain, but may be
63
-------
due to toxicokinetic differences between bolus versus continuous exposure, with bolus exposures
resulting in higher concentrations delivered to cells than with drinking water exposures.
6.1.4. Reproductive Effects and Risks to Children
Exposure of animals to chloroform during pregnancy can also result in reproductive or
developmental toxicity. However, available studies indicate that these effects occur only at the
same or higher doses as those which cause effects on the dam (U.S. EPA, 1998d), suggesting that
most of the effects are secondary to maternal toxicity. No studies were located that demonstrate
that the fetus is more sensitive to chloroform toxicity than the mother. This is supported by
findings that the enzyme responsible for chloroform metabolism (CYP2E1) is low or absent in
the fetus (Hakkola et al., 1998).
6.1.5. Mode of Toxicity
The toxicity of chloroform on liver, kidney, and nasal mucosa is clearly related to the
ability of tissues to metabolize chloroform. This is based on the finding that toxicity occurs in
those tissues that have the greatest ability to metabolize chloroform, and that toxicity can be
increased or decreased by agents that increase or decrease the activity of the metabolic enzymes,
respectively. Similarly, there are clear differences between genders, strains, and species in their
relative sensitivity to chloroform, and these differences in toxicity correlate with differences in
metabolic capacity. For example, renal toxicity is usually more severe in males than females,
apparently because the cytochrome P450 chiefly responsible for the metabolism of chloroform
(CTP2E1) is induced by testosterone. No data were located on the mode of action or the role of
metabolism in the neurological effects of chloroform.
The molecular mechanism by which chloroform metabolism results in cellular toxicity is
not certain, but is probably due mainly to the formation of phosgene as a result of oxidative
metabolism. Phosgene is highly reactive and can bind with and inactivate cellular molecules.
This mode of toxicity is supported by the finding that glutathione protects against the toxic
effects of chloroform and that toxicity becomes manifest only after glutathione levels have been
depleted (Brown et al., 1974; Stevens and Anders, 1981; Docks and Krishna, 1976; ILSI, 1997).
Oxidative metabolism of chloroform also produces hydrochloric acid, which may contribute to
the toxic effect.
6.1.6. Characterization of Human Carcinogenic Potential
There are no data from studies in humans that are adequate to directly assess the potential
carcinogenicity of chloroform to humans.
In animals, chloroform has been shown to cause increased incidence of liver and kidney
tumors in several species by several exposure routes. However, this carcinogenic response
occurs only at high dose levels that result in cytotoxicity, and the weight of evidence indicates
that carcinogenic responses observed in animals are associated with regenerative hyperplasia that
occurs in response to cytolethality (U.S. EPA, 1998c; U.S. EPA, 1998d). In mouse liver,
numerous studies have demonstrated the association of tumor and cytotoxicity (e.g., Larson et al.,
64
-------
1994b,c, 1995a). In mouse kidney, ILSI (1997) reviewed the available data on cytolethality and
carcinogenicity and concluded that in the proximal tubule in the male mouse kidney there was a
strong association between the occurrence of chloroform-induced cytotoxicity and regenerative
hyperplasia, and the development of tumors. More recently, Hard et al. (2000) did a histological
reevaluation of the original kidney tissue preparations from the various time points of the 2-year
chloroform drinking water bioassay in male rats and established the presence of renal tubule cell
alterations consistent with chronically persistent cytotoxicity and cell turnover at dose levels
associated with renal tubule tumor induction. Such changes were confirmed also in the NCI oral
gavage bioassay. Based on the correlation, ILSI (1997) concluded that the data indicate that
phosgene and other metabolites formed via the oxidative route of metabolism are responsible for
the cytotoxic effects and subsequent cell proliferation and development of tumors in the kidney
following exposures to chloroform. Because the carcinogenicity of chloroform is secondary to
cytotoxicity which has a threshold, and because chloroform and its metabolites are not strongly
genotoxic and are unlikely to cause cancer by a genotoxic mechanism (Lohman et al., 1992; ILSI,
1997; Golden et al., 1997; WHO, 1998), a nonlinear approach is more appropriate for low-dose
extrapolation.
Under the Proposed Guidelines for Carcinogen Risk Assessment (U.S. EPA, 1996a, U.S.
EPA, 1999), chloroform is likely to be carcinogenic to humans by all routes of exposure under
high-dose conditions that lead to cytotoxicity and regenerative hyperplasia in susceptible tissues
(U.S. EPA, 1998c,d). Chloroform is not likely to be carcinogenic to humans by any routes of
exposure at a dose level that does not cause cytotoxicity and cell regeneration. This weight-of-
evidence conclusion is based on observations that chloroform is absorbed by all routes of
exposure, and studies in animals exposed by both oral and inhalation pathways, which indicate
that sustained or repeated cytotoxicity with secondary regenerative hyperplasia precedes, and is
probably a causal factor for, hepatic and renal neoplasia. This conclusion is supported by the
finding that chloroform is not a strong mutagen and is not likely to cause cancer through a
genotoxic mode of action. There are no epidemiological data specific to chloroform and, at
most, equivocal epidemiological data related to drinking water exposures that cannot necessarily
be attributed to chloroform amongst multiple other disinfection byproducts.
6.2. DOSE-RESPONSE
Data from studies of chloroform toxicity in humans are not sufficient to support
derivation of reliable quantitative risk estimates. Therefore, quantitative estimates of risk as a
result of low-level chronic exposure to chloroform are based on animal experiments.
6.2.1. OralRfD
The most reliable study for the derivation of the chronic oral RfD is the report by
Heywood et al. (1979). In this study, dogs were exposed to chloroform in toothpaste base for 7.5
years. Chloroform doses of 15 mg/kg/day resulted in an increase in the severity of fatty cysts in
liver. Dose-response data from this study were evaluated using the BMD approach, and the most
reliable estimate of the lower bound on the benchmark dose (BMDL10) was 1.0 mg/kg/day. The
RfD was derived from this value by application of an uncertainty factor of 100 to account for
65
-------
extrapolation from animals to humans and for potential variability in sensitivity between humans.
The resulting RfD is 0.01 mg/kg/day.
The overall confidence in the RfD is medium. The RfD is based on the increase in
severity of "fatty cysts" in the livers of dogs exposed for 7.5 years to chloroform in a toothpaste
base. Although fatty cysts are not a common endpoint, fat accumulation in the liver is a
common and characteristic effect of chloroform exposure (see Roe et al., 1979; Jorgenson and
Rushbrook, 1980; Jorgenson et al., 1982; DeAngelo et al., 1995; Thompson et al., 1974),
supporting the view that fatty cysts are an authentic and lexicologically relevant endpoint. One
important aspect of the study is that these fatty cysts occurred at doses much lower than those
that were required to produce fat accumulation or cytotoxicity in rodents, suggesting either that
the dog is a more sensitive species than rodents or that the long-term exposure period allowed the
detection of effects that were not observed in shorter term studies. Because chloroform was
administered in a toothpaste base that contained glycerol and small quantities of natural oils, it is
plausible that these constituents may have potentiated the effects of the chloroform. However,
this is not judged to be a likely explanation for the effect. On the basis of these considerations,
confidence in the principal study is rated as medium. The confidence in the remainder of the
database on chloroform toxicity is medium. Chloroform has been evaluated in a number of
chronic and reproductive/developmental studies, and data are sufficient to conclude that effects
on liver are the critical (most sensitive) effect.
6.2.2. Inhalation RfC
[RESERVED]
6.2.3. Oral Cancer Risk
In accord with the Proposed Guidelines for Carcinogenic Risk Assessment (U.S. EPA,
1996a), when available evidence indicates that a carcinogenic response is secondary to another
toxicity that has a threshold, the MOE analysis performed for toxicity may be the same as is done
for a noncancer endpoint. This is the case for chloroform. That is, available evidence indicates
that chloroform-induced carcinogenicity is secondary to cytotoxicity and regenerative
hyperplasia, and that doses below the RfD do not result in cytolethality (and hence are unlikely to
result in increased risk of cancer). Accordingly, the RfD developed above (1E-02 mg/kg/day) for
protection against noncancer effects (including cytolethality and regenerative hyperplasia) can
also be considered protective against increased risk of cancer.
6.2.4. Inhalation Cancer Risk
[RESERVED]
66
-------
7. REFERENCES
Ade, P; Guastadisegni, C; Testai, E; et al. (1994) Multiple activation of chloroform in kidney
microsomes from male and female DBA/2J mice. J Biochem Toxicol 9:289-295 (as cited in
ILSI, 1997).
Ames, BN; Gold, LS. (1991a) Too many rodent carcinogens: mitogenesis increases
mutagenesis. Science 249:970-971.
Ames, BN; Gold, LS. (1991b) Mitogenesis, mutagenesis and animal cancer tests. Chemically
induced cell proliferation. Implications for risk assessment. Proceedings of the Chemically
Induced Cell Proliferation Conference, Austin, Texas, November 29-December 2, 1989. In:
Butterworth, BE; Slaga, TJ; Farland, W; et al., eds. New York: Wiley-Liss, pp. 1-20.
ATSDR. (1997). Toxicological Profile for Chloroform (Update). Agency for Toxic Substances
and Disease Registry. Atlanta, GA.
Baeder, C; Hoffman, T. (1988) Initial submission: inhalation embryotoxicity study of
chloroform in Wistar rats (final report) with attachment and cover letter dated 02/21/92. Pharma
Res Toxicol Pathol. Conducted for Occidental Chem. Corp. U.S. EPA/OTS Public Files,
Document Number: 88-920001208.
Baeder, C; Hoffman, T. (1991) Initial submission—chloroform: supplementary inhalation
embryotoxicity study in Wistar rats (final report) with attachments and cover letter dated
12/24/91. NTIS/OTS0535017. EPA/OTS Doc#8-920000566. Study title: Chloroform:
supplementary inhalation embryotoxicity study in Wistar rats. By C. Baeder and T. Hoffmann.
September 12, 1991. Performed by Hoechst Aktiengesellschaff, Germany, Sponsored by Hoechst
AG and Dow Europe SA. Report No. 91.0902.
Blancato, JN; Chiu, N. (1994) Use of pharmacokinetic models to estimate internal doses from
exposure. Water contaminants and health. In: Wang, GM, ed. New York: Marcel Deckker, pp.
217-239.
Bomski, H; Sobolweska, A; Strakowski, A. (1967) Toxic damage of the liver by chloroform in
chemical industrial workers. As cited in U.S. EPA (1994d). Arch Gewerbepathol Gewerbehy
24:127-134.
Borzelleca, JF; O'Hara, TM; Gennings, C; et al. (1990) Interactions of water contaminants: 1.
Plasma enzyme activity and response surface methodology following gavage administration of
carbon tetrachloride and chloroform or trichloroethylene singly and in combination in the rat.
Fundam Appl Toxicol 14:477-490 (as cited in U.S. EPA, 1994d).
Boutelet-Bochan, H; Huang, Y; Juchau, MR. (1997) Expression of CYP2E1 during
embryogenesis and fetogenesis in human cephalic tissues: implications for the fetal alcohol
syndrome. Biochem Biophys Res Commun 238:443-447.
67
-------
Bove, FJ; Fulcomer, MC; Klotz, JB; et al. (1995) Public drinking water contamination and birth
outcomes. Am JEpidemiol 141:850-862.
Brennan, RJ; Schiestl, RH. (1998) Chloroform and carbon tetrachloride induce
intrachromosomal recombination and oxidative free radicals in Saccharomyces cerevisiae.
Mutat Res 397:271-278.
Brown, BR, Jr; Sipes, IG; Sagalyn, MA. (1974) Mechanisms of acute hepatic toxicity:
chloroform, halothane, and glutathione. Anesthesiology 41:554-561 (as cited in ILSI, 1997).
Brusick, D. (1986) Genotoxicity in cultured mammalian cells produced by low pH treatment
conditions and increased ion concentration. Environ Mutagen 8:879-886.
Brusick, DJ; Ashby, J; de Serres, FJ; et al. (1992) A method for combining and comparing short-
term genotoxicity test data. Mutat Res 266:1-6 (as cited in ILSI, 1997).
Bull, RJ; Brown, JM; Meierhenry, EA; et al. (1986) Enhancement of the hepatotoxicity of
chloroform in B6C3F1 mice by corn oil: implications for chloroform carcinogenesis. Environ
Health Perspect 69:49-58.
Butterworth, B; Smith-Oliver, T; Earle, L; et al. (1989) Use of primary cultures of human
hepatocytes in toxicology studies. Cancer Res 49:1075-108.
Butterworth, BE; Templin, MV; Constan, AA; et al. (1998) Long-term mutagenicity studies
with chloroform and dimethylnitrosamine in female lad transgenic B6C3F1 mice. Environ Mol
Mutagen 31:248-256.
Callen, DF; Wolf, CR; Philpot, RM. (1980) Cytochrome P-450 mediated genetic activity and
cytotoxicity of seven halogenated aliphatic hydrocarbons in Saccharomyces cerevisiae. Mutat
Res 77:55-63 (as cited in U.S. EPA, 1994d).
Cantor, KP; Hoover, R; Hartge, RP; et al. (1985) Drinking water source and bladder cancer; a
case-control study. Water chlorination chemistry, environmental impact and health effects. Vol.
5. In: Jolley, RL; Bull, RJ; Davis, WP; et al., eds. Chelsea MI: Lewis Publishers, pp. 145-152.
Cantor, KP; Lunch, CF; Hildesheim, M; et al. (1998) Drinking water source and chlorination
byproducts. I. Risk of bladder cancer. Epidemiology 9:21-28.
Carpenter, SP; Lasker, JM; Raucy, JL. (1996) Expression, induction, and catalytic activity of
the ethanol-inducible cytochrome P450 (CYP2E1) in human fetal liver and hepatocytes. Mol
Pharmacol 49:260-268.
Chalken, PJR; Hickish, DE; Bedford, J. (1958) Chronic chloroform intoxication. BrJIndMed
15:243-249.
68
-------
Cianflone, DJ; Hewitt, WR; Villeneuve, DC; et al. (1980) Role of biotransformation in the
alterations of chloroform hepatotoxicity produced by kepone and mirex. Toxicol Appl
Pharmacol 53:140-149 (as cited in U.S. EPA, 1994d).
Colacci, A; Bartoli, S; Bonora, B; et al. (1991) Chloroform bioactivation leading to nucleic
acids binding. Tumori 77:285-290 (as cited in U.S. EPA, 1994d).
Constan, AA; Sprankle, CS; Peters, JM; et al. (1999) Metabolism of chloroform by cytochrome
P450 2E1 is required for induction of toxicity in the liver, kidney, and nose of male mice.
Toxicol Appl Pharmacol 160:120-126.
Corley, RA; Mendrala, AL; Smith, FA; et al. (1990) Development of a physiologically based
pharmacokinetic model for chloroform. Toxicol Appl Pharmacol 103:512-527.
Correa, P. (1996) Morphology and natural history of cancer precursors. In: Cancer
epidemiology and prevention. Schottenfield, D; Fraumeni, JF; eds. New York: Oxford
University Press.
Crebelli, R; Benigni, R; Franekic, J; et al. (1988). Induction of chromosomes malsegregation by
halogenated organic solvents in Aspergillus nidulans: unspecific or specific mechanism? Mutat
Res 201:401-411 (as cited in U.S. EPA, 1994d).
Cresteil, T; Beaune, P; Leroux, JP; et al. (1979) Biotransformation of chloroform by rat and
human liver microsomes: in vitro effect on some enzyme activities and mechanisms of
irreversible binding to macromolecules. Chem-Biol Interact 24:153-165 (as cited in ILSI, 1997).
Culliford, D; Hewitt, HB. (1957) The influence of sex hormone status on the susceptibility of
mice to chloroform-induced necrosis of the renal tubules. J Endocrinol 14:381-393 (as cited in
U.S. EPA, 1994d).
Davis, ME. (1992) Dichloroacetic acid and trichloroacetic acid increase chloroform toxicity. J
Toxicol Environ Health 37:139-148 (as cited in U.S. EPA, 1994d).
DeAngelo, A. (1995) Evaluation of the ability of chloroform administered in the drinking water to
enhance renal carcinogenesis in male F344 rats (letter summary from A. DeAngelo to N. Chiu,
October 1995) (as cited in U.S. EPA, 1997).
De Bruyn, WJ; Shorter, JA; Davidovits, P; et al. Uptake of haloacetyl and carbonyl halides by
water surfaces. Environ Sci Technol 29:1179-1185.
Dees, C; Travis, C. (1994) Hyperphosphorylation of P53 induced by benzene, toluene, and
chloroform. Cancer Lett 84:117-123 (as cited in U.S. EPA, 1998c).
Diaz-Gomez, MI; Castro, JA. (1980) Covalent binding of chloroform metabolites to nuclear
proteins: No evidence for binding to nucleic acids. Cancer Lett 9:213-218.
69
-------
DiRenzo, AB; Gandolfi, AJ; Sipes, IG. (1982) Microsomal bioactivation and covalent binding
of aliphatic halides to DNA. Tox Lett 11:243-252.
Docks, EL; Krishna, G. (1976) The role of glutathione in chloroform induced hepatotoxicity.
Exp Mol Pathol 24:13-22 (as cited in ILSI, 1997).
Doyle, TJ; Sheng, W; Cerhan, JR; et al. (1997) The association of drinking water source and
chlorination by-products with cancer incidence among postmenopausal women in Iowa: a
prospective cohort study. Am J Public Health 87:7.
Eschenbrenner, AB; Miller, E. (1945) Induction of hepatomas in mice by repeated oral
administration of chloroform, with observations on sex differences. J Natl Cancer Inst 5:251 -255
(as cited in U.S. EPA, 1994d).
Fox, TR; Schumann, AM; Watanabe, PG; et al. (1990) Mutational analysis of the H-RAS
oncogene in spontaneous C57BL/6 x C3H/HE mouse liver tumors and tumors induced with
genotoxic and nongenotoxic hepatocarcinogens. Cancer Res 50:4014-4019 (as cited in U.S.
EPA, 1998c).
Freedman, M; Cantor, KP; Lee, NL; et al. (1997) Bladder cancer and drinking water: a
population-based case-control study in Washington County, Maryland (United States). Cancer
Causes and Controls 8:738-744.
Fry, BJ; Taylor, T; Hathway, DE. (1972) Pulmonary elimination of chloroform and its
metabolite in man. Arch Int Pharmacodyn 196:98-111 (as cited in ILSI, 1997).
Fujie, K; Aoki, T; Wada, M. (1990) Acute and subacute cytogenetic effects of the
trihalomethanes on rat bone marrow cells in vivo. Mutat Res 242:111-119 (as cited in U.S. EPA,
1994d).
Gallagher, MD; Nuckols, JR; Stallones, L; et al. (1998) Exposure to trihalomethanes and
adverse pregnancy outcomes. Epidemiology 9:484-489.
Gemma, S; Faccioli, S; Chieco, P; et al. (1996) In vivo CHC13 bioactivation, toxicokinetics,
toxicity, and induced compensatory cell proliferation in B6C3F1 male mice. Toxicol Appl
Pharmacol 141:394-402.
Gocke, E; King, M-T; Eckhardt, K; et al. (1981) Mutagenicity of cosmetics ingredients licensed
by the European communities. Mutat Res 90:91-109 (as cited in U.S. EPA, 1994d).
Gollapudi, BB; Yano, BL; Day, SJ; et al. (1999) Response of the transgenic P53+/- mouse 26-
week carcinogenicity assay to chloroform. SOT 1999 Annual Meeting, The Toxicologists,
Abstract #1740, page 369.
Gopinath, C; Ford, EJH. (1975) The role of microsomal hydroxylases in the modification of
chloroform hepatotoxicity in rats. Br J Exp Pathol 56:412-422 (as cited in U.S. EPA, 1994d).
70
-------
Gualandi, G. (1984) Genotoxicity of the free-radical producers CC14 and lipoperoxidation in
Aspergillus nidulans. MutatRes 136:109-114.
Hakkola, J; Raunio H; Purkunen R; et al. (1996) Detection of cytochrome P-450 gene expression
in human placenta in first trimester of pregnancy. Biochem. Pharmacol. 52: 379-383.
Hakkola, J; Pelkonen, O; Pasanen, M; et al. (1998) Xenobiotic-metabolizing cytochrome P450
enzymes in the human feto-placental unit: role in intrauterine toxicity. Crit Rev Toxicol 28:35-
72.
Hanasono, GK; Witschi, H; Plaa, GL. (1975) Potentiation of the hepatotoxic responses to
chemicals in alloxan-diabetic rats. Proc Soc Exp Biol Med 149:903-907 (as cited in U.S. EPA,
1994d).
Hard, GC, Wolf, DC. (1999) Re-evaluation of the chloroform 2-year drinking water bioassay in
Osborne Mendel rats indicates that sustained renal tubule injury is associated with renal tumor
development. Toxicol Sci 48 (1-S): Abstr 140, 30.
Hard, GC; Boorman, GA; Wolf, DC. (2000) Re-evaluation of the 2-year chloroform drinking
water carcinogenicity bioassay in Osborne-Mendel rats supports chronic renal tubule injury as the
mode of action underlying renal tumor response. Toxicol Sci 53:237-244.
Henderson, CJ; Scott, AR; Yang, CS; et al. (1989) Testosterone-mediated regulation of mouse
renal cytochrome P-450 isoenzymes. Biochem J 278:499-503 (as cited in ILSI, 1997).
Herren-Freund, SL; Pereira, MA. (1986) Carcinogenicity of by-products of disinfection in
mouse and rat liver. Environ Health Perspect 69:59-65 (as cited in ILSI, 1997).
Hewitt, WR; Miyajima, H; Cote, MG; et al. (1979) Acute alteration of chloroform-induced
hepato- and nephrotoxicity by mirex and kepone. Toxicol Appl Pharmacol 48:509-527 (as cited
in U.S. EPA, 1994d).
Hewitt, WR; Palmason, C; Masson, S; et al. (1990) Evidence for the involvement of organelles
in the mode of action of ketone-potentiated chloroform-induced hepatotoxicity. Liver 10:35-48
(as cited in U.S. EPA, 1994d).
Heywood, R; Sortwell, RJ; Noel, PRB; et al. (1979) Safety evaluation of toothpaste containing
chloroform: III. Long-term study in beagle dogs. J Environ Pathol Toxicol 2:835-851.
Hildesheim, ME; Cantor, KP; Lynch, CF; et al. (1998) Drinking water sources and chlorination
byproducts: risk of colon and rectal cancers. Epidemiology 9(l):29-35.
Hong, JY; Pan, J; Ning, SM; et al. (1989) Molecular basis for the sex-related difference in renal
N-nitrosodimethylamine demethylase in C3H/HeJ mice. Cancer Res 49:2973-2979 (as cited in
ILSI, 1997).
71
-------
Howard, PH; Meylan, WM. (1997) Handbook of physical properties of organic chemicals.
Boca Raton, FL: Lewis Publishers.
Ilett, KI; Reid, WD; Sipes, IG; et al. (1973) Chloroform toxicity in mice: correlation of renal
and hepatic necrosis with covalent binding of metabolites to tissue macromolecules. Exp Mol
Pathol 19:215-229 (as cited in ILSI, 1997).
International Life Sciences Institute (ILSI). (1997) An evaluation of EPA's proposed guidelines
for carcinogen risk assessment using chloroform and dichloroacetate as case studies: report of an
expert panel. Washington, DC: ILSI Health and Environmental Sciences Institute. November
1997.
IPCS. (2000). International Programme on Chemical Safety. Environmental Health Criteria 216.
Disinfectants and Disinfection Byproducts. Geneva: WHO.
Jagannath, DR; Vultaggio, DM; Brusick, DJ. (1981) Genetic activity of forty-two coded
compounds in the mitotic gene conversion assay using Saccharomyces cerevisiae strain D4. In:
Evaluation of short-term tests for carcinogens; report of the International Collaborative Program.
Jo, WK; Weisel, CP; Lioy, PJ. (1990) Routes of chloroform exposure and body burden from
showering with chlorinated tap water. Risk Anal 10:575-580 (as cited in U.S. EPA, 1994d).
Jorgenson, TA; Rushbrook, CJ. SRI International. (1980) Effects of chloroform in the drinking
water of rats and mice: ninety-day subacute toxicity study. Menlo Park, CA: United States
Environmental Protection Agency. Contract No. 68-03-2616. Publication No. EPA-600/1-80-
030.
Jorgenson, TA; Rushbrook, CJ; Jones, DCL. (1982) Dose-response study of chloroform
carcinogenesis in the mouse and rat: status report. Environ Health Perspect 46:141-149.
Jorgenson, TA; Meierhenry, EF; Rushbrook, CJ; et al. (1985) Carcinogenicity of chloroform in
drinking water to male Osborne-Mendel rats and female B6C3F1 mice. Fundam Appl Toxicol
5:760-769.
Kassinova, GV; Kovaltsova, SV; Marfin, SV; et al. (1981) Activity of 40 coded compounds in
differential inhibition and mitotic crossing-over assays in yeast. Environ Health Criteria 163.
King, WD; Marrett, LD. (1996) Case control study of water sources and bladder cancer.
Cancer Causes Control 7:596-604.
Kirkland, DJ; Smith, KL; Van Abbe, NJ. (1981) Failure of chloroform to induce chromosome
damage or sister chromatid exchanges in cultured human lymphocytes and failure to induce
reversion in Escherichia coli. Food Cosmet Toxicol 19:651-656 (as cited in U.S. EPA, 1994d).
72
-------
Klaunig, JE; Ruch, RJ; Pereira, MA. (1986) Carcinogenicity of chlorinated methane and ethane
compounds administered in drinking water to mice. Environ Health Perspect 69:89-95 (as cited
inlLSI, 1997).
Kramer, MD; Lynch, DF; Isacson, P; et al. (1992) The association of waterborne chloroform
with intrauterine growth retardation. Epidemiology 3(5):407-413.
Land, PC; Owen, EL; Linde, HW. (1981) Morphologic changes in mouse spermatozoa after
exposure to inhalational anesthetics during early spermatogenesis. Anesthesiology 54:53-56 (as
cited in U.S. EPA, 1994d).
Larson, JL; Wolf, DC; Butterworth, BE. (1993) Acute hepatotoxic and nephrotoxic effects of
chloroform in male F-344 rats and female B6C3F1 mice. Fundam Appl Toxicol 20:302-315 (as
cited in ILSI, 1997).
Larson, JL; Sprankle, CS; Butterworth, BE. (1994a) Lack of chloroform-induced DNA repair in
vitro and in vivo in hepatocytes of female B6C3F1 mice. Environ Mol Mutagen 23:132-136.
Larson, JL; Wolf, DC; Butterworth, BE. (1994b) Induced cytotoxicity and cell proliferation in
the hepatocarcinogenicity of chloroform in female B6C3F1 mice: Comparison of administration
by gavage in corn oil vs. ad libitum in drinking water. Fundam Appl Toxicol 22:90-102.
Larson, JL; Wolf, DC; Butterworth, BE. (1994c) Induced cytolethality and regenerative cell
proliferation in the livers and kidneys of male B6C3F1 mice given chloroform by gavage.
Fundam Appl Toxicol 23:537-543.
Larson, JL; Wolf, DC; Butterworth, BE. (1995a) Induced regenerative cell proliferation in liver
and kidneys of male F-344 rats given chloroform in corn oil by gavage or ad libitum in drinking
water. Toxicology 95:73-86 (as cited in ILSI, 1997).
Larson, JL; Wolf, DC; Mery, S; et al. (1995b) Toxicity and cell proliferation in the liver, kidney
and nasal passages of female F344 rats induced by chloroform administered by gavage. Food
Chem Toxicol 33:443-456 (as cited in ILSI, 1997).
Larson, JL; Templin, MV; Wolf, DC; et al. (1996) A 90-day chloroform inhalation study in
female and male B6C3F1 mice: Implications for cancer risk assessment. Fundam Appl Toxicol
30:118-137.
LeCurieux, F; Gauthier, L; Erb, F; et al. (1995) Use of the SOS chromotest, the Ames-
fluctuation test and the newt micronucleus test to study the genotoxicity of four trihalomethanes.
Mutagenesis 10:333-341 (as cited in U.S. EPA, 1998f).
Lewis, RJ. (1993) Hawley's condensed chemical dictionary. 12th ed. New York: Van
Nostrand Reinhold, pp. 266-267.
73
-------
Li, LH; Jiang, XZ; Liang, YX. (1993) Studies on the toxicity and maximum allowable
concentration of chloroform. Biomed Environ Sci 6(2):179-186.
Liang, JC; Hsu, TC; Henry, JE. (1983) Cytogenetic assays for mitotic poisoning: The
grasshopper embryo system for volatile liquids. Mutat Res 113:467-479 (as cited in U.S. EPA,
1994d).
Lilly, MZ. (1992) Examination of the hepatic and renal toxicity from concurrent oral exposure
to chloroform and trichloroethylene. M.S. Thesis, University of North Carolina at Chapel Hill
(as cited in U.S. EPA, 1994d).
Lohman, PHM; Mendelsohn, ML; Moore, DH, II; et al. (1992) A method for comparing and
combining short-term genotoxicity test data: The basic system. Mutat Res 266:7-25 (as cited in
ILSI, 1997).
McConnell, G; Ferguson, DM; Pearson, CR. (1975) Chlorinated hydrocarbons and the
environment. Endeavor 34:13-18 (as cited in U.S. EPA, 1994d).
McGeehin, MA; et al. (1993) Case-control study of bladder cancer and water disinfection
methods in Colorado. Am J Epidemiol 138:492-501.
Mehta, RD; Von Borstel, RC. (1981) Mutagenic activity of forty two encoded compounds in the
haploid yeast reversion assay, strain XVI8514C. In: Evaluation of short-term tests for
carcinogens: report of the International Collaborative Program. Prog Mutat Res 1:414-423.
Mery, S; Larson, JL; Butterworth, BE; et al. (1994) Nasal toxicity of chloroform in male F-344
rats and female B6C3F1 mice following a 1-week inhalation exposure. Toxicol Appl Pharmacol
125:214-227
Mirsalis, J; Tyson, K; Butterworth, B. (1982) Detection of genotoxic carcinogens in the vivo-in
vitro hepatocyte DNA repair assay. Environ Mutagen 4:553-562.
Mitchell, A; Myhr, B; Rudd, C; et al. (1988) Evaluation of the L5178Y mouse lymphoma cell
mutagenesis assay: intra-laboratory results for sixty-three coded chemicals tested at SRI
International. Environ Mol Mutagen 12(Suppl 13):37-101.
Mohla, S; Ahir, S; Ampy, FR. (1988) Tissue specific regulation of renal n-
nitrosodimethylamine-de-methylase activity by testosterone in BALB/c mice. Biochem
Pharmacol 37:2697-2707 (as cited in ILSI, 1997).
Moore, L; Ray, P. (1983) Enhanced inhibition of hepatic microsomal calcium pump activity by
CC14 treatment of isopropanol-pretreated rats. Toxicol Appl Pharmacol 71:54-58 (as cited in
U.S. EPA, 1994d).
74
-------
Morimoto, K; Koizumi, A. (1983) Trihalomethanes induce sister chromatid exchanges in
human lymphocytes in vitro and mouse bone marrow cells in vivo. Environ Res 32:72-79 (as
cited in U.S. EPA, 1994d).
Munson, AE; Sain, LE; Sanders, VM; et al. (1982) Toxicology of organic drinking water
contaminants: trichloromethane, bromodichloromethane, dibromochloromethane and
tribromomethane. Environ Health Perspect 46:117-126 (as cited in ILSI, 1997).
Murray, FJ; Schwetz, BA; McBride, JG; et al. (1979) Toxicity of inhaled chloroform in
pregnant mice and their offspring. Toxicol Appl Pharmacol 50:515-522.
Nagano, K; Nishizawa, T; Yamamoto, S; et al. (1998) Inhalation carcinogenesis studies of six
halogenated hydrocarbons in rats and mice. In Advances in the Prevention of Occupational
Respiratory Diseases. Chiyotani, K; Hosoda, Y; Aizawa, Y, eds. Elsevier Science B.V.
Nakajima, T; Elovaara, E; Park, SS; et al. (1995a) Immunochemical detection of cytochrome
P450 isozymes induced in rat liver by n-hexane, 2-hexanone and acetonyl acetone. Arch Toxicol
55:542-547 (as cited in ILSI, 1997).
Nakajima, T; Elovaara, E; Okino, T; et al. (1995b) Different contributions of cytochrome P450
2E1 and P450 2B1/2 to chloroform hepatotoxicity in rat. Toxicol Appl Pharmacol 133:215-222
(as cited in ILSI, 1997).
National Academy of Sciences (NAS). (1987) Drinking Water and Health. Disinfectants and
Disinfectant Byproducts. Volume 7. Trihalomethanes Disinfection and Disinfection By-
Products. Washington, DC: National Academy Press.
National Cancer Institute (NCI). (1976) Report on carcinogenesis bioassay of chloroform.
Bethesda, MD: National Cancer Institute.
National Toxicology Program (NTP). (1988) Chloroform reproduction and fertility assessment
in CD-I mice when administered by gavage. Report by Environmental Health Research and
Testing, Inc., Lexington, KY, to National Toxicology Program, NTP- 89-018. NTIS PB89-
148639.
Noort, D; Hulst, AG; Fidder, A; et al. (2000) In vitro adduct formation of phosgene with
albumin and hemoglobin in human blood. Chem Res Toxicol 13:719-726.
Oesterle, D; Demi, E. (1985) Dose-dependent promoting activity of chloroform in rat liver foci
bioassay. Cancer Lett 29:59-63 (as cited in ILSI, 1997).
Palmer, AK; Street, AE; Roe, FJC; et al. (1979) Safety evaluation of toothpaste containing
chloroform: II. Long-term studies in rats. J Environ Pathol Toxicol 2:821-833.
Pegram, RA; Andersen, ME; Warren, SH; et al. (1997) Glutathione S-transferase mediated
mutagenicity of trihalomethanes in Salmonella typhimurium: contrasting results with
75
-------
bromodichloromethane and chloroform. Toxicol Appl Pharmacol 144:183-188 (as cited in U.S.
EPA, 1998e).
Pereira, MA. (1994) Route of administration determines whether chloroform enhances or
inhibits cell proliferation in the liver of B6C3F1 mice. Fundam Appl Toxicol 23:87-92.
Pereira, MA; Chang, LW. (1981) Binding of chemical carcinogens and mutagens to rat
hemoglobin. Chem-Biol Interact 33:301-305.
Pereira, MA; Lin, LC; Lippitt, JM; et al. (1982) Trihalomethanes as initiators and promoters of
carcinogenesis. Environ Health Perspect 46:151-156 (as cited in ILSI, 1997).
Pereira, MA; Chang, LW; Ferguson, JL; et al. (1984) Binding of chloroform to the cysteine of
hemoglobin. Chem-Biol Interact 51:115-124.
Perocco, P; Prodi, G. (1981) DNA damage by haloalkanes in human lymphocytes cultured in
vitro. Cancer Lett 13:213-218.
Phoon, WH; Goh, KT; Lee, LT; et al. (1983) Toxic jaundice from occupational exposure to
chloroform. Med J Malaysia 38:31-34 (as cited in U.S. EPA, 1998d).
Pitot, HC; Beer, DG; Hendrich, S. (1987) Multistage Carcinogenesis of the Rat Hepatocyte. In:
Nongenotoxic mechanisms in carcinogenesis. Butterworth, BE; Slaga, TJ, eds. Banbury Report
25. Cold Spring Harbor Laboratory, pp. 41-54.
Pohl, LR; Bhooshan, B; Whittaker, NF; et al. (1977) Phosgene: a metabolite of chloroform.
Biochem Biophys Res Comm 79:684-691.
Pohl, LR; Krishna, G. (1978) Deuterium isotope effect in bioactivation and hepatotoxicity of
chloroform. Life Sci 23:1067-1072 (as cited in ILSI, 1997).
Pohl, LR; Martin, JL; George, JW. (1980) Mechanism of metabolic activation of chloroform by
rat liver microsomes. Biochem Pharmacol 29:3271-3276 (as cited in ILSI, 1997).
Pohl, LR; Branchflower, RV; Highet, RJ; et al. (1981) The formation of diglutathionyl
dithiocarbonate as a metabolite of chloroform, bromotrichloromethane, and carbon tetrachloride.
Drug Metab Dispos 9:334-339.
Pohl, LR; George, JW; Satoh, H. (1984) Strain and sex differences in chloroform-induced
nephrotoxicity. Different rates of metabolism of chloroform to phosgene by the mouse kidney.
Drug Metab Dispos 12:304-308 (as cited in ILSI, 1997).
Potter, CL; Chang, LW; DeAngelo, AB; et al. (1996) Effects of four trihalomethanes on DNA
strand breaks, renal hyaline droplet formation and serum testosterone in male F-344 rats. Cancer
Lett 106:235-242 (as cited in U.S. EPA, 1998f).
76
-------
Rapson, WH; Nazar, MA; Butsky, W. (1980) Mutagenicity produced by aqueous chlorination
of organic compounds. Bull Environ Contam Toxicol 24:590-596 (as cited in U.S. EPA, 1994d).
Reddy, TV; Daniel, FB; Lin, EL; et al. (1992) Chloroform inhibits the development of
diethylnitrosamine-initiated, phenobarbital-promoted gamma-glutamyltranspeptidase and
placental form glutathione S-transferase-positive foci in rat liver. Carcinogenesis 13:1325-1330
(as cited in ILSI, 1997).
Reis, L, Smith M et al. (1999). Cancer incidence and survival among children and adolescents:
United States SEER Program 1975-1995. Bethesda, MD, National Cancer Institute, SEER
Program. Available on line at http://seer.cancer.gov/Publications/PedMono
Reitz, RH; Fox, TR; Quast, JF. (1982) Mechanistic considerations for carcinogenic risk
estimation: Chloroform. Environ Health Perspect 46:163-168 (as cited in U.S. EPA, 1994d).
Reitz, RH; Mendrala, AL; Corley, RA; et al. (1990) Estimating the risk of liver cancer
associated with human exposures to chloroform using physiologically based pharmacokinetic
modeling. Toxicol Appl Pharmacol 105:443-459 [Erratum 1991, 110:536].
Robbiano, L; Meretoe, E; Migliazza Morando, A; et al. (1998) Increased frequency of
micronucleated kidney cells in rats exposed to halogenated anaesthetics. Mutat Res 413:1-6.
Roe, FJC; Palmer, AK; Worden, AN; et al. (1979) Safety evaluation of toothpaste containing
chloroform: I. Long-term studies in mice. J Environ Pathol Toxicol 2:799-819.
Ronis, MJJ; Lindros, KO; Ingelman-Sundberg, M. (1996) The CYP2E subfamily. In:
Cytochromes P450: Metabolic and toxicological aspects. loannides, C, ed. Boca Raton, FL:
CRC Press, pp. 211-240.
Ruddick, JA; Villeneuve, DC; Chu, I. (1983) A teratological assessment of four trihalomethanes
in the rat. J Environ Sci Health 18(3):333-349.
SAB. 2000. Science Advisory Board Report. See http://www.epa.gov/sab/fiscalOO.htm.
Scientific Advisory Board (SAB). (2000) Review of the EPA's Draft Chloroform Risk
Assessment. U.S. Environmental Protection Agency. EPA-SAB-EC-00-009. April 2000.
Sasaki, T; Suzuki, MK; Noda, T; et al. (1998) Mutagenicity study of carbon tetrachloride and
chloroform with microbial mutagenicity test and rat liver micronucleus test. Abstract. P-018. J
Toxicol Sci 23 (suppl. II):305.
Salamone, MF; Heddle, JA; Katz, M. (1981) Mutagenic activity of 41 compounds in the in vivo
micronucleus assay, in: Evaluation of short-term tests for carcinogens: Report of the international
collaborative program. Prog Mutat Res 1:686-697.
77
-------
San Agustin, J; Lim-Sylianco, CY. (1978) Mutagenic and clastogenic effects of chloroform.
Bull Phil Biochem Soc 1:17-23 (as cited in U.S. EPA, 1994d).
Schneider, W; Diller, W. (1991) Phosgene. In: Ullmann's encyclopedia of industrial chemicals.
Fifth, completely revised edition. Elvers, B; Hawkins, S; Schulz, G, eds.
Schulte-Hermann, R; Bursch, W; Kraupp-Grasi, B; et al. (1993) Cell proliferation and
apoptosis in normal liver and preneoplastic foci. Environ Health Perspect 101 (Suppl 5):87-90.
Schwetz, BA; Leong, BJK; Gehring, PJ. (1974) Embryo- and fetotoxicity of inhaled chloroform
in rats. Toxicol Appl Pharmacol 28:442-451.
Shelby, MD; Witt, KL. (1995) Comparison of results from mouse bone marrow chromosome
aberration and micronucleus tests. Environ Mol Mutagen 25:302-313 (as cited in U.S. EPA,
1998f).
Simmon, VF; Kauhanen, K; Tardiff, RG. (1977) Mutagenic activity of chemicals identified in
drinking water. In: Progress in genetic toxicology. Scott, D; Bridges, DA; Sobels, eds.
Elsevier/North Holland: Biomedical Press, pp. 249-258 (as cited in U.S. EPA, 1994d).
Sipes, IG; Krishna, G; Gillette, JR. (1977) Bioactivation of carbon tetrachloride, chloroform
and bromotrichloromethane: role of cytochrome P-450. Life Sci 20:1541-1548 (as cited in ILSI,
1997).
Smith, AE; Gray, GM; Evans, JS. (1995) The ability of predicted internal dose measures to
reconcile tumor bioassay data for chloroform. Regul Toxicol Pharmacol 21:339-351.
Smith, JH; Hook, JB. (1983) Mechanism of chloroform nephrotoxicity: II. In vitro evidence for
renal metabolism of chloroform in mice. Toxicol Appl Pharmacol 70:480-485 (as cited in ILSI,
1997).
Smith, JH; Hook, JB. (1984) Mechanism of chloroform nephrotoxi city: III. Renal and hepatic
microsomal metabolism of chloroform in mice. Toxicol Appl Pharmacol 73:511-524 (as cited in
ILSI, 1997).
Smith, JH; Maita, K; Sleight, SD; et al. (1984) Effect of sex hormone status on chloroform
nephrotoxicity and renal mixed function oxidase in mice. Toxicology 30:305-316 (as cited in
ILSI, 1997).
Smith, MT; Loveridge, N; Wills, ED; et al. (1979) The distribution of glutathione in the rat
liver lobule. Biochem J 182:103-108 (as cited in ILSI, 1997).
Sobti, RC. (1984) Sister chromatid exchange induction potential of the halogenated
hydrocarbons produced during water chlorination. Chromosome Information Service No. 37, pp.
17-19 (as cited in U.S. EPA, 1994d).
78
-------
Song, BJ; Gelboin, HV; Park, SS; et al. (1986) Complementary DNA and protein sequences of
ethanol-inducible rat and human cytochrome P-450s. J Biol Chem 261:16689-16697.
Sprankle, CS; Larson, JL; Goldsworthy, SM; et al. (1996) Levels of myc, fox, Ha-ras, met and
hepatocyte growth factor mRNA during regenerative cell proliferation in female mouse liver and
male rat kidney after a cytotoxic dose of chloroform. Cancer Lett 101:97-106 (as cited in U.S.
EPA, 1998c).
Stanford Research Institute. (1978) Teratology and acute toxicology of selected chemical
pesticides administered by inhalation. Prepared by Newell, W, and Dilley, JV for U.S. EPA
Office of Research and Development, HERL, Research Triangle Park, NC. EPA Contract No.
68-02-1751.
Stevens, JL; Anders, MW. (1981) Metabolism of haloforms to carbon monoxide. IV. Studies
on the reaction mechanism in vivo. Chem-Biol Interact 37:365-374 (as cited in ILSI, 1997).
Sturrock, J. (1977) Lack of mutagenic effects of halothene or chloroform on cultured cells using
the azaguanine test system. Br J Anaesth 49:207-210 (as cited in U.S. EPA, 1994d).
Taylor, DC; Brown, DM; Kuble, R; et al. (1974) Metabolism of chloroform: II. A sex difference
in the metabolism of 14C-chloroform in mice. Xenobiotica 4:165-174 (as cited in U.S. EPA,
1994d).
Templin, MV; Larson, JL; Butterworth, BE; et al. (1996a) A 90-day chloroform inhalation
study in F-344 rats: Profile of toxicity and relevance to cancer studies. Fundam Appl Toxicol
32:109-125.
Templin, MV; Jamison, KC; Wolf, DC; et al. (1996b) Comparison of chloroform-induced
toxicity in the kidneys, liver, and nasal passages of male Osborne-Mendel and F-344 rats.
Cancer Lett 104:71-78 (as cited in ILSI, 1997).
Templin, MV; Constan, AA; Wolf, DC; et al. (1998) Patterns of chloroform-induced
regenerative cell proliferation in BDF1 mice correlate with organ specificity and dose-response
of tumor formation. Carcinogenesis 19:187-193.
Thompson, DJ; Warner, SD; Robinson, VB. (1974) Teratology studies on orally administered
chloroform in the rat and rabbit. Toxicol Appl Pharmacol 29:348-357.
Topham, JC. (1980) Do induced sperm-head abnormalities in mice specifically identify
mammalian mutagens rather than carcinogens? Mutat Res 74:379-387 (as cited in U.S. EPA,
1994d).
Tumasonis, CF; McMartin, DN; Bush, B. (1987) Toxicity of chloroform and
bromodichloromethane when administered over a lifetime in rats. J Environ Pathol Toxicol
Oncol 7:55-64.
79
-------
Tyson, CA; Prather, KH; Story, DL; et al. (1983) Correlations of in vitro and in vivo
hepatotoxicity for five haloalkanes. Toxicol Appl Pharmacol 70:289-302 (as cited in ILSI,
1997).
Uehleke, H; Werner, T; Greim, H; et al. (1977) Metabolic activation of haloalkanes and tests in
vitro for mutagenicity. Xenobiotica 7:393-400 (as cited in U.S. EPA, 1994d).
Ueno, T; Gonzalez, FJ. (1990) Transcriptional control of the rat hepatic CYP2E1 gene. Mol
CellBiol 10:4495-4505.
Umeno, M; Song, BJ; Kozak, C; et al. (1988) The rat P450IIE1 gene: complete intron and exon
sequence, chromosome mapping, and correlation of developmental expression with specific 5'
cytosine demethylation. J Biol Chem 263:4956-4962.
U.S. EPA. (1985) Health assessment document for chloroform. Final report. Research Triangle
Park, NC: Office of Research and Development. EPA/600/8-84-004F (as cited in U.S. EPA,
1994d).
U.S. EPA. (1986a) Guidelines for carcinogen risk assessment. Federal Register 51(185):33992-
34003.
U.S. EPA. (1986b) Guidelines for the health risk assessment of chemical mixtures. Federal
Register 51(185):34014-34025.
U.S. EPA. (1986c) Guidelines for mutagenicity risk assessment. Federal Register
51(185):34006-34012.
U.S. EPA. (1988) Recommendations for and documentation of biological values for use in risk
assessment. EPA 600/6-87/008, NTIS PB88-179874/AS, February 1988.
U.S. EPA. (1991) Guidelines for developmental toxicity risk assessment. Federal Register
56(234):63798-63826.
U.S. EPA. (1994a) Interim policy for particle size and limit concentration issues in inhalation
toxicity: notice of availability. Federal Register 59(206):53799.
U.S. EPA. (1994b) Methods for derivation of inhalation reference concentrations and
application of inhalation dosimetry. Research Triangle Park, NC: Environmental Criteria and
Assessment Office. EPA/600/8-90/066F.
U.S. EPA. (1994c) Peer review and peer involvement at the U.S. Environmental Protection
Agency. Signed by the U.S. EPA Administrator, Carol M. Browner, dated June 7, 1994.
U.S. EPA. (1994d) Final draft for the drinking water criteria document on trihalomethanes.
Prepared for Health and Ecological Criteria Division, Office of Science and Technology,
80
-------
Washington, DC, under EPA Contract No. 68-C2-0139 by Clement International Corporation.
April 8, 1994.
U.S. EPA. (1995) Use of the benchmark dose approach in health risk assessment. U.S.
Environmental Protection Agency, Office of Research and Development. EPA/63 0/R-94/007.
U.S. EPA. (1996a) Proposed guidelines for carcinogen risk assessment. Federal Register
61(79):17960-18011.
U.S. EPA. (1996b) Reproductive toxicity risk assessment guidelines. Federal Register
61(212):56274-56322.
U.S. EPA. (1997) Summary of new data on trihalomethanes (TFDVIS) for the notice of
availability. Draft.
U.S. EPA. (1998a) Guidelines for neurotoxicity risk assessment. Federal Register
63(93):26926-26954.
U.S. EPA. (1998b) Science Policy Council handbook: peer review. Prepared by the Office of
Science Policy, Office of Research and Development, Washington, DC. EPA/1 OO-B-98-001.
U.S. EPA. (1998c) Health risk assessment/characterization of the drinking water disinfection
byproduct chloroform. Prepared for Health and Ecological Criteria Division, Office of Science
and Technology, Washington, DC. Prepared by Toxicology Excellence for Risk Assessment,
Cincinnati, OH, under Purchase Order No. 8W-0767-NTLX. November 4, 1998.
U.S. EPA. (1998d) National primary drinking water regulations: disinfectants and disinfection
byproducts. Notice of data availability; proposed rule. 40 CFR Parts 141-142:15674-15692.
March 31, 1998.
U.S. EPA. (1998e) Addenda to criteria documents for the notice of availability (NOA).
Trihalomethanes.
U.S. EPA. (1998f) National primary drinking water regulations: disinfectants and disinfection
byproducts. Final Rule. Federal Register 63(241):69390-69476. Wed., Dec. 16, 1998.
USEPA. (1999) Guidelines for Carcinogenic Risk Assessment. Review Draft. July 1999. US
Environmental Protection Agency, Risk Assessment Forum.
Van Abbe, NJ; Green, TJ; Richold, M; et al. (1982) Bacterial mutagenicity studies on
chloroform in vitro. Food Chem Toxicol 20:557-561 (as cited in U.S. EPA, 1994d).
Varma, MM; Ampy, FR; Verma, K; et al. (1988) In vitro mutagenicity of water contaminants in
complex mixtures. J Appl Toxicol 8:243-248 (as cited in U.S. EPA, 1994d).
81
-------
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
238:476-83.
Vogel, EW; Nivard, MJM. (1993) Performance of 181 chemicals in a drosophila assay
predominantly monitoring interchromosomal mitotic recombination. Mutagenesis 8:57-81 (as
cited in U.S. EPA, 1994d).
Vorce, RL; Goodman, JL (1991) Hypomethylation of ras oncogenes in chemically induced and
spontaneous B6C3F1 mouse liver tumors. J Toxicol Environ Health 34:367-384 (as cited in U.S.
EPA, 1998c).
Voronin, VM; Litvirov, NN; Kazachkov, VI. (1987) Carcinogenicity of chloroform in the
mouse. VoprOnkol 33:81-85.
Waller, K; Swan, SH; DeLorenze, G; et al. (1998) Trihalomethanes in drinking water and
spontaneous abortion. Epidemiology 9:134-140.
Wang, PY; Kaneko, T; Sato, A; et al. (1995) Dose- and route-dependent alteration of
metabolism and toxicity of chloroform in fed and fasting rats. Toxicol Appl Pharmacol 135:119-
126.
Wecher, RA; Scher, S. (1982) Bioassay procedures for identifying genotoxic agents using light
emitting bacteria as indicator organisms. In: Luminescent assays: perspectives in endocrinology
and clinical chemistry. Seno, M; Pazzagli, M, eds. New York: Raven Press, pp. 109-113.
Wennborg, H; Bodin, L; Vainio, H; et al. (2000) Pregnancy outcome of personnel in Swedish
biomedical research laboratories. J Occup Environ Med 42:438-446.
White, AE; Takehisa, S; Eger, El; et al. (1979) Sister chromatid exchanges induced by inhaled
anesthetics. Anesthesiology 50:426-430 (as cited in U.S. EPA, 1994d).
World Health Organization (WHO). (1998) Guidelines for drinking-water quality. Second
edition. Addendum to volume 2. Health criteria and other supporting information. Chloroform.
Geneva: World Health Organization, pp. 255-275.
Wingo, PA; long, T; Bolden, S. (1995) Cancer statistics 1995. CA Cancer J Clin 45:8-30.
Withey, JR; Collins, BT; Collins PG. (1983) Effect of vehicle on the pharmacokinetics and
uptake of four halogenated hydrocarbons from the gastrointestinal tract of the rat. J Appl
Toxicol 3:249-253 (as cited in U.S. EPA, 1994d).
Yamamoto, S; Aiso, S; Ikawa, N; et al. (1994) Carcinogenesis studies of chloroform in F344
rats and BDF1 mice. Abstract. In: Proceedings of the Fifty-Third Annual Meeting of the
Japanese Cancer Association.
82
-------
APPENDIX A. EXTERNAL PEER REVIEW-
SUMMARY OF COMMENTS AND DISPOSITION
The Toxicological Review of Chloroform was derived from "Health Risk
Assessment/Characterization of the Drinking Water Disinfectant Byproduct Chloroform" and the
Mode of Action Analysis for the Carcinogenicity of Chloroform (SAB review draft). Both
documents have gone through internal and external reviews.
External Peer Reviewers and Affiliations
James A. Swenberg, D.V.M., Ph.D., University of North Carolina
Lorenz Rhomberg, Ph.D., Gradient Corporation
R. Julian Preston, Ph.D., Chemical Industry Institute of Toxicology
External peer reviewers' comments and the disposition of their recommendations are as
follows.
Summary & Response of External Reviewer Comments
Health Risk Assessment/Characterization of the Drinking Water Disinfection Byproduct
Chloroform, March 13, 1998
Comments below are either actual text or paraphrases from the reviewers' comments.
Editorial revisions and specific suggestions were invariably made in response to comments by the
reviewers. These comments and responses are not reviewed here.
James A. Swenberg, D.V.M., Ph.D.
University of North Carolina
Section, Comment, and Response
2.1.1. Toxicokinetics
The summary should state more definitively that the oxidative pathway is responsible for
most metabolism. Excretion of chloroform by the lungs is dose dependent, with greater amounts
excreted unchanged at high doses.
Response: Summary was changed as suggested.
2.1.4. Mechanisms of Toxicity
The section on formation of DNA adducts needs revision. DNA adducts have not been
identified with chloroform. Very small amounts of "covalent binding" have been reported.
A-l
-------
These are not the same. They may only represent metabolic incorporation from the C-l pool.
The reductive pathway is associated with free radical formation, which leads to oxidative stress
and lipid peroxidation. Such responses are highly nonlinear and are enhanced by conditions that
deplete cellular defenses such as glutathione. Under normal conditions, a cell is well equipped to
detoxify these free radicals. The Melnick paper further confuses the issue by including
brominated compounds that are generally recognized as genotoxic and that have identifiable
DNA adducts.
Response: Several clarifying changes were made in the text on these issues.
2.2.2. Quantification of Carcinogenic Effects
This section is rather confusing compared to the ILSI document. I do not understand the
factor of 3 for slope, nor the factor of 10 for severity. The cancer endpoints are associated with
toxicity and are expected results for the cause and effect. If cytotoxicity is used instead of
tumors, a factor of 10 and the use of the ED 10 may be reasonable. This can be coupled with a
factor of 10 for intrahuman variability and a factor of 3 to 10 for interspecies variation, providing
a margin of exposure (MOE) between 300 and 1,000. The ILSI document uses a factor of 30.
This should be discussed. This section of the document needs to be more transparent and its
content presented as a range throughout the document.
Response: Text was changed to add clarity but the range of MOE was not specified.
Rather, a single value of 1,000-fold was chosen.
3.1.2. Strengths and Weaknesses
This section is very brief. In comparison with the ILSI document, it is much less
informative and less transparent. There is no discussion of inconsistencies in cell proliferation
data and strain-specific deficiencies in the data sets.
Response: The text was changed only slightly here. Reference is made to the ILSI work.
This text could be greatly expanded but such effort would increase the overall length
beyond that desired for this risk characterization text.
3.1.3. Key Conclusions, Assumptions, and Defaults
Again, this section is very brief and less informative than the ILSI document. The last
part of paragraph 1 is incorrect in calling the DNA damage that results from reductive
metabolism "direct." It is associated with free radicals and is fully expected to be nonlinear and
highly correlated with depletion of cellular defense mechanisms.
Response: Changes were made to correct the first paragraph. This section was
combined with the weight of evidence section.
3.1.4. Weight of Evidence
Once again, the document should not refer to the free radical mechanism for DNA
damage as "direct" and it should not be stated to be linear as a function of dose. I know of no
data that demonstrate this, but there are many data sets that speak against it.
A-2
-------
Response: Changes were made to correct this section. This section was combined with
that described above.
3.1.7. Alternative Conclusions
This may lead to high-dose genotoxicity from free radicals that could contribute to
chloroform's carcinogenicity. This pathway is recognized to be a minor one that would not be
expected to contribute to low-dose carcinogenicity.
Response: Text was changed to reflect the concept of minor pathway.
3.2.1. Overall Conclusions
This section automatically embraces 7 jig/kg-day as the MOE-based dose. This uses a
factor of 3,000. In my opinion, this should be given as a range using a MOE. ILSI would go as
low as 30. Thus, this section should be broadened to better reflect a range of MOE.
Response: The range of a MOE was not specified, but an increase in transparency was
attempted. An overall MOE of 1,000 was chosen as best representative of the underlying
data.
3.2.2. Human Susceptibility
This section addresses only differences in exposure. It also needs to address genetic
polymorphisms that change metabolism. The interhuman factor of 10 has been used for MOE
and should be put in context with this section.
Response: Section was changed as suggested.
3.3.1. Overall Conclusions
Again, this section uses only the MOE of 3,000. The range needs to be more inclusive.
Response: Again, the range of a MOE was not specified, but an increase in transparency
was attempted. An overall MOE of 1,000 was chosen as best representative of the
underlying data.
3.3.7. Significant Issues and Uncertainties
As stated several times in this review, tumors would not be expected "to be evoked in a
linear fashion by gene mutations from free radicals via reductive metabolism." Such
mechanisms are expected to be nonlinear and associated with high doses that deplete cellular
defense mechanisms.
The MOE should be discussed as a range, not as 3,000. This number is highly
conservative and not well supported by the data presented.
Response: The section was revised to address the comments on the expected lack of
low-dose linearity with tumors. Again, the range of a MOE was not specified, but an
increase in transparency was attempted. An overall MOE of 1,000 was chosen as best
representative of the underlying data.
A-3
-------
3.3.8. Alternative Approaches
Again, the free radical mechanism is not expected to be linear.
Response: Section was revised to reflect this comment.
Lorenz Rhomberg, Ph.D.
Newton, MA
Comment:
Critical discussion of key matters not included in the risk characterization text, such as
chloroform's metabolism, the role of corn oil vehicle, the agent's genotoxic potential, and the
central hypothesis that animal tumors are secondary to cytotoxicity and compensatory cell
proliferation in the target tissues. In sum, the risk characterization does not really document the
Agency's weighing of the evidence on these key questions.
Response: The commentor is correct in that the text does not go into detail regarding key
matters. However, this text is intended to be a characterization of the risk, geared more to
risk managers. Previous manuscripts, appropriately referenced in this text, are available
for a fuller explication of the key matters that were only summarized here.
Comment:
The risk characterization document contains no data. Although it need not be comprehensive, it
would seem important to present (in tabular form) the key data sets upon which the quantitative
risk assessment calculations are based, as well as other data sets that stand as the most prominent
alternatives. The administered doses and tumor counts of the key Jorgenson male rat kidney
tumor data are reported differently in Chiu et al. (1996) and in the ILSI document, for instance,
and it is important that the specific version used in the final calculations (and the reasons for its
variation from other versions) be documented.
Response: Tumor counts were added for the key studies. The chosen data set can be
viewed on EPA's IRIS site (www.epa.gov/iris), and was appropriately cited.
Comment:
Similarly, there is no documentation of risk assessment calculations, either in dose metric
calculation or in dose-response analysis. The citation for the key dose-response analysis (p. 19) to
"personal correspondence" with Paul Pinsky of EPA's National Center for Environmental
Assessment is not adequate documentation for so central a calculation. It makes it impossible for
the reader (and for me as a reviewer) to verify the calculations or to see what specific approaches
were taken (for instance, regarding such matters as lower bound determination, allowance for
concurrent mortality, and degree of model).
Response: This document is the original source of the data set used for the cancer
analysis. Citation of a personal communication with the EPA scientist that did this
analysis is appropriate.
A-4
-------
Comment:
Elsewhere (p. 34) it is implied that the cross-species equivalency of doses was assumed to be
based on surface area scaling (i.e., mg/kg2/3/day equivalency), although the calculations are said
to follow the draft Proposed Guidelines, which mandate mg/kg3/4/day as an oral dosing default.
The document should show key calculations so that such matters can be checked and the
calculation's basis understood.
Response: The commentor is correct. The 3/4 power of body weight was used in the
assessment and is appropriately acknowledged in a footnote to the text.
Comment:
Even though reviews of the literature and the state of the science on areas such as epidemiology,
animal bioassays, metabolism and pharmacokinetics, and mechanisms of toxicity need not be
included in the risk characterization per se, a succinct summary of the main issues (as opposed to
the main studies and their findings) would seem to be in order. In the case of chloroform, for
epidemiology the main issues are attribution of water chlorination effects to specific disinfection
byproducts and the confounding of exposure to chlorine-disinfected drinking water with other
factors of potential importance to cancer risks; for animal bioassays they are the inconsistencies
among studies, the dependence on route of administration, the assessment of cytotoxicity among
bioassay subjects, and the potential bearing of minor tumor responses other than those in kidney
and liver; for metabolism the issue is the extent of reductive metabolism in vivo; for
pharmacokinetics the issues are the incomplete understanding of the role of corn oil, the
characterization of metabolism in humans, and the question of the most appropriate internal dose
measure for understanding the relation of tissue toxic effects as a function of dosing regime; for
toxicity mechanisms the issues are the possibility of significant genotoxicity in the target organs
for carcinogenicity, the mechanistic role, if any, of reductive metabolism, the correlation of
cytotoxicity and tumors in the bioassays, the dose-response patterns for cytotoxicity, the nature of
the evidence that cytotoxicity is necessary for chloroform's tumorigenesis (and not just ancillary),
and the link of quantitative measures of cell proliferation to quantitative changes in cancer risk.
Response: Several of these issues were further discussed as suggested by the
commentor, but others were given only cursory treatment, owing in part to the overall
limitations of this text as a risk management tool.
Comment:
The current draft risk characterization refers to some of these issues (although not all of them),
but it tends to do so simply by citing the existence of the arguments. As a result, when it comes
time to make choices of analytical approaches, the reader is left without a strong sense of why the
Agency is choosing the path it takes or how much confidence it invests in its chosen approach.
For example, the key section on Mechanism of Toxicity (pp. 13-17) mentions various relevant
studies and notes their differing implications, but it takes no stand on the issue of whether
chloroform's observed animal carcinogenicity should be considered secondary to (and causally
dependent on) cytotoxicity in the target organs. Yet immediately afterward, and without any
discussion, the document states (p. 19), "In view of the weight of evidence that chloroform may
induce tumors by a nonlinear mechanism (ILSI, 1997), a margin of exposure approach for dose
response analysis [sic]." (One can easily imagine this incomplete sentence having arisen from a
A-5
-------
hesitancy between finishing it with "is chosen" versus "is considered," with neither being chosen
in the end.)
Response: Attempts were made to clarify the Agency's position throughout this text.
Comment:
In my view, something much bolder is needed, even if the outcome is to boldly declare one's
uncertainty. As I argue below, I think the Agency would err if it put all its confidence in the
existing "nonlinear" dose-response method, but if it is to do so, it should more explicitly declare
its reasoning for why such a course is appropriate. After all, it reverses the earlier Agency stance.
As it stands, the conclusions are simply (albeit hesitantly) declared without clearly stated reasons.
Response: Statements were made more boldly throughout the text that the nonlinear
approach to the assessment of chloroform's carcinogenic risk at low doses was consistent
with more of the data than were alternative hypotheses.
Comment:
A better discussion of corn oil gavage is warranted because the differences in cytotoxicity and in
tumorigenesis between drinking water and oil gavage experiments are critical to choosing the
Jorgenson study and to arguments about the most appropriate internal dosimeter. On p. 6 it is
noted that chloroform is actually absorbed more readily from aqueous vehicles, and the reference
to the role of first-pass metabolism seems inappropriate because it applies equally to all oral
exposures. To me, the magnitude of the corn oil effect seems puzzling, and confidence in other
mechanistic explanations of cytotoxicity hinges on the strength of the dose-rate explanation.
This needs to be noted as an issue.
Response: EPA (1987) has discussed this issue and decided to use the Jorgenson study.
This text merely confirms the previous decision. This discussion is highlighted in EPA's
IRIS (EPA, 1998).
Comment:
Even for drinking water contamination, exposure by inhalation can be a major route of uptake (as
noted briefly in the document's exposure section). Yet elsewhere the issue of inhalation is not
mentioned, despite the existence of the positive Japanese inhalation study. Is the earlier EPA
inhalation unit risk for chloroform obviated, or does it stand in the absence of an explicit
revision? How are inhalation exposures, and combined inhalation and oral exposures, to be
assessed dosimetrically?
Response: The inhalation study is not yet published, but these issues should again be
considered after it is.
Comment:
There are several questions about metabolism that bear further discussion. No clear statement is
given regarding the weight to be given to the possibility of reductive metabolism in vivo to
dichloromethyl radicals. What is the evidence that this occurs in vivo? If the free radicals are
thought to be potentially genotoxic, where is the evidence of such genotoxicity? (After all, the
A-6
-------
genotoxicity tests are not specific to damage caused by oxidative metabolism; the only question
is whether metabolic patterns typical of the in vivo situation are examined. Why is there no
unscheduled DNA synthesis in the liver when animals are exposed in vivo, for instance?)
The impact of having two P450 isozymes with activity toward chloroform is not examined. If
CYP2B 1/2 are active at high doses, what does this say about relative high- and low-dose
metabolic activation?
Response: The section on metabolism was revised to further discuss the weight of
evidence for the reductive metabolic pathway. Also, text was added to further discuss the
P450 metabolism.
Comment:
The document tends to equate reductive metabolism with genotoxicity (and hence linear
dose-response approaches) and oxidative metabolism with cytotoxicity (and nonlinear
approaches). It implies that, if one doubts the operation of reductive metabolism, nonlinear
approaches are indicated. This seems inappropriate for several reasons. First, as noted, the in
vivo genotoxicity data do not refer to which reactive moiety may be causing effects, and the lack
of effects speaks equally to the genotoxic potential of metabolically generated free radicals and
metabolically generated phosgene. Second, free radicals attack a variety of cellular
macromolecules, notably lipids, and they may serve as agents of cytotoxicity as well as phosgene.
This effect may, however, have different sensitivity to the rate of metabolism, an argument that
plays importantly in the choice of internal dosimeters. Third, genotoxicity is not the only reason
to rely on a linear approach to dose-response analysis, and lack of genotoxicity should not be
deemed a sufficient reason to undertake only an MOE approach.
Response: Text changes were made throughout the document in order to break the
unduly rigorous connection between reductive metabolism and linearity, and oxidative
metabolism and nonlinearity.
Comment:
As noted above, the key studies should have the data presented in tabular form. It should be clear
how issues such as concurrent mortality are handled. In particular, the Jorgenson study has some
issues with early mortality.
Response: The data from the cancer bioassays were presented in the test. This text
relied on EPA's IRIS file for current interpretation of these bioassays.
Comment:
At one point, EPA was making much of tumor responses in the Jorgenson study other than the
kidney tumors in rats (e.g., possible elevation of mouse lymphomas). The current assessment
presumes that only mouse liver and rat kidney tumors are at issue. A clear statement of the basis
for choosing which tumor responses warrant investigation is in order.
Response: These issues have been discussed previously in EPA texts. No need existed
to repeat the earlier arguments.
A-7
-------
Comment:
Characterization should have a concise summary (probably tabular) of the results on target organ
doses (examining the contending dose metrics) in different species and organs following
chloroform administration by different routes, as implied by the pharmacokinetic modeling. This
will aid in examining the question of how species differences in response are explained by
dosimetric differences.
Response: PBPK modeling was not used in this text, other than to reference the ILSI
report. Thus, a table of these values was not considered necessary. Furthermore, this text
is a risk characterization document, geared more for review by risk managers. Technical
details are appropriately referenced in other previous work.
Comment:
Similarly, a table should summarize the results of target organ cytotoxicity in the animal
bioassays to aid in examining the question of the correlation of cytotoxicity and tumorigenicity.
Also, graphs of the results of the series of CUT studies on labeling index at different exposures
(similar to those shown in Chiu et al., 1996) should be shown. These should be x-y graphs, not
bar graphs, so that the shape of the labeling index response curve with exposure is evident.
Response: These data are interesting and cited as part of the ILSI report. However, they
are not used here and would only add text to an already "full" risk characterization
document.
Comment:
The equation of genotoxicity with the reductive pathway evident in the document is misguided in
my view, as argued earlier.
Response: Text was changed to reflect this concern.
Comment:
From the foregoing, I imagine it is evident that I feel that the "nonlinear" approach as set out in
the Proposed Guidelines is too readily invoked and that it poorly accommodates what we know
and do not know about the carcinogenic process, even when that process is thought to be
secondary to a nonlinear biological precursor effect such as cytotoxicity. In criticizing the
application of this approach to chloroform, I am not invoking genotoxicity nor even denying
cytotoxicity as the key mode of action.
The rationale behind the nonlinear method is that at some moderately low doses the dose-
response curve for tumors becomes sufficiently steep in its decline with exposure level that
smaller exposures will be without meaningful risk. If human exposures are well below those
associated with tumors in animals, it is supposed that one can be assured that the intervening
steep drop is sufficient to render the low exposures safe. The unanswered question is how far
below is "well below," and what evidence can be adduced to judge the adequacy of the size of
the margin-of-exposure?
A-8
-------
Logically, the approach should be to examine the dose-response pattern for the biological effect
to which carcinogenesis is deemed secondary, and which is presumed to have an actual or
practical threshold. In practice, the method assumes that this point of steep decline to trivial risk
levels will be found somewhere slightly below the ED10 for tumors.
The risk characterization document accepts this presumption quite readily; it calculates a point of
departure as the ED 10 (or LED 10) for rat kidney tumors from the Jorgenson drinking water
study, and then concentrates its discussion to the size of uncertainly factors that might be applied.
Actual dose-response patterns are not examined, but they are presumed to be sufficiently
nonlinear. Also, the metabolic activation of chloroform is stated as "nonlinear as a function of
dose" (p. 28). A plot of these data does not necessarily lead to this conclusion.
In sum, these data seem to provide a poor basis for invoking a method that presumes that risk is
highly nonlinear just below the ED 10 for tumors. This shows that a good deal more effort needs
to be put into defining an appropriate point of departure, and that the adequacy of the point
chosen needs to be examined and defended in a way that has not yet been done.
Response: A number of interesting and important issues are raised in this comment that
apply to chloroform, but perhaps more importantly, to the ongoing revisions of the cancer
guidelines of EPA. Specific text changes were made in the chloroform text to downplay
the nonlinearity with dose, especially in the area of metabolism. However, the
recommendation to use an MOE approach rather than a linear extrapolation is still
retained with chloroform, because, of the two available methods, this seems to best fit the
available data. This judgment is consistent with EPA's 1996 guidelines.
Comment:
I question whether the uncertainty factor approach is appropriate at all. Nonetheless, one can
examine the document's reasoning for the size of the various factors.
Three-fold factor for slope
Shallower slopes warrant larger margins, but how big should such a margin be? Allowing
threefold below an ED 10 for slope (p. 19) with a probit slope of 0.62 leads to a dose that has a
projected risk of 2.5%. (A 10-fold factor only goes down to 0.3%, as opposed to the linear
method's 1.0%). Because this is the only factor that is designed to address the nonlinearity and
drop to acceptable doses, it would seem that in the present case a factor of 100 might be just
marginally enough (i.e., going down to a 10"4 risk on the fitted nonlinear probit curve).
The point of the nonlinear method is to avoid extrapolating a fitted curve down to low-dose. But
threefold below the ED10 is still within the observed range (it is about at the level of the second
dose group in the Jorgenson study) and is in the range where increasing tumor risk is empirically
evident.
Tenfold factor for severity ofendpoint
A-9
-------
I do not understand the logic of this factor. It has not been the usual practice to assume site
concordance and to presume that the potential human endpoint will be for the same tumor in
terms of histological type and malignant aggressiveness. If the argument is that the rat tumors
were frequently benign, is it the presumption that a factor needs to be included in case the human
tumors are less so? If the rat tumors were all malignant, would this factor then be smaller (on the
premise that human tumors could only be less malignant, not more)? I also do not understand
why the "closeness" of the cytotoxicity and tumorigenicity curves (an odd concept in itself, as
one is continuous and the other quantal) leads to a lesser factor.
Tenfold factor intrahuman variability
This presumably means among-human variability (rather than intraindividual variability). When
assessing population risk, one should presumably not use this factor, because it allows for
vulnerability in individual risk in some people, but in a population high individual risks on the
part of some will be balanced by lower than average risks in others. The relative CYP2E 1
activity is a poor basis for judging such variability, because it is not established that metabolic
activation is proportional to this activity (although the PBPK models could examine this
question).
Tenfold factor persistence
Because the animal studies are for lifetime exposure, it is not clear what a correction for
persistence is for. Presumably, this is covered by the dosimetry assumptions about equivalently
toxic dose rates in humans and animals.
Tenfold factor interspecies variation
This has the same ambiguity that the analogous factor has in noncancer risk assessment. Is the
factor an adjustment for an expected sensitivity difference (as implied by the discussion of the
surface area adjustment to doses as obviating it), or is it an allowance for the possibility that for
this particular agent, humans are particularly susceptible owing to some cause not covered by the
dosimetry or other adjustments that are made? I note again that the 2/3-power adjustment
mentioned is at odds with the methodology specified in the Guidelines Proposal. The reference
to assumed kinetic differences is inappropriate, as the factor is applied to account for general
differences in the pace of physiological processes, including those that control the
pharmacodynamic response.
The result of applying these factors is that the "acceptable" dose is only 40-fold higher
than the dose producing a 10"6 risk under the linear method. Given that the allowance for slope is
clearly too small, this raises the question of whether the nonlinear method is
meaningfully different from the linear method. It differs mostly in relying on ill-defined
arguments for how large the margin-of-exposure must be in order to be deemed safe.
Response: A number of interesting and important issues are also raised in this comment
that apply to chloroform, but perhaps more importantly, to the ongoing revisions of the
cancer guidelines of EPA. In part based on this comment, specific text changes were
A-10
-------
made in the chloroform text to downplay the numerous discussions of factors with
various pieces of the data. The end result was the choice of a 1,000-fold MOE. This
choice is consistent with the overall amount of available data for chloroform, but also
accounts for potential sensitive populations. The use of MOE is new in EPA dose-
response assessment for carcinogens. Improvements to the process are likely to occur as
additional chemicals are considered for this procedure.
Comment:
It seems unfortunate not to explore the implications of internal dose measures more thoroughly.
A lot has been done in this arena, and an interesting analysis was conducted in the ILSI Panel
document. In that case, I had objections to the use of the ED 10 as a point of departure for similar
reasons that I raise above. Nonetheless, the dose-response analysis shows consistency among
different data sets when internal doses are used, allowing more confidence in the description of
the relationship in the observed range. I refer to my comments on the ILSI document on this
matter.
Response: This risk characterization text was intended to synthesize previous work into
a coherent picture for risk managers, and not to enhance the technical work of the ILSI or
other previous EPA deliberations. Although our understanding of the chloroform
database still can be improved, sufficient data are available to develop a nonlinear
approach to the estimation of low-dose cancer risk for chloroform within EPA's new
cancer guidelines.
Comment:
The case of chloroform is complex, difficult, and challenging. It is a formidable task to bring
rigorous discussion of the many issues into a single, readable, useful risk characterization. The
authors of the risk characterization have made a good try, but they have been hampered by the
lack of defined stances on the issues and clearly articulated bases for analytical approaches
among the source material. The risk characterization is supposed to summarize and
communicate these judgments, not to make them de novo. I am sympathetic with the challenge
faced by the authors, but I am compelled to review the document according to what ideally could
be accomplished, and what ideally should be covered in a fully developed risk assessment for a
compound with a database as rich as that of chloroform.
Many of my comments on the nonlinear method and the margin-of-exposure stem from doubts
about the method as set out in the Proposed Guidelines, especially regarding the wisdom of using
an ED 10 as a point of departure.
I do not disagree markedly with the basic conclusions of the risk characterization regarding the
likely levels of human risk from exposures as actually experienced, although I feel that the
invocation of marked non linearity between the animal response levels and the human exposures is
somewhat misleading. I would feature both the linear and nonlinear approaches, with a fuller
discussion of the ways in which each of these may be misleading if taken too literally. It is a
shame that the risk characterization has not examined the attempt by Evans and coworkers (1994,
Regul Toxicol Pharmacol 20:15-36) to create a distributional approach to chloroform potency
assessment, based on expert judgment regarding the relative credence to be put on a variety of
A-ll
-------
analytical approaches, including linear and nonlinear, administered doses versus internal
pharmacokinetic doses, cytotoxicity and genotoxicity, and other factors. This paper showed that
no single approach suffices, but that a collection of approaches, each weighted by its judged
likelihood to be appropriate, captures the full weight of evidence and span of possibilities.
I do feel that the document would substantially profit from more discussion of the pros and cons
of various approaches and a more forthright discussion of how the Agency views each of the
contentious and problematic issues. The document should show more explicitly what was done
and should discuss more thoroughly why things were done in that way.
Response: These comments were used to change the text in several ways in order to
create an approximation of what the commentor intended, within the overall scope and
intent of a risk characterization document.
R. Julian Preston, Ph.D.
Chemical Industry Institute of Toxicology
Comment:
As discussed, there are no data that indicate that chloroform induces tumors in humans.
However, there is sufficient evidence showing that chloroform can induce tumors in rodents (rats
and mice), but that the particular response can vary with route of exposure, administration
vehicle, and strain of rodent. Thus, establishing which data set is the most appropriate one for use
in estimating cancer risk to humans is not particularly easy. The report utilizes the incidence of
renal tumors in male rats exposed to chloroform in drinking water (Jorgenson et al., 1985). The
justification is not very clearly laid out, although it would appear to be an appropriate choice,
especially absent a published report of the data from Matsushima et al. (1994).
Response: Text was improved to justify the chosen study. Further justification can be
found on EPA's IRIS (U.S. EPA, 1998).
Comment:
Although evidence for relevant effects in humans suggesting carcinogenic effects of chloroform
is not strong, the fact that chloroform induces male liver tumors in female Wistar rats and kidney
tumors in BDF1 male mice and Osborne-Mendel rats is suggestive of the potential for human
carcinogenicity in a qualitative sense. At the same time the variability of response according to
strain, especially absence in some strains, is worthy of somewhat more discussion in the report.
In particular, there is a need to discuss the choice of data set for modeling more detail.
Response: These points were further discussed in the text as indicated.
Comment:
The very weak genotoxicity or absence of genotoxicity argues quite effectively that the pathway
to a tumor involves the indirect production of mutations, bearing in mind that mutations have to
be formed by some mechanism in order that a tumor can develop via its multistep process.
A-12
-------
Response: Text was enhanced to address this issue. Background mutations were
suggested as one source of initiated cells.
Comment:
The discussion of mode of action in the report is generally clear and accurate. The rather close
relationship between regenerative cell proliferation as a result of cytotoxicity and the induction of
tumors suggests a link between the two. However, it needs to be made clearer that cell
proliferation is perhaps a necessary step but clearly not a sufficient step by itself for tumor
induction. Something else is needed, albeit weak genotoxicity, oxidative DNA damage, other
secondarily produced DNA damages, or even an error-prone replication net result would seem to
result in an overall dose response for tumors that is nonlinear and perhaps reflected by the curve
for cell proliferation. This avoids the concern that is often expressed that tumors do not always
arise when cell proliferation is increased (and perhaps also the corollary that tumors arise absent
cell proliferation) by chloroform treatment.
Response: Text was added to clarify the distinction between the induction and the
subsequent growth of tumors induced by the cell proliferation. Text was also added to
further enhance the issue of oxidative DNA damage.
Comment:
This is a clearly written, concise report that follows a defined path to arrive at a risk assessment
for human renal tumors from drinking water exposure. The choice of the rodent cancer data for
the assessment needs to be justified more clearly. The mode of action needs to describe a process
whereby tumors will result, given that cancer is a genetic disease. The Report successfully
follows an approach consistent with EPA's 1996 Proposed Guidelines for Carcinogen
Assessment.
Response: The choice and justification of rodent model was more clearly described;
reference was made to EPA's IRIS where further support is shown. Statements were
added to more clearly describe the cancer process as one that involves several steps, one
of which is genetic damage.
A-13
-------
Mode of Action SAB Review
(Letter from the Assistant Administrator of the Office of Water to the interim chair)
Dr. Morton Lippmann, Interim Chair
Science Advisory Board
Dr. Mark Utell, Co-Chair
Dr. Richard Bull, Co-Chair
Chloroform Risk Assessment Review Subcommittee
Science Advisory Board
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avenue, N.W. (1400A)
Washington, DC 20460
Subject: Review of the EPA's Draft Chloroform Risk Assessment
Dear Dr. Lippmann, Dr. Utell, and Dr. Bull:
The Agency appreciates the efforts of the Science Advisory Board (SAB) in conducting a
review of the cancer risk assessment for chloroform in October 1999. We requested a two-part
review of the assessment. One part was to address issues about the Carcinogen Risk Assessment
Guidelines. The second part was to address mode of action data and epidemiologic information.
The SAB reported its findings on the first part of the review questions on December 15,
1999, and a full report, including the second part, on April 28, 2000. The Agency appreciates
these helpful and timely actions. This response is to the Report as it addresses the following
questions as quoted in the Report:
In the draft chloroform risk assessment document, are the conclusions as to the
following issues adequately supported by the analyses presented in the health risk
assessment/characterization (as supported by the ILSI report) and the framework
analysis?
1) chloroform's mode of action
2) consideration of a nonlinear approach to dose-response,
and the possibility that mutagenesis might play a role in
the carcinogenic response.
3) the relationship of low-dose pathology to the doses that
induce tumors
4) epidemiologic evidence on chlorinated drinking water
as to the carcinogenicity of chloroform, including
comment on any conclusion to be drawn from the
epidemiologic data about mode of action.
A-14
-------
Does the assessment of children's risk for chloroform appropriately
address the risk concerns, including ontogeny of drug metabolizing
enzymes, given the data available?
The Office of Water agrees with the recommendations in the report as to the chloroform
assessment and is incorporating them in a new Toxicological Review that is being developed in
support of the Agency's Integrated Risk Information System. The enclosed summary addresses
the SAB's major overall findings and recommendation.
The Agency looks forward to further dialogue with the SAB as we move forward in our risk
assessments supporting implementation of the Safe Drinking Water Act.
Sincerely,
J. Charles Fox
Assistant Administrator
Enclosure
A-15
-------
Response to Science Advisory Board Recommendations
In October 1999 the Chloroform Risk Assessment Review Subcommittee of the Science
Advisory Board met to consider the Office of Science and Technology health assessment of
chloroform. Summaries of the major parts of the subcommittee's advice and our responses
follow. The documents reviewed were a final hazard and dose-response characterization
document and a draft mode-of-action framework analysis.
1. The Subcommittee agreed with EPA that sustained or repeated
cytotoxicity with secondary regenerative hyperplasia in the liver and/or
the kidney of rats and mice precedes, and is probably a causal factor
for, hepatic and renal neoplasia. Some members of the subcommittee
were concerned about possible mutagenic activity, and the
subcommittee recommended that the risk assessment further address
the possible role of mutagenicity as a mode of action.
The Office of Water will include a more complete analysis of mutagenic potential in its
upcoming Toxicological Profile of Chloroform.
2. The subcommittee concluded that a nonlinear margin-of-exposure
approach is scientifically reasonable for the liver tumor response
because of the strong role cytotoxicity appears to play. In contrast, the
application of the standard linear approach to the liver tumor data is
likely to substantially overstate the low-dose risk. In addition, there is
considerable question about this response because it is not produced
when chloroform is administered to mice in drinking water.
For the kidney response — because sustained cytotoxicity plays a clear role in the rat —
margin-of-exposure is a scientifically reasonable approach. Most members of the
subcommittee thought that genotoxicity might possibly contribute to low-dose response
in this organ, while some thought it unlikely.
The Office of Water will consider and address this possibility in the Toxicological
Review in conjunction with discussions of metabolism and mutagenicity.
3. The subcommittee concluded that the extensive epidemiologic
evidence relating drinking water disinfection (specifically chlorination)
with cancer has little bearing on the determination of whether
chloroform is a carcinogen. It added recommendations for discussion
of endpoints and the potential meaning of these data to the assessment of
chloroform.
The Office of Water notes that the hazard and dose-response assessment document
reviewed did not contain the complete analysis of epidemiologic studies and the population-
attributable risk analysis. The latter were separately provided to the subcommittee. The
A-16
-------
opportunity will be taken to provide the recommended contextual discussion drawn from these
documents to the Toxicological Review on Chloroform.
4. The subcommittee found that the draft mode-of-action document
addressed children's risks quite adequately, based on the scientific
information currently available. The major conclusions were believed
correct, the role of CYP2E1 should be expressed as important, but its
definitive role in the developing human or (other) mammal has yet to
be confirmed. Nevertheless, the subcommittee report discussed
several areas of the discussion of children's risk that can be improved,
including exposure latency and transplacental and transmamillary
exposure.
The Office of Water will address the needed improvements in its Toxicological Review.
As the advice on some issues appears to be applicable beyond the chloroform assessment and
carry implications for Agency guidance documents, the advice will be discussed with the EPA
Risk Assessment Forum.
A-17
-------
APPENDIX B
QUANTITATIVE DOSE-RESPONSE MODELING
Bl: NON-CANCER BENCHMARK DOSE ANALYSIS
B2: CANCER DOSE-RESPONSE MODELING
(see attached electronic file modeling.zip)
B-l
------- |