EPA/635/R-00/001
TOXICOLOGICAL REVIEW
OF
1,3-DICHLOROPROPENE
(CAS No. 542-75-6)
In Support of Summary Information on the
Integrated Risk Information System (IRIS)
May 2000
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.
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.
11
-------
CONTENTS—TOXICOLOGICAL REVIEW FOR 1,3-DICHLOROPROPENE
(CAS No. 542-75-6)
FOREWORD v
AUTHORS, CONTRIBUTORS, AND REVIEWERS vi
1. INTRODUCTION 1
2. CHEMICAL AND PHYSICAL INFORMATION RELEVANT TO ASSESSMENTS 2
3. TOXICOKINETICS RELEVANT TO ASSESSMENTS 3
3.1. ABSORPTION 3
3.2. DISTRIBUTION 5
3.3. METABOLISM 6
3.4. EXCRETION 7
4. HAZARD IDENTIFICATION 10
4.1. STUDIES IN HUMANS-EPIDEMIOLOGY, CASE REPORTS, CLINICAL
CONTROLS 10
4.2. PRECHRONIC AND CHRONIC STUDIES AND CANCER BIO AS SAYS IN
ANIMALS-ORAL AND INHALATION 14
4.2.1. Inhalation Studies 14
4.2.2. Oral Studies 21
4.3. REPRODUCTIVE/DEVELOPMENTAL STUDIES-ORAL AND INHALATION . . 29
4.4. OTHER STUDIES 32
4.4.1. Acute Toxicity 32
4.4.2. Neurotoxicity 33
4.4.3. Mutagenicity 33
4.4.4. Mechanistic Studies 36
4.5. SYNTHESIS AND EVALUATION OF MAJOR NONCANCER EFFECTS AND
MODE OF ACTION (IF KNOWN)—ORAL AND INHALATION 37
4.5.1. Inhalation Studies 38
4.5.2. Oral Studies 39
4.6. WEIGHT-OF-EVIDENCE EVALUATION AND CANCER CLASSIFICATION-
SYNTHESIS OF HUMAN, ANIMAL, AND OTHER SUPPORTING EVIDENCE,
CONCLUSIONS ABOUT HUMAN CARCINOGENICITY, AND LIKELY
MODE OF ACTION 39
4.7. SUSCEPTIBLE POPULATIONS 42
4.7.1. Possible Childhood Susceptibility 42
4.7.2. Possible Gender Differences 42
5. DOSE-RESPONSE ASSESSMENTS 43
5.1. ORAL REFERENCE DOSE 43
5.1.1. Choice of Principal Study and Critical Effect—With Rationale
and Justification 43
5.1.2. Methods of Analysis—Benchmark Dose Analysis 44
iii
-------
CONTENTS (continued)
5.1.3. RfD Derivation—Including Application of Uncertainty Factors (UF)
and Modifying Factors (MF) 45
5.2. INHALATION REFERENCE CONCENTRATION (RfC) 45
5.2.1. Choice of Principal Study and Critical Effect—With Rationale
and Justification 45
5.2.2. Methods of Analysis—Benchmark Concentration Analysis 46
5.2.3. RfC Derivation—Including Application of Uncertainty Factors (UF)
and Modifying Factors (MF) 47
5.3. CANCER ASSESSMENT 48
5.3.1. Oral Exposure—Choice of Study/Data With Rationale and Justification 48
5.3.2. Inhalation Exposure—Choice of Study/Data With Rationale and
Justification 52
6. MAJOR CONCLUSIONS IN THE CHARACTERIZATION OF
HAZARD AND DOSE RESPONSE 54
6.1. HUMAN HAZARD POTENTIAL 54
6.2. DOSE RESPONSE 55
6.2.1. Noncancer Dose-Response Assessment 55
6.2.2. Cancer Dose-Response Assessment 56
7. REFERENCES 57
APPENDIX A. DOSE-RESPONSE CALCULATIONS 62
APPENDIX B. EXTERNAL PEER REVIEW—SUMMARY OF COMMENTS
AND DISPOSITION 72
IV
-------
FOREWORD
The purpose of this Toxicological Review is to provide scientific support and rationale
for the hazard and dose-response assessment in IRIS pertaining to chronic exposure to 1,3-
dichloropropene. It is not intended to be a comprehensive treatise on the chemical or
toxicological nature of 1,3-dichloropropene.
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 Risk Information Hotline at 202-566-1676.
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Chemical Manager/Author
The role of the chemical manager is to develop the draft document with input from
internal and external reviewers. The final document reflects the Agency consensus position on
the health effects of the chemical and does not necessarily reflect the opinion of the chemical
manager.
Judy A. Strickland. Ph.D., D.A.B.T.
National Center for Environmental Assessment
Office of Research and Development
Research Triangle Park, NC
Contributor
Karen A. Hogan, M.S.
National Center for Environmental Assessment
Office of Research and Development
Reviewers
This document and summary information on IRIS have received peer review both by EPA
scientists and by independent scientists external to EPA. Subsequent to external review and
incorporation of comments, this assessment has 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, Planning, and Evaluation; and the Regional Offices.
Internal EPA Reviewers
Karl Baetcke, Ph.D.
Health Effects Division
Office of Pesticide Programs
William Burnam, Ph.D.
Health Effects Division
Office of Pesticide Programs
Vicki Dellarco, Ph.D.
Health Effects Division
Office of Pesticide Programs
Julie T. Du, Ph.D.
Health and Ecological Criteria Division
vi
-------
Office of Science and Technology
Office of Water
Stanley B. Gross, Ph.D., D.A.B.T., C.I.H.
Health Effects Division
Office of Pesticide Programs
E.M. Kenyon, Ph.D.
Experimental Toxicology Division
National Health & Environmental Effects Research Laboratory
Nancy McCarroll, Ph.D.
Health Effects Division
Office of Pesticide Programs
Robert E. McGaughy, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
Alberto Protzel, Ph.D.
Health Effects Division
Office of Pesticide Programs
Esther Rinde, Ph.D.
Health Effects Division
Office of Pesticide Programs
RoyL. Smith, Ph.D.
Emission Standards Division
Office of Air Quality Planning and Standards
External Peer Reviewers
James Bruckner, Ph.D.
College of Pharmacy
University of Georgia
C. Clifford Conaway, Ph.D.
Division of Carcinogenesis & Molecular Epidemiology
American Health Foundation
James E. Klaunig, Ph.D.
Pharmacology & Toxicology Department
Indiana University School of Medicine
vn
-------
Summaries of the external peer reviewers' comments and the disposition of their
recommendations are in Appendix B.
Vlll
-------
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
carcinogenicity 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
noncancer 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 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 1,3-
dichloropropene 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: 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 Carcinogen Risk
Assessment (U.S. EPA, 1996a), Guidelines for Reproductive Toxicity Risk Assessment (U.S.
EPA, 1996b), and Guidelines for Neurotoxicity Risk Assessment (U.S. EPA, 1998a);
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
Concentration Issues 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, 1995); 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.
1
-------
Literature search strategies employed for this compound were based on the Chemical
Abstracts Service Registry Number (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 RELEVANT TO ASSESSMENTS
1,3-Dichloropropene is known as 3-chloroallyl chloride, alpha-chloroallyl chloride,
gamma-chloroallyl chloride, 3-chloropropenyl chloride, 1,3-dichloropropylene, alpha, gamma-
dichloropropylene, 1,3-dichloro-l-propene, DCP, and Telone II®. Some relevant physical and
chemical properties of 1,3-dichloropropene are listed below (Hazardous Substances Data Base
[HSDB], 1998).
CASRN: 542-75-6
Empirical formula: C3H4C12
Molecular weight: 110.98
Vapor pressure: 3.7Paat20°C
Density: 1.220 at 25°C
Boiling point: 108°C
Water solubility: 0.15%
Log Kow: 1.36 (cis isomer); 1.41 (trans isomer)
Conversion factor: 1 ppm = 4.54 mg/m3 at 25°C; 1 mg/m3 = 0.22 ppm
At room temperature, 1,3-dichloropropene is a colorless to straw-colored liquid with a
sharp, sweet, penetrating, chloroform-like odor (HSDB, 1998). It is miscible in most organic
solvents and evaporates easily.
1,3-Dichloropropene is used extensively in agriculture as a preplanting fumigant,
primarily for the control of nematodes affecting the roots of plants. It is one of the few
remaining relatively inexpensive fumigants currently available; the registrations of similar
fumigants, e.g., l,2-dibromo-3-chloropropane (DBCP) and ethylene dibromide (EDB), have been
suspended for most agricultural uses. Therefore, 1,3-dichloropropene is an agriculturally
important pesticide, with a high annual volume of use throughout the United States and abroad.
Commercial formulations of 1,3-dichloropropene (Telone II® soil fumigant) contain mixtures of
cis (Z) and trans (E) isomers. The older formulations of technical-grade 1,3-dichloropropene
called Telone n® contained approximately 89% cis- and trans-1,3-dichloropropene, 2.5% 1,2-
dichloropropene, 1.5% of a trichloropropene isomer, and 1% epichlorohydrin (HSDB, 1998).
Since about 1988, formulations of have replaced epichlorohydrin, the stabilizing agent, with
epoxidized soybean oil. Commercial formulations that contain other dichloropropenes or
dichloropropanes and other chemicals include D-D, Di-Trapex, and Vorlex.
The presence of other active ingredients and stabilizers in commercial formulations is a
confounder in establishing the toxicity of 1,3-dichloropropene. Pure dichloropropene may
-------
contain confounding chemicals as well, as it is unstable when exposed to heat and oxygen or heat
and light (Watson et al., 1987). Degradation products will also form at room temperature if
dichloropropene is stored for several weeks in the presence of oxygen (Watson et al., 1987).
3. TOXICOKINETICS RELEVANT TO ASSESSMENTS
1,3-Dichloropropene toxicokinetics in humans appear to be similar to those observed in
rodents. Inhalation studies with both humans and animals have shown that 1,3-dichloropropene
vapors are readily absorbed, conjugated with glutathione (GSH) via glutathione S-transferase
(GST), and rapidly excreted in the urine as N-acetyl-(S-3-chloroprop-2-enyl)cysteine (3CNAC),
a mercapturic acid metabolite (see Figure 1). Thus, the major metabolic pathway for 1,3-
dichloropropene leads to its detoxification and excretion. Ingestion studies in animals have
demonstrated that the toxicokinetics of oral exposures are similar to those of inhalation
exposures. 1,3-Dichloropropene is unlikely to accumulate in the body.
3.1. ABSORPTION
Stott and Kastl (1986) studied the inhalation pharmacokinetics of technical-grade 1,3-
dichloropropene by exposing male Fischer 344 (F344) rats to mean vapor concentrations of 30,
90, 300, and 900 ppm (136, 409, 1,363, and 4,086 mg/m3, respectively) for 3 hours. These air
concentrations produced vapor uptakes of 147, 307, 880, and 1,810 nanomoles per minute, and
corresponding absorption fractions of 82%, 65%, 66%, and 62%, respectively. Based upon the
uptake of dichloropropene vapors, average amounts of dichloropropene absorbed by rats over the
3-hour exposure period were approximately 14, 29, 85, and 171 mg/kg in the 136, 409, 1,363,
and 4,086 mg/m3 exposure groups, respectively. Even though the rate of uptake increased with
increasing exposure, the increase was not linear at higher concentrations. The decrease in vapor
uptake at higher concentrations was associated with an exposure-related depression in ventilatory
frequency, which was statistically significant at 409 mg/m3 and higher. Stott and Kastl (1986)
indicate that Alarie (1973) observed exposure-related depression in ventilatory frequency with
numerous respiratory irritants; they suggest that 1,3-dichloropropene is a respiratory irritant.
The major site of absorption of inhaled 1,3-dichloropropene in the rat is the lung rather
than the nasal mucosa (Stott and Kastl, 1986). The localized uptake of vapors in rats exposed to
90 or 150 ppm (409 or 682 mg/m3, respectively) was examined by surgically isolating the upper
and lower respiratory tract. The lower respiratory tract absorbed approximately 50% of inhaled
dichloropropene vapors whereas the upper respiratory tract absorbed only 11%-16% of vapors.
Total absorption rates were approximately 73% and 79% at 409 and 682 mg/m3, respectively.
In 1992, Waechter et al. showed that absorption of 1,3-dichloropropene from inhalation
exposure in humans was similar to absorption in rats (Stott and Kastl, 1986). Six male
volunteers were exposed to 1 ppm (4.54 mg/m3)l commercial Telone If8 (50.6% cis isomer,
Calculated using conversion of 1 ppm = 4.54 mg/m3 at 25° C.
3
-------
Minor
H O
\ /\
C-C
/
Cl
H
H2O
\
CH2CI
H Cl
ii
OHC-C-C-H
I i
OH H
1, 3-dichloropropene epoxide 3-chloro-2-hydroxy-propanal
Cytochrome P450
Cl - CH - CH - CH2 - Cl
1 , 3-dichloropropene
Primary
GSH-S-transferase +GSH
Cl
Cl
N-
Cl-
i '
o
ii
NH - C - CH2-CH2-CH -COOH
l l
- CH = CH - CH2 -S - CH2 - CH NH2
i r
- CH = CH - CH2 - S - CH2 -
i
C = O
l
NH - CH2 - COOH
NH - COCH3
CH2
COOH
acetyl - S - (3-chloroprop-2-enyl)
cysteine (3CNAC)
1 O NH - COCH,
ii
l
CH = CH - CH2 - S - CH2 - CH2
3CNAC sulfoxide
l
COOH
/-roposeo
+H20
-HCI
1 '
Cl - CH = CH - CH2 - OH
1-chloroallyl alcohol
alcohol
dehydrogenase
i '
O
ii
Cl - CH = CH - CH
1-chloroacrolein
I aldehyde
| dehydrogenase
9 1
HC-CH2-COOH
malonyl semi-aldehyde
1 ^ CQ
Acetyl CoA
i
1
TCA cycle
1 n NH - COCH,
Cl - CH = CH - CH2 - S - CH2 - CH2
l
O
3CNAC sulfone
COOH
Figure I. Metabolic pathways for 1,3-dichloropropene. Derived from Waechter
and Kastl (1988) and Schneider et al. (1998a).
-------
45.2% trans isomer) for 6 hours. The absorption of cis-l,3-dichloropropene was 72%-80%
while the absorption of trans-1,3 -dichloropropene was 77%-82%. A similar percentage of
absorption (i.e., 82%) was found in rats exposed to 136 mg/m3 1,3-dichloropropene for 3 hours
(Stott and Kastl, 1986). Waechter et al. (1992) also found that the concentration of 1,3-
dichloropropene in expired air plateaued in the first hour of exposure and declined rapidly
postexposure to nondetectable levels in less than an hour.
Stott et al. (1998) showed that the pharmacokinetics of 1,3-dichloropropene
microencapsulated in a starch-sucrose matrix were similar to those of neat 1,3-dichloropropene
administered by gavage to F344 rats. Female rats were simultaneously gavaged with 25 mg/kg
neat 13C-dichloropropene and 25 mg/kg microencapsulated 12C-dichloropropene in a corn oil
suspension. Blood concentrations were measured at various intervals by gas
chromatography/mass spectrometry (GC/MS). The absorption half-life of neat dichloropropene
was 2.5 minutes for the cis isomer and 2.7 minutes for the trans isomer, while the absorption
half-life for encapsulated dichloropropene was 1.3 minutes for the cis isomer and 2.3 minutes for
the trans isomer. The elimination half-lives from blood were also similar. The cis/trans
elimination half-lives for the alpha phase were 6.1 ± 0.9/6.2 ± 0.5 minutes for neat
dichloropropene and 6.9 ± 1.6/6.6 ± 0.6 minutes for encapsulated dichloropropene. The cis/trans
elimination half-lives for the beta phase were 32 ± 18/22 ± 5 minutes for neat dichloropropene
and 20 ± 8/29 ± 5 minutes for encapsulated dichloropropene.
3.2. DISTRIBUTION
In the study by Stott and Kastl (1986) that examined the inhalation pharmacokinetics of
technical-grade 1,3-dichloropropene in rats, the time required for blood concentrations of cis-
and trans-dichloropropene to plateau was dependent on the exposure concentration. In the 136
and 409 mg/m3 groups, blood concentration plateaued after 1 hour of exposure. In the 1,363
mg/m3 group, 2-3 hours were required. Blood levels of dichloropropene in the 4,086 mg/m3
group did not reach a steady state during the 3-hour exposure period. In general, higher blood
levels of trans-dichloropropene than cis-dichloropropene were observed in rats at all vapor
concentrations, even though the cis:trans ratio of the technical-grade dichloropropene used in the
study was 1.2:1. These findings are consistent with the faster metabolism of the cis isomer.
The human pharmacokinetic study by Waechter et al. (1992) also found that blood
concentrations plateaued rapidly at low exposures for most subjects. In five of the six subjects,
blood levels of 1,3-dichloropropene reached an apparent plateau within 1 hour of exposure to
4.54 mg/m3. Blood levels in the remaining subject increased throughout the exposure. Also, as
in the rat study by Stott and Kastl (1986), trans-1,3-dichloropropene reached higher blood
concentrations than did cis-l,3-dichloropropene, even though the cis:trans ratio of the technical-
grade dichloropropene used in the study was 1.1 to 1.
Distribution studies in rats after gavage administration of 1,3-dichloropropene indicate
that the forestomach, glandular stomach, kidney, liver, and bladder are primary organs of
distribution compared with fat, skin, blood, and remaining carcass. Dietz et al. (1985) studied
the amount of 14C-activity in these tissues of male F344 rats and B6C3F1 mice 48 hours after a
-------
single oral dose of 14C-l,3-dichloropropene (1 or 50 mg/kg to rats and 1 or 100 mg/kg to mice).
At 1 mg/kg, forestomach and bladder had the highest 14C-activities in both species and were
followed by liver, kidney, and glandular stomach. 14C-activity in the remaining tissues was much
less. At the high dose, forestomach and kidney had the highest 14C-activities in both species. In
rats, these were followed by glandular stomach, liver, and bladder; in mice, they were followed
by liver, fat, bladder, and glandular stomach. Because of the rapid metabolism and excretion of
1,3-dichloropropene, described in Sections 3.3 and 3.4, the 14C-activities measured 48 hours after
single doses actually represent metabolized dichloropropene rather than the parent compound.
3.3. METABOLISM
Climie et al. (1979) determined that GSH conjugation is the major metabolic pathway of
cis-l,3-dichloro[14C]propene after oral administration to female Wistar rats (see Figure 1). A
hepatic GST catalyzes the conjugation of cis-l,3-dichloropropene with GSH. The conjugate is
further metabolized to a mercapturic acid, cis-3CNAC, and is excreted in the urine. This
metabolite accounted for 92% of the 0 to 24-hour cumulative urinary radioactivity. In vitro
metabolic studies using rat liver preparations showed that the trans isomer is degraded similarly,
but at a much slower rate (four to five times more slowly).
On the basis of pharmacokinetic studies, Dietz et al. (1985) concluded that the major
pathway of metabolism and detoxification of both cis and trans isomers of 1,3-dichloropropene
occurred via conjugation with GSH in male F344 rats and B6C3F1 mice after oral
administration. Additionally, 3CNAC and its sulfone derivative were identified as the two major
urinary metabolites (Figure 1). No parent compound was detected in the urine. 1,3-
Dichloropropene undergoes substantial first-pass metabolism that follows linear
pharmacokinetics over an oral gavage dose range of 1-100 mg/kg for mice and 1-50 mg/kg for
rats (Dietz et al., 1985). Waechter et al. (1992) showed that the major metabolites after
inhalation in humans, cis- and trans-3CNAC, were identical to those found in rats after oral
exposure.
Although the major metabolic pathway of 1,3-dichloropropene is conjugation by GSH,
Schneider et al. (1998a) found that epoxidation of 1,3-dichloropropene is a minor metabolic
pathway in mouse liver at ~LD50 doses. Schneider et al. (1998a) administered either 350 mg/kg
individual isomer or 700 mg/kg cis/trans-l,3-dichloropropene to male Swiss-Webster mice by
intraperitoneal (ip) injection and then measured epoxide formation in the liver at various times
up to 150 minutes later. GC/MS measurements showed that 1,3-dichloropropene concentrations
in the liver peaked about 10 minutes after treatment and then decayed through apparent first-
order kinetics, with half-lives of 36 minutes for the cis isomer and 50 minutes for the trans
isomer. Epoxide concentrations were approximately two orders of magnitude lower than those
of the parent compound. In in vitro experiments, Schneider et al. (1998a) demonstrated that
conjugation of 1,3-dichloropropene with GSH decreases epoxide formation in mouse liver.
Dietz et al. (1985) showed that oral administration of 1,3-dichloropropene depletes tissue
levels of nonprotein sulfhydryls (NPSH), an indication of GSH levels. A single gavage dose of
0, 1, 5, 25, 50, or 100 mg/kg 14C-l,3-dichloropropene was administered to male F344 rats and
-------
B6C3F1 mice. NPSH levels in selected tissues were measured 2 hours after administration.
Significant depression of NPSH levels, to 51%-17% of control values, were observed in the
forestomachs of rats and mice dosed with 25-100 mg/kg. Less severe, dose-dependent
depression, to 70%-45% of control for the same doses, also occurred in the glandular stomach.
Depression of NPSH levels in the liver was less than that in the forestomach and glandular
stomach, with the rat liver about twice as sensitive as the mouse liver. There were no statistically
significant changes in NPSH levels in the kidney or urinary bladder of either rats or mice. The
no-observed-effect levels (NOELs) for NPSH depletion in forestomach were 1 mg/kg for rats and
5 mg/kg for mice. Covalent binding of 1,3-dichloropropene to macromolecules in the
forestomach, glandular stomach, liver, kidney, and urinary bladder in male F344 rats and male
B6C3F1 mice was also studied by Dietz et al. (1985). A single gavage dose of 14C-1,3-
dichloropropene of 1, 50, or 100 mg/kg was administered, and binding was measured after 2
hours. Macromolecular covalent binding increased with increasing dose. Binding was highest in
the forestomach and glandular stomach and lowest in the liver, kidney, and bladder, findings that
correlated with the magnitude of tissue-specific decreases in NPSH.
3.4. EXCRETION
When Hutson et al. (1971) administered 2.53-2.70 mg of either cis- or trans-1,3-dichloro-
[2-14C]propene by gavage to Carworth Farm E rats, 80%-90% of the radiolabel was eliminated in
the feces, urine, or expired air during the first 24 hours of the experiment. Within 24 hours,
80.7% of the administered cis isomer and 56.5% of the trans isomer were eliminated in the urine.
Approximately 3.9% of the cis isomer and 23.6% of the trans isomer were recovered in expired
air as [14C]carbon dioxide (see proposed pathway in Figure 1). A small amount (l%-4%) of
unmetabolized 1,3-dichloropropene was exhaled directly. After 4 days, about 1% of the
administered dose of each isomer was found in the carcass. Thus, the rat retains little ingested
1,3-dichloropropene after oral administration.
Dietz et al. (1984a) administered by gavage 1 or 50 mg/kg 14C-cis, trans-1,3-
dichloropropene to male rats and 1 or 100 mg/kg to male B6C3F1 mice. Elimination was
measured by the amount of radioactivity present in expired air, urine, and feces. Urinary
excretion was the predominant route of elimination during the 48 hours after dosing, accounting
for 51%-61% of the administered dose in rats and 63%-79% in mice. Feces and expired carbon
dioxide contained about 18% and 5%, respectively, of the administered radioactivity in rats, and
15% and 14%, respectively, of the administered radioactivity in mice. Only 2%-6% of the
original dose remained in the carcasses at the end of 48 hours. The predominant urinary
metabolite, identified as 3CNAC, confirmed the earlier findings by Climie et al. (1979). The
sulfoxide or sulfone derivative of 3CNAC was tentatively identified as (an)other metabolite(s).
Dietz et al. (1984a) calculated a urinary excretion half-life of approximately 5.5 hours for
dichloropropene in rats and mice.
In the Stott and Kastl (1986) study in which F344 male rats were exposed to 136, 409,
1,363, and 4,086 mg/m3 1,3-dichloropropene for 3 hours, a pronounced rapid elimination phase
was observed in all rats exposed to 1,363 mg/m3 or less. In this initial phase, the half-life of cis-
dichloropropene was calculated at 3-5 minutes for animals exposed to 136, 409, and 1,363
7
-------
mg/m3 and increased to more than 14 minutes for animals exposed to 4,086 mg/m3. Rats
exposed to trans-dichloropropene had a longer first-phase elimination half-life, averaging 6
minutes for the 136, 409, and 1,363 mg/m3 groups and 27 minutes for the 4,086 mg/m3 group.
Following this first phase, both cis- and trans-dichloropropene exhibited a second, slower and
longer phase of elimination in rats exposed to 1,363 or 4,086 mg/m3, roughly 25 to 43 minutes,
independent of isomer or exposure concentrations. The initial phase of elimination primarily
represents the redistribution of dichloropropene from blood to tissues, whereas the second phase
of elimination is determined by the rate of metabolism. Disproportionately large increases in
blood levels at the end of exposure occurred in rats exposed to 4,086 mg/m3 cis-dichloropropene
and 1,363 and 4,086 mg/m3 trans-dichloropropene, which indicated nonlinear elimination at high
exposures. The longer half-lives and disproportionately higher blood levels at the higher doses
suggest that metabolism was saturated. The data also indicate that elimination of
dichloropropene at lower exposure levels is mediated primarily via metabolism and not via
simple exhalation of the parent compound, a result consistent with Hutson et al. (1971).
In the human study by Waechter et al. (1992), urinary excretion of 1,3-dichloropropene
was an apparent first-order process at an inhalation exposure of 4.54 mg/m3 for 6 hours. The
elimination half-lives for the initial phase were 4.2 ±0.8 hours for the cis isomer and 3.2 ± 0.8
hours for the trans isomer, whereas the half-lives for the terminal phase were 12.3 ± 2.4 hours
(cis isomer) and 17.1 ± 6 hours (trans isomer).
Fisher and Kilgore (1988a) conducted a series of experiments to assess the relationship
among dichloropropene inhalation, GSH reduction in tissues, and serum lactate dehydrogenase
(LDH) activity. These studies demonstrate that very high inhalation exposures of
dichloropropene are required to produce significant decreases in GSH in all target organs except
nasal tissue. It should be noted that the technical-grade dichloropropene used in this study and
others by Fisher and Kilgore (1988b, 1989) contained epoxidized soybean oil as the stabilizing
agent instead of epichlorohydrin. Male Sprague-Dawley rats were exposed to 1,3-
dichloropropene for 1 hour at average concentrations of approximately 0, 2, 5, 33, 306, 771, 954,
or 1,716 ppm (0, 8, 20, 150, 1,390, 3,504, 4,334, or 7,791 mg/m3, respectively)2 for
determination of GSH levels in the heart, kidney, liver, lung, and testes; 0, 5, 31, 70, or 222 ppm
(0, 24, 143, 320, or 1,012 mg/m3, respectively)2 for measurement of GSH levels in nasal tissue;
and 0, 25, 660, or 2,277 ppm (0, 113, 2,995, or 10,336 mg/m3, respectively)2 for assessment of
lung dry weight/wet weight and serum LDH activity. The principal tissue affected by low
exposure concentrations was nasal tissue. GSH in nasal tissue decreased in a concentration-
dependent fashion to 27% at 24 mg/m3, 23% at 143 mg/m3, 18% at 320 mg/m3, and 12% at 1,012
mg/m3, compared to control values. Lung GSH also was decreased but remained relatively
constant at approximately 70% of control values at all exposure concentrations up to 4,334
mg/m3 and showed no evidence of further depletion. In the heart and testis, no significant
reductions were observed at concentrations up to 4,334 mg/m3. At the next highest dose of 7,791
mg/m3, however, lung, heart, and testis GSH levels were decreased significantly, whereas kidney
GSH levels were not. Liver GSH content showed an exposure-related decrease only between
3,504 and 7,791 mg/m3. There were no changes in any lung weight parameters (wet weight,
Calculated using conversion of 1 ppm = 4.54 mg/m3 at 25° C.
-------
percent wet weight/body weight, relative dry weight/wet weight, or dry weight/body weight) for
animals sacrificed either 2 or 6 hours postexposure. Serum LDH activity measured 6 hours after
dichloropropene exposure did not exhibit any significant changes except for a decrease at 10,336
mg/m3.
The dose-dependency of GSH metabolism was evaluated in another study by Fisher and
Kilgore (1988b). Male Sprague-Dawley rats were exposed to technical-grade dichloropropene
vapors, nose only, for 1 hour at concentrations up to 789 ppm (3,582 mg/m3)3 dichloropropene.
Urine was collected for 24 hours postexposure for measurement of the urinary mercapturic acid
metabolite of cis-dichloropropene, cis-3CNAC. The amount of urinary cis-3CNAC exhibited a
concentration-dependent increase in rats exposed from 0 to 284 ppm (1,289 mg/m3)3. However,
the amount of cis-3CNAC in the urine remained constant at exposures from 1,289 to 3,582
mg/m3, a finding consistent with saturation of metabolism at higher doses of dichloropropene as
suggested by the Stott and Kastl (1986) study.
Fisher and Kilgore (1989) postulated that dichloropropene entering the rat via inhalation
exposure is rapidly transformed into its GSH conjugate, (S-3-chloroprop-2-enyl)GSH (GSCP),
which is the precursor to the mercapturic acid metabolite 3CNAC (Figure 1), and that GSCP can
be measured in blood over time. In an initial range-finding study using male Sprague-Dawley
rats, the blood level of GSCP did not significantly change when measured at 15, 30, 45, and 60
minutes during a 1-hour exposure to 610 ppm (2,769 mg/m3)3 technical-grade dichloropropene.
In the main study, the concentrations of GSCP in blood following inhalation exposures to 78,
155, and 404 ppm (354, 704, and 1,834 mg/m3, respectively)3 dichloropropene were examined
using equations for both monophasic and biphasic decay. GSCP was detected in the blood of
rats at all concentrations. No significant differences were found between the regression lines of
the mathematical equations expressed as either monophasic or biphasic decay at any exposure
concentration. Thus, these results could fit either a one- or two-compartment model of
elimination. Moreover, no significant differences were found between the regression lines for
any exposure concentrations, which indicates that the elimination of GSCP was independent of
exposure concentration. The apparent half-life of GSCP was 17 hours. These findings suggest
that the formation of GSCP may occur by dose-independent mechanisms, or that the mechanisms
responsible for formation of GSCP may be saturated at the exposure levels selected for the
experiment. Alternatively, on the basis of the results of Stott and Kastl (1986), an initial rapid
phase of GSCP elimination may have been present but not detected in the Fisher and Kilgore
(1989) study because the first blood samples following exposure were collected after the initial,
rapid elimination phase had already occurred. The significance of these findings is unclear.
Two biological monitoring studies in humans have demonstrated that there is a dose-
dependent relationship between respiratory occupational exposure to 1,3-dichloropropene and
excretion of the urinary mercapturic acids, cis- and trans-3CNAC (van Welie et al., 1991). In
this study of 12 male workers in the flower bulb industry, exposure to cis- and trans-1,3-
dichloropropene measured by personal air samplers ranged from 0.3 to 18.9 mg/m3 during 1- to
11-hour shifts. Urinary excretion of 3CNAC followed first-order elimination kinetics following
Calculated using conversion of 1 ppm = 4.54 mg/m3 at 25° C.
9
-------
exposure. Urinary elimination half-lives of 5.0 ±1.2 hours for cis-3CNAC and 4.7 ±1.3 hours
for trans-3CNAC were not statistically different. The calculated coefficient of variation
indicated that the elimination half-lives of cis- and trans-3CNAC were quite consistent among
individuals. These human half-life values were similar to those reported by Waechter et al.
(1992) for a clinical study (4.2 ± 8 hours for cis and 3.2 ± 0.8 hours for trans) and to those found
by Dietz et al. (1984b) in rats and mice (i.e., 5.5 hours), van Welie et al. (1991) found high
correlations r2 = 0.93) between respiratory 8-hour time-weighted average (TWA) exposures to
cis- and trans-dichloropropene and cumulative urinary excretion of cis- and trans-3CNAC. cis-
Dichloropropene yielded three times more 3CNAC than trans-dichloropropene, consistent with
differences in the rate of metabolism between the isomers. Approximately 45% and 14%,
respectively, of air levels of cis- and trans-dichloropropene were excreted as mercapturic acid
metabolites (i.e., 3CNAC).
In a second study of 15 male applicators in the flower bulb industry (Osterloh et al.,
1989), the relationship between air concentrations of dichloropropene and urinary levels of
3CNAC and N-acetylglucosaminidase (NAG) was assessed. The release of NAG, a renal tubular
enzyme, may indicate low-level subclinical tissue injury. Breathing zone samples were
analyzed for dichloropropene. Urine samples were collected before and at various times after
application and analyzed for 3CNAC and NAG. Dichloropropene exposure concentrations were
0.26-939 mg/m3 with a mean of 2.56 mg/m3. Exposure durations from 120 to 697 minutes
yielded dichloropropene exposure rates of 62-3,700 mg/m3/min. The 24-hour urinary excretion
of 3CNAC was 0.50-9.17 mg, with a mean of 2.57 mg. Twenty-four-hour 3CNAC and NAG
urinary levels correlated well with dichloropropene exposure concentrations (r2 = 0.854). The
correlation increased when next-morning urine samples were used (r2 = 0.914). There was also a
correlation between next-morning 3CNAC levels and 24-hour urinary concentrations of NAG.
The overall mean excretion of NAG was 2.63 milliunits/mg creatinine. Four subjects had NAG
values in a clinically abnormal range, above 4 milliunits/mg. Nine subjects had increases in
NAG that averaged 25% higher than their baseline values. The authors concluded that excretion
of urinary 3CNAC is correlated with dichloropropene exposures. The investigators indicated
that NAG levels above the normal range in four workers suggest that a subclinical nephrotoxicity
may be associated with dichloropropene exposure. Alternatively, this increase may have been an
adaptive response in the kinetics of detoxification and excretion of 1,3-dichloropropene. Follow-
up urinary measurements were not conducted to evaluate whether levels returned to the normal
clinical range following cessation of exposure for 24 hours or longer.
4. HAZARD IDENTIFICATION
4.1. STUDIES IN HUMANS—EPIDEMIOLOGY, CASE REPORTS, CLINICAL
CONTROLS
4.1.1. Bousema, MT; Wiemer, GR; Van Joost; TH. (1991) A classic case of sensitization to
DD-95®. Contact Derm 24(2): 132-133
DD-95® is a nematocide containing approximately 95% 1,3- dichloropropene. A case
report is described in which a 44-year-old male process operator at a pesticide plant developed an
10
-------
acute bullous dermatitis on the dorsae of both feet. This event, first reported in August 1988,
recurred in September 1989. The operator was interviewed by medical personnel and recalled (a)
soiling his shoes with DD-95® about 10 days before he developed dermatitis the first time, and
(b) soiling his shoes again with the same compound 1 day before the dermatitis reappeared in
September 1989. The patient was subsequently patch-tested with DD-95® at 2.0%, 1.0%, 0.5%,
0.1%, 0.03%, and 0.005% and responded positively to all concentrations up to 3 days following
challenge. A control group of 20 volunteers were similarly tested at a concentration of 0.05%
DD-95®, but none showed any positive symptoms. The authors report that this is the third case
study that has reported an association between contact dermatitis and occupational dermal
exposure to DD-95®. The authors suggest that there is a small but distinct subgroup of
individuals working with pesticides who develop an allergic reaction upon dermal contact with
DD-95® and other pesticides containing mainly 1,3-dichloropropene.
4.1.2. Hernandez, AF; Martin-Rubi, JC; Ballesteros, JL; et al. (1994) Clinical and
pathological findings in fatal 1,3-dichloropropene intoxication. Hum Exp Toxicol
13(5):303-306
A 27-year-old previously healthy male, working on a friend's farm, accidentally drank an
unknown quantity of dichloropropene from a receptacle containing clear fluid because he thought
it was water. After he realized it was not water, he vomited. He was taken to the hospital
emergency department 2 hours later with acute gastrointestinal distress, sweating, tachypnoea,
tachycardia, hypovolemic disturbance, and lividity on both legs. Over the next several hours, the
patient developed respiratory and cardiac hypotension; metabolic acidosis; elevation of blood
glucose and subsequent insulin-resistant hyperglycemia; abdominal wall distension and other
gastrointestinal disturbances, including increased peritoneal fluid amylase levels, indicative of
pancreatic disorder. These acutely toxic effects progressed to extensive bilateral interstitial and
alveolar infiltration consistent with adult respiratory distress syndrome, as well as hemodynamic,
gastrointestinal, liver, and kidney deterioration. Death from multiple organ failure occurred 38
hours following admission to the hospital. The autopsy revealed edematic brain and lungs,
bloody-clear fluid in the abdominal cavity, congested spleen and lungs, and hemorrhagic exudate
from the stomach mucosa. Histological examination of the stomach revealed congestion of the
gastric mucosal vessels, autolysis, scattered mucosal erosions, and small foci of mononuclear
inflammatory cells. Histological examination of the liver showed architectural anomalies,
autolysis, sinusoidal congestion and small biliary thrombi.
Toxicological identification of the ingested compound by GC/MS confirmed that Telone H8
(cis- and trans-1,3-dichloropropene) was the fatal agent. The initial body-burden fluid
concentrations of the toxicant at the time of the emergency admission were 1.1 • mol/L in blood
and 0.2 • mol/L in urine.
11
-------
4.1.3. Kezic, S; Monster, AC; Verplanke, AJW; et al. (1996) Dermal absorption of cis-1,3-
dichloropropene vapor: human experimental exposure. Hum Exp Toxicol 15(5):396-399
Because cis-l,3-dichloropropene-based pesticides account for a large majority of the
total pesticides used in the Dutch flower bulb industry, the objective of this study was to estimate
the significance of skin absorption of cis-l,3-dichloropropene vapor compared with inhalation
uptake. Under controlled clinical exposure conditions, five adults (four males and one female)
were dermally exposed on 1,277 cm2 of the forearm and hand to 86 mg/m3 cis-1,3-
dichloropropene for 45 minutes. Urine samples were collected before and after the exposure
sessions and analyzed by GC/MS to determine levels of creatinine and cis-3CNAC. Urinary
excretion of cis-3CNAC peaked during the first hour after exposure and declined thereafter, with
an average half-life of 6 hours. The total amount of cis-3CNAC excreted over 24 hours averaged
48 (ig. The mean total uptake of the cis-l,3-dichloropropene was estimated at 67 j^g. The
subject with the highest urinary excretion of cis-3CNAC (90.4 |j.g) complained of skin irritation
experienced as tingling. The estimated permeability constant of 0.8 cm/hour for cis-1,3-
dichloropropene is consistent with permeability constants reported for two other dihalomethane
vapors in rats, dibromomethane and bromochloromethane.
To compare dermal absorption of cis-l,3-dichloropropene with absorption via inhalation,
Kezic et al. (1996) used data from inhalation studies by Waechter et al. (1992) and van Welie et
al. (1991). The data were normalized for identical exposure conditions: 8 hours exposure to 5
mg/m3 of cis-l,3-dichloropropene. A number of assumptions were made, including (a) dermal
uptake from the forearm surface area can be linearly extrapolated to the whole-body surface area;
(b) the permeability constant and metabolism of 1,3-dichloropropene are concentration-
independent over the range of concentrations used in the three studies, and (c) co-exposure to
trans-1,3-dichloropropene in the inhalation studies did not affect the human toxicokinetics of cis-
1,3-dichloropropene.
The results of these analyses demonstrate that dermal absorption upon vapor exposure to
dichloropropene does occur but is relatively minor in terms of total internal dose when compared
with inhalation. Dermal exposure to cis-l,3-dichloropropene vapors was estimated to account
for approximately 2%-4% of the total uptake under conditions of combined inhalation and
whole-body dermal exposures. However, in certain acute high-exposure situations, such as
accidental releases, dermal absorption as a route of entry may be more significant.
4.1.4. Markovitz, A; Crosby, WH. (1984) Chemical carcinogenesis. A soil fumigant,
1,3-dichloropropene, as possible cause of hematologic malignancies. Arch Intern Med
144:1409-1411
This study examined case reports of hematologic neoplasms following intoxicating
exposure to 1,3-dichloropropene. In two firefighters, histiocytic (non-Hodgkin's) lymphomas
appeared simultaneously several years after both men were exposed to 1,3-dichloropropene at the
site of a chemical spill. In both cases, the cancers were refractory to standard regimens of
treatment and the subjects died within a few months of one another. Six other firefighters,
simultaneously exposed, did not develop any malignancies. Leukemia developed in a farmer a
12
-------
few months after the right side of his head was exposed to 1,3-dichloropropene during soil
application of the chemical (the hose he was using had a leak). This exposure occurred for 30
days. The farmer suffered from a smoldering leukemia until, after a second series of daily
exposures 1 year later, the leukemia became extremely aggressive. He died of pneumonia in the
hospital during treatment for leukemia.
4.1.5. Nater JP; Gooskens, VHJ. (1976) Occupational dermatosis due to a soil fumigant.
Contact Dermatitis 2(4):227-229
The aim of this study was to determine whether occupational dermatitis resulting from
direct contact with 1,3-dichloropropene was due to an allergic or a primary irritant reaction.
Three cases of occupational skin contact with a common nematocide soil fumigant, D-D mixture,
were examined. The mixture contained 1,3-dichloropropene, 1,2-dichloropropane, and
epichlorohydrin. Patient 1 received two 1-week exposures 1 year apart and developed an itching
erythematous rash. Patient 2 developed the rash after a single exposure. Patient 3 was employed
spraying pesticides on a daily basis for 10 years between September and January. After 7 years
he developed dermatitis on his arms, face, and ears, which subsided upon avoidance of the
nematocide. Patch testing was performed with D-D, other preparations of 1,3-dichloropropene,
and 1,2-dichloropropane at 1% in acetone (a concentration producing no reaction in five
volunteers) and with the 20 standard allergens of the International Contact Dermatitis Research
Group. Patch testing of all 1,3-dichloropropene preparations produced allergic reactions in
patient 1 (with spongiosis, lymphocyte infiltration, and migration) but not in patients 2 or 3. No
patients reacted positively to 1,2-dichloropropane. The results indicate that 1,3-dichloropropene
is a primary irritant, as demonstrated by the occupational dermatitis in patients 2 and 3, but also
that 1,3-dichloropropene can cause a contact allergic reaction, as demonstrated by the positive
patch test in patient 1.
4.1.6. Brouwer, EJ; Evelo, CTA; Verplanke, AJW; et al. (1991) Biological effect
monitoring of occupational exposure to 1,3-dichloropropene: effects on liver and renal
function and on glutathione conjugation. Br J Ind Med 48(3): 167-172
This study examined the liver and kidney effects of subchronic exposure to 1,3-
dichloropropene in employees of the Dutch flower bulb industry. The cohort consisted of 14
commercial applicators who used 1,3-dichloropropene in soil fumigation operations in the
Bollenstreek region of the Netherlands. Venous blood and spot urine samples were collected
from the subjects at the start of the bulb culture season in July and after the season ended in
October. Possible hepatotoxicity was assessed by determining serum activities of alanine
aminotransferase, aspartate aminotransferase, lactate dehydrogenase, Y-glutamyltranspeptidase,
alkaline phosphatase, and total serum bilirubin. Kidney function was evaluated by measuring
serum P2-microglobulin and creatinine, urinary albumin, P2-microglobulin, retinol binding
protein, p -galactosidase, and alanine aminopeptidase. Blood GSH concentration and erythrocyte
GST activity were determined to evaluate the effect on blood GSH conjugation capacity.
13
-------
Data from the environmental monitoring study indicated that the fumigators were
exposed to TWA concentrations of 1.9-18.9 mg/m31,3-dichloropropene. The Dutch standard of
5 mg/m3 was exceeded about 30% of the exposure time. A decrease in serum total bilirubin
concentration was the only parameter of liver function to be significantly affected by 1,3-
dichloropropene. Urine albumin and retinol binding protein concentration were significantly
increased and serum creatinine concentration was significantly decreased by the end of the
spraying season. Blood GSH concentration and erythrocyte GST activity were also significantly
decreased. The authors felt that a subclinical nephrotoxic effect due to exposure to 1,3-
dichloropropene over a spraying season could not be ruled out. Alternately, changes in serum
chemistry and urine analysis parameters may have been adaptive responses to detoxification and
elimination of 1,3-dichloropropene. The serum chemistry and urine analysis parameters of the
exposed workers were not evaluated subsequently to assess whether the observed alterations
returned to normal values. The decreases in GSH and GST values indicate that GSH conjugation
is involved in 1,3-dichloropropene elimination and likely detoxification.
4.1.7. Hayes, WJ. (1982) Pesticides studied in man. Baltimore: Williams and Wilkins, pp.
139-171
In a collision between two trucks, a tank carried by one truck ruptured and spilled
approximately 4,542 L 1,3-dichloropropene. An estimated 80 people were exposed to vapors.
The most common signs and symptoms were headaches in six people, irritation of mucous
membranes in five people, dizziness in five people, and chest discomfort in four people. Three
persons became unconscious at the scene of the accident. Of 41 persons tested, 11 had slightly
elevated serum glutamic oxaloacetic transaminase and/or glutamic pyruvic transaminase values.
Within 48-72 hours, values reverted to normal in eight people, but five still had slightly elevated
serum glutamic oxaloacetic transaminase.
4.2. PRECHRONIC AND CHRONIC STUDIES AND CANCER BIOASSAYS IN
ANIMALS—ORAL AND INHALATION
4.2.1. Inhalation Studies
4.2.1.1. Parker, CM; Coaste, WB; Voelker, RW. (1982) Subchronic inhalation toxidty of 1,3-
dichloropropene/l,2-dichloropropene (D-D) in mice and rats. J Toxicol Environ Health
9:899-910
Groups of F344 rats and CD-I mice (28/sex/group) were exposed to vapors of D-D at
nominal concentrations of 0, 5, 15, or 50 ppm (0, 22.7, 68.1, or 227 mg/m3)4 6 hours/day, 5
days/week for either 6 (10/sex/group) or 12 weeks (19/sex/group). The D-D formulation
contained 25% cis-l,3-dichloropropene; 27% trans-l,3-dichloropropene; 29% 1,2-
dichloropropane; and minor amounts of 3,3-dichloropropene, 2,3-dichloropropene, and other
related chlorinated hydrocarbons. Clinical symptoms, body and organ weights, hematology,
4Calculated using conversion of 1 ppm = 4.54 mg/m3 at 25° C.
14
-------
serum chemistry, and urine analysis parameters were examined, and gross and histopathology
were performed at sacrifice.
No clinical signs of toxicity were observed in either species. There were no statistically
significant differences in mean body weights in either species at either 6-week or 12-week
terminal sacrifices. White blood cell counts were significantly decreased at 12 weeks in the male
mice exposed to 68.1 mg/m3 and in female mice exposed to 227 mg/m3 D-D. Glutamic pyruvic
transaminase activity was significantly decreased in the 68.1 and 227 mg/m3 groups of female
mice at 12 weeks. Other histologic, serum chemistry, and urinalysis parameters were either
transiently altered or showed no changes that were dose-related or outside normal ranges. In rats,
relative kidney weights in females and relative liver weights in males were significantly
increased after 12 weeks of exposure to 227 mg/m3 D-D. In male mice, statistically significant
decreases were observed in relative testis weight at 6 weeks, but not at 12 weeks, in the 227
mg/m3 group. Absolute and relative liver weights in male mice were statistically increased at 12
weeks in both the 22.7 and 227 mg/m3 groups, but not in the 68.1 mg/m3 group. The
toxicological significance of the increased liver weights is questionable because control male
mice had lower than normal liver weights. All other mean organ weights and relative weights
were within normal ranges.
The only gross pathological change observed in any treated animals during the study was
an increased incidence of enlarged peribronchial lymph nodes, with no accompanying
histopathology, in all exposed mice after 6 weeks. The only treatment-related histopathology,
which occurred at 12 weeks, was a slight to moderate diffuse hepatocyte enlargement in male
mice exposed to 227 mg/m3 D-D (12/21 treated vs. 4/18 controls). In females, a similar but
equivocal increase was also observed at 227 mg/m3 (6/18 treated vs. 1/18 controls). No
treatment-related gross pathology or histopathology was observed in the respiratory tracts of
either rats or mice.
Thus, the exposure-related effects in this study occurred after 12 weeks of exposure to
227 mg/m3 D-D. Female rats exhibited increased relative kidney weights whereas male rats and
mice exhibited increased relative liver weights. In the absence of histopathologic changes such
as degeneration or necrosis, or functional deficits as exhibited by abnormal serum or urine
analyses, these changes in organ weights are considered adaptive rather than adverse. Therefore,
the no-observed-adverse-effect-level (NOAEL) for both rats and mice is 227 mg/m3 D-D and 118
mg/m3 1,3-dichloropropene (D-D was 52% 1,3-dichloropropene). There is no lowest-observed-
adverse-effect-level (LOAEL).
4.2.1.2. Stott, WT; Young, JT; Calhoun, LL; et al (1988) Subchronic toxicity of inhaled
technical-grade 1,3-dichloropropene in rats and mice. Fundant Appl Toxicol 11:207-220
Male and female F344 rats and B6C3F1 mice (10/sex/group) were exposed to vapors of
technical-grade dichloropropene for 6 hours/day, 5 days/week for 13 weeks at nominal
concentrations of 0, 10, 30, 90, or 150 ppm (0, 45.4, 136, 409, or 681 mg/m3, respectively). The
technical-grade formulation contained 90.9% 1,3-dichloropropene, 2.4% 1,2-dichloropropene,
1.2% epichlorohydrin, and an unnamed quantity of mixed isomers of chlorohexane,
chlorohexene, and trichloropropene. The only treatment-related clinical effects observed during
15
-------
the study were a transient brown discoloration of the muzzle fur of rats exposed to 681 mg/m3
immediately following exposure, and a strong mercaptan odor associated with the coats and urine
of all rats and mice exposed to 409 or 681 mg/m3 of the formulation. There were no treatment-
related differences in survival.
The body weights of male and female rats were depressed in an exposure-related manner,
but were lexicologically significant (i.e., > 10%) only at the 681 mg/m3 exposure level.
Consistent with decreases in body weight, there were numerous changes in organ weight
parameters. In males, the mean absolute weights of kidney, liver, brain, and heart were
statistically decreased in the 681 mg/m3 group, but the mean relative weights of these organs
were increased in the 409 and 681 mg/m3 groups. Relative testis weights of males in the 681
mg/m3 group were also increased. In females in the 681 mg/m3 group, absolute weights were
decreased in the brain, heart, liver, and thymus gland, whereas relative weights were increased in
the brain, heart, and kidney. The study authors attributed the decreased absolute organ weights
and increased relative weights to the fact that rats were fasted prior to sacrifice, and stated that
the findings were consistent with reduced body weights and proportionately lower body fat
content and nonparenchymal cell mass in the high-dose animals, especially females.
A dose- and duration-related decrease in body weight was also observed in male and
female mice. Again, the body weight decrease reached toxicological significance only at the 681
mg/m3 exposure. As with rats, the diminished growth rate and reduced body weights of mice in
the 409 and 681 mg/m3 exposure groups were reflected in decreases in the absolute weights of
the heart, kidney, liver, brain, and thymus glands. Because mice were not fasted prior to
sacrifice, relative heart, kidney, and liver weights of male mice in the 409 and 681 mg/m3 groups
were also decreased. Relative brain weights of male mice in the 681 mg/m3 group were
increased. Female mice in the 681 mg/m3 group had decreased relative liver and thymus weights;
those in the 409 and 681 mg/m3 groups had increased relative kidney weights. No pathological
or histopathologic findings were observed in any of these organs in either rats or mice. The
authors concluded that the observed organ weight changes were likely due to body weight
decreases associated with nutritional changes and a general nonspecific effect of high-dose
exposures to dichloropropene.
Clinical chemistry and hematologic changes in the 409 and 681 mg/m3 groups of both
species were generally consistent with decreased body weights and associated nutritional status.
There were no treatment-related changes in urinalysis parameters. The only observable gross
pathology differences in the tissues of treated animals relative to controls were (a) a decrease in
the amount of abdominal fat in female rats in the 681 mg/m3 group and (b) a decrease in the size
of the thymus in male mice exposed to 681 mg/m3.
In rats, histopathologic changes were observed in the following organs or organ systems:
(a) mild hyperplasia of the nasal respiratory epithelium in all male and female rats exposed to
409 or 681 mg/m3 and in 2/10 male rats exposed to 136 mg/m3; (b) slight to very slight
degeneration of the olfactory epithelium in all male and female rats in the 681 mg/m3 group and 1
female in the 409 mg/m3 group; (c) incomplete development of the uteri and uterine tissue
hypoplasia in 7/10 female rats in the 681 mg/m3 group, consisting of a significant decrease in the
cross-sectional diameter of the uterine horns and a decrease in the number and stage of
development of the uterine glands (examination of the ovaries revealed normal development in
16
-------
graffian-follicles as well as corpora lutea); and (d) significant atrophy of mesenteric adipose
tissue in females exposed to 681 mg/m3.
In mice, exposure-related histopathology of the nasal mucosa was similar to that observed
in rats, consisting of slight to very slight degeneration of the olfactory neuroepithelium and
hyperplasia of respiratory epithelium in all male mice exposed to 409 and 681 mg/m3 and in 9/10
female mice at these exposures. These animals also had small focal areas of respiratory
metaplasia, a condition in which the damaged sensory olfactory epithelium is replaced by ciliated
respiratory epithelium identical to that lining the remainder of the nasal cavity and respiratory
tract. The urinary bladders of 7/10 and 6/10 female mice in the 409 and 681 mg/m3 groups,
respectively, exhibited large confluent areas of moderate hyperplasia of the transitional
epithelium. Mild aggregates of lymphoid cells in the subepithelial tissues were found to be
associated with these areas of hyperplasia in about half the affected mice. Lymphoid aggregates
without epithelial hyperplasia were also present in 9/10 female mice exposed to 136 mg/m3.
The organ weight, hematology, clinical chemistry, and gross pathology findings observed
in this study in animals of both species appeared to be secondary to, or associated with, the
decreases in body weight gain and terminal body weight. The most significant treatment-related
histopathologic findings occurred in the nasal mucosa of both sexes of rats and mice and in the
urinary bladders of female mice. The significant decrease in the size and development of the
uterus of female rats in the 681 mg/m3 group appears to have resulted from the marked growth
retardation in these animals (20% decrease in body weight by the end of the study), rather than
from a direct effect of high-dose exposure to dichloropropene vapors.
The results of this study identify a subchronic NOAEL of 45.4 mg/m3 and a LOAEL of
136 mg/m3 technical-grade 1,3-dichloropropene based on degenerative changes in the nasal
mucosa of both sexes of rats and mice. Because the technical-grade formulation was 90.9% 1,3-
dichloropropene, the NOAEL is 41.3 mg/m3 and the LOAEL is 123.6 mg/m3 1,3-
dichloropropene.
4.2.1.3. Lomax, LG; Stott, WT; Johnson, KA; et al (1989) The chronic toxicity and
oncogenicity of inhaled technical-grade 1,3-dichloropropene in rats and mice. Fundam Appl
Toxicol 12:418-431
Male and female F344 rats and B6C3F1 mice (50/sex/dose/level) were exposed via
whole-body chamber inhalation to 0, 5, 20, or 60 ppm (0, 22.7, 90.8, or 272 mg/m3, respectively)
technical-grade dichloropropene vapors for 6 hours/day, 5 days/week for 2 years. The technical-
grade formulation consisted of 92.1% 1,3-dichloropropene, 0.7% 1,2-dichloropropane, mixtures
of hexanes and hexadienes, and approximately 2% epoxidized soybean oil. Two satellite groups
of rats and mice (10/sex/dose group) were also exposed to dichloropropene for 6 and 12 months,
respectively. Standard protocols for chronic toxicity and carcinogenicity bioassays were
followed.
No clinical signs indicative of toxicity were observed in treated animals throughout the
study, and there were no significant differences in survival. Mean body weights of male and
female rats exposed to 272 mg/m3 were decreased 3%-5% during the first 11-15 months of
17
-------
treatment, but they reverted to normal during the remainder of the 2-year study. The body
weights of male and female mice exposed to 272 mg/m3 dichloropropene were also depressed. In
males, 3%-9% decreases were noted throughout the study. In females, decreases of 2%-l 1% in
the 272 mg/m3 group occurred only during the first 5 months of the study. Body weight
depression in rats and mice was not considered to be lexicologically significant.
There were no treatment-related changes in hematology, clinical chemistry, and urine
analysis parameters in any of the treated groups in either rats or mice. Weights of the brain,
heart, kidney, liver, and testis in treated rats did not differ significantly from control animals.
The mean relative kidney and liver weights of male mice in the 272 mg/m3 exposure group were
slightly lower (10%-15%) than mean control values at all exposure intervals (6, 12, and 24
months). At 90.8 mg/m3, male mice had statistically significant decreases in relative liver and
kidney weights after 12 months of exposure, but not after 24 months. Small but statistically
significant changes in relative heart, testis, and brain weights in male mice and in absolute heart
and brain weights in female mice were observed in the 272 mg/m3 group following one or two of
the three exposure periods. Organ weight changes in mice were sporadic and small and, with the
exception of the liver and kidneys, were not associated with organ histopathology. Thus, most
organ weight changes were considered to be due to decreased total body weights of the mice
and/or normal biological variability, and without toxicological significance.
4.2.1.3.1. Nonneoplastic changes. In rats, gross pathological changes were not detected in
either males or females. No significant increases in nasal histopathologic effects were observed
in the 22.7 or 90.8 mg/m3 groups. Increased incidences were observed in both sexes exposed to
272 mg/m3 for 24 months, but not after exposure for 6 or 12 months. The incidences of these
lesions at 24 months are shown in Table 1. The microscopic changes were located in the
olfactory mucosa covering the upper portions of the nasal cavity, nasal septum, and turbinates
and were characterized by. (a) unilateral or bilateral decreased thickness of the olfactory
epithelium due to degenerative changes; (b) erosions of the olfactory epithelium; and (c) fibrosis
beneath the affected olfactory epithelium, primarily in the ecto- and/or endoturbinates.
Gross pathological examination of mice showed an increase in lung masses in male mice
exposed to 272 mg/m3 compared with controls. Statistically significant, treatment-related
morphological changes in the urinary bladder were noted in females exposed to 90.8 and
exposure concentration and duration. Changes occurred in one 272 mg/m3 and in males exposed
to 272 mg/m3. The incidence of these changes increased with female mouse exposed to 90.8
mg/m3 for 12 months and in several females exposed to 90.8 mg/m3 for 24 months. Nearly half
the female mice exposed to 272 mg/m3 were affected at 6 months, and nearly all were affected at
12 and 24 months. The incidence of this lesion at 24 months is shown in Table 2.
Microscopically, the urinary bladders of both sexes exhibited hyperplasia characterized
by diffuse, uniform thickening of the transitional epithelium. In females, the hyperplasia was
accompanied by inflammation in 6/48 and 8/45 cases in the 90.8 and 272 mg/m3 groups,
respectively. For both sexes, the hyperplasia increased in frequency and severity with
concentration, and in females exposed to 272 mg/m3, the hyperplasia increased with exposure
duration. The incidence of hyperplasia was not statistically higher than in controls in the 22.7
mg/m3 groups exposed for 24 months.
18
-------
Table 1. Incidence" of selected noncancer effects in rats from Lomax et al. (1989)
Lesion
Decreased thickness
of olfactory
epithelium
Erosions of olfactory
epithelium
Fibrosis of
submucosa, olfactory
epithelium
Males
0
mg/m3
0/50
0/50
0/50
22.7
mg/m3
1/50
0/50
0/50
90.8
mg/m3
1/50
1/50
0/50
272
mg/m3
20/50
15/50
6/50
Females
0
mg/m3
0/50
0/50
0/50
22.7
mg/m3
0/50
0/50
0/50
90.8
mg/m3
0/50
0/50
0/50
272
mg/m3
15/50
6/50
2/50
1 Compared with number examined.
Table 2. Incidence3 of selected cancer and noncancer effects in mice from
Lomax et al. (1989)
Lesion
Hypertrophy/
hyperplasia of
respiratory
epithelium (slight)
Degeneration of
olfactory epithelium
(slight)
Epithelial
hyperplasia of
urinary bladder
Bronchioalveolar
adenoma
Males
0
mg/m3
5/50
1/50
4/48
9/50
22.7
mg/m3
1/50
0/50
7/48
6/50
90.8
mg/m3
4/50
1/50
11/48
13/50
272
mg/m3
48/50
48/50
37/47
22/50
Females
0
mg/m3
4/50
0/50
1/47
4/50
22.7
mg/m3
4/50
0/50
4/46
3/50
90.8
mg/m3
28/50
1/50
21/48
5/50
272
mg/m3
49/50
45/50
44/45
3/50
a Compared to number examined. For tumor incidence, mice that died before the first tumor appeared were omitted
from the number at risk.
19
-------
Exposure-related histopathologic effects in nasal tissues in both sexes of mice consisted
of hypertrophy and hyperplasia of the respiratory epithelium and/or degeneration of the olfactory
epithelium. In all cases, these changes were graded as "slight," involved approximately 10% or
less of the total respective epithelium, and did not progress in severity or extent of distribution
from one time period to the next. After 24 months of exposure, nearly all males and females in
the 272 mg/m3 group exhibited nasal mucosa histopathology (Table 2), as did most female mice
in the 90.8 mg/m3 group.
Additional microscopic changes noted in mice in the 272 mg/m3 group were (a) focal
hyperplasia and hyperkeratosis in the forestomach of 8/50 males exposed for 24 months; (b)
decreased vacuolation of renal proximal tubular epithelial cells in males exposed for 24 months;
and (c) decreased hepatocyte vacuolation in males exposed for 6 and 12, but not 24, months, and
in females exposed for 24 months.
Based on female mouse urinary bladder and nasal epithelial histopathology, this study
identifies aNOAEL of 22.7 mg/m3 and a LOAEL of 90.8 mg/m3 technical-grade
dichloropropene for mice. The NOAEL for nasal epithelial histopathology from exposure to
technical-grade dichloropropene in rats is 90.8 mg/m3, while the LOAEL is 272 mg/m3. Because
the technical- grade formulation was 92.1% 1,3-dichloropropene, the mouse NOAEL/LOAEL is
20.9 mg/m3/83.6 mg/m3 1,3-dichloropropene and the rat NOAEL/LOAEL is 83.6 mg/m3/250.5
mg/m3 1,3-dichloropropene.
4.2.1.3.2. Neoplastic lesions. In rats, there were no statistically significant increases in primary,
benign, or malignant tumors in either males or females exposed to dichloropropene for 24
months compared with concurrent or historical controls. Despite degenerative changes in the
nasal mucosa of rats and hyperplasia/hypertrophy of nasal epithelium in mice, no nasal tumors
were noted in either species. In mice, a statistically significant increase in the incidence of
benign lung tumors (i.e., bronchioalveolar adenomas) was observed in males in the 272 mg/m3
group after 24 months of exposure (see Table 2 for incidences). In spite of the prolonged
hyperplastic response observed in the transitional epithelium of the urinary bladder, no dose-
related tumorigenic responses were observed in either male or female mice. One adenoma and
two carcinomas were diagnosed in female mice exposed to 90.8 mg/m3 technical-grade
dichloropropene for 24 months; however, no bladder tumors were observed in male mice at any
dose or in female mice in the 272 mg/m3 group.
A NOAEL of 90.8 mg/m3 and a LOAEL of 272 mg/m3 technical-grade dichloropropene
was observed for neoplastic effects, i.e., bronchioalveolar adenomas in male mice. Since the
technical-grade formulation was 92.1% 1,3-dichloropropene, the NOAEL was 83.6 mg/m3 and
the LOAEL was 250.5 mg/m3 1,3-dichloropropene. The NOAEL for tumorigenic effects in rats
was 272 mg/m3 technical-grade dichloropropene, or 250.5 mg/m3 1,3-dichloropropene. There
was no LOAEL because tumors were not observed in rats.
20
-------
4.2.2. Oral Studies
4.2.2.1. Haut, KT; Stebbins, KE; Johnson, KA; et al. (1996) Subchronic toxicity of ingested
1,3-dichloropropene in rats and mice. Fundam Appl Toxicol 32:224-232
Male and female F344 rats and B6C3F1 mice (10/sex/dose) were given 0, 5, 15, 50, or
100 mg/kg/day (rats) or 0, 15, 50, 100, or 175 mg/kg/day (mice) racemic 1,3-dichloropropene in
their diets for 13 weeks. Satellite groups of rats (10/sex/dose) from the control and 100
mg/kg/day groups were retained for observation for 4 weeks following treatment in order to
examine recovery.
1,3-Dichloropropene was administered in the diet by encapsulating it into a starch/sucrose
microsphere matrix and mixing the microcapsules into rodent chow. The microcapsule
technology provided a means by which oral toxicity data in rodents could be obtained using a
contemporary Tel one formulation (95.8% 1,3-dichloropropene) and a nonbolus oral dosing
procedure. Although stability and mixability studies on the microcapsule formulation showed
that (a) it was stable for several years at room temperature, (b) it was stable in rodent chow for at
least 21 days postmixing, and (c) it provided a homogenous mixture with chow, no in-cage
stability studies were performed to determine the stability of the microencapsulated formulation
under actual feeding conditions. As discussed in Section 3.1, Stott et al. (1998) showed that the
bioavailability of microencapsulated 1,3-dichloropropene is similar to that of neat 1,3-
dichloropropene when both materials are given by gavage in corn oil. Analysis of the test diets
during the study determined that all animals received approximately the targeted doses of 1,3-
dichloropropene. Empty microspheres were mixed with rodent feed to serve as the appropriate
control. Dietary administration of 1,3-dichloropropene is preferable over bolus dosing for the
assessment of ingestion toxicity.
There were no treatment-related clinical signs of toxicity in rats or mice at any dose level
over the course of the study. Food consumption in rats was consistently decreased for high-dose
groups of males and females and occasionally depressed at lower doses relative to control values.
Food consumption in treated mice was generally unchanged and only occasionally depressed at
the higher dose levels. A dose-related statistically significant decrease in body weight, compared
with controls, was observed in male rats at doses of 15 mg/kg/day and higher, in female rats at 50
mg/kg/day and higher, and in male and female mice at all doses. Decreases in body weight
became lexicologically significant (i.e., > 10%) in male rats at 50 mg/kg/day, in female rats at
100 mg/kg/day, and in mice at 100 mg/kg/day.
Although several organ weights (absolute and/or relative) in treated animals exhibited
statistically significant differences compared with controls, the changes were consistent with
decreases in body weight and were not considered lexicologically significant. Similarly,
statistically significant differences in hematology, clinical chemistry, and urine analysis
parameters, with the possible exception of alkaline phosphatase levels (in male rats), were either
not dose-related or were consistent with the lower body weights and reduced nutritional status of
affected animals. Thus, these differences were not considered to be lexicologically significant.
21
-------
At necropsy, no gross pathology was observed in treated animals. Mild basal cell
hyperplasia and a slight prominence of mononuclear cells (consisting of endothelial, fibroblast,
and inflammatory cells) in the proximity of the basement membrane of the forestomach were
noted in all treated male and female rats in dose groups of 15 mg/kg/day and higher. After 4
weeks of recovery, animals in the 100 mg/kg/day group (the only treated group continued
through recovery) continued to exhibit basal cell hyperplasia; however, the severity and
incidence were diminished compared with those observed immediately following cessation of
treatment. These findings were attributed to localized portal-of-entry irritant effects on the
forestomach. There were no histopathologic changes in the glandular stomach.
Histopathologic changes were noted in the livers of male mice at doses of 15 mg/kg/day
and higher, and the changes consisted of a slight decrease in hepatocellular size, congruent with
decreased liver weight and decreased body weight. Therefore, this finding was not considered to
be lexicologically significant. Decreased vacuolation of tubular epithelial cells of the kidney was
observed in male mice in the highest dose group (175 mg/kg/day). There were no
histopathologic changes in the liver or the kidney in female mice.
On the basis of the results of this 13-week oral study, the NOAEL for rats is 5 mg/kg/day
and the LOAEL is 15 mg/kg/day, based on the irritant effect manifested by mild basal cell
hyperplasia of the nonglandular stomach in both sexes. For mice, the NOAEL is 50 mg/kg/day
and the LOAEL is 100 mg/kg/day, based on a lexicologically significant decrease in body weight
in both sexes.
4.2.2.2. Stott, WT; Johnson, KA; Jeffries, TK; et al (1995) Telone II® soil fumigant: two-
year chronic toxicity/oncogenicity study in Fischer 344 rats. Prepared by Dow Chemical
Company, Midland, ML Study #M-003993-0311
Male and female F344 rats (50/sex/dose) were administered a microencapsulated
formulation of Telone n® (96% 1,3-dichloropropene and no epichlorohydrin) in the diet at doses
of 0, 2.5, 12.5, or 25 mg/kg/day for 24 months. Satellite groups of rats (10/sex/dose) were
administered Telone n® for 12 months. Standard bioassay data, including body weight, food
consumption, clinical chemistry, hematology, urine analysis, organ weights, pathology, and
histopathology, were collected. Ophthalmologic examinations were conducted at the start of the
study and prior to necropsy. All animals were observed at least twice daily. Clinical
examinations were conducted at least once weekly.
There is some concern about the bioavailability of the microencapsulated formulation of
1,3-dichloropropene used in this study. In the bioavailability study conducted by Stott et al.
(1998), microencapsulated 1,3-dichloropropene in corn oil was coadministered with neat 1,3-
dichloropropene in corn oil by gavage. Although conjecture may be made, the experiment did
not delineate how the absorption of microencapsulated 1,3-dichloropropene in feed compares
with gavage dosing. In addition, no in-cage stability tests were conducted so there is no
assurance that the loaded microcapsules were stable during use.
22
-------
No treatment-related effects were observed for either mortality or clinical signs in either
sex. Body weights were significantly decreased in a dose-dependent manner in treated animals.
Decreases were greater than 10% for both sexes at 25 mg/kg/day. In the 12.5 mg/kg/day group,
average body weight decreases for males and females were 5% and 8%, respectively. Food
consumption was statistically decreased and averaged 11% and 7% less than controls in males at
12.5 and 25 mg/kg/day, respectively, and 5% less than controls for females at 25 mg/kg/day.
Thus, the observed decrement in body weights is at least partially attributable to reduced food
consumption.
Decreased absolute adrenal, heart, and liver weights and higher relative brain and kidney
weights were observed in males at 25 mg/kg/day. Females at 25 mg/kg/day had statistically
significant decreases in adrenal and liver weights and increases in relative brain, heart, kidney,
and liver weights compared with controls. No consistent treatment-related changes in
hematologic, clinical chemistry, and urine analysis parameters were observed in any of the dosed
groups.
4.2.2.2.1. Nonneoplastic changes. The only histopathology observed was in the forestomach,
which exhibited mild basal cell hyperplasia of the mucosal lining characterized by increased
cytoplasmic basophilia and increased number of cell layers in the basilar portion of the mucosa.
Basal cell nuclei were oval in shape. There was a slight prominence of mononuclear cells at the
basement membrane consisting of endothelial, fibroblast, and inflammatory cells. The
forestomach hyperplasia is believed to be a manifestation of chronic irritation, which is
consistent with the observation of primary dermal irritation (Nater and Gooskens, 1976) and
other portal-of-entry effects from 1,3-dichloropropene exposure (Haut et al., 1996; Breslin et al.,
1989; Lomax et al., 1989; Linnett et al., 1988; Stott et al., 1988). Forestomach lesions were
noted in both sexes of rats fed microencapsulated dichloropropene at > 15 mg/kg/day for 13
weeks (Haut et al., 1996). At 12 months of dietary exposure in the present study, forestomach
lesions occurred in half the animals ingesting 12.5 mg/kg/day and in almost all the animals
ingesting 25 mg/kg/day. Table 3 shows the incidence of forestomach hyperplasia at 24 months.
The incidence of hyperplasia was statistically increased at 12.5 and 25 mg/kg/day. There were no
indications of basal cell hyperplasia in the 2.5 mg/kg/day group. Other histopathologic
changes in the liver and kidney were not considered to be treatment related, as they appeared to
be age related and/or secondary to decreased body weight.
On the basis of results of this study, the NOAEL and LOAEL for noncancer effects are
2.5 and 12.5 mg/kg/day, respectively, based on chronic irritation manifested by an increased
incidence of forestomach hyperplasia in rats of both sexes. The co-critical effect of body weight
decrease, which was less sensitive, occurred at 25 mg/kg/day.
4.2.2.2.2. Neoplastic lesions. No statistically significant incidence of malignancies was
observed in rats of either sex. A trend test identified an increased incidence of benign liver cell
tumors, (i.e., hepatocellular adenomas), in both sexes of rats at 24 months but not at 12 months
of exposure. At 24 months, the incidence of benign tumors was statistically increased by
pairwise comparison only in males at 25 mg/kg/day (see Table 3 for incidences). A single
nonfatal hepatocellular carcinoma was observed in a male rat in the 25 mg/kg/day group.
23
-------
Table 3. Incidence" of cancer and noncancer effects in rats from Stott et al. (1995)
Lesion
Basal cell
hyperplasia in
forestomach
Hepatocellular
adenoma/
carcinoma
Males
0
mg/kg/
day
3/50
2/49
2.5
mg/kg/
day
3/50
1/50
12.5
mg/kg/
day
20/50
6/50
25
mg/kg/
day
30/50
10/49
Females
0
mg/kg/
day
0/50
0/49
2.5
mg/kg/
day
1/50
0/50
12.5
mg/kg/
day
20/50
0/50
25
mg/kg/
day
37/50
4/50
a 50 rats started in each group. For tumor incidence, rats that died before the first tumor appeared were omitted
from the number at risk.
Females at 25 mg/kg/day showed a statistically significant decrease in the incidence of benign
mammary gland fibroadenomas (0/50 vs. 8/50 for controls).
The NOAEL and LOAEL for tumors are 12.5 and 25 mg/kg/day, respectively, based on a
statistically significant increase in benign hepatocellular adenomas. In this study, the suspicion
that the rats may not have received adequate dosing of 1,3-dichloropropene (due to the lack of in-
cage stability studies) is allayed by the presence of liver adenomas and forestomach hyperplasia
seen in an earlier gavage study (NTP, 1985).
4.2.2.3. Redmond, JM; Stebbins, KE; Stott, WT. (1995) Telone II® soilfumigant: two-year
dietary chronic toxicity/oncogenicity study in B6C3F1 mice—Final Report. Prepared by Dow
Chemical Company, Midland, ML Study #M-003993-032
Male and female B6C3F1 mice (50/sex/dose) were administered a microencapsulated
formulation of Telone n® (95.8% 1,3-dichloropropene and no epichlorohydrin) in the diet at dose
levels of 0, 2.5, 25, or 50 mg/kg/day for 24 months. Satellite groups of mice (10/sex/dose) were
administered Telone n® for 12 months. Standard bioassay data, including body weights, food
consumption, clinical chemistry, hematology, urine analysis, organ weights, pathology, and
histopathology, were collected. Ophthalmologic examinations were conducted at the start of the
study and prior to necropsy. All animals were observed at least twice daily. Clinical
examinations were conducted at least weekly.
No treatment-related clinical signs were observed in any treatment groups. Body weights
and body weight gains were significantly decreased in a dose-dependent manner in 25 and 50
mg/kg/day groups. Male body weights in those groups were more than 10% less than controls
(11%-23%), while female body weights were 7%-9% lower than controls. The body weight
decreases may be at least partially explained by the 6% lower food consumption in those groups.
There were no statistically or biologically significant weight changes for either sex at 2.5
mg/kg/day.
24
-------
Decreased absolute heart, liver, and kidney weights and higher relative brain, heart,
kidney, and testes weights were observed in 25 and 50 mg/kg/day males. Females in those
groups had statistically significant decreases in heart, kidney, and liver weights. Higher relative
brain weight, compared with concurrent controls, was observed in females treated with 50
mg/kg/day. Changes in organ weights were considered to be secondary to decreased body
weights. No consistent treatment-related changes in hematologic, clinical chemistry, and urine
analysis parameters were observed in any of the treated groups.
4.2.2.3.1. Nonneoplastic changes. Treatment-related pathology and histopathology were not
observed in any of the groups except for a decrease in hepatocyte size in 50 mg/kg/day males at
12 months but not at 24 months of treatment. These effects were considered to be secondary to
decreased body weights.
On the basis of the results of this study, the NOAEL and LOAEL for noncancer effects
are 2.5 and 25 mg/kg/day, respectively, based on statistically and lexicologically significant
decreases in body weight for male mice.
4.2.2.3.2. Neoplastic lesions. No increase in tumor incidence was observed in any of the dose
groups. Thus, the NOAEL for cancer effects is 50 mg/kg/day. There is no LOAEL because no
tumors were detected.
The NOAEL/LOAEL for noncancer and cancer in this study may be uncertain. Because
in-cage stability tests were not conducted, there is no assurance that the loaded microcapsules
were stable during use. In addition, the mice did not exhibit the cancer (urinary bladder tumors)
and noncancer effects (urinary bladder hyperplasia, forestomach hyperplasia, and
hydronephrosis) seen in an earlier gavage study (NTP, 1985; see Section 4.2.2.4). The incidence
of lung tumors (combined bronchioalveolar adenoma and carcinoma) in the two studies are
similar for similar doses, however. Control rates for females were 2/50 in NTP (1985) and 3/50
in Redmond et al. (1995), whereas those for males were 1/50 in NTP (1985) and 6/50 in
Redmond et al. (1995). Half of the control male mice in NTP (1985) died at 1 year from events
unrelated to chemical exposure. Incidence rates for the 50 mg/kg groups were 13/50 in NTP
(1985) and 11/50 in Redmond et al. (1995) for males and 4/50 in NTP (1985) and 5/50 in
Redmond et al. (1995) for females.
4.2.2.4. National Toxicology Program (NTP). (1985) Toxicology and carcinogenesis studies
of Telone If (technical-grade 1,3-dichloropropene containing 1% epichlorohydrin as a
stabilizer) in F344/N rats andB6C3Fl mice (gavage studies). U.S. Department of Health and
Human Services Technical Report Series No. 269
Yang, RSH; Huff, JE; Boor man, GA; et al. (1986) Chronic toxicology and carcinogenesis
studies of Telone II® by gavage in Fischer-344 rats and B6C3F1 mice. J Toxicol Environ
Health 18:377-392
Toxicology and carcinogenesis ingestion studies of Telone II® (88%-90% 1,3-
dichloropropene, 2.5% 1,2-dichloropropane, 1.5% trichloropropene isomer, and 1%
epichlorohydrin) were conducted by administering the commercial-grade formulation in corn oil
25
-------
by gavage to groups of 52 male and 52 female F344/N rats at doses of 0, 25, or 50 mg/kg and to
groups of 50 male and 50 female B6C3F1 mice at doses of 0, 50, or 100 mg/kg, three times
weekly for 104 weeks. Ancillary studies were conducted in which additional dose groups
containing five male and five female rats were killed after receiving Tel one II® for 9, 16, 21, 24,
or 27 months. At study termination, there were no lexicologically significant changes in body
weight in either species. However, 25 vehicle control mice died from myocarditis during weeks
48-51. Survival in treated rats was comparable to that in vehicle controls. Survival of female
mice was statistically lower in the 100 mg/kg group.
4.2.2.4.1. Nonneoplastic lesions. In rats, increases in hyperplastic lesions of the basal layer of
the squamous epithelium in the forestomach were observed in both sexes at 25 and 50 mg/kg.
These lesions were duration dependent and were seen as early as 9-16 months after treatment
began. The incidence for males was significantly increased in the 2-year and ancillary studies
combined at both doses, whereas the incidence for females was significant only at 50 mg/kg.
Table 4 shows the incidence for these effects in the 2-year study. Stott et al. (1995) also
observed these forestomach lesions in both sexes of rats receiving >12.5 mg/kg/day 1,3-
dichloropropene in the diet.
Three types of nonneoplastic changes were observed in mice. Epithelial hyperplasia of
the forestomach was statistically significant for females at 100 mg/kg but not for males in any
treated group (see Table 5 for incidences). The incidence of transitional epithelial hyperplasia of
the urinary bladder (see Table 5) was also observed with statistical significance in both sexes at
50 mg/kg and 100 mg/kg. Such lesions in the urinary bladder have also been noted in inhalation
studies. Hyperplasia of the transitional epithelium of the urinary bladder was observed in female
mice exposed to > 409 mg/m3 technical-grade dichloropropene for 13 weeks (Stott et al., 1988)
and to 90.8 mg/m3 for 2 years (Lomax et al., 1989). In the chronic study (Lomax et al., 1989),
male mice were affected at 272 mg/m3 technical-grade dichloropropene. The third nonneoplastic
change found in mice (NTP, 1985) was an increased incidence of hydronephrosis exhibited by
female mice in the 100 mg/kg group.
The LOAEL for rats, based on hyperplastic lesions of the forestomach, is 25 mg/kg.
Averaging the exposure over 7 days yields a LOAEL of 10.7 mg/kg/day. This study does not
identify a NOAEL. For mice, the LOAEL is 50 mg/kg, based on epithelial hyperplasia in the
urinary bladder. Averaging the exposure over 7 days yields a LOAEL of 21.4 mg/kg/day. There
is no NOAEL for mice.
4.2.2.4.2. Neoplastic lesions. In rats, an increase in the incidence of forestomach tumors,
mainly benign tumors, was observed (see Table 4 for incidences). In males there was a
statistically significant increase in the 50 mg/kg group for squamous cell papilloma and
squamous cell papillomas and carcinomas combined. In female rats, a statistically significant
increase was only observed for squamous cell papillomas at 50 mg/kg when the 2-year and
ancillary studies were combined. Although the nonneoplastic lesions of the forestomach
developed within 1 year of exposure, the neoplastic lesions did not appear until 24 months after
exposure began. The incidence of forestomach tumors at 25 mg/kg was similar to controls for
both sexes.
26
-------
Table 4. Incidence3 of selected cancer and noncancer effects in rats from NTP
(1985) 2-year study
Lesion
Basal cell or epithelial
hyperplasia of forestomach
Squamous cell
papilloma/carcinoma of
forestomach
Liver neoplastic nodule
Hepatocellular carcinoma
Males
0
mg/kg/day
2/52
1/49
1/49
0/49
25
mg/kg/day
5/52
1/48
6/48
0/48
50
mg/kg/day
13/52
13/50
7/50
1/50
Females
0
mg/kg/day
1/52
0/47
6/46
0/46
25
mg/kg/day
0/52
2/45
6/42
0/42
50
mg/kg/day
16/52
3/48
10/49
0/49
a 52 rats started in each group.
the number at risk.
For tumor incidence, rats that died before the first tumor appeared were omitted from
Table 5. Incidence3 of selected cancer and noncancer effects in mice from NTP
(1985) 2-year study
Lesion
Basal cell or epithelial
hyperplasia of
forestomach
Squamous cell
papilloma/carcinoma of
forestomach
Bronchioalveolar
adenoma/carcinoma
Urinary bladder epithelial
hyperplasia
Urinary bladder
transitional cell
carcinoma
Males
0
mg/kg/day
0/52
0/37
1/22
0/50
0/50
50
mg/kg/day
0/50
2/47
13/40
9/50
0/50
100
mg/kg/day
4/50
3/49
12/44
18/50
2/50
Females
0
mg/kg/day
1/50
0/50
2/50
2/50
0/50
50
mg/kg/day
1/50
1/50
4/50
15/50
8/50
100
mg/kg/day
21/50
4/47
8/47
19/47
21/47
a 50 mice started in each group. For tumor incidence, mice that died before the first tumor appeared were omitted
from the number at risk.
27
-------
Neoplastic nodules, classified as nodular hyperplasia and described as "small focal
lesions causing only minimal compression, with little or no cytologic atypia in livers, or with
toxic or anoxic hepatic changes," (NTP, 1985) were noted in the livers of male rats. In the
current classification scheme, neoplastic nodules are classified as adenomas (Maronpot et al.,
1986). The increased incidence was statistically significant for the 25 and 50 mg/kg doses. One
male rat in the 50 mg/kg group exhibited a hepatocellular carcinoma after 2 years of exposure.
In light of the occurrence of a liver carcinoma, the biological significance of the neoplastic liver
nodules is increased. There were no statistically significant increases in liver tumors in female
rats. Stott et al. (1995) also observed liver adenomas in male rats receiving 25 mg/kg/day 1,3-
dichloropropene in the diet.
In mice, the most lexicologically significant neoplastic finding was a dose-related
statistically significant increase in the incidence of transitional cell carcinoma of the urinary
bladder in females in both 50 and 100 mg/kg dose groups (see Table 5 for incidences). Two
males in the 100 mg/kg group also developed transitional cell carcinoma of the bladder, but the
incidence was not statistically significant. Urinary bladder carcinoma was not observed in the 2-
year feeding study of Redmond et al. (1995).
An increase in the incidence of bronchioalveolar adenomas in the lung was statistically
significant in female mice at 100 mg/kg (see Table 5 for incidences). One carcinoma was found
in the 50 mg/kg group but not in the 100 mg/kg group. In male mice, a statistically significant
increase in the incidence of bronchioalveolar adenomas was observed at 50 mg/kg but not at 100
mg/kg. Two additional males in the 50 mg/kg group and three additional males in the 100 mg/kg
group were diagnosed with bronchioalveolar carcinoma. Thus, the combined incidences of lung
adenomas and carcinomas in male mice reached statistical significance for both 50 and 100
mg/kg groups. A significant increase in the incidence of bronchioalveolar adenomas was also
observed in male mice exposed to 272 mg/m3 technical-grade dichloropropene by inhalation for
24 months (Lomax et al., 1989).
Forestomach tumors were also observed in mice. The incidences for both males and
females were statistically significant at 100 mg/kg (see Table 5 for incidences).
From the rat study, NTP (1985) concluded that there was clear evidence of
carcinogenicity in male rats, based on the combined incidences of squamous cell papillomas and
carcinomas of the forestomach and the increased incidence of liver adenoma. In female rats,
there was some evidence of carcinogenicity, based on the increased incidence of squamous cell
papillomas of the forestomach. However, NTP (1985) recognized that epichlorohydrin, a
stabilizer present in Telone II®, may be partially responsible for the hyperplasia and squamous
cell papilloma/carcinoma, at least in rat forestomach. NTP states this is plausible because the
same types of lesions were found by Konishi et al. (1980) in a drinking water study with Wistar
rats, and because the local exposure of the stomach to epichlorohydrin may have reached a
concentration similar to that administered by Konishi et al. (1980). Subsequent to the NTP
(1985) study, a 2-year gavage study with epichlorohydrin was published (Wester et al., 1985).
Wester et al. (1985) observed a 28% incidence of forestomach papilloma/carcinoma in male
Wistar rats and a 15% incidence in females given 1.4 mg/kg/day epichlorohydrin. In
comparison, NTP (1985) found a 22% incidence in males and a 10% incidence in females at 2.1
28
-------
mg/kg/day (50 mg/kg/day Telone II® x 0.1 epichlorohydrin x 3 days/7 days). The chronic
feeding study by Stott et al. (1995), which did not include epichlorohydrin, reported forestomach
hyperplasia in rats but no carcinomas or papillomas.
With regard to the mouse studies, NTP concluded that the male mouse study was
inadequate for investigation of carcinogenicity because of the greatly reduced survival in the
vehicle control group. In females, however, there was clear evidence of carcinogenicity, based
on the increased incidence of transitional cell carcinoma of the urinary bladder, a very rare form
of rodent cancer. Supporting evidence for carcinogenicity of Telone II® in female mice included
the increased incidences of alveolar/bronchiolar adenomas of the lung and combined squamous
cell papillomas and/or carcinomas of the forestomach (not statistically significant) at the highest
dose, 100 mg/kg. Chronic toxicity of Telone n® was evidenced by hyperplasia of the
forestomach in both sexes of rats and mice, and epithelial hyperplasia of the urinary bladder in
male and female mice. Based on the serial-sacrifice (ancillary) study, development of both
hyperplasia and carcinogenicity of the forestomach in rats was dependent on exposure duration.
On the basis of forestomach and liver neoplasms in rats, and urinary bladder and lung
neoplasms in mice, the LOAEL for cancer in the NTP (1985) study is 21.4 mg/kg (50 mg/kg/day
x 3 days/7 days). The NOAEL for rats is 10.7 mg/kg (25 mg/kg/day x 3 days/7 days). There is
no NOAEL for mice.
4.3. REPRODUCTIVE/DEVELOPMENTAL STUDIES—ORAL AND INHALATION
4.3.1. Breslin, WJ; Kirk, HO; Streeter, CM; et al. (1989) 1,3-Dichloropropene:
two-generation inhalation reproduction study in Fischer 344 rats. Fundam Appl Toxicol
12:129-143
The reproductive and developmental effects of inhaled technical-grade 1,3-
dichloropropene were studied using F344 rats. The formulation was 92% 1,3-dichloropropene,
2% epoxidized soybean oil, and unknown amounts of chlorinated and unchlorinated alkanes and
alkenes. The F0 generation animals (30/sex/group) were exposed via whole-body inhalation to 0,
10, 30, or 90 ppm (0, 45.4, 136, or 409 mg/m3, respectively)5 1,3-dichloropropene for 6
hours/day, 5 days/week for 10 weeks before mating and for 6 hours/day, 7 days/week during
mating, gestation, and lactation. Weaned Ft rats were subjected to the same dosing regimen.
The animals were evaluated for fertility, pup survival, length of gestation, litter size, pup body
weight, pup sex ratio, gross pathology, and histologic alterations.
No effects in any animals were noted at 45.4 or 136 mg/m3. At 409 mg/m3, males in the
F0 and Fx generations exhibited a statistically significant decrease in body weight compared to
controls, but the decrease was less than 10% and is not considered to be lexicologically
significant. Gross and histologic examinations were conducted on all F0 and Fx adults and on
randomly selected Flb and F2b weanlings. 1,3-Dichloropropene exposure had no significant effect
Calculated using conversion of 1 ppm = 4.54 mg/m3 at 25° C.
29
-------
on either behavior or clinical appearance of the animals. In both adults and litters, no
lexicologically significant changes in mating and fertility indices, including cohabitation time
required for mating, gestation length, litter size, pup survival, and pup body weights, were
observed. There were no increases in either physical or behavioral abnormalities of the pups.
Parental toxicity was observed only at 409 mg/m3 and consisted of histopathological changes of
the nasal mucosa of the adult male and female rats. The alterations consisted of slight focal
hyperplasia of the respiratory epithelium and/or focal degenerative changes of the olfactory
epithelium. The nasal mucosa histopathology resulting from inhalation exposure to 1,3-
dichloropropene has been observed in other high-dose inhalation exposure studies and is most
likely due to a localized irritant effect (Linnett et al., 1988; Stott et al., 1988).
These results demonstrate that 1,3-dichloropropene is not a reproductive or
developmental toxicant via inhalation in a two-generation reproduction study with F344 rats at
exposures as high as 409 mg/m3. The NOAEL for reproductive/developmental toxicity is 376
mg/m3 because the formulation was 92% 1,3-dichloropropene. There is no LOAEL for
reproductive/developmental toxicity. The NOAEL and LOAEL for parental toxicity are 125 and
376 mg/m3, respectively, based on nasal histopathology.
4.3.2. Linnett, SL; Clark, DG; Blair, D; et al. (1988) Effects of subchronic inhalation of
D-D (l,3-dichloropropene/l,2-dichloropropene) on reproduction in male and female rats.
Fundam Appl Toxicol 10:214-223
The reproductive toxicity of D-D was determined in a single-generation study with Wistar
rats. Groups of 30 male rats of proven fertility and 24 virgin females were exposed by inhalation
to nominal concentrations of 0, 10, 30, or 90 ppm (0, 45.4, 136, or 409 mg/m3, respectively)6
D-D for 6 hours/day, 5 days/week for 10 weeks. The major components of D-D are cis-1,3-
dichloropropene (28.1% weight/weight), trans-l,3-dichloropropene (25.6% weight/weight), and
1,2-dichloropropene (25.6% weight/weight). Minor components include 2,3-dichloropropene,
3,3-dichloropropene, 1,2,3-trichloropropane, trichloropropene, and allyl chloride. The fertility of
20 males per exposure level (male fertility subgroup) was evaluated at intervals during and after
treatment by mating them with unexposed females. On the 12th day following confirmed
mating, each female was killed and examined according to standard indices of mating and
fertility, including numbers of corpora lutea, uterine implantation and uterine resorption sites,
and percent preimplantation and postimplantation losses. Males were sacrificed 5 weeks
postexposure and given standard toxicological evaluations, including semen analysis.
The fertility of 15 females per exposure level (female fertility subgroup) was assessed by
mating them with unexposed proven males at the end of the 10-week treatment period and
allowing them to deliver a litter. Exposure was not continued during gestation because this study
was designed to evaluate the effects of D-D on female libido, estrus cycling, and indices of
mating, conception, gestation, and fertility, rather than effects on fetal development. All females
6Calculated using conversion of 1 ppm = 4.54 mg/m3 at 25° C.
30
-------
in this subgroup were sacrificed 7 weeks postexposure and evaluated according to standard
toxicological guidelines. The remaining 9 females and 10 males from each treatment group
(toxicological subgroup) were not used in fertility assessment but were sacrificed immediately
postexposure for standard toxicological evaluation, including semen analysis for males.
No treatment-related effects were observed in any of the mating, fertility, fecundity, and
reproductive pathology/histopathology endpoints, including sperm morphology and estrus
cycling. Male rats exposed to 409 mg/m3 exhibited a statistically significant decrease in body
weight, but the decrease was < 10% and is not considered to be lexicologically significant.
Males also exhibited a statistically significant increases in relative kidney and liver weights;
however, there were no associated changes in clinical chemistry or urine analysis parameters, or
in organ pathology or histopathology. Unlike other studies of repeated inhalation exposure to
high doses of 1,3-dichloropropene (i.e., Breslin et al., 1989; Stott et al., 1988), no histopathology
of the nasal turbinates was found at 409 mg/m3. This may be due to the fact that D-D contained
only 57% (w/w) 1,3-dichloropropene, whereas other 1,3-dichloropropene-based fumigants (such
as Telone IT8) contain about 90% (w/w) dichloropropene.
Under the conditions of this well-conducted rat study, there was no evidence of
reproductive toxicity associated with inhalation exposure to D-D at doses up to 409 mg/m3. The
NOAEL for reproductive toxicity was 233 mg/m3 1,3-dichloropropene (409 mg/m3 D-D with
57% 1,3-dichloropropene). There was no LOAEL.
4.3.3. Hanley, TR, Jr; John-Greene, JA; Young, JT; et al. (1988) Evaluation of the effects
of inhalation exposure to 1,3-dichloropropene on fetal development in rats and rabbits.
Fundam Appl Toxicol 8:562-570
Technical-grade 1,3-dichloropropene (90.1% w/w) was evaluated for its potential effects
on embryonal and fetal development in F344 rats and New Zealand white rabbits. Groups of 30
bred rats and 25-31 inseminated rabbits were exposed via inhalation to 0, 20, 60, or 120 ppm (0,
91, 272, or 545 mg/m3, respectively) 1,3-dichloropropene for 6 hours/day during gestation days
6-15 (rats) or 6-18 (rabbits). Control groups of 30 rats and 29 rabbits were exposed to filtered
room air in a manner similar to treated groups.
In rats, maternal body weight gain was depressed in all exposed groups. Significant
depression in food consumption was observed in all exposed groups, along with a significant
decrease in water consumption in rats exposed to 545 mg/m3. However, there were no consistent
or dose-related effects on any of the following reproductive parameters: implantations,
resorptions, litter size, fetal body weight, and fetal length. Although pregnancy rates in the 91
and 272 mg/m3 groups were lower than the control rate, with a statistically significant decrease at
272 mg/m3, these findings were not considered to be lexicologically significant because (a) there
was no consistent dose-effect (the pregnancy rate at 545 mg/m3 was higher than the control group
rate), and (b) pregnancy rates in the 91 and 272 mg/m3 groups were within the historical control
range of the laboratory.
-------
External, visceral, and skeletal examination of the pups revealed no major abnormalities
or malformations. A slight but statistically significant increase in delayed ossification of the
vertebra central was observed among fetuses in the 545 mg/m3 exposure group compared with
controls. This increase was considered to have little toxicological significance because it was
judged to be secondary to the significant maternal toxicity observed among females in the 545
mg/m3 group.
Because no significant developmental effects were detected, the NOAEL for
developmental toxicity in rats was 490 mg/m3 (90% of 545 mg/m3 formulation) and there was no
LOAEL. Based on decreased body weight, the NOAEL for maternal toxicity in rats was 245
mg/m3 and the LOAEL was 490 mg/m3 1,3-dichloropropene (both corrected for 90% 1,3-
di chl oropropene).
In rabbits, statistically significant exposure-related decreases in maternal weight gain
were observed at 272 and 545 mg/m3. There were no treatment-related adverse effects on
pregnancy rate, implantation and resorption rates, preimplantation losses, litter size, or fetal
measurements among any of the exposed groups. External, visceral, and skeletal examination of
the pups did not show evidence of treatment-related abnormalities or malformations.
Statistically significant decreases in the incidence of two minor skeletal variants among the
exposed groups (delayed ossification of the hyoid in the high-dose group, and the presence of
cervical spurs in the low- and high-dose groups) were considered to be indicative of the normal
variability among rabbit pups and thus not lexicologically significant.
The NOAEL for developmental toxicity in rabbits was 490 mg/m3 (90% of 545 mg/m3)
as no effects were detected. There was no LOAEL. Based on decreased body weight, the
NOAEL for maternal toxicity in rabbits was 82 mg/m3 (90% of 91 mg/m3), with a LOAEL of
245 mg/m3 (90% of 272 mg/m3).
The weight and strength of evidence from three well-conducted reproductive/
developmental toxicity studies in two species indicates that 1,3-dichloropropene is not a
reproductive or developmental toxicant.
4.4. OTHER STUDIES
4.4.1. Acute Toxicity
Oral LD50s range from 140 to 710 mg/kg 1,3-dichloropropene for rats and 300-640
mg/kg for mice (U.S. EPA, 1998c). The LC50 for Telone II® for a 4-hour exposure in rats was
904 ppm (4,104 mg/m3)7 for females and between 846 and 990 ppm (3,841 and 4,495 mg/m3)7
for males (Streeter et al., 1987).
Calculated using conversion of 1 ppm = 4.54 mg/m3 at 25° C.
32
-------
4.4.2. Neurotoxicity
A single oral dose of 3,500 mg/kg in dogs caused staggering, partial central nervous
system (CNS) depression, and death within 24 hours (U.S. EPA, 1980).
4.4.3. Mutagenicity
Early in vitro mutagenicity testing of 1,3-dichloropropene using the Ames Salmonella
test usually elicited positive results for mutagenicity (e.g., Vithayathil et al., 1983; Stolzenberg
and Hine, 1980; Haworth et al., 1983). However, in 1984, Talcott and King demonstrated that
preparations of 1,3-dichloropropene assayed in vitro for mutagenic activity contained direct-
acting mutagenic polar impurities. Four commercial preparations of 1,3-dichloropropene were
tested for mutagenic activity before and after silicic acid chromatography. All samples were
positive before purification and negative afterwards. Polar impurities were isolated from one
preparation and tested positive for mutagenicity in the Ames Salmonella test. Talcott and King
(1984) regenerated a mixture of mutagenic polar impurities by refluxing a purified preparation
for 6 hours and then analyzed the mixture using GC/MS. Although the mixture was too complex
to be characterized completely, two known mutagens, epichlorohydrin and l,3-dichloro-2-
propanol, were tentatively identified. NTP (1985) and Watson et al. (1987) confirmed the
findings that purified 1,3-dichloropropene is not mutagenic in the Ames Salmonella test. Watson
et al. (1987) also identified two additional trace impurities: cis- and trans-2-chloro-3-
(chloromethyl)oxiranes. In addition, Watson et al. (1987) reported that purification by gas
chromatography can produce mutagenic trace impurities. Thus, the weight of evidence of these
data suggests that the mutagenic activity of 1,3-dichloropropene preparations in earlier bacterial
tests was likely due to mutagenic polar impurities and not to 1,3-dichloropropene.
Although purified 1,3-dichloropropene was not directly mutagenic, Watson et al. (1987)
observed mutagenic activity after the addition of washed microsomes from rat liver.
Mutagenicity was abolished, however, when GSH at normal physiological concentrations (5
mM) was added to the bacterial culture. Watson et al. (1987) have suggested that cis-1,3-
dichloropropene undergoes mono-oxygenase-dependent bioactivation to mutagenic metabolites
only in the absence of GSH. These findings are consistent with the results of Greedy et al.
(1984), which showed that GSH eradicated the microbial mutagenicity of both isomers of 1,3-
dichloropropene. Thus, microbial assays show that physiological concentrations of GSH provide
efficient protection against the mutagenic activity of 1,3-dichloropropene and associated trace
impurities.
Neudecker and Henschler (1986) used enzyme inhibitors to determine whether rat liver
enzymes (i.e., S9) metabolize allylic chloropropenes, such as 1,3-dichloropropene, via
epoxidation or via cleavage of the allylic chlorine, which subsequently forms the allylic
chloroalcohol, the aldehyde, and then acrylic acid. The investigators distinguished these
pathways by measuring mutagenicity in Salmonella TA100. Addition of SKF525, an inhibitor of
microsomal oxygenase that prevents formation of 1,3-dichloropropene epoxide, or 1,1,1-
trichloropropene-2,3-oxide, an inhibitor of epoxide hydrolase that prevents metabolism of the
epoxide, had no effect on mutagenicity. However, addition of cyanamide, an inhibitor of
33
-------
aldehyde dehydrogenase that prevents metabolism of the aldehyde formed from 1,3-
dichloropropene, clearly increased mutagenic activity. The addition of GSH markedly reduced
mutagenicity. The authors hypothesized that in the absence of GSH, S9 metabolically activates
1,3-dichloropropene by hydrolysis to chloroalcohols that subsequently oxidize to 3-
chloroacrolein (hydrolytic-oxidative pathway) and then to the respective acrylic acid.
In contrast to the results of Neudecker and Henschler (1986), Schneider et al. (1998a)
found that epoxidation of 1,3-dichloropropene is a minor metabolic pathway in mouse liver.
Relatively stable 1,3-dichloropropene epoxides were measured by GC/MS in mouse liver after in
vitro addition of 1,3-dichloropropene to microsomes and after in vivo administration of LD50
doses (i.e., 350 or 700 mg/kg) of 1,3-dichloropropene. 3-Chloroacrolein, a metabolite postulated
by Neudecker and Henschler (1986), was not observed in studies by Schneider et al. (1998a).
Schneider et al. (1998a) also showed that conjugation of 1,3-dichloropropene with GSH
decreases epoxide formation. The authors showed that cis and trans epoxides are mutagenic in
the Salmonella TA100 assay. The addition of GSH to the assay, with or without GST,
diminished the mutagenicity of cis-l,3-dichloropropene epoxide, the most potent isomer, and
obliterated the mutagenicity of trans-1,3-dichloropropene epoxide. The investigators postulated
that the epoxides or their decomposition products (i.e., 3-chloro-2-hydroxypropanal) are
responsible for the mutagenicity of 1,3-dichloropropene in the presence of liver enzymes.
Martelli et al. (1993) investigated the cytotoxicity and genotoxicity of 1,3-
dichloropropene in cultured Chinese hamster lung, i.e., V79 cells, and in hepatocytes from male
Sprague-Dawley rats. DNA fragmentation was significantly increased in a dose-dependent
manner in V79 cells, which cannot metabolize 1,3-dichloropropene, after 1 hour incubation with
subtoxic concentrations (1.8-5.6 mM) of 1,3-dichloropropene. This result is inconsistent with
the Salmonella assays that showed no genotoxic activity without metabolic activation (Talcott
and King, 1984; NTP, 1985; Watson et al., 1987). During an experiment to determine the time
course for DNA repair, DNA lesions in V79 cells were only partially repaired 24 hours after
removal of 1,3-dichloropropene. Subtoxic concentrations (0.18-0.56 mM) did not produce DNA
fragmentation after 20 hours' incubation. Thus, in V79 cells, it appears that DNA fragmentation
due to subtoxic concentrations of 1,3-dichloropropene was successfully repaired. However, rat
hepatocytes, which have an intact metabolizing system, were more sensitive to DNA
fragmentation. DNA fragmentation produced by 0.18-1 mM 1,3-dichloropropene in rat
hepatocytes was reduced by both GSH and inhibition of cytochrome P450 activity with
metapyrone. This experiment showed that the protective effect of GSH in bacterial assays
(Watson et al., 1987; Greedy et al., 1984; Neudecker and Henschler, 1986) also applies to
mammalian cells, and contradicts the finding of Neudecker and Henschler (1986) that
metabolism by cytochrome P450 has no role in the mutagenicity of 1,3-dichloropropene.
Ghia et al. (1993) examined the genotoxic activity of 1,3-dichloropropene using a battery
of short-term in vivo tests. Male Sprague-Dawley rats were administered doses of 1,3-
dichloropropene ranging from 62.5 mg/kg to 250 mg/kg by either a single oral gavage or a single
ip injection. Animals were pretreated with either buthionine-sulfoximine (BSO) or diethyl-
maleate (DEM) to reduce GSH levels, or with methoxsalen (MS) to inhibit cytochrome P450.
DNA fragmentation, unscheduled DNA synthesis (UDS), and micronucleus (MN) frequency
were quantitated. A dose-dependent increase in DNA fragmentation was most pronounced in the
34
-------
liver (site for tumors at 25 mg/kg/day in Stott et al. [1995] and NTP [1985] at 50 mg/kg) and
stomach mucosa (site for tumors at 50 mg/kg in NTP [1985]) and occurred to a lesser extent in
the kidney. No DNA fragmentation occurred in the lung, bone marrow, or brain, which are sites
where no tumors were detected in Stott et al. (1998) or NTP (1985). Partial repair was observed
after 24 hours. Reduction of GSH levels with BSO or DEM pretreatment did not affect DNA
fragmentation in the liver, but that was explained by the fact that neither BSO nor DEM
increased depletion of liver GSH over that caused by dichloropropene alone. The inhibition of
cytochrome P450 with MS reduced the frequency of DNA fragmentation in the liver as shown by
Martelli et al. (1993) in rat hepatocytes. Despite the fact that the 125 mg/kg dose administered
was 5 times higher than that of Stott et al. (1995) and 2.5 times higher than that of NTP (1985),
there was no evidence of DNA repair induction in UDS assays. In addition, no statistically
significant increases in micronucleated polychromatic erythrocytes (PCE) in bone marrow
(consistent with the absence of DNA fragmentation) and spleen or in micronucleated hepatocytes
were observed at the same dose.
The authors concluded that DNA fragmentation in vivo correlated well with 1,3-
dichloropropene carcinogenic activity in the rat liver and stomach mucosa observed by Stott et al.
(1995) and NTP (1985), respectively; however, the doses used by Ghia et al. (1993) were at least
2.5 times those producing liver tumors in Stott et al. (1995) and 1.25 times those producing
forestomach tumors in NTP (1985). In addition, even at the high doses used by Ghia et al.
(1993), the genotoxicity results of the rat hepatocyte DNA repair assay and the MN assay of bone
marrow, spleen, and liver cells were negative.
Von der Hude et al. (1987) assessed the genotoxicity of several halogenated short-chain
hydrocarbons, including cis- and trans-l,3-dichloropropene, using the in vitro sister chromatid
exchange (SCE) test in the Chinese hamster V79 cell line. Without S9 activation, 0.1-0.4 mM
1,3-dichloropropene showed a dose-dependent increase in the frequency of SCE. Higher
concentrations were required to induce significant SCE frequencies with 1,3-dichloropropene
compared with other short-chain chlorinated hydrocarbons tested. The observed increase in SCE
was abolished by the addition of rat liver S9 mix. These results are inconsistent with those of
Watson et al. (1987), which showed mutagenic activity of purified 1,3-dichloropropene after the
addition of S9, but not without S9. Moreover, Von der Hude et al. (1987) used a formulation
purified by gas chromatography, and as established by Watson et al. (1987), impurities due to
such "purification" have mutagenic activity. Thus, the positive response to 1,3-dichloropropene
in this assay was probably caused by mutagenic impurities rather than dichloropropene.
Kevekordes et al. (1996) tested a number of pesticides for clastogenic and aneugenic
properties in (a) an in vivo mouse bone marrow MN test and (b) an in vitro SCE assay using
human lymphocytes in the presence or absence of rat liver S9. 1,3-Dichloropropene by gavage
significantly increased the frequency of micronucleated PCE in the bone marrow cells of female
mice at the two highest doses tested (187 and 234 mg/kg), whereas no increase in PCE was
observed in male mice at doses up to 280 mg/kg/day.
With and without S9 activation, the frequency of SCE in cultured human lymphocytes
was statistically increased compared with controls, but only at the highest dose tested (100 fiM).
In the discussion of these findings, the authors point out that 1,3-dichloropropene formulations
35
-------
are likely to contain a number of mutagenic impurities (Kevekordes et al., 1996). Therefore, the
mutagenic activity cannot necessarily be attributed to 1,3-dichloropropene.
1,3-Dichloropropene does not produce dominant lethal mutations in Wistar or F344 rats
or New Zealand white rabbits, as evidenced by the absence of embryonic or fetal deaths in
inhalation studies by Hanley et al. (1988) and Linnett et al. (1988).
Valencia et al. (1985) evaluated 1,3-dichloropropene for its potential to induce sex-linked
recessive lethal mutations in Drosophila melanogaster, using a standard NTP protocol. Canton-
S wild-type males were treated with concentrations of 1,3-dichloropropene that resulted in
approximately 30% mortality. Following treatment, males were mated individually to three
harems of virgin females to produce three broods for analysis. 1,3-Dichloropropene produced
sex-linked recessive lethal mutations in males at 5,570 ppm administered by feeding. However,
it appears that this dose was cytotoxic, so the results are of questionable validity for assessing
genotoxic potential.
Stott et al. (1997a) studied the in vitro binding potential of 11 mM 14C-1,3-
dichloropropene to calf thymus DNA in the presence or absence of rat liver S9 and in the
presence of S9 + GSH. The measurement of DNA adducts exhibited a significant amount of
variation and showed no significant difference in control and treated groups.
4.4.4. Mechanistic Studies
Stott et al. (1997b) conducted a series of studies to elucidate the potential mechanisms of
tumorigenicity of 1,3-dichloropropene in male B6C3F1 mice and F344 rats. The selection of
dose, sex, species, and route of administration was based on the tumors seen in 2-year oral and
inhalation bioassays with rats and mice. In the oral study, hepatocellular adenomas were
observed in male rats fed 25 mg/kg/day 1,3-dichloropropene (Stott et al., 1995). In the inhalation
study, bronchioalveolar adenomas were observed in male mice at 272 mg/m3 (Lomax et al.,
1989). Urinary bladder tumors were noted in female mice gavaged with 50 mg/kg 1,3-
dichloropropene 3 times/week and in male mice at 100 mg/kg (NTP, 1985). Nonneoplastic
bladder effects in mice were observed at 25 mg/kg thrice weekly by gavage (NTP, 1985) and at
90.8 mg/m3 by inhalation (Lomax et al., 1989).
Stott et al. (1997b) gavaged male rats with 0, 5, 12.5, 25, or 100 mg/kg/day 1,3-
dichloropropene, 5 days/week for 3, 12, or 26 days. Male mice were exposed to whole-body
inhalation concentrations of 0, 10, 30, 60, or 150 ppm for 6 hours/day, 5 days/week for 3, 12, or
26 days. The following mechanistic endpoints were evaluated:
1. GSH levels in rat liver and mouse lung: GSH protects against tissue injury,
cytotoxicity, and mutagenicity.
2. Levels of DNA replication as determined by increased regenerative cell
proliferation in rat liver and mouse epithelia from urinary bladder and bronchiole:
Increases in this measure indicate cytotoxicity, cytolethality, and compensatory
36
-------
cell division and thus provide evidence to support a nongenotoxic mode of
tumorigenic action.
3. Rates of apoptosis in rat liver and mouse epithelia from urinary bladder and
bronchiole: Changes in apoptosis may be associated with compensatory cell
regeneration and/or a disruption of normal cellular functioning and would support
a nongenotoxic mode of action.
4. Adduct formation in rat liver and mouse lung measured by the 32P-Post-Labeling
assay: The formation of DNA adducts usually demonstrates that the test
compound is interacting directly with genetic material and thus indicates
genotoxicity.
Results from this study (Stott et al., 1997b) included a dose-dependent decrease in tissue
GSH levels. Although liver GSH levels increased back to control levels by the end of the
exposure period (26 days), GSH levels in mouse lung did not. Both tissues showed a rebound
(greater than control levels) in GSH levels when animals exposed for 11 days were tested 24
hours after dosing was terminated. No changes were noted in either cell proliferation or
apoptosis rates in rat liver, mouse lung, or urinary bladder epithelia. In addition, no unique DNA
adduct formation or increase in the incidence of normally occurring adducts was found in rat
liver or mouse lung.
The authors concluded that these studies provide scientific support to a weight-of-
evidence conclusion that tumorigenesis associated with high-dose ingestion or inhalation of 1,3-
dichloropropene is nongenotoxic in etiology and is not dependent on (a) enhanced cell
proliferation, (b) depressed rates of apoptosis, or (c) increased or unique DNA adduct formation.
However, these mechanistic studies did not identify a mechanism of action for tumor formation.
Neither the genotoxic nor the nongenotoxic mechanisms tested elicited positive results. The
GSH studies, which showed that 1,3-dichloropropene at doses used in chronic bioassays depletes
GSH in target organs, were consistent with GSH protection against cytotoxicity and
tumorigenicity by conjugating with 1,3-dichloropropene. Bacterial assays (Watson et al., 1987;
Greedy et al., 1984; Neudecker and Henscher, 1986) and in vitro mammalian assays (Martelli et
al., 1993) have also shown that GSH protects against genotoxic effects.
4.5. SYNTHESIS AND EVALUATION OF MAJOR NONCANCER EFFECTS AND
MODE OF ACTION (IF KNOWN)—ORAL AND INHALATION
Despite the high potential for occupational exposures of agricultural workers, reports of
adverse health effects reflect only relatively mild effects. Dermatitis is the only effect noted in
humans after repeated occupational exposures, and only two case studies have been published.
Exposure to high concentrations in cases of chemical spills, however, can produce severe toxicity
manifested by a dose-related range of acute neurotoxic symptoms. Accidental ingestion of large
quantities of 1,3-dichloropropene has been fatal.
No epidemiologic data on the chronic health effects of 1,3-dichloropropene could be
found. In chronic and subchronic high-dose animal studies, histopathologic changes have been
noted in target organs along the portals of entry (e.g., forestomach for oral administration; nasal
37
-------
mucosa and lung for inhalation) and/or in organs involved in the metabolism (liver) and
excretion of conjugated metabolites (e.g., urinary bladder and kidney). It should be noted that
early studies used technical-grade mixtures of 1,3-dichloropropene that contained 2%-10%
impurities and stabilizing agents. Epichlorohydrin, a stabilizer in older formulations, has
produced forestomach lesions in rodent studies, and it is likely that epichlorohydrin contributed
to these effects in the NTP (1985) study. More recent formulations of commercial 1,3-
dichloropropene have replaced epichlorohydrin with epoxidized soybean oil (Lomax et al., 1989;
Breslin et al., 1989). Nonetheless, commercial formulations may still contain a number of
potentially toxic impurities whose presence may contribute to toxic effects.
Neither reproductive nor developmental toxicity has been observed in a well-conducted
two-generation study in rats or in developmental studies in rats and rabbits at maternal inhalation
concentrations up to 376 mg/m31,3-dichloropropene (Breslin et al., 1989; Linnett et al., 1988;
Hanley et al., 1988). Even concentrations that produced parental toxicity (i.e., decreased body
weight and/or nasal histopathology) did not produce reproductive or developmental effects
(Hanley et al., 1988; Breslin et al., 1989).
The toxicokinetics of 1,3-dichloropropene are reasonably well understood. 1,3-
Dichloropropene is rapidly absorbed and quickly conjugated with GSH into mercapturic acids
(Climie et al., 1979; Dietz et al., 1985; Waechter and Kastl, 1988; Waechter et al., 1992), which
are rapidly excreted in the urine. The extent of epoxidation, a minor metabolic pathway
identified at ~LD50 doses in mice, is reduced by conjugation of 1,3-dichloropropene with GSH.
1,3-Dichloropropene does not bioaccumulate in target tissue to any significant degree (Hutson et
al., 1971; Dietz et al., 1984a). Relatively high repeated exposures to 1,3-dichloropropene are
required to significantly deplete GSH in target organs, with the exception of nasal tissue.
Nonlinear kinetics consistent with saturation of GSH-mediated conjugation systems have been
reported at exposure levels of 1,363-4,086 mg/m3 in rats (Fisher and Kilgore, 1988a,b).
Pharmacokinetic studies have demonstrated that reductions in GSH due to repeated
administration of 1,3-dichloropropene occur over a range of doses (22.7-7,786 mg/m3 by
inhalation and 12.5-100 mg/kg orally), that significant depletion occurs in most tissues only at
high doses, and that GSH levels rebound upon cessation of exposure (Stott et al., 1997b). Thus,
it appears likely that toxicity is associated with depletion of GSH. Based on in vitro studies and
biological monitoring of workers exposed to 1,3-dichloropropene vapors, human toxicokinetics
and metabolism by GSH conjugation appear to be similar to those in rodents.
4.5.1. Inhalation Studies
The olfactory and/or nasal epithelium is a primary target organ for both rats and mice
inhaling vaporized formulations of technical-grade 1,3-dichloropropene.
Histopathology of the nasal mucosa was observed in male and female rats (Lomax et al.,
1989) exposed to 272 mg/m3 1,3-dichloropropene (with epoxidized soybean oil as the stabilizing
agent) for 24 months, but not after exposure for 6 or 12 months, or to lower doses. Although
similar histopathology is produced in the mouse nasal mucosa at 90.8 mg/m3, the severity of the
response is less (Lomax et al., 1989). The other target organ for inhalation exposure is the mouse
38
-------
urinary bladder (Lomax et al., 1989). Morphological changes in the urinary bladder occurred in
females exposed to 90.8 or 272 mg/m3 and in males exposed to 272 mg/m3. Microscopically, the
urinary bladders exhibited extensive hyperplasia and some inflammation. No effects were
observed in rats or mice at 22.7 mg/m31,3-dichloropropene.
4.5.2. Oral Studies
In early studies (NTP, 1985; Yang et al., 1986), oral gavage was employed as the means
of administration, and a formulation of 1,3-dichloropropene containing epichlorohydrin was used
as the test substance. 1,3-Dichloropropene produced forestomach hyperplasia in rats and mice.
Other target organs included the mouse urinary bladder (epithelial hyperplasia), rat liver
(neoplastic nodule formation), and mouse kidney (hydronephrosis).
When the method of oral administration of 1,3-dichloropropene was changed to feeding
(Haut et al., 1996; Stott et al., 1995; Redmond et al., 1995), forestomach lesions still occurred in
rats at 12.5 mg/kg/day and higher, but compared with the NTP (1985) study, the severity of
hyperplasia was reduced. Other targets identified in the NTP gavage study—mouse forestomach,
urinary bladder, kidney, and rat liver—exhibited no histopathology in the feeding studies (Stott et
al., 1995; Redmond et al., 1995). Differences in histopathology between the NTP (1985) and the
feeding studies may be due to the method of compound administration (daily dietary exposure
vs. concentrated bolus dosing). Other investigators have shown that oral gavage increases blood
levels of toxicant and toxicity compared with the same dose administered by gastric infusion
over 2 hours (Sanzgiri et al., 1995). The decrease in the number of target organs in the feeding
studies may also be because of the absence of epichlorohydrin in the feeding formulation. In the
mouse dietary study (Redmond et al., 1995), there is some uncertainty as to whether the mice
received the intended dose because of the absence of cancer (urinary bladder tumors) and
noncancer effects (urinary bladder hyperplasia, forestomach hyperplasia, and hydronephrosis)
seen in the NTP (1985) study. However, the incidences of lung tumors (combined
bronchioalveolar adenoma and carcinoma) in the two studies are similar for similar doses. The
other major toxic effect in the feeding studies was reduced body weight at the higher doses in
both rats and mice.
4.6. WEIGHT-OF-EVIDENCE EVALUATION AND CANCER CLASSIFICATION-
SYNTHESIS OF HUMAN, ANIMAL, AND OTHER SUPPORTING EVIDENCE,
CONCLUSIONS ABOUT HUMAN CARCINOGENICITY, AND LIKELY MODE OF
ACTION
The only evidence associating carcinogenicity in humans to 1,3-dichloropropene
exposures is three case studies in which two firemen and one farmer were accidentally exposed
to acute high doses and subsequently developed blood cancers (non-Hodgkin's lymphoma and
leukemia). Case reports are often anecdotal or highly selective, but they may identify an
association when there are unique features such as uncommon tumors (U.S. EPA, 1996a). Non-
Hodgkin's lymphoma and leukemia, however, are not rare cancers. In 1979, the same year the
reported lymphomas were diagnosed, the age-adjusted incidence of non-Hodgkin's lymphoma in
39
-------
the United States for males, 11.8 per 100,000 population, was between that of skin cancer,
8.9/100,000, and stomach cancer, 13.6/100,000 (Ries et al., 1998). In 1975, the year the reported
leukemia was diagnosed, 13.6/100,000 cases were reported in males. These case studies do not
provide a basis for inferring a causal association between exposure to 1,3-dichloropropene and
blood cancers because the possibility of confounding factors has not been considered or ruled out
(U.S. EPA, 1987). Additionally, animal bioassays do not suggest that the hematopoietic system
is a target organ of 1,3-dichloropropene carcinogenicity.
Two-year animal bioassays indicate that 1,3-dichloropropene is carcinogenic at relatively
high doses. Feeding studies in rodents by Stott et al. (1995) found a late-onset increase in the
incidence of benign hepatocellular adenomas (with one hepatocarcinoma) in male rats at the
highest dose tested, 25 mg/kg/day. No treatment-related tumors were observed in female rats or
in male or female mice fed up to 50 mg/kg/day. The thrice weekly gavage study by NTP (1985)
found significant incidences of bronchioalveolar, forestomach, and urinary bladder tumors in
mice at 50 mg/kg and forestomach and liver tumors in rats at 25 mg/kg. With the exception of
the urinary bladder tumors in mice, most tumors were benign. In rats at 50 mg/kg, four
carcinomas were observed in the forestomach and one was observed in the liver. In mice, eight
carcinomas in urinary bladder and three in bronchioalveolar areas were observed at 50 mg/kg
while two were observed in the forestomach at 100 mg/kg. Although the NTP study was rejected
for RfD development by EPA (IRIS, online 10/1/90), because the thrice-weekly high-dose
gavage regime was not well designed to study chronic toxicity, the data do show that 1,3-
dichloropropene is a carcinogen at relatively high bolus doses. Current test guidelines
recommend seven times weekly gavage, but indicate that five times/week is acceptable (U.S.
EPA, 1998d). NTP acknowledged that the epichlorohydrin used as a stabilizer in Telone II® may
be partially responsible for the squamous cell papillomas and carcinomas, at least in the rat
forestomach, as hyperplasia, papilloma, and carcinoma were found in the forestomachs of rats in
an epichlorohydrin drinking water study (Konishi et al., 1980). The chronic feeding study by
Stott et al. (1995), which did not include epichlorohydrin, found forestomach hyperplasia in rats
but no carcinomas or papillomas.
In chronic inhalation bioassays, a statistically significant increase in the incidence of
benign lung adenomas was observed in male mice only at the highest exposure of 272 mg/m3
dichloropropene, but no malignancies were observed (Lomax et al., 1989). The tumors occurred
with late onset as they were observed after 24 months of exposure, but not after 6 or 12 months.
No tumors were reported for female mice or for male or female rats. Despite the dose-dependent
hypertrophy and hyperplasia of the nasal respiratory epithelium and/or degeneration of the
olfactory epithelium in rats at the highest exposure of 272 mg/m3, no animals developed tumors
in the nasal mucosa. Mice also exhibited these effects, but all cases were graded as "slight"
histopathologic changes, involved approximately 10% or less of the total respective epithelium,
and did not progress in severity or distribution from one exposure duration to the next. Most
animals exposed to 272 mg/m3 for 6, 12, or 24 months exhibited nasal histopathology. For the
24-month exposures, the incidence of nasal histopathology was significant in female mice at 90.8
mg/m3 and in male mice at 272 mg/m3 dichloropropene.
The lack of tumorigenesis in the rat nasal mucosa may be due to the relatively low uptake
of toxic vapors in this tissue (Stott and Kastl, 1986) and the protective action of GSH. Uptake is
40
-------
much higher in the rat lung than in the nasal mucosa. Additionally, whereas GSH is depleted in a
dose-dependent manner in the nasal mucosa, depletion appears to be dose-independent in the
lung. Decreases of up to 70% of control values are maintained across a wide range of dose levels
(Fisher and Kilgore, 1988a). The relatively low uptake and rapid detoxification of inhaled 1,3-
dichloropropene by GSH in the nasal mucosa appear to be sufficient to protect against
carcinogenicity, but not toxicity, along the primary portal of entry. In the rat lung, neither
toxicity nor carcinogenicity was observed.
The mutagenicity and genotoxicity of 1,3-dichloropropene have been extensively studied
in both in vitro and in vivo assays. Early bacterial studies demonstrated that 1,3-dichloropropene
was mutagenic in a variety of test systems in the absence of metabolic activation. Although later
studies showed that these findings were due to mutagenic impurities in the 1,3-dichloropropene
formulation, even purified 1,3-dichloropropene produced mutations in the presence of S9.
Bacterial reversions were prevented, however, by the addition of physiological concentrations of
GSH.
In the absence (verified or assumed) of mutagenic impurities, 1,3-dichloropropene has
produced mixed results in mammalian in vitro and in vivo genotoxicity studies. Although the
positive studies indicate that 1,3-dichloropropene can be mutagenic, the relevance of these
studies to mammalian tumor formation is uncertain owing to the high concentrations or doses
used. The lowest concentration used in in vitro studies, -0.1 mM, is still two orders of
magnitude higher than that found in rat blood after high acute doses of 1,3-dichloropropene. The
peak blood level detected after a 3-hour exposure of rats to 409 mg/m3 1,3-dichloropropene (the
highest concentration in the 2-year chronic bioassay by Lomax et al. [1989] was 227 mg/m3) was
0.004 mM 1,3-dichloropropene (Stott and Kastl, 1986). The peak blood level detected in rats
after a 25 mg/kg gavage with 1,3-dichloropropene (highest dietary dose administered by Stott et
al., 1995) was approximately 0.0027 mM 1,3-dichloropropene (Stott et al., 1998). Even the
lowest doses used in in vivo genotoxicity tests (62.5 mg/kg in rats by Ghia et al., 1993) were
more than twice those used in the chronic bioassays (Stott et al., 1995). Although several high-
concentration and high-dose genotoxicity studies have shown that 1,3-dichloropropene is
mutagenic, the relevance of these studies to tumor formation in chronic rodent bioassays is
uncertain because of the lack of information about the relative sensitivity of the test systems.
However, the weight of the evidence in the short-term studies suggests that 1,3-dichloropropene
is mutagenic.
Although the major metabolic pathway of 1,3-dichloropropene is conjugation by GSH
and subsequent excretion in the urine, Schneider et al. (1998a) found that epoxidation of 1,3-
dichloropropene is a minor metabolic pathway in mouse liver at ~LD50 doses. The doses
administered were 3.5-7 times the maximum dose given to mice in the NTP (1985) study and
7-14 times those given to mice in the feeding study of Redmond et al. (1995). Schneider et al.
(1998a) showed that the epoxides were mutagenic in bacterial assays and that the mutagenicity
was decreased (cis-epoxide) or abolished (trans-epoxide) by the addition of GSH. The
investigators also demonstrated that conjugation of 1,3-dichloropropene with GSH decreases
epoxide formation in mouse liver. The authors postulated that the epoxides or their
decomposition products are responsible for the mutagenicity of 1,3-dichloropropene in the
presence of liver enzymes and showed that the epoxides bind to deoxyguanosine in vitro
41
-------
(Schneider et al., 1998b). Stott et al. (1997a,b), however, found no evidence of DNA adduct
formation in vivo after subchronic exposures to tumorigenic doses of 1,3-dichloropropene. It is
possible that GSH effectively scavenged 1,3-dichloropropene in the subchronic studies and that
lifetime exposures to high doses of 1,3-dichloropropene eventually lead to significant GSH
depletion and lack of protection from the genotoxic metabolites. 1,3-Dichloropropene may be
nongenotoxic at low-dose exposures that do not interfere significantly with normal function of
GSH, but bioassay data showing the protective effect of GSH against tumor formation are
lacking.
Although the available human data are inadequate, under EPA's proposed cancer risk
assessment guidelines (1996), the weight of evidence indicates that 1,3-dichloropropene is
clearly a rodent carcinogen and is "likely to be carcinogenic to humans." This characterization is
based on tumors observed in chronic animal bioassays for both inhalation and oral routes of
exposure. Although the chronic dietary and inhalation bioassays suggest that tumors may not
occur at low doses, a nonlinear mechanism of tumor formation is not supported by mechanistic
data. In fact, the mutagenic properties of 1,3-dichloropropene suggest a genotoxic mechanism of
action. The mutagenic properties and the absence of data to support a nonlinear mechanism of
tumor formation require the quantitative assessment to default to a linear model. Under current
EPA (1987) cancer risk assessment guidelines, 1,3-dichloropropene is characterized as a class
"B2," probable human carcinogen, with little or no evidence for carcinogenicity in humans and
sufficient evidence in animals. This classification is based on observations of tumors in F344
rats (forestomach, liver) and B6C3F1 mice (forestomach, urinary bladder, and lung) at high bolus
doses, observations of benign liver tumors in F344 rats at lower dietary doses, and the formation
of mutagenic epoxide metabolites at high, ~LD50, doses.
4.7. SUSCEPTIBLE POPULATIONS
4.7.1. Possible Childhood Susceptibility
There are no human studies that indicate the relative sensitivity of children and adults to
the toxic effects of 1,3-dichloropropene. Although no animal studies have examined the effect of
1,3-dichloropropene exposure on juvenile animals, well-conducted studies in rats and rabbits
show no evidence of developmental toxicity (Hanley et al., 1988; Linnett et al., 1988; Breslin et
al., 1989) even at doses that caused maternal toxicity. On the basis of these results, it is unlikely
that 1,3-dichloropropene causes developmental toxicity in humans, but its effects on children are
unknown.
4.7.2. Possible Gender Differences
There are no human data that suggest that gender differences in toxicity or tumorigenicity
might occur as a result of exposure to 1,3-dichloropropene. In chronic animal studies, the female
mouse was more sensitive to the urinary bladder toxicity induced by inhalation exposure to 1,3-
dichloropropene, but male mice exhibited bronchioalveolar adenomas while female mice did not
(Lomax et al., 1989). Inhalation exposure also produced mild kidney histopathology in female
42
-------
mice and mild kidney and liver histopathology in male mice (Lomax et al., 1989). In a feeding
study, male mice also exhibited a decrease in body weight when females did not (Redmond et al.,
1995). In a rat feeding study, only males exhibited liver adenomas (Stott et al., 1995), but both
sexes had neoplastic liver nodules in a gavage study (NTP, 1985). However, the relevance of
gender differences in rodents to those in humans is unknown.
5. DOSE-RESPONSE ASSESSMENTS
5.1. ORAL REFERENCE DOSE (RfD)
5.1.1. Choice of Principal Study and Critical Effect—With Rationale and Justification
There are no chronic human studies suitable for dose-response assessment, but there are
four chronic studies for orally administered 1,3-dichloropropene in rats and mice: two gavage
bioassays by NTP (1985) and two dietary studies (Stott et al., 1995; Redmond et al., 1995).
Limitations of the gavage studies include the thrice-weekly dosing regime. Current test
guidelines recommend seven times weekly, but indicate that five times/week is acceptable (U.S.
EPA, 1998d). The NTP (1985) study was rejected by EPA for RfD development (IRIS, online
10/1/90) because the dosing regimen (high doses by gavage three times/week) was not well
designed to study chronic toxicity. Another problem with the NTP (1985) study is that the
dichloropropene formulation contained epichlorohydrin, which NTP acknowledged as a possible
contributor to tumorigenic effects in the forestomach. In addition, gavage administration is much
less relevant to human exposure than dietary administration. Gavage administration delivered a
single bolus dose, but human exposure would be similar to dietary intake, which occurs at
intervals throughout the course of a day. The dietary studies, however, lack information about
the in-cage stability of the food mixture. The absence of such information leaves doubt as to the
actual dose received by the animals.
The rat dietary study of Stott et al. (1995) is the most appropriate choice of a principal
study for derivation of toxicity values for nonneoplastic effects because adverse effects were seen
at lower doses than in the mouse dietary study (Redmond et al., 1995). Although statistically and
lexicologically significant decrements in body weight were reported in rats (Stott et al., 1995)
and in male mice (Redmond et al., 1995) at 25 mg/kg/day, no significant pathology or
histopathology was observed in mice of either sex at any dose. In rats, a statistically significant
increase in the incidence of forestomach histopathology was observed at 12.5 and 25 mg/kg/day
for both sexes (see Table 6 for incidences). The histopathology consisted of mild basal cell
hyperplasia of the mucosal lining and was characterized by (a) a prominence of the basal layers
of the mucosa due to increased cytoplasmic basophilia and (b) an increased number of cell layers
in the basal portion of the mucosa. The hyperplasia was graded as very slight or slight.
Of the two major effects, body weight decrease and forestomach hyperplasia, data from
the most sensitive effect, forestomach hyperplasia, were used to develop the RfD, as hyperplasia
occurred at lower doses. The forestomach hyperplasia is a manifestation of chronic irritation and
is consistent with the observation of primary dermal irritation (Nater and Gooskens, 1976) and
43
-------
Table 6. Incidence of forestomach histopathology in male F344 rats
Administered
dose (mg/kg/day)
0
2.5
12.5
25
Forestomach
histopathology
(animal incidence)
3/100
4/100
40/100
67/100
other portal-of-entry effects from 1,3-dichloropropene exposure (Haut et al., 1996; Breslin et al.,
1989; Lomax et al., 1989; Linnett et al., 1988; Stott et al., 1988). The irritant effects of 1,3-
dichloropropene on the stomach in humans are verified by a case report of gastric mucosal
erosion produced by a human poisoning incident (Hernandez et al., 1994; see Section 4.1.2). The
lack of chronic irritation (i.e., forestomach hyperplasia) or body weight decrease at 2.5
mg/kg/day defines the study NOAEL. The LOAEL is 12.5 mg/kg/day. No adjustment for
exposure duration is necessary because 1,3-dichloropropene was administered daily in the diet
for 2 years.
5.1.2. Methods of Analysis—Benchmark Dose Analysis
The incidence of treated animals with forestomach histopathology is a quantitative
measure of toxicity amenable to benchmark dose (BMD) analysis. BMD analysis was chosen
because it uses the entire dose-response curve to identify the point of departure, it does not
depend upon dose spacing, and it is sensitive to the number of animals used in the study. The
data available met the suggested criteria (U.S. EPA, 1995) of at least three dose levels, with two
doses eliciting a greater than minimum and less than maximum response.
The seven statistical models for dichotomous data from U.S. EPA's Benchmark Dose
Software Version 1. Ib were used to identify the model that best fit the dose-response curve
(Appendix A). The best model was chosen by eliminating all models that did not have a
statistically significant goodness-of-fit (p>0.05). The remaining models were then ranked by
best visual fit of the data, especially for the lower doses, as observed in the graphical output of
the Benchmark Dose Software. The model with the best visual fit and a statistically significant
goodness-of-fit was used to estimate the BMD10 (maximum likelihood estimate at 10% risk) and
the BMDL10 (95% lower confidence limit on the BMD10).
The results for gamma, multistage, and Weibull models were statistically significant for
goodness-of-fit. The gamma model was chosen because the visual fit at low doses was the best
of the three models. The gamma model yielded a BMD10 of 5.07 mg/kg/day and a BMDL10 of
3.38 mg/kg/day (Appendix A).
44
-------
5.1.3. RfD Derivation—Including Application of Uncertainty Factors (UF) and Modifying
Factors (MF)
Uncertainty factors (UFs) are applied to account for uncertainties in extrapolation from
rodent bioassay data to human exposure conditions, for unknown variability in human
sensitivities, for data deficiencies, and for other factors. The default uncertainty factor of 10 for
interspecies extrapolation is applied because there are no data on the relative sensitivity of rats
and humans to stomach irritation. Because there are no data documenting the nature and extent
of variability in human susceptibilities to 1,3-dichloropropene, the default uncertainty factor of
10 is also applied to protect sensitive human subpopulations. The database for 1,3-
dichloropropene is substantial and includes studies of genotoxicity, mode of action,
pharmacokinetics, reproductive and developmental toxicity, systemic toxicity, and cancer.
Therefore, no additional UFs or MFs are needed.
The BMD10 and BMDL10 are divided by a total UF of 100 to yield the RfD.
BMD10 = 5.07 - 100 = 0.05 mg/kg/day
BMDL10 = 3.38 - 100 = 0.03 mg/kg/day
Thus, the RfD derived from the BMDL10 is 0.03 mg/kg/day.
5.2. INHALATION REFERENCE CONCENTRATION (RfC)
5.2.1. Choice of Principal Study and Critical Effect—With Rationale and Justification
Lomax et al. (1989), the only chronic inhalation bioassay for 1,3-dichloropropene, was
chosen as the principal study because it was well designed and well conducted and used both rats
and mice. The two potential critical effects in this study are histopathology of the respiratory
epithelium in the nasal tract in rats and mice and hyperplasia and inflammation in the urinary
bladder in mice. Although nasal tract histopathology was observed in both genders of rats
exposed to 272 mg/m3, the female mouse was more sensitive with increased incidences of
histopathology at 90.8 mg/m3. The nasal histopathology was characterized by hypertrophy and
hyperplasia of the respiratory epithelium and/or degeneration of the olfactory epithelium.
Urinary bladder hyperplasia also occurred at 90.8 mg/m3 in female mice and at 272 mg/m3 in
males. Microscopically, the urinary bladders of both sexes exhibited hyperplasia characterized
by diffuse, uniform thickening of the transitional epithelium. In females, the hyperplasia was
accompanied by inflammation in 20%-30% of affected animals. Generally, the hyperplasia
increased in severity with increasing concentration, and in the high-dose female group, with
increasing exposure duration.
Nasal histopathology was chosen as the most relevant critical effect because it was also
found in subchronic studies of rats or mice (Stott et al., 1988; Breslin et al., 1989) and because it
was reported in humans exposed to 1,3-dichloropropene (Markovitz and Crosby, 1984). Table 7
shows the incidences for nasal histopathology in female mice. The lack of any such effect at 3.7
45
-------
Table 7. Incidence of nasal histopathology in female B6C3F1 mice
Administered
dose (mg/m3)
0
22.7
90.8
272
Adjusted
administered dose
(mg/m3)3
0
3.7
14.9
44.7
Nasal hypertrophy/
hyperplasia
4/50
4/50
28/50
49/50
Correction for purity of formulation concentration (92%) and correction for
intermittent exposure to continuous exposure: 22.7 mg/m3 x 0.92 x 6/24 hrs x
5/7 days = 3.7 mg/m3.
mg/m3, adjusted for purity and continuous exposure duration, defines the NOAEL. The LOAEL,
adjusted for purity and continuous exposure duration, is 14.9 mg/m31,3-dichloropropene.
5.2.2. Methods of Analysis—Benchmark Concentration Analysis
Benchmark concentration (BMC) analysis was chosen because it uses the entire dose-
response curve to identify the point of departure, it does not depend upon dose spacing, and it is
sensitive to the number of animals used in the study. The data available met the suggested
criteria of at least three dose levels with two doses eliciting a greater than minimum and less than
maximum response (U.S. EPA, 1995).
The seven statistical models for dichotomous data from U.S. EPA's Benchmark Dose
Software Version 1. Ib were applied to the incidence data for the adjusted administered doses
(see Appendix A). The best model fit was determined by eliminating all models that did not
have a statistically significant goodness-of-fit (p>0.05). The remaining models were then ranked
by best visual fit of the data, especially for the lower doses, as observed in the graphical output of
the Benchmark Dose Software. The model with statistically significant goodness-of-fit and best
visual fit was used to estimate the BMC at 10% risk and the 95% lower confidence limit of the
BMC, the BMCL,
•'lO-
The gamma, logistic, multistage, Weibull, and quantal-quadratic models provided
statistically significant fits (see Appendix A). The gamma model was the best fit overall because
it provided the best visual fit.
3.66 mg/m3 (Appendix A).
This model yielded a BMC10 of 5.91 mg/m and a BMCL10 of
1,3-Dichloropropene is a Category 2 gas (U.S. EPA, 1994b) because it is not highly
reactive or water soluble and it produces both respiratory (nasal histopathology) and remote
effects (urinary bladder histopathology). For Category 2 gases, adjustment of animal exposure to
human equivalent concentrations (HECs) is based on algorithms for Category 1 or Category 3
gases, depending upon whether the major effect is respiratory or systemic. Because the critical
target was the nasal mucosa, algorithms for extrathoracic effects for Category 1 gases are used to
46
-------
adjust animal exposure concentrations of 1,3-dichloropropene to HECs (U.S. EPA, 1994b). The
HEC for a Category 1 gas is derived by multiplying the animal BMC10 and BMCL10 by an
interspecies dosimetric adjustment for gas respiratory effects in the extrathoracic area of the
respiratory tract, according to the following calculation (U.S. EPA, 1994b):
RGDR(ET) = (MVa/Sa)/(MVh/Sh)
where:
RGDR(ET) = regional gas dose ratio for the extrathoracic area of the respiratory tract
MVa = animal minute volume (mouse = 0.041 L/min)
MVh = human minute volume (13.8 L/min)
Sa = surface area of the extrathoracic region in the animal (mouse = 3 cm2)
Sh = surface area of the extrathoracic region in the human (200 cm2).
Using default values, the RGDR(ET) = (0.041/3)7(13.8/200) = 0.014/0.069 = 0.198. The animal
BMC10 and BMCL10 are then multiplied by 0.198 to yield the HECs for these values:
BMC10HEC= BMC10 x 0.198 = 5.91 x 0.198= 1.17 mg/m3
BMCL10HEC = BMCL10 x 0.198 = 3.66 x 0.198 = 0.725 mg/m3.
5.2.3. RfC Derivation—Including Application of Uncertainty Factors (UFs) and Modifying
Factors (MFs)
UFs are applied to account for uncertainties in extrapolation from rodent bioassay data to
human exposure conditions, for unknown variability in human sensitivities, for data deficiencies,
and for other factors. Historically, UFs were applied as values of 10 in a multiplicative fashion
(Dourson and Stara, 1983). Recent EPA practice, however, also includes use of a partial UF
such as 101/2 (U.S. EPA, 1994) under conditions where toxicokinetics and mechanistic
information are available, or data are available on the nature and extent of human variability, or
prior interspecific adjustment has already been conducted (e.g., using pharmacokinetic or
dosimetric scaling).
For long-term rodent bioassays, the default UFs for interspecies extrapolation and within-
species variability are each 10. Half of that factor, 101/2, or 3, reflects the pharmacokinetic
component of uncertainty and half represents the pharmacodynamic component of uncertainty.
The toxicokinetics of 1,3-dichloropropene are reasonably well understood and do not involve
bioaccumulation. Instead, 1,3-dichloropropene is rapidly conjugated via GSH-mediated systems
to mercapturic acids and excreted in the urine. The toxicokinetics in rats and humans are similar.
The calculation of an HEC adjustment reduces the uncertainty associated with interspecies
variation. Therefore, the use of a UF of 3, instead of the default UF of 10, is more than justified
for interspecies extrapolation. There are no data documenting the nature and extent of variability
in human susceptibility; therefore, the default UF of 10 is used for within-species variation. The
database is substantial and includes studies of pharmacokinetics, reproductive and developmental
toxicity, systemic toxicity, mechanism of action and mutagenicity/genotoxicity. Therefore, no
additional UFs or MFs are needed.
47
-------
The BMC10 and BMCL10 are divided by a total uncertainty factor of 30 to yield the RfC
for noncancer effects:
BMC10HEC= 1.17 mg/m3 + 30 = 0.039 mg/m3
BMCL10HEC = 0.725 mg/m3 + 30 = 0.024 mg/m3
Thus, the RfC derived from the BMCL10 mc is 0.02 mg/m3.
5.3. CANCER ASSESSMENT
As discussed in Section 4.6, human data are inadequate for assessment of the potential
human carcinogenicity of 1,3-dichloropropene. The human data on 1,3-dichloropropene, which
consist of anecdotal reports of three cases of cancer, cannot be used to infer a causal association
with 1,3-dichloropropene exposure because the possibility of confounding factors has not been
considered or ruled out (U.S. EPA, 1987).
Animal carcinogenicity data are sufficient to provide a quantitative assessment of the
potential human carcinogenicity of 1,3-dichloropropene. The weight of evidence for both the
oral and inhalation carcinogenicity of 1,3-dichloropropene indicates that this compound is
carcinogenic in animals. A gavage study in rodents (NTP, 1985) indicates that 1,3-
dichloropropene at relatively high bolus doses is carcinogenic at multiple sites (forestomach and
liver in rats, and forestomach, urinary bladder, and lung in mice). On the other hand, dietary or
inhalation administration only produced tumors in target organs with extensive geriatric changes
(rat liver adenomas in Stott et al. [1995]) and/or high background incidences of benign tumors
(mouse lung adenomas in Lomax et al. [1989]), and only at the highest doses tested.
EPA cancer risk assessment guidelines (U.S. EPA, 1996a and 1987) recommend a linear
quantitative cancer assessment for 1,3-dichloropropene because there is evidence that 1,3-
dichloropropene is a mutagen. To support a nonlinear assessment, the guidelines require the
identification of a nonlinear mode of tumor formation. Although GSH is hypothesized to protect
against tumor formation, which would result in a nonlinear dose-response, this hypothesis is not
supported by the available mechanistic data. Thus, in the absence of definitive data for a
nonlinear mechanism of tumor formation, a linear approach is taken for the cancer dose-response
assessment. The linear approach assumes that a straight line best represents the shape of the dose
response from the point of departure to the origin.
5.3.1. Oral Exposure—Choice of Study/Data With Rationale and Justification
All the chronic studies (NTP, 1985; Stott et al., 1995; Redmond et al., 1995) for orally
administered 1,3-dichloropropene were relatively well conducted, but each study has distinct
limitations. Limitations of the NTP gavage studies (1985) include the bolus dosing and the
thrice-weekly rate of administration. Current test guidelines recommend seven times weekly, but
indicate that five times/week is acceptable (U.S. EPA, 1998d). The NTP (1985) study was
rejected by EPA for RfD development (IRIS, online 10/1/90) because the dosing regimen (high
doses by gavage three times/week) was not well designed to study chronic toxicity. Another
48
-------
problem with the NTP (1985) study is that the dichloropropene formulation contained
epichlorohydrin, which NTP acknowledged as a possible contributor to tumorigenic effects in the
forestomach. In addition, gavage administration is less relevant to human exposure than is
dietary administration. However, the dietary studies, which used a microencapsulation
technique, did not provide data on the in-cage stability of the food mixture. The absence of such
information leaves uncertainty as to the actual dose received by the animals. In the absence of a
single best study, both the NTP (1985) and Stott et al. (1995) studies will be used for the
quantitative cancer assessment.
NTP (1985) reported forestomach squamous cell papilloma and carcinoma in male F344
rats at 50 mg/kg technical-grade 1,3-dichloropropene thrice weekly. Rats exhibited a significant
incidence of liver adenomas at 25 mg/kg, and one carcinoma was observed at 50 mg/kg.
Incidences of liver tumors and forestomach tumors were not statistically significant in female
rats. At 50 mg/kg, female B6C3F1 mice gavaged thrice weekly exhibited statistically significant
transitional cell carcinoma of the urinary bladder, while male mice displayed bronchioalveolar
adenoma/carcinoma. Both sexes of mice exhibited significant incidences of forestomach
papilloma/carcinoma and bronchioalveolar adenoma/carcinoma at 100 mg/kg.
In the dietary study, Stott et al. (1995) showed that F344 rats exposed to up to 25
mg/kg/day 1,3-dichloropropene developed late-onset benign liver tumors (see Table 8). One
nonfatal hepatocellular carcinoma was also observed. A small, nonsignificant increase in
Table 8. Incidence of tumors in chronic bioassays
Administered
dose
(mg/kg/event)a
0
2.5
12.5
25
25
50
50
100
Human
equivalent dose
(mg/kg/day)b
0
0.65
3.22
2.75
6.31
2.88
5.4
5.81
Hepatocellular
adenoma/
carcinoma:
male rats
(NTP, 1985)
1/49
—
—
6/48
—
—
8/50
—
Urinary bladder
carcinoma:
female mice
(NTP, 1985)
0/50
—
—
—
—
8/50
—
21/47
Hepatocellular
adenoma/
carcinomas:
male rats
(Stott et al., 1995)
2/49
1/50
6/50
—
10/49
—
—
—
"Daily doses for dietary study (Stott et al., 1995); dose per gavage for NTP (1985) study.
bAdministered doses averaged over 7 days/week (if necessary) and adjusted to human equivalent doses by
multiplying by (animal body weight/human body weight)1'4 and the % 1,3-dichloropropene in the formulation (92%
for NTP [1985] and 96% for Stott et al. [1995]).
- Dose not used.
49
-------
hepatocellular adenomas was also observed in female rats, but the increased incidence was
within the historical control range. No tumors were observed in a 2-year dietary study with mice
exposed to up to 50 mg/kg/day 1,3-dichloropropene (Redmond et al., 1995) in the diet, but as
explained earlier, dosing may have been inadequate.
The tumor data chosen for the quantitative assessment are shown in Table 8. Since there
was concordance in rat liver tumors in both the gavage (NTP, 1985) and feeding studies (Stott et
al.,1995), these tumors were chosen for quantitative assessment. The forestomach tumor data in
rats and mice in the NTP (1985) study were not chosen due to the confounding effects of
epichlorohydrin in the formulation and because the tumors did not appear in the feeding studies
(Stott et al., 1995; Redmond et al., 1995). The male mouse tumor data (bronchioalveolar
adenoma/carcinoma) are unacceptable for quantitative assessment because the control group
survival was inadequate owing to early deaths attributed to myocarditis. Although the urinary
bladder tumors in female mice in the gavage study (NTP, 1985) were not observed in the feeding
study (Redmond et al., 1995), these data were chosen for quantitative assessment
because transitional cell carcinoma of the bladder is a rare tumor and because the dosing for mice
in the feeding study may have been inadequate, as it was not verified by in-cage stability
measurements.
5.3.1.1. Dose Conversion and Dose-Response Analysis
Thrice-weekly gavage doses (NTP, 1985) were converted to an average daily dose by
multiplying by 3 times/week and dividing by 7 days/week. In accordance with cancer risk
assessment guidelines (U.S. EPA, 1996a), daily doses from both NTP (1985) and Stott et al.
(1995) were adjusted to human equivalent doses by dividing by (human body weight/animal
body weight)174 using 70 kg as the human body weight and the final body weights for the animal
weights. Doses were also adjusted for the purity of the formulation (see Appendix A, II).
Oral cancer potency factors were calculated from each set of tumor data in Table 8 using
recommendations from both the proposed cancer risk assessment guidelines (U.S. EPA, 1996a)
and the existing cancer risk assessment guidelines (U.S. EPA, 1987). The multistage model for
extra risk from EPA's Benchmark Dose Software, Version Lib, was used for analysis in
accordance with the proposed guidelines. Human equivalent doses and tumor incidences in
Table 8 were used to calculate the point of departure, the 95% lower confidence limit of the ED10
(LED10) (U.S. EPA, 1996a) The cancer slope factor (i.e., risk at 1 mg/kg/day) was estimated by
drawing a straight line from the point of departure to the origin, thus, the cancer slope =
0.1/LED10 (see Table 9). For analysis by the existing guidelines, the GLOB AL86 linearized
multistage model for extra risk was applied to the same data to determine the slope at 1
mg/kg/day. For both analyses, the unit risk for drinking water was calculated by multiplying the
cancer slope factor by 1/70 kg, 2 L/day and 0.001 (for conversion of mg to fig). Risk-specific
concentrations corresponding to 10"4, 10"5, and 10"6 risk were calculated by dividing risk level by
unit risk. Table 9 shows the oral cancer potency results from the multistage model (proposed
guidelines) and Table 10 shows the results for the linearized multistage model (existing
guidelines).
50
-------
Table 9. Multistage cancer potency calculations
Parameter
LED10
Slope factor
(mg/kg/day J1
Drinking water unit
risk
(risk per • g/L)
10-4risk
lO'5 risk
10'6 risk
Hepatocellular adenoma/
carcinoma:
male rats
(NTP, 1985)
2 mg/kg/day
5E-2
1E-6
7E+1 • g/L
7EO • g/L
7E-1 • g/L
Urinary bladder
carcinoma:
female mice
(NTP, 1985)
1 mg/kg/day
1E-1
3E-6
4E+1 • g/L
4EO • g/L
4E-1 • g/L
Hepatocellular adenoma/
carcinoma:
male rats
(Stott et al., 1995)
2 mg/kg/day
4E-2
1E-6
8E+1 • g/L
8EO • g/L
8E-1 • g/L
Table 10. Linearized multistage cancer potency calculations
Parameter
Slope factor
(mg/kg/day)"1
Drinking water unit
risk
(risk per • g/L)
ID'4 risk
ICT5 risk
10'6 risk
Hepatocellular adenoma/
carcinoma:
male rats
(NTP, 1985)
5E-2
2E-6
7E+1 • g/L
7EO • g/L
7E+1 • g/L
Urinary bladder
carcinoma:
female mice
(NTP, 1985)
1E-1
3E-6
4E+1 • g/L
4EO • g/L
4E-1 • g/L
Hepatocellular adenoma/
carcinoma:
male rats
(Stott et al., 1995)
5E-2
1E-6
8E+1 • g/L
8EO • g/L
8E-1 • g/L
The cancer slope factors calculated by the linearized multistage model ranged from
5E-2 to 1E-1 (mg/kg/day)"1 and the cancer slope factors from the multistage model ranged from
4E-2 to 1E-1 (mg/kg/day)"1. Although there was little difference in the results between the two
models, the cancer slope factors calculated from the linearized multistage model are
recommended because the proposed cancer guidelines have not been finalized. Because there is
less uncertainty in the delivered dose for the NTP (1985) study, the slope factor of 1E-1
(mg/kg/day)"1 from the mouse bladder tumor data is recommended.
51
-------
5.3.2. Inhalation Exposure—Choice of Study/Data With Rationale and Justification
The critical study for assessment of inhalation cancer potency is the study by Lomax et al.
(1989) in which rats and mice were exposed to up to 272 mg/m3 of 1,3-dichloropropene for 2
years. This was a well-designed and well-conducted bioassay that followed standard guidelines.
Epoxidized soybean oil replaced epichlorohydrin as the stabilizing agent in this formulation of
1,3-dichloropropene, which eliminated a potentially confounding effect of epichlorohydrin. The
study by Lomax et al. (1989) is the only 2-year inhalation bioassay available. Confidence in this
The only neoplastic response observed in any species and sex was an increased incidence
of benign lung tumors (bronchioalveolar adenomas), with late onset, in male mice at 272 mg/m3
(see Table 11). The incidence in the high-dose group exceeded the range of historical control
rates among mice in the same laboratory. No statistically significant incidence of tumors was
found in male or female rats at any exposure level.
5.3.2.1. Dose Conversion and Dose-Response Analysis
The administered dose was adjusted for purity and for continuous exposure duration as
shown in the notes for Table 11. The adjustment to convert animal exposure concentrations to
HECs for a Category 2 gas depends upon the critical target. For the critical effect of
bronchioalveolar adenoma, algorithms for the thoracic effects of Category 1 gases are used to
adjust animal exposure concentrations of 1,3-dichloropropene to HECs (U.S. EPA, 1994b). The
HEC for a Category 1 gas is derived by multiplying the duration-adjusted concentrations by an
interspecies dosimetric adjustment for gas:respiratory effects in the tracheobronchial and
Table 11. Incidence of bronchioalveolar adenomas in male
mice exposed to 1,3-dichloropropene via inhalation
Administered
dose (mg/m3)
0
22.7
90.8
272
Purity and
duration
adjusted dosea
(mg/m3)
0
3.7
15
45
Human
equivalent
concentration15
(mg/m3)
0
11.9
48.2
144.4
Tumor
incidence
9/50
6/50
13/50
22/50
Correction for purity of formulation concentration (92%) and correction for intermittent exposure to
continuous exposure: 22.7 mg/m3 x 0.92 x 6/24 hours x 5/7 days = 3.7 mg/m3.
bCorrection for thoracic effects using RGDR(TH) of 3.21 as described in the text.
52
-------
pulmonary (i.e., thoracic) regions of the lung, according to the following calculation (U.S. EPA,
1994b):
RGDR(TH) = (MVa/Sa)/(MVh/Sh)
where
RGDR(TH) = regional gas dose ratio for the thoracic area of the lung
MVa = animal minute volume (mouse = 0.041 L/min)
MVh = human minute volume (13.8 L/min)
Sa = surface area of the thoracic region of the animal lung (mouse = 503.5 cm2)
Sh = surface area of the thoracic region of the human lung (543,200 cm2).
Using default values, RGDR(TH) = (0.041/503.5)7(13.8/543,200) = 3.21.
Purity- and duration- adjusted animal concentration x 3.21 = HEC value:
3.7 mg/irf x 3.21 = 11.9 mg/m3.
Inhalation unit risk factors were calculated from the tumor data in Table 11 using
recommendations from both the proposed cancer risk assessment guidelines (U.S. EPA, 1996a)
and the existing cancer risk assessment guidelines (U.S. EPA, 1987). The multistage model for
extra risk from EPA's Benchmark Dose Software, Version 1. Ib, was used for analysis in
accordance with the proposed guidelines. HECs and tumor incidences in Table 11 were used to
calculate the point of departure, the 95% lower confidence limit of the EC10 (LEC10) (U.S. EPA,
1987) The cancer slope factor, or unit risk (i.e., risk at 1 • g/m3), was estimated by multiplying
the LEC10 by 1,000 to convert mg to • g, and then drawing a straight line from the point of
departure to the origin. Thus, the unit risk = 0.1/(LEC10 x 1,000) Concentrations corresponding
to doses yielding 10"4, 10"5, and 10"6 risk levels were calculated by dividing risk level by unit risk.
Table 12 shows the inhalation cancer potency results for both the multistage (proposed
guidelines) and linearized multistage (existing guidelines) analyses. The air unit risks for both
the multistage and linearized multistage model were 4E-6 (• g/m3)"1.
Table 12. Inhalation cancer potency results
Parameter
Point of departure, 95% LCL of ED10
Air unit risk (95% UCL on risk at 1
•g/m3)
Concentration at 10"4 risk
Concentration at 10"5 risk
Concentration at 10"6 risk
Multistage
24 mg/m3
4E-6 (• g/m3)-1
2E+1 • g/m3
2EO • g/m3
2E-1 • g/m3
53
-------
6. MAJOR CONCLUSIONS IN THE CHARACTERIZATION OF
HAZARD AND DOSE RESPONSE
6.1. HUMAN HAZARD POTENTIAL
1,3-Dichloropropene is a colorless to straw-colored liquid with a sharp, sweet,
penetrating, chloroform-like odor (HSDB, 1998). It is miscible in organic solvents and
evaporates easily. 1,3-Dichloropropene is used extensively in agriculture as a preplanting
fumigant for the control of nematodes. Commercial formulations, including Tel one, D-D,
Di-Trapex, and Vorlex, contain mixtures of cis and trans isomers, other chloropropenes,
chloropropanes, and stabilizers.
In both humans and animals, 1,3-dichloropropene is rapidly absorbed, conjugated with
GSH to form water-soluble mercapturic acids, and quickly excreted in the urine. GSH reduces
the formation of mutagenic epoxides, which are produced through a minor metabolic pathway at
high (~LD50) doses of dichloropropene. 1,3-Dichloropropene does not bioaccumulate in target
tissues. In vitro and in vivo mutagenicity and genotoxicity tests have yielded mixed results.
The only repeated exposure human toxicity data for 1,3-dichloropropene are case study
data showing that direct contact produced dermatitis. Accidental high-dose poisoning following
chemical spills or accidental releases has caused a dose-related range of acute neurotoxic
symptoms. Accidental ingestion of large quantities of 1,3-dichloropropene has also been
reported to be fatal. In chronic studies with animals, 1,3-dichloropropene produces
histopathology at the portal of entry or in organs involved in excretion of metabolites.
Specifically, inhalation exposure produces nasal histopathology in mice and rats and urinary
bladder hyperplasia in mice (Lomax et al., 1989). Mild hyperplasia of the forestomach was
observed in rats ingesting 1,3-dichloropropene with their feed (Stott et al., 1995). No
lexicologically significant effects were noted in reproductive (Breslin et al., 1989) or
developmental toxicity studies with rats and rabbits (Hanley et al., 1988).
There is no evidence associating carcinogenicity in humans to 1,3-dichloropropene
exposures. Three case studies in which men accidentally exposed to acute high doses
subsequently developed blood cancers cannot be used to infer a causal association with 1,3-
dichloropropene exposure because the possibility of confounding factors has not been considered
or ruled out (U.S. EPA, 1987).
Chronic gavage studies in animals (NTP, 1985) yielded forestomach squamous cell
papilloma and carcinoma in rats and mice at bolus doses of 50 mg/kg, but no tumors were
observed in the glandular stomach, which is more relevant to humans. Liver tumors were
observed in male rats, and urinary bladder tumors were observed in female mice. In chronic
dietary studies with rats, 1,3-dichloropropene produced an increased incidence of benign
hepatocellular adenomas and one nonfatal hepatocellular carcinoma in male rats at the highest
dose tested, 25 mg/kg/day (Stott et al., 1995). No tumors were observed in mice (Redmond et
al., 1995). In chronic inhalation studies with rats and mice, 1,3-dichloropropene significantly
increased the incidence of benign bronchioalveolar tumors in the lungs of male mice at the
highest dose tested, 272 mg/m3, but produced no tumors in rats (Lomax et al., 1989).
54
-------
Several high-concentration and high-dose genotoxicity studies have shown that 1,3-
dichloropropene is mutagenic in the presence of metabolizing enzymes and that GSH protects
against the mutagenic effects (Greedy et al., 1984; Watson et al., 1987; and Schneider et al.,
1998a,b). Because there is no evidence for a nongenotoxic mode of action, the mode of action of
1,3-dichloropropene is assumed to be DNA toxicity.
Under the proposed cancer risk assessment guidelines (U.S. EPA, 1996a), the weight of
evidence for evaluation of cancer hazard strongly suggests that 1,3-dichloropropene is likely to
be carcinogenic in humans. Although its tumorigenic action is dose-dependent in chronic animal
bioassays, positive evidence of mutagenicity and the lack of mode-of-action data to support a
nonlinear mechanism requires the quantitative cancer assessment to assume a linear dose
response.
Under EPA's (1987) cancer risk assessment guidelines, 1,3-dichloropropene would be
characterized as a class B2, probable human carcinogen, i.e., inadequate data in humans,
sufficient data in animals. This characterization is supported by observations of tumors in F344
rats (forestomach, liver) and B6C3F1 mice (forestomach, urinary bladder, and lung) at high bolus
doses, observations of liver tumors in F344 rats at lower dietary doses, and the formation of
mutagenic epoxide metabolites at high doses.
A recent International Agency for Research on Cancer Working Group (IARC, 1999)
report also concluded that there was sufficient evidence in animals to determine that 1,3-
dichloropropene is possibly carcinogenic to humans. In addition, 1,3-dichloropropene is listed
under California's Safe Drinking Water and Toxic Enforcement Act of 1986 (Proposition 65) as
a chemical known to the State to cause cancer (OEHHA, 1998).
6.2. DOSE RESPONSE
The quantitative estimates of human risk as a result of low-level chronic exposure to 1,3-
dichloropropene are based on high-dose animal experiments because no human data exist.
6.2.1. Noncancer Dose-Response Assessment
Noncancer ingestion potency estimates were derived from a 2-year chronic bioassay
(Stott et al., 1995) in which chronic irritation, exhibited by mild histopathology of the rat
forestomach, was the critical effect. Significantly reduced body weight, which occurred in both
rats and mice at higher doses, was the co-critical effect. The RfD of 0.03 mg/kg/day was
calculated using BMD analysis and the application of UF = 100 to extrapolate from rats to
humans and to account for within species variability among humans.
The overall confidence in the oral RfD is high. The confidence in the principal study is
high. The study was well designed and well conducted and followed standard guidelines for
chronic bioassays. Results from a subchronic ingestion study by Haut et al. (1996) are consistent
with the findings in the 2-year bioassay. The confidence in the database, judged here as high, is
55
-------
much improved from the earlier version on IRIS (10/1/90) because of the availability of three
new dietary bioassay studies (Stott et al., 1995; Haut et al., 1996; Redmond et al., 1995) as well
as new studies on metabolism and genotoxicity. Well-conducted studies on reproductive and
developmental toxicity used inhalation as the route of administration. However, sufficient
toxicokinetics are available to show that 1,3-dichloropropene is well absorbed by all routes.
The RfC, the daily inhalation exposure to the human population that is likely to be
without an appreciable risk of deleterious effects during a lifetime, is 0.02 mg/m3. The RfC was
derived from the 2-year inhalation bioassay by Lomax et al. (1989). Histopathology was
observed in the nasal epithelium of both rats and mice. The female mouse was identified as the
most sensitive sex and species because statistically significant increases in the incidence of nasal
histopathology were observed at lower doses than in the male mouse or in either gender of rat.
The overall confidence in the RfC is high. The inhalation toxicity potency values are
based on the findings of a well-conducted 2-year bioassay (Lomax et al. 1989). The results of
the 2-year bioassay are supported by the findings from a 90-day subchronic inhalation study
(Stott et al., 1988). The overall confidence in the database is high because of the availability of
90-day and 2-year bioassays, as well as studies on reproductive and developmental toxicity,
toxicokinetics, ingestion toxicity, and genotoxicity.
The UF used to extrapolate from rat to human and to account for within-species
variability among humans is 30. Justification for the use of a partial UF for extrapolation from
rats to humans is threefold: (a) toxicokinetics between humans and rats (see Section 3), based on
absorption, metabolism, and excretion studies, are similar; (b) 1,3-dichloropropene does not
bioaccumulate; and (c) interspecies dosimetric adjustments for pharmacokinetic differences were
made when the HEC was calculated.
6.2.2. Cancer Dose-Response Assessment
Human data are inadequate for assessment of the potential human carcinogenicity of 1,3-
dichloropropene. Animal data from exposures indicate that 1,3-dichloropropene is carcinogenic.
A chronic gavage study in rodents indicates that 1,3-dichloropropene is carcinogenic when high
bolus doses are administered. Although chronic feeding bioassays indicated that
dichloropropene's tumorigenic action is dose-dependent, the lack of mode-of-action data to
support a nonlinear mechanism requires the quantitative cancer assessment to assume a linear
dose response to derive oral and inhalation cancer potency values. The linear approach assumes
that a straight line best represents the shape of the dose response from the point of departure, at
the lower end of the range of experimental observation, to lower doses.
Cancer potencies were calculated from chronic dietary, gavage, and inhalation data using
both proposed and existing guidelines, with similar results. The oral cancer slope factors were 5
x 10'2 to 1 x 10'1 (mg/kg/day)-1. The slope factor of 1 x 10'1 (mg/kg/day)4 from the NTP (1985)
study is recommended because there is less uncertainty in the delivered dose in that study. The
California Environmental Protection Agency included forestomach tumors and calculated similar
cancer slope factors, 3.4 x 10"2 to 9.1 x 10"2 (mg/kg/day)"1, from the same studies (OHHEA,
56
-------
1999). The inhalation unit risk for the current assessment is 4 x 10"6 ( Hg/m3) 4. An inhalation
unit risk of 1.6 x 10"5 ([Ig/m3) 4 is derived from the inhalation slope factor reported by the
California Environmental Protection Agency (Department of Pesticide Regulation, 1994). The
different interspecies dosimetry adjustment used by the two agencies is responsible for the
difference in unit risks.
Confidence in the database is medium to high. Major database uncertainties are the
importance of 1,3-dichloropropene's mutagenic potential in a whole-animal system and the
precise mechanism of tumorigenic action. The results from short-term mutagenicity assays of
the parent compound are mixed, and although 1,3-dichloropropene is metabolized to mutagenic
epoxides at ~LD50 doses, the extent of epoxide formation in vivo at the low doses characteristic
of chronic exposure is unknown. In vitro assays indicated that the presence of GSH decreases
epoxide formation and abolishes or greatly reduces the mutagenic response. Thus, the linear
quantitative assessment provides a very conservative estimate of cancer potency.
7. REFERENCES
Alarie, Y. (1973) Sensory irritation of the upper airways by airborne chemicals. Toxicol Appl Pharmacol 24:279-
297.
Bousema, MT; Wiemer, GR; Van Joost, TH. (1991) A classic case of sensitization to DD-95®. Contact Dermatitis
24(2):132-133.
Breslin, WJ; Kirk, HO; Streeter, CM; etal. (1989) 1,3-Dichloropropene: two-generation inhalation reproduction
study in Fischer 344 rats. Fundam Appl Toxicol 12:129-143.
Brouwer, EJ; Evelo, CTA; Verplanke, AJW; et al. (1991) Biological effect monitoring of occupational exposure to
1,3-dichloropropene: effects on liver and renal function and on glutathione conjugation. Br J Ind Med 48(3): 167-
172.
Climie, IJG; Hutson, DH; Morrison, BJ; et al. (1979) Glutathione conjugation in the detoxification of (Z)-1,3-
dichloropropene (a component of the nematocide D-D) in the rat. Xenobiotica 9:149-156.
Greedy, CL; Brooks, TM; Dean, BJ; et al. (1984) The protective action of glutathione on the microbial mutagenicity
of the Z- and E-isomers of 1,3-dichloropropene. Chem Biol Interact 50(l):39-48.
Department of Pesticide Regulation. (1994) Interim risk assessment of 1,3-dichloropropene for the proposed use of
Telone II* (1994-1995). Department of Pesticide Regulation, California Environmental Protection Agency.
Dietz, FK; Hermann, EA; Ramsey, JC. (1984a) The pharmacokinetics of 14C-1,3-dichloropropene in rats and mice
following oral administration. Toxicologist 4:147(Abstr. No. 585).
Dietz, FK; Dittenber, DA; Kirk, HD; et al. (1984b) Non-protein sulfhydryl content and macromolecular binding in
rats and mice following oral administration of 1,3-dichloropropene. Toxicologist 4:147 (Abstr. 586).
Dietz, F; Hermann, E; Kastl, P; et al. (1985) 1,3-Dichloropropene: pharmacokinetics, effect on tissue non-protein
sulfhydryls, and macromolecular binding in Fischer-344 rats and B6C3F1 mice following oral administration. The
Dow Chemical Company, Midland, MI. No. 86-870023122.
57
-------
Dourson, ML; Stara, JF. (1983) Regulatory history and experimental support of uncertainty (safety) factors. Regul
Toxicol Pharmacol 3:224-238.
Fisher, GD; Kilgore, WW. (1988a) Mercapturic acid excretion by rats following inhalation exposure to 1,3-
dichloropropene. Fundam Appl Toxicol 112:300-307.
Fisher, GD; Kilgore, WW. (1988b) Tissue levels of glutathione following acute inhalation of 1,3-dichloropropene.
JToxicolEnvironHealth 23(2):171-182.
Fisher GD; Kilgore, WW. (1989) Pharmacokinetics of S-[3-chloropropene-2-enyl]glutathione in rats following
acute inhalation exposure to 1,3-dichloropropene. Xenobiotica 19(3):269-278.
Ghia, M; Robbiano, L; Allavena, A; et al. (1993) Genotoxic activity of 1,3-dichloropropene in a battery of in vivo
short-term tests. Toxicol Appl Pharmacol 120:120-125.
Hanley, TR, Jr.; John-Greene, JA; Young, JT; et al. (1988) Evaluation of the effects of inhalation exposure to
1,3-dichloropropene on fetal development in rats and rabbits. Fundam Appl Toxicol 8:562-570.
Haut, KT; Stebbins, KE; Johnson, KA; et al. (1996) Subchronic toxicity of ingested 1,3-dichloropropene in rats and
mice. Fundam Appl Toxicol 32:224-232.
Haworth, S; Lawlor, T; Mortelmans, K; et al. (1983) Salmonella mutagenicity testing for 250 chemicals. Environ
Mutagen Suppl 1:3-142.
Hayes, WJ. (1982) Pesticides studied in man. Baltimore: Williams and Wilkins, pp. 139-171.
Hernandez, AF; Martin-Rubi, JC; Ballesteros, JL; et al. (1994) Clinical and pathological findings in fatal 1,3-
dichloropropene intoxication. Hum Exper Toxicol 13(5):303-306.
HSDB (Hazardous Substances Data Base). On-line. 1998.
Hutson, DH; Moss, JA; Pickerine, BA; et al. (1971) The excretion and retention of components of the soil
fumigant D-D and their metabolites in the rat. Food Cosmet Toxicol 8:677-680.
IARC Working Group. (1999) 1,3-dichloropropene. In: Re-evaluation of some organic chemicals, hydrazine and
hydrogen peroxide. Lyon, France: International Agency for Research on Cancer; pp. 933-945. (IARC monographs
on the evaluation of carcinogenic risks to humans: v. 71, part 3).
Kevekordes, S; Gebel, T; Pav, K; et al. (1996) Genotoxicity of selected pesticides in the mouse bone-marrow
micronucleus test and in the sister-chromatid exchange test with human lymphocytes in vitro. Toxicol Lett 89:35-
42.
Kezic, S; Monster, AC; Verplanke, AJW; et al. (1996) Dermal absorption of cis-1,3-dichloropropene vapor:
human experimental exposure. Hum Exper Toxicol 15(5):396-399.
Konishi, Y; Kawabata, A; Denda, A; et al. (1980) Forestomach tumors induced by orally administered
epichlorohydrin in male Wistar rats. Gann 71:922-923.
Linnett, SL; Clark, DG; Blair, D; et al. (1988) Effects of subchronic inhalation of D-D
(l,3-dichloropropene/l,2-dichloropropene) on reproduction in male and female rats. Fundam Appl Toxicol
10:214-223.
Lomax, LG; Stott, WT; Johnson, KA; et al. (1989) The chronic toxicity and oncogenicity of inhaled technical-
grade 1,3-dichloropropene in rats and mice. Fundam Appl Toxicol 12:418-431.
58
-------
Markovitz, A; Crosby, WH. (1984) Chemical carcinogenesis. A soil fumigant, 1,3-dichloropropene, as possible
cause of hematologic malignancies. Arch Intern Med 144:1409-1411.
Maronpot, RR; Montgomery, CA; Boorman, GA; et al. (1986) National Toxicology Program nomenclature for
hepatoproliferative lesions of rats. Toxicol Pathol 14:163-273.
Martelli, A; Allavena, A; Ghia, M; et al. (1993) Cytotoxic and genotoxic activity of 1,3-dichloropropene in
cultured mammalian cells. Toxicol Appl Pharmacol 120:114-119.
Nater JP; Gooskens, VHJ. (1976) Occupational dermatosis due to a soil fumigant. Contact Dermatitis 2(4):227-
229.
National Research Council. (1983) Risk assessment in the Federal Government: managing the process.
Washington, DC: National Academy Press.
NTP (National Toxicology Program). (1985) Toxicology and carcinogenesis studies of Telone II® (technical-grade
1,3-dichloropropene containing 1% epichlorohydrin as a stabilizer) in F344/N rats and B6C3F1 mice (gavage
studies). U.S. Dept. of Health and Human Services, Technical Report Series No. 269.
Neudecker, T; Henschler, D. (1986) Mutagenicity of chloroolefins in the Salmonella/ mammalian microsome test.
MutatRes 170:1-10.
OEHHA. (1999) Public health goal for 1,3-dichloropropene in drinking water. Office of Environmental Health
Hazard Assessment, California Environmental Protection Agency. February 1999.
Osterloh, JD; Wang, R; Schneider, S; et al. (1989) Biological monitoring of dichloropropene: air concentrations,
urinary metabolite, and renal enzyme excretion. Arch Environ Health 44(4):207-213.
Parker, CM; Coate, W; Voelker, R. (1982) Subchronic inhalation toxicity of
l,3-dichloropropene/l,2-dichloropropene (D-D) in mice and rats. J Toxicol Environ Health 9:899-910.
Redmond, JM; Stebbins, KE; Stott, WT. (1995) Telone II® soil fumigant: two-year dietary chronic
toxicity/oncogenicity study in B6C3F1 mice - final report. Dow Chemical Company, Midland, MI. Study # M-
003993-032.
Sanzgiri, UY; Kim, HJ; Muralidhara, S; et al. (1995) Effect of route and pattern of exposure on the
pharmacokinetics and acute hepatotoxicity of carbon tetrachloride. Toxicol Appl Pharmacol 134:148-154.
Schneider, M; Quistad, GB; Casida, JE. (1998a) 1,3-Dichloropropene epoxides: intermediates inbioactivation of
the promutagen 1,3-dichloropropene. Chem Res Toxicol 11:1137-1144.
Schneider, M; Quistad, GB; Casida, JE. (1998b) N2,7-Bis(l-hydroxy-2-oxopropyl)-2'-deoxyguanosine: identical
noncyclic adducts with 1,3-dichloropropene epoxides and methylglyoxal. Chem Res Toxicol 11:1536-1542.
Stolzenberg, SJ; Hine, CH. (1980) Mutagenicity of 2- and 3-carbon halogenated compounds in the
Salmonella/mammalian-microsome test. Environ Mutagen 2(l):59-66.
Stott, WT; Kastl, PE. (1986) Inhalation pharmacokinetics of technical-grade 1,3-dichloropropene in rats. Toxicol
Appl Pharmacol 85(3): 332-341.
Stott, WT; Young, JT; Calhoun, LL; et al. (1988) Subchronic toxicity of inhaled technical-grade
1,3-dichloropropene in rats and mice. Fundam Appl Toxicol 11:207-220.
59
-------
Stott, WT; Johnson, KA; Jeffries, TK; et al. (1995) Telone II® soil fumigant: two-year chronic
toxicity/oncogenicity study in Fischer 344 rats. The Dow Chemical Company: Midland, Michigan. Study # M-
003993-0311.
Stott, WT; Miller, TJ; Wardynski, AK. (1997a) 1,3-Dichloropropene: in vitro DNA binding. The Dow Chemical
Company.
Stott, WT; Gollapudi, BB ; Clements, CM; etal. (1997b) 1,3-Dichloropropene: mechanisms of action. The Dow
Chemical Company.
Stott, WT; Gilbert, JR; McGuirk, RJ; et al. (1998) Bioavailability and pharmacokinetics of microencapsulated
12,3-dichloropropene in rats. Toxicol Sci 41:21-28.
Streeter, C; Battjes, J; Lomax, L. (1987) Telone II® soil fumigant: An acute vapor inhalation study in Fischer 344
rats. The Dow Chemical Company. Midland, Michigan.
Talcott RE; King, J. (1984) Mutagenic impurities in 1,3-dichloropropene preparations. J Natl Cancer Inst
U.S. EPA. (Environmental Protection Agency). (1980) Ambient water quality criteria document:
dichloropropenes/dichloropropenes (Draft), p. C-12.
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. (1987) Risk assessment guidelines of 1986 (EPA/600/8-87/045, dated August 1987).
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, dated December 5, 1991. 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. EPA/600/8-90/066F. October 1994.
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. (1995) Use of the benchmark dose approach in health risk assessment. EPA/630/R-94/007, February
1995.
U.S. EPA. (1996a) Proposed guidelines for carcinogen risk assessment, Notice, 1996. Federal Register
61(79):17960-18011.
U.S. EPA. (1996b). Guidelines for reproductive toxicity risk assessment, dated October 31, 1996. Federal Register
61(212):56274-56322.
60
-------
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/100/B-98-001.
U.S. EPA. (1998c) Ambient water quality criteria for the protection of human health. 1,3-dichloropropene (1,3-
DCP), Draft. Office of Water. EPA/822/R-98-005.
U.S. EPA. (1998d) Health effects test guidelines, OPPTS 870.4200, carcinogenicity, prevention, pesticides and
toxic substances. EPA/712/C-98-211.
Valencia, R; Mason, JM; Woodruff, RC; et al. (1985) Chemical mutagenesis testing in Drosophila. III. Results of
48 coded compounds tested for the National Toxicology Program. Environ Mutagen 7:325-348.
van Welie, RTH; van Duyn, P; Brouwer, DH; et al. (1991) Inhalation exposure to 1,3-dichloropropene in the Dutch
flower-bulb culture. Part II. Biological monitoring by measurement of urinary excretion of two mercapturic acid
metabolites. Arch Environ Contain Toxicol 20(1):6-12.
Vithayathil, AJ; McClure, C; Myers, JW. (1983) Salmonella/microsome multiple indicator mutagenicity test.
MutatRes 121(l):33-37.
Von der Hude, W; Scheutwinkel, M; Gramlich, U; et al. (1987). Genotoxicity of three-carbon compounds
evaluated in the SCE test in vitro. Environ Mutagen 9:401-410.
Waechter, J; Kastl, P. (1988) 1,3-Dichloropropene: pharmacokinetics and metabolism in Fischer 344 rats following
repeated oral administration. Dow Chemical Company, Midland, MI.
Waechter, JM; Brzak, KA; McCarty, LP; et al. (1992) 1,3-Dichloropropene (Telone II® soil fumigant): inhalation
pharmacokinetics and metabolism in human volunteers (internal report). Dow Chemical Company, Midland, MI.
Watson, PW; Brooks, TM; Huckle, KR; et al. (1987) Microbial mutagenicity studies with (Z)-l,3-dichloropropene.
Chem-Biol Interact 61:17-30.
Wester, PW; Van der Heijden, CA; Bisschop, A; et al. (1985) Carcinogenicity study with epichlorohydrin (CEP)
by gavage in rats. Toxicology 36:325-339.
Yang, RSH; Huff, JE; Boorman, GA; et al. (1986). Chronic toxicology and carcinogenesis studies of Telone II® by
gavage in Fischer-344 rats and B6C3F1 mice. Toxicol Environ Health 18:377-392.
61
-------
APPENDIX A
DOSE-RESPONSE CALCULATIONS
I. NONCANCER DATA
A. Oral Exposure/RfD (Stott et al., 1995)
Forestomach histopathology in F344 rats
Administered dose
(mg/kg/day)
0
2.5
12.5
25
Incidence of
(forestomach
histopathology)
3/100
4/100
40/100
67/100
1. Determined BMDL10 (95% lower confidence limit [LCL] of BMD10) using U.S. EPA
Benchmark Dose Software Version Lib, 1998. Selected models had statistically significant
goodness-of-fit statistics (p-value>0.05) and best visual fit, particularly at low doses. The
gamma model was chosen because of better visual fit than the multistage or Weibull models.
BMP results: forestomach histopathology in F344 rats
Model
Gamma
Logistic
Multistage
Probit
Quantal-linear
Quantal-quadratic
Weibull
Chi-square
goodness-of-fit
^-value
0.2435
0.0028
0.0677
0.0100
0.0439
0.0306
0.1589
/7-value
>0.05
X
X
X
Visual
rank
1
o
J
2
BMD10
(mg/kg/day)
5.07
7.25
4.6
6.75
2.71
7.17
4.82
BMDL10
(mg/kg/day)
3.38
6.31
2.87
5.9
2.3
6.57
3.22
62
-------
2. Derived RfD by applying necessary UFs to BMDL10. UF = 10 for interspecies extrapolation
x 10 for intraspecies extrapolation =100:
BMD10 = 5.07 - 100 = 0.05 mg/kg/day
BMDL10 = 3.38 - 100 = 0.03 mg/kg/day
The RfD derived from the BMDL10 is 0.03 mg/kg/day.
B. Inhalation Exposure/RfC (Lomax et al., 1989)
1. Converted formulation doses to 1,3-dichloropropene doses. Duration adjusted to 24
hours/day, 7 days/week.
Duration and purity-adjusted concentration = (% 1,3-dichloropropene in commercial
formulation) x 6/24 hrs x 5/7 days = 22.7 x 0.92 x 6/24 hrs x 5/7 = 3.7 mg/m3.
Incidence of nasal histopathology in female B6C3F1 mice
Administered dose
(mg/m3)
0
22.7
90.8
272
Adjusted dose3 (mg/m3)
0
3.7
14.9
44.7
Incidence of nasal
hypertrophy/
hyperplasia
4/50
4/50
28/50
49/50
"Exposure duration and purity adjustments.
2. Determined BMCL10 (95% LCL of BMC10) using U.S. EPA Benchmark Dose Software
Version Lib, 1998. Selected model with statistically significant goodness-of-fit and best visual
fit, particularly at low doses. The gamma, logistic, multistage, quantal-quadratic and Weibull
models provided statistically significant fits. Of these, the gamma model was chosen because it
provided the best visual fit at low doses.
63
-------
BMC results: nasal hypertrophy/hyper
Model
Gamma
Logistic
Multistage
Probit
Quantal-linear
Quantal-quadratic
Weibull
Chi-square
goodness-of-fit
/7-value
0.4654
0.0614
0.1066
0.0219
0.0085
0.1312
0.2202
/7-value
>0.05
X
X
X
X
X
plasia in female B6C3F1 mice
Visual
rank
1
4
5
2
o
6
BMC1(j
(mg/m3)
5.91
5.43
5.01
5.23
1.92
6.34
4.97
BMCL.O
(mg/m3)
3.66
4.38
2.63
4.36
1.54
5.43
3.14
3. Applied adjustment for HEC for a Category 1 gas to BMC10 and BMCL10. The HEC for a
Category 1 gas is derived by multiplying the animal BMC10 and BMCL10 by an interspecies
dosimetric adjustment for extrathoracic effects according to the following calculation (U.S. EPA,
1994b):
RGDR(ET) = (MVa/Sa)/(MVh/Sh)
where
RGDR(ET) = regional gas dose ratio for the extrathoracic area of the respiratory tract
MVa = animal minute volume (mouse = 0.041 L/min)
MVh = human minute volume (13.8 L/min)
Sa = surface area of the extrathoracic region in the animal (mouse = 3 cm2)
Sh = surface area of the extrathoracic region in the human (200 cm2).
Using default values, the RGDR(ET) = (0.041/3)7(13.8/200) = 0.014/0.069 = 0.198.
The animal BMC10 and BMCL10 are then multiplied by 0.198 to yield the HECs of these
values:
BMC10HEC = BMC10 x 0.198 = 5.91 x 0.198 = 1.17 mg/m3
BMCL10HEC= BMCL10 x 0.198 = 3.66 x 0.198 = 0.725 mg/m3
4. Derived RfC by applying necessary UFs to BMC10HEC and BMCL10HEC.
UF = 3 for interspecies extrapolation x 10 for intraspecies extrapolation = 30
BMC10HEC= 1.17 mg/m3 + 30 = 0.039 mg/m3
BMCL10HEC = 0.725 mg/m3 - 30 = 0.024 mg/m3
The RfC is 0.02 mg/m3.
64
-------
II. CANCER DATA
A. Oral exposure (Stott et al., 1995; ad libitum in feed; male rats, hepatocellular
adenoma/carcinoma in male rats)(NTP, 1985; gavage; hepatocellular adenoma/carcinoma
in male rats and urinary bladder carcinoma in female mice)
1. Scaled administered doses to human equivalent doses by multiplying administered dose *
(animal BW/human BW)1/4 x 0.96 (purity). Human weight = 70 kg.
Incidence of hepatocellular adenoma/carcinoma in F344 rats
Administered dose
(mg/kg/day)
0
2.5
12.5
25
Final mean
animal weight
(kg)
0.384
0.374
0.364
0.335
Human equivalent
dose (mg/kg/day)
0
0.65
3.22
6.31
Incidence of
hepatocellular
adenoma3
2/49
1/50
6/50
10/49
a Fifty rats started in each group. Because hepatocellular adenomas and carcinomas were not observed until
study termination, rats who died before day 365 were excluded from the analysis: one control rat (day 195)
and one high-dose rat (day 187).
2. Fit data to multistage model from U.S. EPA's Benchmark Dose Software version 1. Ib.
Multistage model results for hepatocellular adenoma/carcinoma
(Stott et al., 1995)
Model
Multistage
Chi-square
goodness-of-fit
/7-value
0.3341"
ED10
(mg/kg/day)
4.02
LED10
(mg/kg/day)
2.25
a Statistically significant fit.
3. For linear assessment per proposed cancer risk assessment guidelines (U.S. EPA, 1996a),
used 95% lower confidence level of ED10 (same as BMD10) from the multistage model for extra
risk as the point of departure. Cancer slope factor (risk at 1 mg/kg/day) was estimated by
drawing a straight line from point of departure to the origin, thus cancer slope = 0.1/BMD10.
Also used GLOBAL86 linearized multistage model per existing cancer risk assessment
guidelines (U. S. EPA, 1987). Unit risk for drinking water = cancer slope factor x 1/70 kg x 2
L/day x 1E-3. Concentrations corresponding to 10"4, 10"5, and 10"6 risk levels were calculated by
dividing risk level by unit risk.
65
-------
Linear assessment results for cancer via oral route hepatocellular
adenoma/carcinoma (Stott et al., 1995)
Parameter
Point of departure, 95% LCL of ED10
Cancer slope factor (95% UCL on
risk at 1 mg/kg/day)
Drinking water unit risk
Concentration at 10"4 risk
Concentration at 10"5 risk
Concentration at 10"6 risk
BMDS value
2.3 mg/kg/day
4.4E-2 (mg/kg/day)-1
1.3E-6per(«g/L)
7.7E+1 • g/L
7.7EO • g/L
7.7E-1 • g/L
GLOBAL86 value
2.3 mg/kg/day
4.6E-2 (mg/kg/day)-1
1.3E-6per(«g/L)
7.7E+1 • g/L
7.7EO • g/L
7.7E-1 • g/L
B. Oral exposure, by gavage three times per week NTP (1985); male rats, hepatocellular
adenoma/carcinoma
1. Scaled administered doses to continuous human equivalent doses by administered dose x (3
gavages/week + 1 days/week) x (animal BW/human BW)1/4 x 0.92 (purity). Human weight =
70kg.
Incidence of hepatocellular adenoma/carcinoma in F344 rats (NTP, 1985)
Administered dose
(mg/kg/day)
0
25
50
Mean final
animal
weight (kg)
0.418
0.423
0.393
Human
equivalent
(continuous)
dose
(mg/kg/day)
0
2.75
5.40
Incidence of
hepatocellular
adenoma/
carcinomaa
1/49
6/48
8/50
a 52 rats started in each group. Since hepatocellular adenomas and carcinomas were not observed until
weeks 106-108, rats who died during the first 52 weeks of the study were excluded from the analysis: three
control rats (weeks 1, 8, and 49), four low dose rats (weeks 2 and 3) and two high dose rats (weeks 12 and
28).
2. Fit data to linear multistage model from U.S. EPA's Benchmark Dose Software
version Lib.
66
-------
Model results for hepatocellular adenoma/carcinoma:
hepatocellular adenoma/carcinoma (NTP, 1985)
Model
Multistage
Chi-square
goodness-of-fit
/7-value
0.54123
ED10
(mg/kg/day)
3.38
LED10
(mg/kg/day)
2.05
1 Statistically significant fit.
3. For linear assessment per proposed cancer risk assessment guidelines (U.S. EPA, 1996a),
used 95% LCL of ED10 (LED10) from the linear multistage model for extra risk as the point of
departure. Cancer slope factor (risk at 1 mg/kg/day) was estimated by drawing a straight line
from point of departure to the origin, thus cancer slope = 0.1/LED10. Also used GLOBAL86
linearized multistage model per existing cancer risk assessment guidelines (U.S. EPA, 1987).
Unit risk for drinking water = cancer slope factor x 1/70 kg x 2 L/day x 1E-3. Concentrations
corresponding to calculating doses yielding 10"4, 10"5, and 10"6 risk levels were calculated by
dividing risk level by unit risk.
Linear assessment results for cancer via oral route: Hepatocellular
adenoma/carcinoma (NTP, 1985)
Parameter
Point of departure, 95% LCL of ED10
Cancer slope factor (95% UCL on risk at
1 mg/kg-day)
Drinking water unit risk
Concentration at 10"4 risk
Concentration at 10"5 risk
Concentration at 10"6 risk
BMDS value
2.0 mg/kg/day
4.9E-2
(mg/kg/day)-1
1.4E-6per(«g/L)
7.1E+1 • g/L
7.1EO«g/L
7.1E-1 • g/L
GLOBAL86 value
2.0 mg/kg/day
5. 1E-2 (mg/kg/day)-1
1.5E-6per(«g/L)
6.7E+1 • g/L
6.7EO • g/L
6.7E-1 • g/L
C. Oral exposure, by gavage three times per week NTP (1985); female mice, urinary
bladder transitional cell carcinomas
1. Scaled administered doses to continuous human equivalent doses by administered dose x (3
gavages/week + 7 days/week) x (animal BW/human BW)1/4 x 0.92 (purity). Human weight =
70kg.
67
-------
Incidence of urinary bladder carcinoma in B6C3F1 mice
Administered
dose (mg/kg/day)
0
50
100
Mean final
animal weight
(kg)
0.035
0.032
0.033
Human
equivalent dose
(mg/kg/day)
0
2.88
5.81
Incidence of
urinary
bladder
carcinomas"
0/50
8/50
21/47
a 50 mice started in each group. Because these carcinomas were first observed in week 75, mice who died
before week 52 were excluded from the analysis: three high-dose mice (weeks 1, 25 and 30).
2. Fit data to linear multistage model from U.S. EPA's Benchmark Dose Software version Lib.
Model results for urinary bladder carcinoma in B6C3F1 mice (NTP, 1985)
Model
Multistage
Chi-square
goodness-of-fit
/7-value
1.0a
ED10
(mg/kg/day)
2.12
LED10
(mg/kg/day)
1.02
Statistically significant fit.
3. For linear assessment per proposed cancer risk assessment guidelines (U.S. EPA, 1996a),
used 95% LCL of ED10 (LED10) from the linear multistage model for extra risk as the point of
departure. Cancer slope factor (risk at 1 mg/kg/day) was estimated by drawing a straight line
from point of departure to the origin, thus cancer slope = 0.1/LED10. Also used GLOBAL86
linearized multistage model per existing cancer risk assessment guidelines (U. S. EPA, 1987).
Unit risk for drinking water = cancer slope factor x 1/70 kg x 2 L/day x 1E-3. Concentrations
corresponding to 10"4, 10"5, and 10"6 risk levels were calculated by dividing risk level by unit risk.
68
-------
Linear assessment results for cancer via oral route urinary
bladder carcinoma (NTP, 1985)
Parameter
Point of departure, 95% LCL
ofED10
Cancer slope factor (95% UCL
on risk at 1 mg/kg/day)
Drinking water unit risk
Concentration at 10"4 risk
Concentration at 10"5 risk
Concentration at 10"6 risk
BMDS value
1.0 mg/kg/day
9.8E-2 (mg/kg/day)-1
2.8E-6 per (• g/L)
3.6E+1 • g/L
3.6EO • g/L
3.6E-1 • g/L
GLOBAL86 value
1.0 mg/kg/day
9.8E-2 (mg/kg/day)-1
2.8E-6 per (• g/L)
3.6E+1 • g/L
3.6EO • g/L
3.6E-1 • g/L
D. Inhalation exposure: Lomax et al., 1989
1. Purity, duration and HEC adjustments.
Purity- and duration-adjusted concentration = exposure concentration x % 1,3-dichloropropene
in commercial formulation x 6/24 hrs x 5/7 days = 22.7 mg/m3 x 0.92 x 6/24 hrs x 5/7 = 3.7
mg/m3.
HEC for a category 1 gas is derived by multiplying the animal exposure concentrations by
dosimetric adjustment for thoracic (tracheobronchial + pulmonary) effects, because tumors were
found in the bronchioalveolar region, according to the following calculation (U.S. EPA, 1994b):
RGDR(TH) = (MVa/Sa) / (MVh/Sh)
where
RGDR(TH) = regional gas dose ratio for the thoracic area of the lung
MVa = animal minute volume (mouse = 0.041 L/min)
MVh = human minute volume (13.8 L/min)
Sa = surface area of the thoracic region in animal (mouse = 503.5 cm2)
Sh = surface area of the thoracic region in human (543,200 cm2).
Using default values, the RGDR(TH) = (0.041/503.5)(13.8/543200) = 3.21.
Purity and duration animal concentration x 3.21 = HEC value:
3.7 mg/m3 x 3.21 = 11.9 mg/m3.
69
-------
Incidence of bronchioalveolar adenomas in male mice exposed
to 1,3- dichloropropene via inhalation (Lomax et al., 1989)
Administered
dose (mg/m3)
0
22.7
90.8
272
Adjusted dosea
(mg/m3)
0
11.9
48.2
144.4
Incidence of
bronchioalveolar
adenomas
9/50
6/50
13/50
22/50
""Correction for purity, duration, HEC.
2. Fit data to linear multistage model from U.S. EPA's Benchmark Dose Software, version Lib.
Model results for bronchioalveolar adenomas in male mice
Model
Multistage
Chi-square
goodness-of-fit
/j-value
0.23973
MLE10
(mg/m3)
47.33
BMC10
(mg/m3)
24.13
Statistically significant fit.
3. For linear assessment per proposed cancer risk assessment guidelines (U.S. EPA, 1996a),
used point of departure, 95% LCL of the EC10 (LEC10) from the linear multistage model for extra
risk. Unit risk (risk at 1 |j.g/m3) was estimated by drawing a straight line from point of departure
to the origin and converting mg to j^g. Thus, unit risk = 0.1/(LEC10 x 1,000). Also used
GLOBAL86 linearized multistage model per existing cancer risk assessment guidelines (U. S.
EPA, 1987). Concentrations corresponding to 10"4, 10"5, and 10"6 risk levels were calculated by
dividing risk level by unit risk.
70
-------
Linear assessment results for cancer via inhalation route
Parameter
Point of departure, 95% LCL of ED10
Air unit risk (95% UCL on risk at 1
•g/m3)
Concentration at 10"4 risk
Concentration at 10"5 risk
Concentration at 10"6 risk
BMDS value
24.1 mg/m3
4.1E-6(«g/irf)-1
2.4E+1 • g/m3
2.4EO • g/m3
2.4E-1 • g/m3
71
-------
APPENDIX B. EXTERNAL PEER REVIEW-
SUMMARY OF COMMENTS AND DISPOSITION
The support document and IRIS summary for 1,3-dichloropropene have undergone both
internal peer review performed by scientists within EPA and a more formal external peer review
performed by scientists outside EPA in accordance with EPA guidance on peer review (U.S.
EPA, 1994). Comments made by the internal reviewers were addressed prior to submitting the
documents for external peer review and are not part of this appendix. The external peer
reviewers were tasked with providing written answers to general questions on the overall
assessment and on chemical-specific questions in areas of scientific controversy or uncertainty.
A summary of significant comments made by the external reviewers and EPA's response to these
comments follows.
(1) General Comments
A. Appropriateness of the critical studies and critical effects for the RfD and RfC
All three reviewers agreed with the critical study (Lomax et al., 1989) and the critical
effect, nasal hyperplasia, for RfC derivation. All reviewers agreed with the selection of the
feeding study of Stott et al. (1995) as the critical study for the RfD; however, one reviewer
questioned the use of forestomach hyperplasia as the critical effect on the basis that the relevance
to humans was marginal. This reviewer requested that more supporting information be added,
e.g., data for chemicals that cause serious lesions in the forestomach of rats and that also cause
adverse effects in the glandular stomach of animals that do not possess a forestomach.
Response to Comment: Even though humans do not have a forestomach, the hyperplasia
is important as a manifestation of chronic irritation and is consistent with other portal-of-entry
effects of dichloropropene (Nater and Gooskens, 1976; Haut et al., 1996; Breslin et al., 1989;
Lomax et al., 1989; Linnett et al., 1988; Stott et al., 1988). Thus, Sections 4.2.2 and 5.1.1 have
been altered to characterize forestomach hyperplasia as a manifestation of chronic irritation.
Evidence of relevance to humans, a case report of gastric mucosal erosion produced by a human
poisoning incident (Hernandez et al., 1994), has also been added to Sections 4.1.2 and 5.1.1.
B. Appropriateness of the uncertainty and modifying factors applied to the RfD and RfC
Two reviewers agreed with the uncertainty and modifying factors applied to both the RfC
and RfD. The dissenting reviewer indicated that an uncertainty factor of < 10 should be used to
protect sensitive subpopulations in the RfD derivation because the document had previously
concluded that children and women were not subpopulations at risk. This reviewer also stated
that the 10-fold uncertainty factor applied to the RfC to protect sensitive subpopulations was not
warranted because human variability to a contact irritant such as 1,3-dichloropropene should be
no more than two- or threefold.
Response to Comment: The intraspecies UF is intended to protect the sick and elderly in
addition to protecting children. In the absence of data to show the range of variability in human
72
-------
response, the UF to protect sensitive subpopulations in the derivation of the RfD remains at 10.
Regarding the analogous UF in the derivation of the RfC, there are no data to support less than
10-fold variability in the responses of humans to inhaled irritants. Thus, the UF to protect
sensitive subpopulations remains at 10.
C. Use of benchmark dose models
One reviewer had reservations about the routine use of benchmark dose models and
recommended also presenting the RfD and RfC as derived by the classical NOAEL method.
Response to Comment: Because the NOAEL method limits the quantitative assessment
to the experimental doses used in the critical study, modeling techniques that use the entire
dose-response curve are preferred for adequate data sets. For 1,3-dichloropropene, the method
used would have made little difference because the points of departure for both methods were
almost identical. The NOAEL for the oral study was 2.5 mg/kg/day whereas the BMDL10 was
3.4 mg/kg/day. The NOAEL for the inhalation study was 3.7 mg/m3 and the BMCL10 was 3.66
mg/m3. The NOAEL for oral exposure was identified in Section 5.1.1 and the NOAEL for
inhalation exposure has been added to Section 5.2.1.
D. Interspecies scaling factor for oral quantitative cancer assessment
One reviewer remarked that the basis for calculating the HED (human body
weight/animal body weight)1'4 was not apparent. The reviewer indicated that major interspecies
determinants would include the levels of reduced GSH and GST activity in the liver and the
ability to regenerate GSH.
Response to Comment: The scaling factor used to determine the HED is recommended
by the proposed cancer risk assessment guidelines (U.S. EPA, 1996a). The reference for the
scaling factor has been added to Section 5.3.1.1. The scaling factor, which represents an
adjustment for metabolic rate across animals of different sizes, is used in the absence of
chemical-specific data on interspecies sensitivity. There are several isoforms of GST that confer
target organ and species sensitivity. Because the literature search for 1,3-dichloropropene did not
find information on the specific GST isoform that catalyzes its metabolism, a more specific
interspecies extrapolation cannot be performed.
(2) Comments on Chemical-Specific Questions
A. Cancer classification
All reviewers agreed with the B2 carcinogen classification.
73
-------
B. Choice of data for the oral quantitative cancer assessment
One reviewer stated that because the NTP (1985) study had serious experimental design
problems, the feeding study (Stott et al., 1995) should be used. This reviewer indicated that the
gavage administration (NTP, 1985) was a serious problem because it was not relevant to human
exposures, which occur intermittently over the course of a day. The reviewer cited a study by
Sanzgiri et al. (1995) to show that gavage administration of carbon tetrachloride produces higher
blood levels and toxicity than a 2-hour gastric infusion of the same dose. Another reviewer
agreed with using both gavage and feeding studies for liver tumors in rats, but remarked that the
use of the mouse bladder cancer data from the NTP (1985) study was questionable because of the
decreased survival in control animals and the presence of epichlorohydrin, a known carcinogen,
in the dichloropropene formulation. The third reviewer remarked that the use of a single data set
alone would not increase the reliability of the oral cancer estimate as long as a linear
dose-response assessment was performed. If a single study were to be used, this reviewer
preferred the study by Stott et al. (1995).
Response to Comment: The variety of opinions on this topic reflect the Agency's
difficulty in choosing the appropriate data set between the feeding study in rats (Stott et al.,
1995) and the gavage study in mice and rats (NTP, 1985). The liver tumors in rats were
important because they were observed in both feeding and gavage studies, but neither study was
clearly better than the other (see Section 5.3.1), so both were used. Transitional cell carcinoma
of the urinary bladder was observed in the mouse gavage study (NTP, 1985), but it did not
appear in the feeding study (Redmond et al., 1995). The mouse gavage data was used because it
was a rare tumor type. The lack of effects, other than decreased body weight, in the mouse
feeding study (Redmond et al., 1995) and the lack of in-cage stability studies of the food mixture
cast doubt that the mice in the feeding study received the intended dose. All three data sets were
used because no single data set was clearly the best.
C. Appropriateness of linear vs. nonlinear quantitative cancer assessment
All reviewers agreed that there was evidence that the dose-response for cancer was
nonlinear; however, two reviewers agreed that the linear assessment was appropriate in the
absence of more definitive mode-of-action data for tumor formation. The other reviewer opined
that cancer formation was clearly a threshold phenomenon and that a nonlinear assessment is
entirely appropriate. The third reviewer noted that two apparent modes of action had been
identified and that both were threshold phenomena. The modes of action cited by this reviewer
were epoxide formation with subsequent reactive metabolites that bind to DNA, and chronic
irritation and cell killing, which results in degenerative changes and regenerative hyperplasia.
Response to Comment: The dose-response for mouse urinary bladder tumors seemed to
be rather linear, but only two doses were used in that study (NTP, 1985). The dose-responses for
rat liver tumors (NTP, 1985; Stott et al., 1995) and for mouse bronchioalveolar adenomas were
nonlinear. Regardless of the shape of the dose-response, both current (U.S. EPA, 1987) and
proposed (U.S. EPA, 1996a) cancer guidelines require defaulting to a linear dose-response
assessment in the absence of definitive mode-of-action data supporting a nonlinear mechanism of
74
-------
tumor formation. Because there was evidence of mutagenicity (Watson et al., 1987; Martelli et
al., 1993; Ghia et al., 1993; Kevekordes et al., 1996, Schneider et al., 1998a) in short-term in
vitro and in vivo tests, the cancer guidelines (U.S. EPA, 1996a; U.S. EPA, 1987) call for a linear
dose-response assessment.
75
------- |