DRAFT - DO NOT CITE OR QUOTE
EPA/635/R-09/010A
www.epa.gov/iris
f/EPA
TOXICOLOGICAL REVIEW
OF
CHLOROPRENE
(CAS No. 126-99-8)
In Support of Summary Information on the
Integrated Risk Information System (IRIS)
September 2009
NOTICE
This document is an External peer review draft. This information is distributed solely for the
purpose of pre-dissemination peer review under applicable information quality guidelines. It has not
been formally disseminated by EPA. It does not represent and should not be construed to represent
any Agency determination or policy. It is being circulated for review of its technical accuracy and
science policy implications.
U.S. Environmental Protection Agency
Washington, DC
September 2009 DRAFT-DO NOT CITE OR QUOTE
-------
DISCLAIMER
1 This document is a preliminary draft for review purposes only. This information is distributed
2 solely for the purpose of pre-dissemination peer review under applicable information quality
3 guidelines. It has not been formally disseminated by EPA. It does not represent and should not be
4 construed to represent any Agency determination or policy. Mention of trade names or commercial
5 products does not constitute endorsement or recommendation for use.
September 2009 « DRAFT - DO NOT CITE OR QUOTE
-------
TABLE OF CONTENTS
LIST OF TABLES v
LIST OF FIGURES viii
LIST OF ABBREVIATIONS AND ACRONYMS ix
FOREWORD xi
AUTHORS, CONTRIBUTORS, AND REVIEWERS xii
1. INTRODUCTION 1-1
2. CHEMICAL AND PHYSICAL INFORMATION 2-1
3. TOXICOKINETICS 3-1
3.1. ABSORPTION 3-1
3.2. DISTRIBUTION 3-1
3.3. METABOLISM 3-2
3.4. ELIMINATION 3-7
3.5. PHYSIOLOGICALLY BASED TOXICOKINETIC MODELS 3-7
4. HAZARD IDENTIFICATION 4-1
4.1. STUDIES IN HUMANS—EPIDEMIOLOGY, CASE REPORTS, CLINICAL CONTROLS... 4-1
4.1.1. Chioroprene Exposure and Cancer Effects 4-1
4.1.1.1. Overview 4-1
4.1.1.2. Individual Occupational Studies 4-1
4.1.1.3. Summary and Discussion of Relevant Methodological Issues 4-15
4.1.2. Chi oroprene Exposure and Noncancer Effects 4-19
4.1.2.1. Acute-, Short-, and Sub chronic-Duration Noncancer Effects 4-19
4.1.2.2. Chronic Noncancer Effects 4-20
4.2. SUBCHRONIC AND CHRONIC STUDIES AND CANCER BIO AS SAYS IN ANIMALS-
ORAL AND INHALATION 4-21
4.2.1. Oral Exposure 4-21
4.2.2. Inhalation Exposure 4-23
4.3. REPRODUCTIVE/DEVELOPMENTAL STUDIES—ORAL AND INHALATION 4-44
4.4. OTHER DURATION- OR ENDPOINT-SPECIFIC STUDIES 4-50
4.4.1. Acute and Subchronic Studies 4-50
4.4.2. Immunotoxicity 4-52
4.5. MECHANISTIC DATA AND OTHER STUDIES IN SUPPORT OF MODE OF ACTION...4-53
4.5.1. Mode-of-Action Studies 4-53
4.5.2. Genotoxicity Studies 4-56
4.5.2.1. Bacterial Mutagenicity Assays 4-57
4.5.2.2. Mammalian Cell Assays 4-58
4.5.2.3. In Vivo Bioassays 4-59
4.5.3. Structural Alerts 4-60
4.6. SYNTHESIS OF MAJOR NONCANCER EFFECTS 4-61
4.6.1. Animal Studies 4-62
September 2009 i" DRAFT - DO NOT CITE OR QUOTE
-------
4.6.1.1. Oral Exposure 4-62
4.6.1.2. Inhalation Exposure 4-62
4.7. EVALUATION OF CARCINOGENICITY 4-65
4.7.1. Summary of Overall Weight of Evidence 4-66
4.7.2. Synthesis of Human, Animal, and Other Supporting Evidence 4-68
4.7.2.1. Human 4-68
4.7.2.1.1. Evidence for Causality 4-70
4.7.2.2. Laboratory Animal 4-72
4.7.3. Mode-of-Action Information 4-73
4.7.3.1. Hypothesized Mode of Action 4-73
4.7.3.2. Experimental Support for the Hypothesized Mode of Action 4-73
4.7.3.3 Conclusions about the Hypothesized Mode of Action 4-76
4.8. SUSCEPTIBLE POPULATIONS AND LIFE STAGES 4-78
4.8.1. Possible Childhood Susceptibility 4-78
4.8.2. Possible Gender Differences 4-79
5. DOSE-RESPONSE ASSESSMENTS 5-1
5. LORAL REFERENCE DOSE (RFD) 5-1
5.2. INHALATION REFERENCE CONCENTRATION (RFC) 5-1
5.2.1. Choice of Principal Study and Critical Effect(s) 5-1
5.2.2. Methods of Analysis 5-3
5.2.3. Exposure Duration and Dosimetric Adjustments 5-4
5.2.4. RfC Derivation—Including Application of Uncertainty Factors 5-7
5.2.5. Previous RfC Assessment 5-8
5.2.6. RfC Comparison Information 5-8
5.3. UNCERTAINTIES IN THE INHALATION REFERENCE CONCENTRATION 5-9
5.4. CANCER ASSESSMENT 5-12
5.4.1. Choice of Study/Data—with Rationale and Justification 5-12
5.4.2. Dose-Response Data 5-12
5.4.3. Dose Adjustments and Extrapolation Methods 5-14
5.4.4. Oral Slope Factor and Inhalation Unit Risk 5-17
5.4.5 Application of Age-Dependent Adjustment Factors 5-21
5.4.6. Previous Cancer Assessment 5-22
5.4.7. Uncertainties in Cancer Risk Values 5-22
6. MAJOR CONCLUSIONS IN CHARACTERIZATION OF HAZARD AND DOSE RESPONSE. 6-1
6.1. HUMAN HAZARD POTENTIAL 6-1
6.2. DOSE RESPONSE 6-4
7. REFERENCES 7-1
APPENDIX A. Summary of External Peer Review and Public Comments and Disposition A-1
APPENDIX B. Benchmark Dose Modeling Results for the Derivation of the RfC B-l
APPENDIX C. Cancer Dose-Response Modeling Results C-l
September 2009 iv DRAFT - DO NOT CITE OR QUOTE
-------
LIST OF TABLES
Table 2-1. Physical properties and chemical identity of chloroprene 2-3
Table 3-1. Tissue-to-air partition coefficients for chloroprene 3-2
Table 3-2. Liver microsomal metabolites as a percentage of 1-butanol internal standard 3-3
Table 3-3. Steriochemical comparison of relative amounts (percentages) of R- and S-
enantiomers of the major chloroprene metabolite (l-chloroethenyl)oxirane from liver
microsomes compared across species, strains, gender, and chloroprene concentration (mM) 3-5
Table 3-4. Kinetic parameters used to describe the microsomal oxidation of chloroprene 3-6
Table 3-5. Metabolic parameters of chloroprene 3-7
Table 3-6. Physiological parameters used for chloroprene PBPK modeling 3-8
Table 4-1. Standardized mortality ratios (SMRs) for the DuPont Louisville Works cohort
relative to general U.S. population rates 4-3
Table 4-2. Standardized mortality ratios (SMRs) for all cancers, liver and lung cancer among
males exposed to chloroprene relative to general Chinese population rates 4-5
Table 4-3. Standardized mortality ratios (SMRs) for selected cancer risks relative to general
population rates of Moscow, Russia 4-7
Table 4-4. Selected relative risk (RRs) estimates for the high exposure group relative to
unexposed factory workers 4-7
Table 4-5. Internal relative risks (RRs) by duration of employment in the high-exposure
category 4-8
Table 4-6. Selected standardized incidence ratios (SIRs) for chloroprene monomer cohort
relative to the general Armenian population 4-9
Table 4-7. Standardized incidence ratios (SIRs) for elevated cancer risks for plant workers
relative to general population rates of Isere, France 4-11
Table 4-8. Standardized mortality ratios (SMRs) at each of four chloroprene production
facilities 4-13
Table 4-9. Relative risks (RRs) for respiratory cancers by cumulative chloroprene exposure 4-14
Table 4-10. Epidemiologic summary results of respiratory system cancers: Overall
standardized mortality ratios (SMRs) and SMRs for intermediate and high chloroprene
exposures relative to external population comparison 4-16
Table 4-11. Epidemiologic summary results of liver/biliary passage cancers: Total cohort
standardized mortality ratios (SMRs) and relative risk (RRs) for intermediate and high
chloroprene exposures 4-17
Table 4-12. Frequency of chromosomal aberrations in lymphocyte culture cells from
chloroprene production workers 4-21
Table 4-13. Tumor incidence in female BD IV rats treated orally with chloroprene (100
mg/kg) on GDI7 and in their progeny treated (50 mg/kg) weekly for life (120 weeks) 4-22
September 2009 v DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-14. Distribution of tumors in female BD IV rats treated orally with chloroprene (100
mg/kg) on GDI? and their progeny treated (50mg/kg) weekly for life (120 weeks) 4-23
Table 4-15. Survival and body weights of rats in the 16-day inhalation study of chloroprene 4-24
Table 4-16. Incidences of selected nonneoplastic lesions in rats in the 16-day inhalation study
of chloroprene 4-25
Table 4-17. Survival and body weights of mice in the 16-day inhalation study of chloroprene 4-26
Table 4-18. Survival and body weights of rats in the 13-week inhalation study of chloroprene 4-26
Table 4-19. Incidences of selected nonneoplastic lesions in rats in the 13-week inhalation
study of chloroprene 4-28
Table 4-20. Survival and body weights of mice in the 13-week inhalation study of chloroprene 4-28
Table 4-21. Incidences of forestomach lesions in mice in the 13-week inhalation study of
chloroprene 4-29
Table 4-22. 2-Year survival probability estimates for F344/N rats chronically exposed (2
years) to chloroprene by inhalation 4-30
Table 4-23. Incidence and severity of non-neoplastic lesions in F344/N rats chronically
exposed (2 years) to chloroprene by inhalation 4-31
Table 4-24. Incidence of neoplasms in F344/N rats chronically exposed (2 years) to
chloroprene by inhalation 4-32
Table 4-25. 2-Year survival probabilities for B6C3F1 mice chronically exposed (2 years) to
chloroprene by inhalation 4-34
Table 4-26. Incidence and severity of non-neoplastic lesions in B6C3F1 mice chronically
exposed (2 years) to chloroprene by inhalation 4-35
Table 4-27. Incidence of neoplasms in B6C3F1 mice chronically exposed (2 years) to
chloroprene by inhalation 4-36
Table 4-28. Survival-adjusted neoplasm rates for mice in the 2-year inhalation study of
chloroprene 4-38
Table 4-29. Selected mean relative organ weights of rats exposed for 24 months and hamsters
exposed for 18 months to chloroprene vapor 4-40
Table 4-30. Incidence, site and type of tumor in selected organs and tissues of rats exposed to
chloroprene for 24 months 4-42
Table 4-31. Incidence, site and type of tumor in selected organs and tissues of hamsters
exposed to chloroprene for 18 months 4-43
Table 4-32. Summary of epididymal spermatozoal and estrous cycle parameters for rats and
mice in the 13-week study of chloroprene 4-45
Table 4-33. Results of teratology and embryotoxicity studies in rats exposed to chloroprene by
inhalation 4-47
Table 4-34. Incidence of anomalies in litters of rats exposed to chloroprene by inhalation 4-48
Table 4-35. Chloroprene-induced mortality in male rats 4-51
Table 4-36. Genotoxicity assays of chloroprene 4-56
September 2009 vi DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-37. Sites of increased incidences of neoplasms in the 2 year inhalation studies of 1,3-
butadiene, isoprene, and chloroprene in rats and mice 4-61
Table 4-38. Summary of animal and human tumor data and weight of evidence descriptor for
chloroprene 4-67
Table 5-1. Incidences of nonneoplastic lesions statistically significantly increased at lowest
exposure concentration resulting from chronic exposure (ppm) to chloroprene 5-3
Table 5-2. Human equivalent concentration estimates for best fitting models of the BMD from
chronic exposure to chloroprene 5-6
Table 5-3. Summary of Uncertainties in the Chloroprene noncancer risk assessment 5-11
Table 5-4. Tumor incidence in female and male B6C3F1 mice exposed to chloroprene via
inhalation 5-13
Table 5-5. Tumor incidence in female and male F344 rats exposed to chloroprene via
inhalation 5-14
Table 5-6. Dose-response modeling summary for male mouse tumor sites associated with
inhalation exposure to chloroprene 5-18
Table 5-7. Dose-response modeling summary for female mouse tumors associated with
inhalation exposure to chloroprene 5-19
Table 5-8. Summary of uncertainties in chloroprene cancer unit risk estimate 5-23
Table B-l. Modeling Results for Alvelolar Epithelial Hyperplasia in Male F344/N Rats B-l
Table B-2. Modeling Results for Alveolar Epithelial Hyperplasia in Female F344/N Rats B-4
Table B-3. Modeling Results for Bronchiolar Hyperplasia in Male B6C3F1 Mice B-7
Table B-4. Modeling Results for Bronchioloar Hyperplasia in Female B6C3F1 Mice B-10
Table B-5. Modeling Results for Histiocytic Cell Infiltration in Male B6C3F1 Mice B-ll
Table B-6. Modeling Results for Olfactory Chronic Inflammation in Male F344/NRats B-ll
Table B-7. Modeling Results for Olfactory Atrophy in Male F344/N Rats B-14
Table B-8. Modeling Results for Olfactory Necrosis in Male F344/N Rats B-17
Table B-9. Modeling Results for Degenerative Nasal Lesions (olfactory atrophy or necrosis) in
Male F344/N Rats B-20
Table B-10. Modeling Results for Splenic Hematopoietic Cell Proliferation in Female B6C3F1
Mice B-23
Table C-l. Summary of human equivalent overall cancer risk values estimated by R/BMDR,
based on male and female mouse tumor incidence (NTP, 1998) C-16
September 2009 vii DRAFT - DO NOT CITE OR QUOTE
-------
LIST OF FIGURES
Figure 2-1. The chemical structure of chloroprene 2-1
Figure 3-1. Proposed metabolism of chloroprene 3-4
Figure 5-1. Points of departure (in mg/m3) for selected endpoints with corresponding applied
uncertainty factors and derived sample RfCs (chosen RfC value is circled) 5-9
Figure B-l. Log-logistic model fit for alveolar epitheliala hyperplasia in male F344/N rats B-2
Figure B-2. Log-logistic model fit for alveolar epithelial hyperplasia in female F344/N rats B-5
Figure B-3. Log-logistic model fit for bronchiolar hyperplasia in male B6C3F1 Mice B-8
Figure B-4. Log-logistic model fit for olfactory chronic inflammation in male F344/N rats B-12
Figure B-5. Logistic model fit for olfactory atrophy in male F344/Nrats B-14
Figure B-6. Logistic model fit for olfactory necrosis in male F344/Nrats B-18
Figure B-7. Logistic model fit for degenerative nasal lesions (olfactory atrophy or necrosis) in
male F344/N rats B-21
Figure B-8. Probit model fit for splenic hematopoietic cell proliferation in female B6C3F1
mice B-24
Figure C-l. Female mice, hemangiomas and hemangiosarcomas in all organs; high dose
dropped,) hemangiosarcomas occurring before termination considered fatal C-l
Figure C-2. Female mice, hemangiomas and hemangiosarcomas in all organs; high dose
dropped, hemangiosarcomas occurring before termination considered fatal. Details below C-2
Figure C-3. Female mice, alveolar/bronchiolar tumors. Details below C-3
Figure C-4. Female mice, hepatocellular adenomas and carcinomas. Details below C-4
Figure C-5. Female mice, forestomach tumors. Details below C-5
Figure C-6. Female mice, Harderian gland tumors. Details below C-6
Figure C-7. Female mice, mammary gland tumors. Details below C-7
Figure C-8. Female mice, skin sarcomas. Details below C-8
Figure C-9. Female mice, Zymbal's gland tumors. Details below C-9
Figure C-10. Male mice, hemangiomas and hemangiosarcomas; hemangiosarcomas occurring
before termination considered fatal. Details below C-10
Figure C-ll. Male mice, alveolar/bronchiolar tumors. Details below C-12
Figure C-12. Male mice, forestomach tumors. Details below C-13
Figure C-13. Male mice, Harderian gland tumors. Details below C-14
Figure C-14. Male mice, renal tubule tumors. Details below C-15
September 2009 viii DRAFT - DO NOT CITE OR QUOTE
-------
LIST OF ABBREVIATIONS AND ACRONYMS
ADAF age dependent adjustment factor
AEH alveolar epithelial hyperplasia
AIC Akaike Information Criterion
ALP alkaline phosphatase
ALT alanine aminotransferase
BMC benchmark concentration
BMCL lower bound on the benchmark concentration
BMD benchmark dose
BMDL lower confidence limit on the benchmark dose
BMDS benchmark dose software
BMR benchmark response
CASRN Chemical Abstracts Service Registry Number
CI confidence interval
CNS central nervous system
CYP cytochrome
DAF dosimetric adjustment factor
DMSO dimethyl sulfoxide
DNA deoxyribonucleic acid
EH epoxide hydralases
EPA U.S. Environmental Protection Agency
GD gestational day
GDH glutamine dehydrogenase
GSH glutathione
HEC human equivalent concentration
IARC International Agency for Research on Cancer
ICD International Classification of Diseases
IPCS International Programme on Chemical Safety
IRIS Integrated Risk Information System
LOAEL lowest-observed-adverse-effect level
LOH loss of heterozygosity
M Molar
MLE maximum likelihood estimate
MOA mode of action
MV minute volume
NCEA National Center for Environmental Assessment
NIOSH National Institute for Occupational Safety and Health
NLM National Library of Medicine
NOAEL no-observed-adverse-effect level
NPSH nonprotein sulfhydryl
NRC National Research Council
NTP National Toxicology Program
OECD Organisation for Economic Cooperation and Development
OR odds ratio
p probability value
PBPK physiologically based pharmacokinetic (model)
PCB polychlorinated biphenyl
September 2009
IX
DRAFT - DO NOT CITE OR QUOTE
-------
POD point of departure
ppm parts per million
PU pulmonary
R level of risk
RBC red blood cell
RfC reference concentration
RfD reference dose
RGDR regional gas dose ratio
RR relative risk
SA surface area
SD standard deviation
SDH sorbitol dehydrogenase
SIR standard incidence ratio
SMR standardized mortality ratio
UCL upper confidence limit
UF uncertainty factor
v/v volume/volume
X2 chi squared
September 2009 x DRAFT - DO NOT CITE OR QUOTE
-------
FOREWORD
1 The purpose of this Toxicological Review is to provide scientific support and rationale for the
2 hazard and dose-response assessment in IRIS pertaining to chronic exposure to chloroprene. It is not
3 intended to be a comprehensive treatise on the chemical or toxicological nature of chloroprene.
4 The intent of Section 6, Major Conclusions in the Characterization of Hazard and Dose
5 Response, is to present the major conclusions reached in the derivation of the reference dose, reference
6 concentration and cancer assessment, where applicable, and to characterize the overall confidence in
7 the quantitative and qualitative aspects of hazard and dose response by addressing the quality of data
8 and related uncertainties. The discussion is intended to convey the limitations of the assessment and to
9 aid and guide the risk assessor in the ensuing steps of the risk assessment process.
10 For other general information about this assessment or other questions relating to IRIS, the
11 reader is referred to EPA's IRIS Hotline at (202) 566-1676 (phone), (202) 566-1749 (fax), or
12 hotline.iris@epa.gov (email address).
September 2009 xi DRAFT - DO NOT CITE OR QUOTE
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
CHEMICAL MANAGER
J.Allen Davis M.S.P.H.
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
AUTHORS
J.Allen Davis, M.S.P.H.
Jeffrey Gift, Ph.D.
Karen Hogan, M.S.
Ines Pagan, D.V.M., Ph.D.
John Stanek, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, D.C.
J. Michael Wright, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH
Aparna M. Koppikar, M.D., Ph.D.
E MAGE Inc.
CONTRACTOR SUPPORT
Karla Thrall, Ph.D.
Anthony Fristachi
Nick Heyer, Ph.D.
Paul Hinderliter, Ph.D.
Jessica Sanford, Ph.D.
Battelle Memorial Institute
September 2009 xii DRAFT - DO NOT CITE OR QUOTE
-------
REVIEWERS
This document has been reviewed by EPA scientists and interagency reviewers from other
federal agencies and White House offices.
INTERNAL EPA REVIEWERS
Ila Cote, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
Lynn Flowers, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
Connie Meacham, M.S.
National Center for Environmental Assessment
Office of Research and Development
John Vandenberg, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
Debra Walsh, M.S.
National Center for Environmental Assessment
Office of Research and Development
Kate Guyton, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
Jennifer Jinot, Ph.D.
National Center for Environmental Assessemnt
Office of Research and Development
September 2009 xiii DRAFT - DO NOT CITE OR QUOTE
-------
1. INTRODUCTION
1 This document presents background information and justification for the Integrated Risk
2 Information System (IRIS) Summary of the hazard and dose-response assessment of chloroprene.
3 IRIS Summaries may include oral reference dose (RfD) and inhalation reference concentration (RfC)
4 values for chronic and other exposure durations, and a carcinogenicity assessment.
5 The RfD and RfC, if derived, provide quantitative information for use in risk assessments for
6 health effects known or assumed to be produced through a nonlinear (presumed threshold) mode of
7 action. The RfD (expressed in units of mg/kg-day) is defined as an estimate (with uncertainty
8 spanning perhaps an order of magnitude) of a daily exposure to the human population (including
9 sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during a
10 lifetime. The inhalation RfC (expressed in units of mg/m3) is analogous to the oral RfD, but provides a
11 continuous inhalation exposure estimate. The inhalation RfC considers toxic effects for both the
12 respiratory system (portal of entry) and for effects peripheral to the respiratory system (extrarespiratory
13 or systemic effects). Reference values are generally derived for chronic exposures (up to a lifetime),
14 but may also be derived for acute (< 24 hours), short-term (> 24 hours up to 30 days), and subchronic
15 (> 30 days up to 10% of lifetime) exposure durations, all of which are derived based on an assumption
16 of continuous exposure throughout the duration specified. Unless specified otherwise, the RfD and
17 RfC are derived for chronic exposure duration.
18 The carcinogenicity assessment provides information on the carcinogenic hazard potential of
19 the substance in question and quantitative estimates of risk from oral and inhalation exposure may be
20 derived. The information includes a weight-of-evidence judgment of the likelihood that the agent is a
21 human carcinogen and the conditions under which the carcinogenic effects may be expressed.
22 Quantitative risk estimates may be derived from the application of a low-dose extrapolation procedure.
23 If derived, the oral slope factor is a plausible upper bound on the estimate of risk per mg/kg-day of
24 oral exposure. Similarly, an inhalation unit risk is a plausible upper bound on the estimate of risk per
25 |ig/m3 air breathed.
26 Development of these hazard identification and dose-response assessments for chloroprene has
27 followed the general guidelines for risk assessment as set forth by the National Research Council
28 (NRC) (1983). EPA Guidelines and Risk Assessment Forum Technical Panel Reports that may have
29 been used in the development of this assessment include the following: Guidelines for the Health Risk
30 Assessment of Chemical Mixtures (U.S. EPA, 1986a), Guidelines for Mutagenicity Risk Assessment
31 (U.S. EPA, 1986b), Recommendations for and Documentation of Biological Values for Use in Risk
32 Assessment (U.S. EPA, 1988), Guidelines for Developmental Toxicity Risk Assessment (U.S. EPA,
33 1991), Interim Policy for Particle Size and Limit Concentration Issues in Inhalation Toxicity (U. S.
34 EPA, 1994a), Methods for Derivation of Inhalation Reference Concentrations and Application of
35 Inhalation Dosimetry (U.S. EPA, 1994b), Use of the Benchmark Dose Approach in Health Risk
36 Assessment (U.S. EPA, 1995), Guidelines for Reproductive Toxicity Risk Assessment (U.S. EPA, 1996),
September 2009 1-1 DO NOT CITE OR QUOTE
-------
1 Guidelines for Neurotoxicity Risk Assessment (U.S. EPA, 1998), Science Policy Council Handbook:
1 Risk Characterization (U.S. EPA, 2000a), Benchmark Dose Technical Guidance Document (U.S. EPA,
3 2000b), Supplementary Guidance for Conducting Health Risk Assessment of Chemical Mixtures (U.S.
4 EPA, 2000c), A Review of the Reference Dose and Reference Concentration Processes (U.S. EPA,
5 2002a), Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005a), Supplemental Guidance for
6 Assessing Susceptibility from Early-Life Exposure to Carcinogens (U.S. EPA, 2005b), Science Policy
7 Council Handbook: Peer Review (U. S. EPA, 2006a), and A Framework for Assessing Health Risks of
8 Environmental Exposures to Children (U.S. EPA, 2006b).
9 The literature search strategy employed for this compound was based on the Chemical
10 Abstracts Service Registry Number (CASRN) and at least one common name. Any pertinent scientific
11 information submitted by the public to the IRIS Submission Desk was also considered in the
12 development of this document. The relevant literature was reviewed through August 2009.
September 2009 1-2 DO NOT CITE OR QUOTE
-------
2. CHEMICAL AND PHYSICAL INFORMATION
1 Beta-chloroprene (C4H5C1) (hereafter referred to as chloroprene) is a volatile, flammable liquid
2 used primarily in the manufacture of polychloroprene or neoprene rubber. The latter is used to make
3 diverse products, such as tires, wire coatings, and tubing. While 90% of chloroprene is used to make
4 neoprene solid, polychloroprene, about 10% is converted to polychloroprene latex, a colloidal
5 suspension of polychloroprene in water (International Agency for Research on Cancer [IARC], 1999).
6 In 1995, there was one commercial producer of chloroprene in the United States; other plants
7 produced chloroprene for on-site use and processing, as a by-product of vinyl chloride production, or
8 as a manufacturing impurity (National Toxicology Program [NTP], 2005). Used almost exclusively to
9 produce polychloroprene, chloroprene is sold to only three U.S. companies for polychloroprene
10 manufacture; less than 20 Ib/yr is sold for research applications. The total estimated production of
11 polychloroprene from 1986 to 1988 was approximately 250 to 300 million Ib (113,000 to 136,000
12 metric tons), and the volume from 1995 to 1996 was approximately 200 to 250 million Ib (90,700 to
13 113,000 metric tons).
14 There are no known natural occurrences of chloroprene in the environment. The main sources
15 of releases to the environment are through effluent and emissions from facilities that use chloroprene
16 to produce polychloroprene elastomers or transport of the product. In 1995, there were 14 facilities
17 reporting releases of chloroprene to the atmosphere, eight of which reported individual atmospheric
18 releases from 2 to 481,871 Ibs (0.0009 to 218.6 Mg). Three plants in Kentucky, Texas, and Louisiana,
19 each reporting atmospheric releases of > 100,000 Ibs, accounted for most of the reported chloroprene
20 releases. One of these sites produced chloroprene, while the other two converted chloroprene to
21 polychloroprene (NTP, 2005). The chemical structure of chloroprene is shown in Figure 2-1.
H2C
c—
c/
H
CH2
Figure 2-1. The chemical structure of chloroprene.
22 The starting material for the synthesis of chloroprene is 1,3-butadiene. Chloroprene is also a
23 structural analogue of isoprene (2-methyl 1,3-butadiene) and resembles vinyl chloride as far as having
24 a chlorine bound to a double-bonded carbon (alkene) backbone. However, chloroprene contains four
25 carbons arranged with two double bonds. The odor of chloroprene is described as pungent and ether-
26 like (National Library of Medicine [NLM], 2008). Chloroprene is volatile and highly reactive; it is not
September 2009 2-1 DRAFT - DO NOT CITE OR QUOTE
-------
1 expected to bioaccumulate or persist in the environment (Organisation for Economic Cooperation and
2 Development [OECD], 1998). Because of its high vapor pressure (215 mm Hg at 25°C), chloroprene
3 is expected to readily volatilize from water and solid surfaces (NTP, 2005). Chloroprene vapor has an
4 estimated ionization potential of 8.95 ± 0.05 eV and an estimated half-life in the atmosphere of less
5 than 20 hours (Grosjean, 1990). Reactions with *OH (to produce formaldehyde), O3, and NO3 are the
6 expected pathways of removal, although no experimental data exist (Grosjean, 1991).
7 Of particular relevance to any toxicological studies involving chloroprene is its propensity to
8 spontaneously oxidize and form dimers and other oxygenated species unless stabilizers or inhibitors
9 are added. Uninhibited chloroprene must be stored under nitrogen at temperatures below 0°C (e.g., -
10 20°C) to prevent spontaneous polymerization. When bulk chloroprene with 5% n-octane added as an
11 internal standard was stored at 55°C for up to 6 hours, dimer content increased by 62% and
12 chloroprene monomer decreased by 22% (NTP, 1996). Because these reaction products, if formed,
13 may themselves account for any observed toxicity, toxicological studies that do not report storage or
14 generation conditions may yield results that are questionable for their relevance to chloroprene
15 monomer. A discussion of the polymerization process has been reported by Lynch (200 la), Kroshwitz
16 and Howe-Grant (1993), Stewart (1971), and Nystrom (1948). Additional information on production
17 and use has been reported by the International Agency for Research on Cancer (IARC, 1999).
18 Structures have been proposed for some of the chloroprene dimers (Stewart, 1971); some dimers result
19 upon reaction at room temperature while others result after prolonged heating.
20 In addition to volatilization, the potential fate of chloroprene that is released to soil is to leach
21 into groundwater. Breakdown via hydrolysis is not likely, as it is only partially soluble in water
22 (CambridgeSoft Corp., 2007). Chloroprene that is released to the water may only moderately adsorb
23 to suspended sediments or particles, and there will be little bioaccumulation in aquatic organisms (Kow
24 = 2.2). The occupational exposure potential to chloroprene is limited to facilities in the U.S., Europe,
25 and Asia where chloroprene is produced and converted to poly chloroprene (Lynch, 200 Ib). The
26 physical and chemical properties of chloroprene are shown in Table 2-1.
September 2009 2-2 DRAFT - DO NOT CITE OR QUOTE
-------
Table 2-1. Physical properties and chemical identity of chloroprene
CHLOROPRENE
CASRN
Synonyms
Registered trade name
Melting point
Boiling point
Density
Vapor pressure
Vapor density
Flashpoint (open cup)
Flammability limits
Water solubility
Other solubilities
Log KQW
Henry's law constant
Odor threshold
Molecular weight
Conversion factors (in air)
Empirical formula
126-99-8
2-chlorobuta-l,3-diene; 2-chlorobutadiene; 2-chloroprene;
alpha-chloroprene; beta-chlorobutadiene; beta-chloroprene;
chlorobutadiene; chloroprene
Neoprene
-130°C
59.4°C
0.956 at 20°C (relative to the density of H2O at 4°C)
215mmHgat25°C
3.0(air= 1)
-15.6°C
4-20% in air
2.115g/Lat25°C
Miscible with ethyl ether, acetone, benzene; soluble in alcohol
2.2
5.6 x 10~2 atm/m3-mol at 25°C
15 ppm(54 mg/m3)
88.54
1 mg/m3 = 0.276 ppm; 1 ppm = 3.62 mg/m3 at 25°C, 760 torr
Cfi5Cl
REFERENCE
NLM (2008)
CambridgeSoft Corp. (2007)
CambridgeSoft Corp. (2007)
NLM (2008)
NLM (2008)
NLM (2008)
NLM (2008)
NLM (2008)
CambridgeSoft Corp. (2007)
NLM (2008)
CambridgeSoft Corp. (2007)
NLM (2008)
OECD (1998)
NLM (2008)
U.S. EPA(2000d)
NLM (2008)
NLM (2008)
CambridgeSoft Corp. (2007)
September 2009
2-3
DRAFT - DO NOT CITE OR QUOTE
-------
3. TOXICOKINETICS
1 No reports are available that address the toxicokinetics of chloroprene in humans by any route
2 of exposure. Limited information is available for animals regarding the absorption and in vivo
3 metabolism of chloroprene. No information regarding tissue distribution of chloroprene from animal
4 studies is available. In vitro studies have been conducted to evaluate the metabolism of chloroprene in
5 lung and liver tissue fractions from rat, mouse, hamster, and humans (Munter et al., 2007a, b, 2003;
6 Himmelstein et al., 2004a, 2001a, 2001b; Cottrell et al., 2001; Summer and Greim, 1980). Hurst and
7 Ali (2007) evaluated the kinetics of R- and S-enantiomers of the chloroprene metabolite
8 (l-chloroethenyl)oxirane in mouse erythrocytes. A physiologically based pharmacokinetic (PBPK)
9 model has been developed to describe changes in chamber chloroprene concentrations during
10 exposures with mice, rats, and hamsters (Himmelstein et al. 2004a, 2004b). No in vivo time-course
11 data for blood or tissue concentration are available for model validation.
3.1. ABSORPTION
12 Quantitative data on the absorption of chloroprene from any route of exposure have not been
13 reported. The Hazardous Substances Data Bank states that chloroprene is "rapidly absorbed by the
14 skin" (Lefaux [1968] as cited in NLM [2008]). Chronic inhalation studies in B6C3F1 mice and
15 F344/N rats suggest that chloroprene has multiple nonneoplastic and neoplastic targets (nose and lung,
16 kidney, forestomach, Harderian gland, skin); therefore, the absorption and systemic distribution via the
17 inhalation route can be inferred (NTP, 1998).
3.2. DISTRIBUTION
18 No quantitative in vivo data on the tissue distribution of chloroprene have been reported. As
19 indicated above, the widespread distribution of chloroprene in vivo following absorption can be
20 inferred from effects in several target organs (NTP, 1998). Himmelstein et al. (2004b) determined
21 partition coefficients for chloroprene in mouse, F344 rat, Wistar rat, and hamster tissues by using the
22 vial equilibration method described by Gargas et al. (1989), as given in Table 3-1. These tissue-to-air
23 ratios suggest that chloroprene will be preferentially distributed in adipose tissue, followed by lung,
24 kidney, liver, and muscle. The partition coefficient values suggest there are no significant species
25 differences expected in tissue distribution of chloroprene.
September 2009 3-1 DRAFT - DO NOT CITE OR QUOTE
-------
Table 3-1. Tissue-to-air partition coefficients for chloroprene
TISSUE
Blood
Lung
Liver
Fat
Muscle
Kidney
TISSUE-TO-AIR PARTITION COEFFICIENTS3
Mouse
7.8 ±0.1
18.6 ±5.1
9.8 ±0.9
135.3 ±1.6
4.6 ±0.8
13.7 ±0.6
F344 rat
7.3 ±0.1
13. 5 ±1.6
11. 5 ±0.3
124.0 ±1.5
4.4 ±0.4
16.7 ±0.6
Wistar rat
8.0 ±0.5
11.2 ±0.5
10.9 ±0.2
126.3 ±1.4
4.0 ±0.3
9.4 ±0.4
Hamster
9.3 ±0.3
9.7 ±0.6
10.5 ±0.5
130.1 ±0.9
5.0 ±0.2
8.2 ±0.3
Humanb
4.5 ±0.1
13.3 ±4.1
10.7 ±1.1
128.9 ±2.7
4.5 ± 1.0
12.0 ±0.9
"Mean ± standard error for three replicates per rodent tissue.
b Human blood values determined for nine replicates (three subjects, three replicates/subject); human tissue partition
coefficient values were derived from rodents with standard error adjusted to account for the proportion of variation from
each set of rodent data.
Source: Himmelstein et al. (2004b).
3.3. METABOLISM
1 The metabolism of chloroprene has been primarily evaluated in vitro with lung and liver tissue
2 fractions from rat, mouse, hamster, and humans (Munter et al., 2007a, 2007b, 2003; Himmelstein et al.,
3 2004a, 2001a, 2001b; Cottrell et al., 2001; Summer and Greim, 1980). In a 1978 review of the older
4 literature, a number of reports suggested that chloroprene forms peroxides that interact with tissue thiol
5 groups and that the disposition of chloroprene is likely similar to vinyl chloride and vinylidene
6 chloride (Haley, 1978). This report was the first to postulate a metabolic profile of chloroprene,
7 including formation of epoxides by cytochrome P450 (CYP450) enzymes that could give rise to
8 aldehydes and eventually form mercapturic acid derivatives.
9 In studies using mouse and human liver microsomes, Bartsch et al. (1979) showed that
10 2-chloro-2-ethynyloxirane and/or (l-chloroethenyl)oxirane could be intermediates in the
11 biotransformation of chloroprene. This was based on the finding that 4-(4-nitrobenzyl)pyridine
12 trapped a volatile metabolite produced during reaction of mouse liver microsomes with chloroprene.
13 Summer and Greim (1980) reported that in vitro incubation of chloroprene with hepatocytes isolated
14 from male Wistar rats produced a concentration-dependent decrease in cellular glutathione (GSH),
15 suggesting a GSH-dependent detoxification pathway. A report by Himmelstein et al. (2001b) was the
16 first to quantitatively identify (l-chloroethenyl)oxirane as an epoxide metabolite of chloroprene and
17 confirmed the identify of the volatile metabolite reported by Bartsch et al. (1979). Himmelstein et al.
18 (2001b) reported that the oxidation of chloroprene to (l-chloroethenyl)oxirane was evident in rodent
19 and human liver microsomes and most likely involved CYP 2E1, as evidenced by nearly complete in
20 vitro inhibition with 4-methylpyrazole hydrochloride. A comparison across species suggested that a
21 greater amount of (l-chloroethenyl)oxirane was present in B6C3F1 mice and F344 rat liver
22 microsomes, followed by the Wistar rat, then humans and hamsters (Table 3-2).
September 2009
3-2
DRAFT - DO NOT CITE OR QUOTE
-------
Table 3-2. Liver microsomal metabolites as a percentage of 1-butanol internal
standard
METABOLITE PEAK3
1
2
o
J
4
5
6
LIVER MICROSOMAL SUSPENSION
B6C3F1 mouse
9.0
0.0
0.8
0.2
0.2
0.6
F344 rat
12.0
0.1
0.3
0.0
0.3
0.4
Wistar rat
4.0
0.1
0.2
0.1
0.0
0.3
Hamster
0.8
0.2
0.8
0.4
0.1
0.3
Human
1.3
0.1
0.3
0.1
0.0
0.1
a Metabolite peak 1 = (l-chloroethenyl)oxirane. Metabolite peaks 2-5 had insufficient signal to obtain meaningful spectral
data. A tentative spectral match for peak 6 was made as 3-chloro-2-butenal.
Source: Himmelstein et al. (2001b).
1 Further metabolism of (l-chloroethenyl)oxirane was observed in time-course evaluations with
2 liver microsomes (Himmelstein et al., 2001b). In mouse liver microsomes, the
3 (l-chloroethenyl)oxirane concentration showed an initial increase over 10 minutes that was followed
4 by a decline, attributable to either epoxide hydrolase-mediated hydrolysis or further oxidative
5 metabolism. Preliminary results indicated that the ranking of (l-chloroethenyl)oxirane hydrolysis in
6 liver microsomes was as follows: hamsters ~ humans > Wistar rats > B6C3F1 mice and F344 rats.
7 Studies by Cottrell et al. (2001) are in agreement with reports from Himmelstein et al. (2001a,
8 200Ib) and further define the structures and stereochemistry of chloroprene metabolites from rodent
9 species and humans by comparison with synthetic reference standards. Based on these studies, the
10 metabolic pathway illustrated in Figure 3-1 was proposed.
September 2009
3-3
DRAFT - DO NOT CITE OR QUOTE
-------
P450
•*• GSH
conjugate
Cl
0
18
19
15
20
Figure 3-1. Proposed metabolism of chloroprene.
Key: 1 = chloroprene; 5a/5b = R- and S-enantiomers of (l-chloroethenyl)oxirane; 21a/b = R- and
S-enantiomers of 3-chlorobut-3-ene-l,2-diol; 4a/4b = R- and S-enantiomers of 2-chloro-2-
ethenyloxirane; 11 = l-hydroxybut-3-en-2-one; 17 = l-hydroxybutan-2-one; 14 = 2-chlorobut-3-
en-l-al; 15 = 2-chlorobut-2-en-l-al; 20 = 2-chloro-butanal; 12 = l-chlorobut-3-en-2-one; 18 = 1-
chlorobutan-2-one; and 19 = l-chloro-2-hydroxy-but-3-ene.
Source: Adapted from Cottrell et al. (2001).
1 Comparing metabolism between species, Cottrell et al. (2001) observed that qualitative profiles
2 of metabolites from liver microsomes obtained from B6C3F1 mice, Sprague-Dawley or F344 rats, and
3 humans were similar. In all species and either gender, (l-chloroethenyl)oxirane was the major
4 metabolite detected. An important difference among species was in the stereoselectivity of the
5 formation of R- and S-enantiomers of (l-chloroethenyl)oxirane (Table 3-3). For liver microsomes
6 from both male and female Sprague-Dawley and F344 rats, there was a distinct enantioselectivity in
7 the mono-epoxidation of chloroprene to preferentially form the R-enantiomer of (1-
8 chloroethenyl)oxirane. A further study by this group (Munter et al., 2003) verified significant
September 2009 3-4 DRAFT - DO NOT CITE OR QUOTE
-------
1 differences between species in the amounts of R- and S-enantiomers of (l-chloroethenyl)oxirane
2 formed in microsomal liver incubations, the order being mouse > rat > human.
Table 3-3. Steriochemical comparison of relative amounts (percentages) of R- and
S-enantiomers of the major chloroprene metabolite (l-chloroethenyl)oxirane from
liver microsomes compared across species, strains, gender, and chloroprene
concentration (mM)
MALE
Chloroprene(mM)
5
10
20
30
40
5
10
20
30
40
5
10
20
30
40
10
20
30
Species/strain"
Sprague-Dawley rat
F344 rat
B6C3F1 mouse
Human
%R
58
62
61
60
64
62
62
62
60
64
48
47
46
47
47
43
43
43
%S
42
38
39
40
36
38
38
38
40
36
52
53
54
53
53
57
57
57
FEMALE
Chloroprene(mM)
10
20
30
40
10
20
30
40
10
20
30
40
10
20
30
Species/strain"
Sprague-Dawley rat
F344 rat
B6C3F1 mouse
Human
%R
56
56
55
59
56
54
53
54
47
45
47
46
43
44
42
%S
44
44
45
41
46
46
47
46
53
55
53
54
57
56
58
J
4
5
6
7
8
9
10
11
12
13
aAverage of three samples per species/strain.
Percentage estimated error ± 1%.
Source: Cottrell etal. (2001)
Himmelstein et al. (2004a) developed a two-compartment closed vial model to describe both
chloroprene and (l-chloroethenyl)oxirane metabolism in liver and lung fractions from rat (two strains,
F344 and Wistar), mouse, hamster, and humans. Estimates for Vmax and Km for oxidation of
chloroprene in liver microsomes ranged from 0.068-0.29 |imol/hour/mg protein and 0.53-1.33 jiM,
respectively. Oxidation (Vmax/Km) of chloroprene in the liver was slightly faster in the mouse and
hamster than in rats or humans (Table 3-4). In lung microsomes, Vmax/Km was much greater for mice
compared with the other species. Conversely, hydrolysis (Vmax/Km) of (l-chloroethenyl)oxirane in
liver and lung microsomes was faster for the human and hamster, than for rat or mouse. Glutathione S-
transferase-mediated metabolism of (l-chloroethenyl)oxirane in cytosolic tissue fractions was
described as a pseudo second-order reaction, with rates ranging from 0.0016-0.0068 hour/mg cytosolic
protein in liver and 0.00056-0.0022 hour/mg in lung.
September 2009
3-5
DRAFT - DO NOT CITE OR QUOTE
-------
Table 3-4. Kinetic parameters used to describe the microsomal oxidation of
chloroprene
TISSUE
Liver
Lung
SPECIES
Mouse
F344 rat
Wistar rat
Hamster
Human
Mouse
F344 rat
Wistar rat
Hamster
Human
ACTIVITY OF MICROSOMAL OXIDATION
^max
0.23
0.078
0.11
0.29
0.068
0.10
—
—
—
-
Km
1.03
0.53
0.84
1.33
0.68
1.5
—
—
—
-
Vmax/Km
224
146
125
218
101
66.7
1.3"
1.3a
1.3a
1.3a
"The apparent rate of lung metabolism, over the range of biologically relevant concentrations tested, was linear and was
estimated as Vmax/Km
Source: Himmelstein et al. (2004a).
1 Hurst and Ali (2007) evaluated the kinetics of R- and S-enantiomers of the chloroprene
2 metabolite, (l-chloroethenyl)oxirane in mouse erythrocytes. These results implied that
3 S-(l-chloroethenyl)oxirane was much more quickly detoxified than the R-enantiomer when incubated
4 with mouse erythrocytes in vitro. The disappearance of S-(l-chloroethenyl)oxirane was blocked when
5 erythrocytes were preincubated with diethyl maleate, indicating dependence on cellular glutathione for
6 rapid removal. The study by Hurst and Ali (2007) suggested that the R-enantiomer of (1-
7 chloroethenyl)oxirane is potentially more toxic because of slower detoxification.
8 The limited in vivo rodent studies support the postulated metabolic pathway for chloroprene.
9 For example, male Wistar rats administered 100 or 200 mg/kg chloroprene by gavage demonstrated a
10 rapid depletion of hepatic GSH and a dose-dependent increase in excreted urinary thioethers
11 (presumably GSH-conjugates), which is consistent with in vitro studies using isolated liver
12 hepatocytes (Summer and Greim, 1980). Pretreatment of rats or hepatocytes with phenobarbital or a
13 polychlorinated biphenyl (PCB) mixture (Clophen A50) to induce the mixed-function oxidase enzymes
14 enhanced the GSH depletion effect.
15 Himmelstein et al. (2004b) conducted closed-chamber gas uptake exposures with rats (Wistar
16 and F344), mice, and hamsters to evaluate metabolism rates with and without metabolic inhibition by
17 using pretreatment with 4-methyl pyrazole. Initial exposure concentrations ranged from 160-240 parts
18 per million (ppm) chloroprene. A PBPK model was used to describe the decrease in chamber
19 chloroprene concentrations over time by using metabolic parameters (Vmax, Km) scaled from in vitro
20 studies (Himmelstein et al., 2004a). The in vitro scaling of total chloroprene metabolism (Table 3-5)
21 was sufficient to explain the in vivo gas uptake data.
September 2009
3-6
DRAFT - DO NOT CITE OR QUOTE
-------
Table 3-5. Metabolic parameters of chloroprene
BIOCHEMICAL PARAMETERSa
Liver
Lung
Vmax (mg/kg-hour)
Km (mg/L)
Vmax/Km (L/kg-hour)
Vmax (mg/kg-hour)
Km (mg/L)
Vmax/Km (L/kg-hour)
SPECIES
Mouse
39.2
0.091
431.0
1.02
0.13
7.67
F344 rat
11.50
0.047
244.0
—
—
0.14
Wistar rat
15.5
0.075
208.0
—
—
0.14
Hamster
42.8
0.118
363.0
—
—
0.14
"Scaled from Himmelstein et al. (2004a) using microsomal protein content.
Source: Himmelstein et al. (2004b).
3.4. ELIMINATION
1 Limited information is available regarding the elimination of chloroprene in rodents. Summer
2 and Greim (1980) administered male Wistar rats 100 or 200 mg/kg chloroprene by gavage and
3 observed a dose-dependent, nonlinear increase in excreted urinary thioethers (presumably glutathione
4 conjugates). The clearance of these thioethers reached a threshold at 24 hours after dosing, indicating
5 that elimination was rapid.
3.5. PHYSIOLOGICALLY BASED TOXICOKTNETIC MODELS
6 Himmelstein et al. (2004b) published a physiologically based toxicokinetic model of
7 chloroprene. Construction of the mathematical model was based on physicochemical, physiological,
8 and metabolic parameters for chloroprene from mouse, rat, hamster, and humans (Table 3-6). The
9 model consisted of distinct compartments for liver and lung, as well as lumped compartments for fat
10 and slowly and rapidly perfused tissues. Metabolism of chloroprene was localized to the lung and
11 liver compartments and described by Michaelis-Menten type saturable kinetics. Although the model
12 was used to estimate the chloroprene concentration in each of the defined compartments, comparisons
13 of model predictions were limited to experimental determinations of chloroprene vapor uptake in
14 closed chambers. Inhibition of uptake was achieved with 4-methyl pyrazole pretreatment, indicating
15 that the decline of chloroprene chamber concentration was due to CYP450 monooxygenase-mediated
16 metabolism. The loss in chamber concentration in the presence of metabolic inhibition represented
17 uptake due to chemical distribution within the animal. A satisfactory model description for inhibition
18 was obtained by setting Vmax to zero for both liver and lung metabolism. No blood or tissue time-
19 course concentration data are available for model validation.
20
September 2009
3-7
DRAFT - DO NOT CITE OR QUOTE
-------
Table 3-6. Physiological parameters used for chloroprene PBPK modeling
PHYSIOLOGICAL PARAMETERS
Body weight (kg)
Ventilation (L/kg-hour)
Cardiac output (L/kg-hour)
Tissue volumes (% body weight)
Liver
Fat
Rapid perfused
Slow perfused
Lung
Blood flow (% cardiac output)
Liver
Fat
Rapid perfused
Slow perfused
SPECIES
Mouse
0.024-0.034
15
15
5.5
5.0
3.5
77.0
0.73
16.1
7.0
51.0
15.0
F344 rat
0.16-0.28
10.5
9
4.0
7.0
5.0
75.0
0.50
18.3
7.0
51.0
15.0
Wistar rat
0.20-0.34
10.5
9
4.0
7.0
5.0
75.0
0.50
18.3
7.0
51.0
15.0
Hamster
0.10-0.18
12
12
4.0
7.0
5.0
75.0
0.50
18.3
7.0
51.0
15.0
Human
NA
NA
NA
2.6
21.4
7.7
56.1
0.76
22.7
5.2
47.2
24.9
Source: Himmelstein et al. (2004b).
September 2009
3-8
DRAFT - DO NOT CITE OR QUOTE
-------
4. HAZARD IDENTIFICATION
4.1. STUDIES IN HUMANS—EPIDEMIOLOGY, CASE REPORTS, CLINICAL CONTROLS
1 Potential for human exposure to chloroprene primarily is via inhalation and perhaps by the
2 dermal route. This section summarizes studies in occupationally-exposed populations from the period
3 of 1978 to 2008.
4.1.1. Chloroprene Exposure and Cancer Effects
4.1.1.1. Overview
4 The NTP (1998, 2005) described chloroprene as reasonably anticipated to be a human
5 carcinogen based on evidence of benign and malignant tumor formation at multiple sites in animals.
6 Evidence in humans for the carcinogenicity was reported to be limited based on consideration of only
7 two occupational epidemiological studies by Pell (1978) and Li et al. (1989). Rice and Boffetta (2001)
8 and briefly examined evidence from five epidemiologic studies (Pell ,1978; Li et al., 1989; Bulbulyan
9 et al., 1998; Bulbulyan et al., 1999; Colonna and Laydevant, 2001). Although several of these earlier
10 epidemiological studies noted suggestive evidence of an association between chloroprene exposure
11 and liver cancer risk, study limitations included possible bias from cohort enumeration, follow-up, and
12 choice of reference population. Other study limitations noted included limited exposure assessment
13 data, low statistical power and the possible confounding by unmeasured co-exposures (Rice and
14 Boffetta, 2001). To date, there have been nine occupational epidemiological studies conducted
15 covering 11 cohorts. This epidemiological database is reviewed in the following section.
4.1.1.2. Individual Occupational Studies
16 Pell (1978) conducted a cohort mortality study in two neoprene (polychloroprene)
17 manufacturing plants of DuPont. The first cohort ("Louisville Works Cohort") consisted of 1,576 male
18 workers identified from a roster of wage roll employees in 1957. All workers who were exposed to
19 chloroprene were followed through December 31, 1974, accruing 26,939 person-years. Workers
20 terminated before June 30, 1957 were excluded and 17 individuals were lost to follow-up. Causes of
21 death were obtained from death certificates and coded according to the 7th and 8th revised editions of
22 the "International Classification of Diseases" (ICD). Worker exposures to chloroprene were classified
23 qualitatively as "high," "moderate," "low," and "varied" based on job description. Statistical analyses
24 were performed using Poisson probability distribution with statistical significance level at p < 0.05.
25 The general U.S. male population and all male DuPont wage roll employees were used as external and
26 internal comparison populations, respectively. The study's primary objective was to examine
27 respiratory system cancer mortality, but mortality from other site-specific cancers was also evaluated.
September 2009 4-1 DRAFT - DO NOT CITE OR QUOTE
-------
1 Among the 193 deaths detected in this cohort, 51 were due to cancer and 16 were due to cancer
2 of the respiratory system. Compared to U.S. rates, the standardized mortality ratios (SMRs) for all-
3 cause mortality, total cancer mortality and respiratory system cancer mortality were 69.0, 96.6, and
4 98.4, respectively. Based on the internal comparison SMRs of 114.0 were detected for total cancer
5 mortality and 109.6 for respiratory system cancer mortality. The internal comparison yielded SMRs of
6 108.7 (15 cases) and 113.2 (12 cases) for respiratory cancer after 15- and 20-year latency periods,
7 respectively. SMRs were lower for the same latency periods when compared with the U.S. general
8 population. Thirteen of the 16 deaths due to respiratory system cancer occurred in smokers, while
9 smoking history was unknown for the other three. Analyses by high exposure occupation did not show
10 any significant change in SMRs or any statistically significant trend when analyzed by years since first
11 exposure. Other cancer deaths that were detected included 19 of the digestive organs (SMR = 142.9
12 using an internal comparison) and seven of the lymphatic and hematopoietic tissues (SMR = 155.6
13 using an internal comparison). All the SMRs observed in this study were not statistically significant
14 based on both internal (DuPont) and U.S. general population mortality rates.
15 These data were reanalyzed by the National Institute for Occupational Safety and Health
16 (NIOSH) using a modified life-table analysis (Leet and Selevan, 1982). Workers were classified into
17 high and low exposure categories based on a classification scheme developed by an industrial
18 hygienist who worked at the plant. Eight hundred and fifty-one workers were allocated to the high
19 exposure group and 823 to the low exposure group, with some workers contributing person-years in
20 both categories when their exposures or job titles changed. A total of 26,304 person-years were
21 accrued, with 13,606 person-years in the high-exposure and 12,644 in the low-exposure category.
22 Compared to U.S. population rates, the overall SMR for the total cohort was 79. Excess deaths were
23 observed for cancers of the digestive system (especially the biliary passages and liver), the lung, and
24 the lymphatic/hematopoietic system. The only statistically significant SMR of biliary passage and
25 liver was based on four cases, three from the high-exposure category (Table 4-1). Of these three
26 deaths, one was due to liver cancer, and the other two to gall bladder cancer. Cancer mortality data
27 were analyzed with respect to latency and duration of exposures stratified into 10-year intervals.
28 Statistically significant trends were not observed in either the latency analysis or the years of presumed
29 chloroprene exposure analysis, but these analyses were based on small numbers.
30 The main limitations of the Pell (1978) study and the NIOSH reanalysis (Leet and Selevan,
31 1982) are lack of smoking history data, lack of adjustment for other potential risk factors, and absence
32 of quantitative exposure information. Exclusion of workers terminated prior to June 30, 1957, might
33 have also resulted in some unidentified cancer deaths that could have been associated with earlier
34 higher exposures. Moreover, as pointed out by Leet and Selevan (1982), the statistical power of the
35 study to detect a significant excess in mortality was low when the sub-cohort analyses were conducted.
September 2009 4-2 DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-1. Standardized mortality ratios (SMRs) for the DuPont Louisville Works
cohort relative to general U.S. population rates.
CAUSE OF DEATH
All Causes
All Cancers
Digestive
Biliary /liver
Trachea, bronchus, lung
Lymphatic, hematopoietic
TOTAL COHORT
CASES, SMR (95% CF)
193, 79 (68-91)
51, 107(80-141)
19, 145(87-227)
4, 571 (156-1463)
17, 106(62-170)
7, 140(56-288)
HIGH EXPOSURE
CASES, SMR (95% CI)
91, 75 (61-92)
25, 107(69-158)
8, 125(54-246)
3, 750(155-2192)
10, 128(61-236)
4, 160(44-410)
LOW EXPOSURE
CASES, SMR (95% CI)
102, 82 (67-100)
26, 107(70-157)
11, 164(82-294)
1, 250(6-1393)
7, 86 (35-178)
3, 120(25-351)
aCI = confidence interval.
Source: Leet and Selevan (1982).
1 Pell (1978) also evaluated a second cohort that originally consisted of 270 males ("Chamber
2 Works Cohort") believed to be exposed between 1931 and 1948 in a neoprene manufacturing facility
3 and followed through December 31, 1974. Follow up was complete for 240 workers. Since historical
4 records were not complete for this cohort, efforts were made to assess exposures for former employees
5 based largely on the recall of other employees. The observation period, during which latency in tumor
6 induction could be analyzed, was 30-40 years from date of first exposure. Examination of mortality
7 following a long latency period was considered a strength of this study.
8 A total number of 55 deaths were observed in this cohort. Study exclusions included thirteen of
9 these deaths occurring prior to 1957 (the starting point of observation assuming a 15-year latency
10 period) and three deaths occurring due to heart disease and malignant melanoma among former
11 laboratory personnel who had little or no exposure. The 39 observed deaths that occurred from 1957 to
12 1974 were slightly more than the 37.7 expected using the DuPont comparison population. The 12
13 observed cancer deaths were also elevated (SMR = 140) but were not statistically significant. There
14 were three deaths due to digestive cancer compared to 2.7 expected and were four deaths due to lung
15 cancer compared to 3.0 expected. The five observed cancers of the urinary system (3 bladder and 2
16 kidney) were significantly elevated (SMR = 300 compared to the DuPont population and SMR = 250
17 compared to the U.S. general population: p < 0.01 for both). The authors attributed the bladder cancers
18 to beta-naphthylamine exposure. Biliary and liver cancers were not examined in this study. Small
19 cohort size, low statistical power, and lack of quantitative exposure data were limitations of this
20 analysis.
21 Li et al. (1989) conducted a cohort mortality study of Chinese employees who worked in one of
22 three shops with chloroprene exposure (a chloroprene monomer workshop, a neoprene workshop, and
23 a laboratory) within a larger chemical plant. A cohort of 1,258 employees who had worked for at least
24 one year of chloroprene-related work prior to June 30, 1980 were identified from an employee roster.
25 The follow-up period for cancer deaths was from July 1, 1969 through June 30, 1983. Cancer mortality
26 was assessed by searching the death registries at the plant's hospital and the police substation; cancer
27 diagnoses were verified by review of medical records at the city general hospitals and cancer hospitals.
September 2009
4-3
DRAFT - DO NOT CITE OR QUOTE
-------
1 Exposures were assigned to occupations based upon measured concentrations in air at work sites and
2 duration of exposure at different sites. When these levels were not available, exposures were
3 determined through interviews with workers and administrators. Exposure assignments took into
4 account movement between exposure areas and were designed to roughly represent time-weighted
5 average exposure values. Follow up was achieved for 1,213 (96%) cohort members (955 males and
6 258 females) and SMRs were calculated using sex- and age-specific mortality in the local area. A total
7 of 721 (75%) males and 131 (51%) females were exposed for more than 15 years, while 131 (14%)
8 males and 9 (3%) females were exposed for more than 25 years. Statistically significant differences (p
9 < 0.005) in exposure to chloroprene were detected in males compared to females based on > 15 years
10 and > 25 years of exposure.
11 Person-years were computed by 5-year categories for the total cohort and for the subgroup
12 starting from July 1, 1969 or when the individual first started working with chloroprene through June
13 30, 1983 for live individuals or until their date of death.1 SMRs were calculated using sex- and age-
14 specific local area rates in 1973-1975. The results presented below are for male workers only as all
15 sixteen cancer deaths were reported among male workers (Table 4-2). The all-cancer SMR for the
16 male workers was 271 (p < 0.01). Of 955 males, 464 (49%) were employed in occupations with high
17 exposures such as maintenance mechanics and monomer/polymer operators. The SMRs for male
18 workers in several high exposure areas were statistically significant for liver and lung cancer mortality.
19 An increased SMR for liver cancer was observed, with four deaths occurring among monomer workers
20 and two deaths occurring in polymer mechanics. Half of the cancers in the monomer shop were
21 primary liver cancers (4 observed, SMR = 482, p < 0.01), with two occurring among the maintenance
22 mechanics (SMR = 1667, p < 0.05).
1 Person-years accrued were not reported in the paper.
September 2009 4-4 DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-2. Standardized mortality ratios (SMRs) for all cancers, liver and lung
cancer among males exposed to chloroprene relative to general Chinese population
rates
NUMBER OF DEATHS/SMR
EXPOSURE AREA
Total cohort
Monomer workshop
Vmylacetylene operator3
Monomer operator
Maintenance mechanic3
Neoprene workshop
Polymer operator3
Final treatment
Maintenance3 mechanic
Laboratory
Quality monitor3
Researcher
ALL CAUSE
16/271*
8/377
O/—
4/450"
4/1,290"
5/176
5/394"
07—
O/—
3/319
1/129
21,176"
LIVER CANCER
6/242
4/482"
O/—
2/465
2/1667"
2/165
O/—
07—
2/357
O/—
O/—
O/—
LUNG CANCER
2/513
1/714
07—
07—
1/5000"
1/556
1/1250
07—
O/—
O/—
Statistical significance: * p < 0.01; ** p < 0.05.
3: High Exposure Area
Source: Lietal. (1989)
1 One of the limitations of the Li et al. (1989) study include insufficient mortality data especially
2 since only three years of local area data were available to calculate SMRs. If these years are not
3 representative of the entire study period, then the SMRs could be biased. For example, if the general
4 population experienced higher mortality during the times periods not examined (i.e., 1969-1972 &
5 1976-1983) then the SMRs would be underestimated due to a lower expected number of deaths. If the
6 mortality was lower during the other time periods, the SMRs would be overestimated. Lack of
7 quantitative exposure information precluded conducting internal analyses by latency or duration of
8 exposure. Additionally, there was no data on alcohol use or smoking history and limited information
9 was available on other potential confounders such as co-exposures to chloroprene oligomers. The
10 authors did consider potential confounding exposures due to benzene and anti-ager D (N-phenyl-Z-
11 naphthylamine) but determined that these exposures were limited and not likely to influence the
12 results.
13 Li et al. (1989) also conducted a case-control study for the entire plant. Of 55 observed cancer
14 deaths, 54 were matched with the same number of non-cancer deaths among plant workers based upon
15 gender, age (± 2 years) and date of death (± 2 years). The authors observed that 16 of the cancer deaths
16 (30%) were among workers exposed to chloroprene compared to only four of the non-cancer deaths
17 (7%), yielding an odds ratio of 13 (p < 0.005). Although the average age at death was 12.7 years less
18 for the exposed cancer cases relative to the unexposed cancer cases (p < 0.001), these findings are
19 limited by lack of data on co-exposures and other potential confounders.
20 Bulbulyan et al. (1998) examined cancer mortality at a Moscow shoe factory with exposures to
21 chloroprene from glue and from poly chloroprene latex (a colloidal suspension of poly chloroprene in
22 water). The cohort consisted of 5,185 workers (4,569 women and 616 men) employed for at least two
September 2009 4-5 DRAFT - DO NOT CITE OR QUOTE
-------
1 years during 1960-1976 at specific production departments (i.e., cutting, fitting, lasting and making,
2 and finishing). Auxiliary departments and management employees were excluded. Work histories
3 were obtained from the personnel department, and subjects were assigned exposure levels based on
4 department and job; industrial hygiene measurements of exposure levels were conducted in the 1970s.
5 The authors provided detailed exposure data by job and department, ranging from a high of 20 mg/m3
6 (gluers in the finishing department) to an intermediate level of 0.4-1 mg/m3 (all other jobs in the
7 finishing department and all jobs in the lasting and making department) to the unexposed (all jobs in
8 the cutting and fitting departments).
9 The authors concluded that the industrial hygiene data were not systematic enough to assign
10 quantitative exposures to each worker since the collection of samples varied by locations and by
11 different years. They, therefore, devised a relative exposure system, where workers in the high-
12 exposure assignment was assigned a level of 10, intermediate-exposure - a level of 1, and unexposed -
13 a level of 0. Cumulative exposures for individual workers were calculated by multiplying years of
14 exposure by the level of exposure, taking into account changes in job and department. In addition,
15 workers were classified by their highest exposure category. The authors considered confounding
16 exposures, including benzene exposures (6-20 ppm) in the high polychloroprene exposure group
17 during the 1950s, but did not adjust for those exposures in their analysis.
18 Mortality follow up was conducted from 1979-1993 which included 70,328 (62,492 in females
19 and 7,836 in males) person-years of observation. Thirty-seven percent of cohort members contributing
20 26,063 (female and male distribution was not provided) person-years were unexposed. Death
21 certificates were acquired from the National Registry Office Card Index and causes of deaths were
22 classified using ICD-9. Mortality rates of general population of Moscow were used for comparison.
23 For the general population, mortality data for five cancers (liver, kidney, bladder, pancreas, and
24 malignant neoplasm of mediastinum and rhabdomyosarcoma of the heart) were only available for
25 1992-1993. Therefore, the number of expected deaths among these sites during 1992-1993 was
26 applied to the entire cohort for the entire period of observation. A Poisson distribution was used to
27 calculate the 95% CIs. One hundred thirty-one (2.5%) workers were lost to follow up. SMRs were
28 calculated for the entire cohort and separately for females and males. Among the total cohort, SMRs
29 were statistically significant for all cancers, liver cancer and leukemia (Table 4-3). Cancer-specific
30 SMRs for liver cancer and leukemia were statistically significant in females but not in males, while the
31 SMR for lung cancer was significant in males only.
September 2009 4-6 DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-3. Standardized mortality ratios (SMRs) for selected cancer risks relative
to general population rates of Moscow, Russia.
CAUSE OF DEATH
All causes
All cancers
Liver cancer
Lung cancer
Leukemia
TOTAL COHORT
CASES, SMR (95% CI)
900; 103 (97-110)
265; 122a (107-137)
10; 240a (110-430)
31; 140(90-200)
13; 190a (100-330)
MEN
CASES, SMR (95% CI)
181; 121 (104-140)
56; 158(119-205)
2; 240 (30-860)
17; 170a (100-270)
2; 190 (20-700)
WOMEN
CASES, SMR (95% CI)
719; 100 (93-107)
209; 115(100-131)
8; 230a (100-460)
14; 110 (60-190)
11; 190a (100-350)
1
2
o
J
4
5
6
7
10
11
12
13
14
15
16
a Statistical significance p < 0.05.
Source: Bulbulyan et al. (1998).
Internal relative risk (RR) analyses (controlling for gender, age, and calendar period) were
conducted for selected cancers by using multivariate Poisson regression models, with trends evaluated
with the Mantel-extension test. Estimates for liver cancer were relatively imprecise since only one
liver cancer death was observed in the no-exposure category (a low number since this category
included 29% of all observed deaths). Stratified analyses by gender were not reported. Internal
analyses comparing the high exposure group to the unexposed resulted in statistically significant RRs
for all causes of death (Table 4-4). Although they were not statistically significant largely due to a
small number of cases, elevated RRs ranging from 2.2-4.9 were detected for leukemia, and cancers of
the liver, kidney and colon.
Table 4-4. Selected relative risk (RRs) estimates for the high exposure group
relative to unexposed factory workers
CAUSE OF DEATH
All causes
Liver cancer
Colon cancer
Kidney cancer
Leukemia
CASES
194
3
8
2
5
HIGH-EXPOSURE RR (95% CI)a
1.23b (1.02-1.49)
4.9 (0.5-47)
2.6 (0.8-7.9)
3.3 (0.3-37)
2.2 (0.6-8.4)
a Reference group is defined as workers with no chloroprene exposure.
b Statistical significance p < 0.05.
Source: Bulbulyan et al., (1998).
Although there were only a few deaths in each group, RR analysis by duration of employment
in the highest exposure categories (1-9 years, 10-19 years, 20+ years) relative to no exposure showed
a significant trend for liver cancer but not for leukemia mortality (Table 4-5). The cumulative
exposure analysis indicated an increased risk of liver cancer mortality based upon six deaths in the
intermediate exposure category (RR = 7.1, 95% CI: 0.8-61) and three deaths in the highest exposure
category (RR = 4.4, 95% CI: 0.4-44). Kidney cancer was increased in all categories but none of the
RRs were statistically significant and no overall trend was observed.
September 2009
4-7
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-5. Internal relative risks (RRs) by duration of employment in the high-
exposure category.
CAUSE OF
DEATH
Liver cancer
Leukemia
1-9 YEARS
CASES; RR (95% CI)
1; 2.7 (0.2-45)
2; 1.3 (0.2-7.3)
10-19 YEARS
CASES; RR (95% CI)
1; 8.3 (0.5-141)
2; 3.4 (0.6-19)
20+ YEARS
CASES; RR (95% CI)
1; 45.0 (2.2-903)
1; 8.8 (0.7-66)
TREND
p = 0.02
p = 0.07
Source: Bulbulyan et al. (1998).
1 The most prominent finding in the Bulbulyan et al. (1998) cohort was 10 deaths occurring from
2 liver cancer. The authors detected 11 deaths (3 in males and 8 in females) due to cirrhosis, a precursor
3 of primary liver cancer, but did not adjust for this as a potential confounder. Increased mortality due to
4 leukemia was observed in all categories for both cumulative exposure and duration of employment
5 (with high exposure) but neither trend was statistically significant. The authors suspected a causal role
6 of chloroprene in the leukemia deaths but could not rule out a possible role of exposure to benzene. A
7 significant increase in lung cancer was observed among males only, which may have been due to
8 confounding by smoking. Potential confounding by smoking could not be examined due to lack of
9 data for this cohort. Pancreatic cancer, which may be smoking related, was also observed in males
10 only. No excess risk for lung cancer was observed in females or in the total cohort. Lack of precise
11 quantitative exposure information, no adjustment for confounding risk factors, and exclusion of deaths
12 prior to 1979 resulting in relatively low statistical power were some of the limitations of this study.
13 Similar to Li et al study, the minimal data on observed deaths for some cancers among the general
14 population may have also resulted in biased SMR values if these years were not representative of
15 mortality during the entire study period.
16 Bulbulyan et al. (1999) conducted a retrospective cohort study of 2,314 workers (1,897 males,
17 417 females) who had been employed in production departments of a chloroprene monomer
18 production plant in Yerevan, Armenia, for at least two months between 1940 and 1988 and were alive
19 as of 1979. Mortality was followed from 1979-1988, and vital status was accessed through the
20 Yerevan Address Bureau. Death certificates were coded by using ICD-9 revision. Sixty-three (3%)
21 individuals were lost to follow-up. Industrial hygiene exposure measurements of chloroprene were
22 available both before and after 1980, when production changes led to a dramatic decrease in exposures.
23 Before 1980, exposures averaged 5.59-69.80 mg/m3 (1.54-19.3 ppm) during the summer and 2.30-
24 249.5 mg/m3 (0.63-68.9 ppm) during the winter. After 1980, for the same seasons, these averages
25 were 0.80-3.60 mg/m3 (0.22-0.99 [summer] ppm) and 0.55-2.10 mg/m3 (0.15-0.58 [winter] ppm),
26 respectively. Work histories were obtained from the personnel department, including the start and end
27 of each job, and from the departments of employment. Relative exposure values were assigned based
28 on either high exposure (production operators: six units before 1980, three units after 1980) or low
29 exposure (other production workers: two units before 1980 and one unit after 1980). Unexposed
30 workers were assigned a relative exposure value score of zero. SMRs and standardized incidence
31 ratios (SIRs) were calculated based on comparison rates for the entire Armenian population, and 95%
September 2009 4-8 DRAFT - DO NOT CITE OR QUOTE
-------
1 CIs were also calculated by using a Poisson distribution assumption. Internal RR estimates were
2 calculated by using multivariate Poisson regression models and adjusting for age, calendar period, and
3 gender.
4 A total of 21,107 person-years were contributed by the study population. There were 20 deaths
5 during the observation period with four due to stomach cancers and three each resulting from liver and
6 lung cancers. The SMR was statistically significant for liver cancer only (SMR = 339, 95% CI 109-
7 1050). Two liver and two lung cancer deaths were identified among males, while one liver cancer
8 death and one lung cancer death were identified in females. No internal comparisons were included in
9 the SMR analysis. Cancer incidence data was examined from 1979-1990 through the Armenian
10 Cancer Registry. Several types of cancers (37 cases) were identified with six liver and six lung cancers
11 (five each in males) being the most prevalent (Table 4-6). The SIRs for liver cancer were statistically
12 significant for the total cohort (SIR = 327, 95% CI 147-727) and for males (SIR = 303, 95% CI 126-
13 727) when stratified by gender. SIRs below 100 were observed for lung cancer among both the total
14 cohort as well as among males only.
Table 4-6. Selected standardized incidence ratios (SIRs) for chloroprene monomer
cohort relative to the general Armenian population.
CANCER TYPE
All cancers
Lung cancer
Liver cancer
OBSERVED
37
6
6
SIR (95% CI)
0.68 (0.49-0.94)
0.53 (0.24-1.19)
3.27a (1.47-7.27)
a Statistical significance p < 0.05.
Source: Bulbulyan et al. (1999).
15 Internal trend analyses of plant workers only showed increasing incidence of liver cancer by
16 duration of employment with a statistically significant relative risk among chloroprene production
17 workers who were employed for more than 20 years (4 cases, SIR =3.45, 95% CI: 1.29-9.20).
18 Evaluation of liver cancer incidence by duration of employment (<1 year, 1-9 years and 10+ years) in
19 the high chloroprene exposure groups resulted in a statistically significant SIR in the 10+ years
20 category (SIR = 6.1; 95% CI: 2.3-16.3). Similar findings were noted when analyzed using cumulative
21 exposure with a statistically significant SIR of 4.9 (95% CI: 2.0-11.7) among the five cases in highest
22 cumulative exposure of 40+ units. All six cases of liver cancer in this study occurred among highly
23 exposed operators. These internal analyses suggested a possible dose-response relationship between
24 chloroprene exposure and liver cancer incidence.
25 The authors discussed the strong healthy worker effect observed in this study. In particular,
26 they suggested that the low SMRs might be due, in part, to potential loss of early cases resulting from
27 not beginning the follow-up period until 1979. In addition to the incomplete enumeration of outcomes
28 among the workers, the authors acknowledged that there might be misclassification as well as
29 incomplete registration of liver cancers in the Armenian registry. Furthermore, although measurements
September 2009 4-9 DRAFT - DO NOT CITE OR QUOTE
-------
1 of chloroprene levels were available, investigators were unable to develop quantitative estimates and
2 assigned exposure units to the workers depending upon their job description. The role of potential
3 confounding by alcohol use and smoking could not be examined due to lack of data. The high
4 incidence (27 in males and 5 in females) of liver cirrhosis, a precursor for liver cancer, is an unlikely
5 confounder as it is likely an intermediate in the causal pathway precluding statistical adjustment. There
6 was also little evidence that several other co-exposures (i.e., vinyl acetate, toluidine, talc, and
7 mercaptans) that were not adjusted for either in mortality or incidence analyses are liver carcinogens.
8 Romazini et al. (1992) investigated cancer mortality in a retrospective cohort study of 660.
9 French chloroprene polymer manufacturing workers (599 males, 61 females) employed for at least two
10 years at a poly chloroprene plant. The follow-up period was from 1966-1989 with 32 observed deaths
11 included in the study; an additional 18 potential study subjects were lost to follow up. No excess
12 mortality was observed compared to regional rates. In a nested case-control study comparing time of
13 employment, the authors found that workers exposed to conditions prior to 1977 had a much higher
14 risk of death compared to those exposed to chloroprene after 1977. Similar to other studies, the small
15 size of this cohort and inability to control for smoking and other potential confounders limited the
16 conclusions that could be drawn from this study.
17 Colonna and Laydevant (2001) conducted a cohort cancer incidence study among 533 males
18 who worked a chloroprene production plant in Isere, France, for at least two years between January
19 1966 (when the plant opened) and December, 1997. Cancer incidence cases were traced through the
20 Isere cancer registry from 1979 (when the registry was founded) through 1997. Workers who died
21 before 1979 or who left the area were not traced (the number of untraced incident cancers was not
22 estimated). Work histories were collected and jobs were classified into low, intermediate, and high
23 chloroprene exposure groups based on estimated exposure of < 2 ppm, 2-5 ppm, and > 5 ppm
24 respectively. Exposure duration was divided into three groups of <_10 years, 11-20 years and > 20
25 years. The cohort was divided into two groups, workers employed prior to 1977 and those employed in
26 1977 or later, based on lower anticipated exposures following significant changes in worker protection.
27 SIRs were calculated using the general population rates of Isere as a reference and confidence intervals
28 were calculated using a Poisson distribution.
29 A total of 7,950 person-years were accrued. Of the 34 incident cancers, 32 occurred prior to
30 1977. There were nine lung cancers, nine cancers of the head and neck (including three laryngeal
31 cancers), and one liver cancer. SIRs were calculated for various cancers including those occurring in
32 the head and neck, larynx, lung, liver and colon/rectum (Table 4-7). With the exception of
33 colon/rectum, all of the SIRs exceeded 100 with most of the cases and higher SIRs noted for earlier
34 periods of first employment (i.e., before 1977).
September 2009 4-10 DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-7. Standardized incidence ratios (SIRs) for elevated cancer risks for plant
workers relative to general population rates of Isere, France.
CANCER TYPE
All Cancers
Head and Neck
Larynx
Lung
Liver
Colon/Rectum
TOTAL COHORT
CASES, SIR (95% CI)
34, 1.26 (0.88-1.77)
9, 1.89 (0.87-3.59)
3,2.43(0.50-7.13)
9, 1.84 (0.84-3.49)
1, 1.36 (0.04-7.63)
2, 0.66 (0.08-2.39)
BEFORE 1977
CASES, SIR (95% CI)
32, 1.46 (1.00-2.06)
8,2.09(0.90-4.11)
3, 2.97 (0.61-8.68)
8, 1.99 (0.86-3.91)
1, 1.64(0.05-9.13)
2,0.79(0.10-2.87)
Source: Colonna and Laydevant, (2001).
1 Although none of the SIRs were statistically significant, a significant trend was observed when
2 the data were analyzed by duration of exposure. Five lung cancers were found in workers with > 20
3 years of exposure (SIR = 2.57), 3 in 11-20 years exposure (SIR = 1.49) and 1 in <_10 years exposure
4 (SIR = 1.06). No significant excesses were observed in head and neck cancer by duration of exposure.
5 No trend was detected for lung cancer incidence in relation to intensity of exposure with SIRs of 4.63
6 (95% CI: 1.27-11.91), 1.25 (95% CI: 0.15-4.51), and 1.23 (95% CI: 0.26-3.61) reported for the low,
7 intermediate and high exposure categories, respectively.
8 Increased lung cancer and laryngeal cancer were observed in this study. Given that smoking is
9 strongly associated with lung cancer, and since seven of the eight lung cancer cases were smokers, the
10 investigators concluded that the lung cancers excess was unlikely to be due to chloroprene exposure.
11 Although smoking and alcohol consumption was discussed as strongly associated with laryngeal
12 cancer, no additional information was provided in the paper. This study found only one incident liver
13 cancer but noted that liver cancer incidence was likely under-estimated due to difficulties in case
14 enumeration. Study limitations included lack of precise quantitative exposure information, low cancer
15 incidence, and reduced power because of elimination of workers who had died or left the area prior to
16 1979.
17 More recently, Marsh et al. (2007a) evaluated mortality patterns of four chloroprene production
18 facilities by using external regional rates and using internal comparisons (Marsh et al., 2007b). This
19 study attempted to address the problems identified with earlier studies by conducting a detailed
20 exposure assessment for both the chloroprene and a potential confounding co-exposure, vinyl chloride
21 monomer (Esmen et al., 2007a, 2007b, 2007c; Hall et al., 2007). As described in detail by Esmen et al.
22 (2007c), a historical review of processes at all four plants led to the assignment of exposures to 257
23 unique tasks. Taking into account shared tasks or rotation between tasks, job title-based exposures to
24 chloroprene were assigned to one of seven categories, including unexposed (< 0.0005 ppm). Vinyl
September 2009
4-11
DRAFT - DO NOT CITE OR QUOTE
-------
1 chloride exposures were assigned to one of five categories, including unexposed (< 0.01 ppm) (Esmen
2 et al., 2007b).
3 Two of the facilities evaluated were in the U.S. - DuPont/Dow plants at Louisville (L),
4 Kentucky and Pontchartrain (P), Louisiana. The third one was in Northern Ireland - the Maydown (M)
5 plant, and the fourth facility was in Grenoble (G), France - the Enichem Elastomer plant. These plant
6 cohorts included 5,507 workers (L), 1,357 workers (P), 4,849 workers (M), and 717 workers (G).
7 Median cumulative exposures to chloroprene at these plants were 18.35, 0.13, 0.084, and 1.01 ppm-
8 years, respectively. Vinyl chloride exposures existed in only two plants, Louisville and Maydown.
9 Their median cumulative vinyl chloride exposures were 1.54 and 0.094 ppm-years, respectively.
10 The study period for the cohorts encompassed 52 (L), 41(M), 39 (P), and 34 (G) years resulting
11 in 197,919, 127,036, 30,660, and 17,057 person-years, respectively (Marsh et al., 2007a). Vital status
12 was assessed using several different sources. A trained nosologist using the ICD codes in effect at the
13 time of death coded the underlying cause of death. A total of 3,002 deaths had occurred during the
14 follow-up period in the chloroprene cohort and cause of death was ascertained for 2,850 individuals
15 (95%). A modified Occupational Cohort Mortality Program was used to conduct statistical analyses.
16 Independent analyses were conducted for the four facilities for total cancer deaths and certain site-
17 specific deaths. Person-years at risk were computed for each individual by race, sex, age group,
18 calendar time, duration of employment, and the time since first employment. SMRs and 95% CIs were
19 calculated for the total cohort and selected sub-cohorts for each plant.
20 All cause combined mortality was significantly reduced (compared to local county rates) for
21 each of the four cohorts (Table 4-8). In addition, each cohort had significantly reduced mortality for all
22 cancers, and the largest cohort, Louisville, had significantly reduced mortality from respiratory
23 cancers. The total number of cancer deaths observed at each of the four plants was 652 (L), 128 (M),
24 34 (P), and 20 (G). Reported lung cancer deaths were 266, 48, 12, and 10, while liver cancer deaths
25 were 17, 1,0, and 1 for L, M, P, and Q respectively. Compared to the local population rates, fewer
26 deaths from liver cancer were observed in the Louisville (SMR = 90, 95% CI: 52-144) cohort than
27 expected. All other sites had no more than one death due to liver cancer. Similar to the healthy worker
28 effect observed in other studies, fewer cancer deaths were were reported in the occupational cohorts
29 when compared to general population estimates. When chloroprene exposed and unexposed workers
30 were analyzed separately, the SMRs for all cancers were all significantly reduced for exposed workers
31 at each plant, while they were generally higher (at or above expected levels for all plants except at
32 Grenoble) for unexposed workers. Given that there were few unexposed workers in these cohorts,
33 these values are unstable and are also difficult to interpret given the healthy worker effect bias.
September 2009 4-12 DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-8. Standardized mortality ratios (SMRs) at each of four chloroprene
production facilities
CAUSE OF
DEATH
All Causes
All Cancers
Respiratory
Cancers
All Cancers:
Exposed
Unexposed
LOUISVILLE
(L) CASES,
SMR (95% CI)
2403
74 (71-77)
652
75 (69-80)
266
75 (66-85)
651
74 (69-80)
1
99(3-551)
MAYDOWN
(M) CASES,
SMR (95% CI)
435
60 (55-67)
128
68 (56-80)
48
79 (58-105)
114
62 (51-75)
14
126 (69-212)
PONTCHARTRAIN
(P) CASES, SMR
(95% CI)
102
53 (43-65)
34
68 (47-95)
12,
62 (32-109)
26
57 (37-84)
8
144 (62-285)
GRENOBLE
(G) CASES,
SMR (95% CI)
62
65 (50-83)
20
59 (36-91)
10
85 (41-156)
15
59 (33-97)
5
61 (20-142)
TOTAL
CASES, SMR
(95% CI)
3002
70 (67-73)
834
73 (68-78)
336
75 (68-74)
806
71 (66-76)
28
108 (72-156)
Source: Marsh et al. (2007a).
1 In their companion paper (Marsh et al., 2007b), the authors conducted internal RR analyses at
2 each of these four plants. Exposure-response trends across categories of exposures (based on quartiles)
3 were examined using a forward stepwise regression modeling approach to adjust for potential
4 confounding. Analyses were conducted by considering 5- and 15-year lagged exposures and using
5 white/blue collar as a surrogate for lifetime smoking. Absolute mortality rates were estimated by
6 calculating exposure category-specific SMRs using external mortality rates. The internal analyses for
7 all cancers showed increasing RRs with duration of exposure (< 10, 10-19, 20+ years) to chloroprene
8 in plants L and M, but a statistically significant trend (p< 0.007) was only noted for Plant M. Relative
9 to less than 10 years of exposure, increased RRs were noted for 10-19 years (RR = 1.53; 95% CI =
10 1.00-2.34) and 20+years (RR= 1.78;95%CI= 1.11-2.84) of exposure. The external comparison
11 consistently showed SMRs less than the internal analysis (and mostly below 1) for both the plants
12 suggestive of bias due to the healthy worker effect.
13 The internal analysis for liver cancer could only be conducted in the Louisville cohort, which
14 contained 17 of the 19 observed deaths and also had the highest chloroprene levels. Despite the limited
15 number of deaths, these data show some evidence of a dose-response effect across the four exposure
16 levels (p = 0.09). Although the individual RRs were not statistically significant, the range for the
17 highest three exposure levels was from 1.9-5.1.
18 The results of the internal analyses for respiratory cancers at the three plants (M, P, G) without
19 worker status adjustment showed higher RRs with increasing cumulative exposure (Table 4-9). The
20 observed trends were not statistically significant but were based on a small number of respiratory
21 cancers. In contrast, the plant with the most cases (L) showed little evidence of an exposure-response
22 relationship. The investigators adjusted for the potential confounding by smoking status in the analyses
23 of lung cancer mortality at Louisville only (due to small numbers at the other plants) using the
September 2009 4-13 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
employment status as a surrogate of blue versus white collar workers. This decision was justified by
the authors based upon this variable being a surrogate for variables associated with smoking such as
education and socio-economic status. It is impossible, however, to discern whether this surrogate
resulted in control for smoking or resulted in an over-adjustment since work status is so highly
correlated with chloroprene exposures.
Table 4-9. Relative risks (RRs) for respiratory cancers by cumulative chloroprene
exposure
PLANT
Louisville (L)
Maydown (M)
Pontchartrain (P)
Grenoble (G)
LEVEL la
(LOWEST)
N
62, Reference
14, Reference
3, Reference
2, Reference
LEVEL 2
N, RR (95% CI)
67, 1.00 (0.71-1.43)
9, 1.65(0.66-4.15)
3, 1.60 (0.20-12.8)
1, 0.61 (0.05-6.76)
LEVEL 3
N, RR (95% CI)
77, 1.32 (0.94-1.88)
12, 1.89 (0.72-4.96)
2,2.90(0.20-34.1)
4, 2.87 (0.35-39.7)
LEVEL 4
N, RR (95% CI)
60, 0.85 (0.58-1.23)
13, 2.28 (0.86-6.01)
4, 2.32 (0.30-21.8)
3,3.14(0.30-48.0)
TREND
p = 0.71
p = 0.10
p = 0.34
p = 0.17
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
"Chloroprene exposure (in ppm years) levels varied by plant: L (<4.7->164.1); M (<0.04->24.5); P (<0.02->16.2); G (O.05-
>23.9).
Source: Marsh et al. (2007b)
The authors also conducted internal analyses of cancer mortality and vinyl chloride exposure
(the primary co-exposure in this study) at the Louisville plant. They found an inverse association
(many of them statistically significant) between risk of both respiratory and liver cancer in relation to
vinyl chloride exposures. In fact, the vast majority of respiratory and liver cancers occurred among
workers who were unexposed to vinyl chloride. If vinyl chloride is a negative confounder of the
association between chloroprene and liver cancer, then the reported association between chloroprene
and liver cancer would be an underestimate of the association adjusted for vinyl chloride. Given this, it
is highly unlikely that confounding by vinyl chloride could explain the associations observed between
chloroprene and these cancers. In addition, the authors reported that there was no correlation between
cumulative exposures to vinyl chloride and chloroprene among these workers.
The Marsh et al. (2007a, 2007b) study is one of the more comprehensive studies to date, largely
due to exposure assessment data which allowed for internal comparisons. Although the authors
concluded that their study provided no evidence of cancer risk associated with chloroprene exposures,
the data on chloroprene exposure and liver cancer risk was not inconsistent with previous findings.
While the authors stated that "chance alone does not appear to be an explanation for the cancer deficits
observed among unexposed workers," they rejected the healthy worker effect as a plausible
explanation. Instead, they postulated "some heretofore unknown selection factors for low cancer
incidence or mortality were operating on the unexposed subjects" (Marsh et al., 2007b). This is
inconsistent with the data that Marsh et al. (2007b) presented on cancer mortality among exposed and
September 2009
4-14
DRAFT - DO NOT CITE OR QUOTE
-------
1 unexposed workers (Table 4-9). It can be concluded that the explanations for observed associations
2 between chloroprene exposure and cancer, especially liver and respiratory cancers, given by Marsh et
3 al. (2007b) are not entirely consistent with the data presented. Despite the study limitations including
4 small numbers, this study is informative given the findings of the internal analysis including the data
5 on cumulative exposure and liver cancer risk.
4.1.1.3. Summary and Discussion of Relevant Methodological Issues
6 Nine studies covering 11 cohorts were reviewed to assess the relationship between exposure to
7 chloroprene and cancer incidence and mortality. Four cohorts had less than 1000 workers, while the
8 remaining cohorts had sample sizes less than 6000. The most consistent finding was excess liver
9 (Bulbulyan et al. 1999, 1998; Li et al., 1989; Leet and Selevan, 1982) and lung/respiratory system
10 (Marsh et al., 2007b; Colonna and Laydevant, 2001; Bulbulyan et al. 1999, 1998; Leet and Selevan,
11 1982; Pell et al., 1978) cancer incidence or mortality (Tables 4-10 and 4-11). The limitations of each of
12 the aforementioned studies are discussed in this section. Most occupational cohort studies are
13 historical in nature gathering human subject information from existing records and going back many
14 years. In general, the constructed databases do not include detailed information on the workers'
15 individual habits (e.g., tobacco use, alcohol consumption) and usually only have limited exposure
16 information. These limitations often limit the ability to control for bias due to confounding variables
17 and to assess the potential for misclassification of exposure.
18 One of the limitations of the occupational epidemiologic studies examining chloroprene
19 exposure is the potential for the healthy worker effect to influence the results. Since occupational
20 studies involve workers who are healthier than the general population, a reduced mortality risk is often
21 observed among these populations when compared to external populations. This potential bias was
22 likely reduced in some studies by using internal comparisons or other study designs such as a nested
23 case-control study.
24 Another concern in these occupational studies is the reliance on death certificates for outcome
25 ascertainment especially in the mortality studies. Although misclassification of cause of death can be
26 minimized by the review of medical records or by histological confirmation, this was not done in any
27 of the studies. Incomplete enumeration of incident cases was another limitation of several of the
28 studies. This may limit the ability to detect associations as it directly reduces statistical power through
29 reduced sample sizes. Outcome misclassification can also bias the measures of associations that were
30 examined, but it is difficult to gauge the potential impact of this bias on the reported findings.
31 Finally, the lack of quantitative exposure assessment is clearly a limiting factor of most
32 occupational studies; however, they still are able to contribute to the overall qualitative weight of
33 evidence considerations. In many cases where exposure data were missing or insufficient to provide
34 quantitative assessments, exposure levels were differentiated based upon job titles and industrial
35 hygiene knowledge of the processes involved. Although measurement error is present in all studies to
September 2009 4-15 DRAFT - DO NOT CITE OR QUOTE
-------
1 varying degrees, there is no evidence that this error differed by outcome (i.e., was non-differential) in
2 these studies. Although there are rare exceptions, non-differential misclassification of workers'
3 exposures due to lack of information usually results in an underestimate of the association between
4 exposure and outcome.
Table 4-10. Epidemiologic summary results of respiratory system cancers: Overall
standardized mortality ratios (SMRs) and SMRs for intermediate and high
chloroprene exposures relative to external population comparison
STUDY
Bulbulyan et al, 1998
Bulbulyan et al., 1999
Colonna and Laydevant, 2001
Leet and Sullivan, 1982
Marsh et al., 2007a,b-Louisville
Marsh et al., 2007a,b-Maydown
Marsh et al., 2007a,b-Pontchartrain
Marsh et al., 2007a,b-Grenoble
TOTAL COHORT
SMR (95% CI)
140 (90-200)
50 (16-155)
184 (84-349)e
106 (62-170)
75 (66-85)
79 (58-105)
62 (32-109)
85 (41-156)
HIGH
EXPOSUREA
SMR (95% CI)
0.8 (0.3-2.4)c'd
123 (26-361)e
128 (61-236)
65 (50-85)^
113 (60-192)^
85 (23-218)d'e
128 (26-373)^
INTERMEDIATE
EXPOSUREA
SMR (95% CI)
1.0 (0.4-2.5)c'd
125(15^51)e
86 (35-178)b
92 (73-115)d'e
97 (50-169)d'e
96 (12-348)d'e
119 (32-304)^
"Relative to Low or Unexposed Groups
bLow Exposure Group
'Relative Risk of Death from Lung Cancer
dCumulative Chloroprene Exposures
Standardized Incidence Ratios
September 2009
4-16
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-11. Epidemiologic summary results of liver/biliary passage cancers: Total
cohort standardized mortality ratios (SMRs) and relative risk (RRs) for
intermediate and high chloroprene exposures
STUDY
Bulbulyan et al., 1998
Bulbulyan et al, 1999
Colonna and Laydevant, 2001
Leet and Sullivan, 1982
Li et al, 1989
Marsh et al., 2007a,b-Louisville
Marsh et al., 2007a,b-Maydown
Marsh et al., 2007a,b-Pontchartrain
Marsh et al., 2007a,b-Grenoble
TOTAL COHORT
SMRA (95% CI)
240(110-430)
339 (109-1050)
136 (4-763)e
571 (156-1463)
482 (N/R,p< 0.01)
90 (52-144)
24(1-134)
56(1-312)
HIGH EXPOSURE8
RR (95% CI)
4.4 (0.4-44)c'd
4.9 (2.02-11.7)d'e
750 (155-2192/
3.3 (0.5, 39.3)d
INTERMEDIATE
EXPOSURE8
RR (95% CI)
7.1 (0.8-61)c'd
2.9 (0.41-20.8)^
250 (6-1393/
5.1 (0.9, 54.5)d
N/R: Not Reported
"Relative to External Population Rates
bRelative to Low or Unexposed Groups
'Relative Risk of Death from Liver Cancer
dCumulative Chloroprene Exposures
Standardized Incidence Ratio
f Standardized Mortality Ratio
Lung Cancer Summary
1 An increased risk of lung cancer incidence and mortality was observed in a few studies
2 (Colonna and Laydevant, 2001; Bulbulyan et al., 1998; Pell et al., 1978; Li et al., 1989), although few
3 statistically significant associations were reported. None of the studies adjusted for smoking because
4 the investigators either did not have this information available or because the majority of their lung
5 cancer cases were observed in smokers. Marsh et al. (2007b) used white/blue collar as a surrogate for
6 smoking habits assuming that blue collar workers smoked more than white collar workers. But due to
7 small number of deaths in white collar workers the authors reportedly only adjusted the lung cancer
8 risk for worker type in the Louisville, Kentucky plant. Since worker pay type is a crude surrogate of
9 smoking status, it is difficult to rule out the potential confounding effects of smoking. Worker pay
10 status is also a marker of chloroprene exposure. Therefore, inclusion of this variable in regression
11 models may result in over-adjustment distorting the relationship between cancer mortality and
12 chloroprene exposure. A few studies noted higher SMRs for lung cancer among workers exposed to
13 chloroprene; however, there was no evidence of an exposure-response relationship across various
14 chloroprene exposure categories.
September 2009
4-17
DRAFT - DO NOT CITE OR QUOTE
-------
Liver Cancer Summary
1 Statistically significant excesses of liver cancers were detected in four of the cohorts that were
2 examined (Bulbulyan et al. 1999, 1998; Li et al., 1989; Leet and Selevan, 1982). Although no
3 statistically significant increase in risk of liver cancer was detected in the Louisville plant (Marsh et
4 al., 2007b), the relative risk increased with increasing cumulative exposures indicating a dose-response
5 trend. In three of the cohorts, there was only one case of liver cancer or mortality from liver cancer
6 (Marsh et al., 2007a, 2007b and Colonna and Laydevant, 2001) detected, while the Pontchartrain
7 cohort study had no reported liver cancer deaths (Marsh et al., 2007b). This precluded meaningful
8 examination especially in the latter studies with more detailed exposure information.
9 Confounding by occupational co-exposures is addressed in some studies but few of these
10 included direct adjustments for the possible confounders. Some studies have selected workers from
11 several different processes where the co-exposures might have been different or non-existent in some
12 processes to help address the potential for confounding. Bulbulyan et al. (1999, 1998) discussed other
13 possible exposures and concluded that confounding was unlikely, since none of the known co-exposure
14 chemicals were known to be associated with liver cancer. Marsh et al. (2007b) conducted a separate
15 analysis with vinyl chloride in the Louisville plant and found that 15 out of 17 liver cancer cases were
16 found in workers that were not exposed to vinyl chloride. The authors also reported that there was no
17 correlation between cumulative exposures to vinyl chloride and chloroprene among these workers.
18 Given these data, it is highly unlikely that confounding by vinyl chloride could explain the association
19 observed between chloroprene and these cancers. No adjustments for other risk factors for liver
20 cancer, such as alcohol consumption, were performed in any of the cohorts observing statistically
21 significant increases in liver cancer mortality. If alcohol consumption was associated with chloroprene
22 exposure this might be a source of residual confounding. Further limitations in these cohorts include
23 the lack of precise quantitative exposure information, limited statistical power to detect effects due to
24 insufficient general population mortality data, and incomplete ascertainment of health outcomes.
25 Studies that relied upon comparisons to external population mortality rates are also susceptible to the
26 healthy worker effect although the potential impact on cancer mortality in these populations is unclear
27 (see above).
28 Primary liver cancer is relatively rare in the U.S. It accounts for approximately 1.3% of new
29 cancer cases and 2.6% of cancer deaths (Jemal et al., 2003). There are few identified chemicals that
30 have been associated with primary liver cancer. The observation of an increased risk of liver cancer
31 mortality is reasonably consistent and there is some evidence of an exposure-response relationship
32 among workers exposed to chloroprene in different cohorts in different continents (i.e. U.S., China,
33 Russia, and Armenia).
September 2009 4-18 DRAFT - DO NOT CITE OR QUOTE
-------
4.1.2. Chloroprene Exposure and Noncancer Effects
4.1.2.1. Acute-, Short-, and Subchronic-Duration Noncancer Effects
1 Nystrom (1948) reported effects associated with the levels (not specified) of chloroprene
2 exposure experienced during the start-up of chloroprene production in Sweden. The author noted a
3 high level of symptoms among workers in two departments, chloroprene distillation and
4 polymerization, in both the pilot plant and early period of regular production. Over the time period
5 from 1944-1997, the author conducted a series of employee medical examinations. In the distillation
6 department of the production plant, 19 of 21 workers (90%) complained of fatigue and pressure or
7 pains over the chest, with much fewer numbers (3-6 employees) complaining of palpitations,
8 giddiness, irritability, and dermatitis. No workers experienced loss of hair.
9 In the polymerization department of the production plant, temporary hair loss affected 11 of 12
10 workers or 90%. The author attributed this to systemic rather than direct skin exposure (which was
11 carefully controlled). Dermatitis was present in four workers (30%), and all other symptoms evaluated
12 were limited to no more than one worker.
13 Guided by animal studies and reports from other companies, Nystrom (1948) evaluated
14 employees for impaired renal and liver function, basal metabolism, and pulmonary and cardiovascular
15 abnormalities by conducting general body examination, clinical chemistry of the urine and blood, and
16 other tests referred to as "special investigations" (including X-rays, electrocardiograms, and
17 hypoxemia and stress tests). The results of these evaluations were reported in an anecdotal manner
18 with no qualitative or quantitative (e.g., statistical significance of results) details. Except for increased
19 symptoms with exercise right after exposure (among distillation department workers), no clear
20 pathologies were observed. In the pilot plant, where exposures were less controlled, Nystrom (1948)
21 noted anemia among exposed workers. The author also observed that, when the workers were
22 educated about the dangers and safety precautions were enforced, the symptoms decreased.
23 In a Russian review of the effects of chloroprene, Sanotskii (1976) noted that medical
24 examinations of chloroprene production workers had found changes in the nervous system, hepatic and
25 renal function, cardiovascular system, and hematology. Assessment of exposures in Russian latex and
26 rubber manufacturing plants showed that chloroprene was the main hazard and that exposures ranged
27 from 1-7 mg/m3 in exposed work areas.
28 One of the studies reported in this review included medical exams of 12 men and 53 women, of
29 whom two-thirds had been employed in a chloroprene production plant for less than 5 years.
30 Cardiovascular examinations found muffled heart sounds in 30 workers, reduced arterial pressure in
31 14, and tachycardia in 9. There was also a reduction in RBC counts, with hemoglobin substantially
32 below the limit of physiological variation. Erythrocytopenia, leucopenia, and thrombocytopenia were
33 observed. Increases in vestibular function disturbance were associated with duration of work.
September 2009 4-19 DRAFT - DO NOT CITE OR QUOTE
-------
1 In another study reviewed by Sanotskii (1976), women aged 19-23 and employed in jobs with
2 chloroprene exposure for 2-4 years had abnormal diurnal variation in arterial pressure, with reduced
3 systolic and diastolic components at the end of the workday when compared with controls. Their pulse
4 rates were considerably higher than those of controls (p < 0.01). Central nervous system (CNS)
5 function was also affected with lengthening of sensorimotor response to visual cues compared with
6 controls. Olfactory thresholds increased with duration of employment.
4.1.2.2. Chronic Noncancer Effects
7 Chronic effects in exposed workers at an electrical engineering plant were also reported in the
8 review by Sanotskii (1976). When compared to 118 unexposed controls, the chloroprene-exposed
9 cohort (143 workers) exhibited an increased incidence of disturbances of spermatogenesis after 6-10
10 years of work and morphological disturbances after 11 years or more. A questionnaire showed that
11 cases of spontaneous abortion in the wives of chloroprene workers occurred more than three times as
12 frequently as in the control group. This study presents interpretational difficulties concerning the level
13 of participation of the exposed workers and their wives, the quantitative interpretation of the reported
14 sperm abnormalities, and the appropriate matching of exposed and control populations. In an earlier
15 evaluation of this study, U.S. EPA (1985) concluded that recall bias associated with a retrospective
16 questionnaire, such as was used in the study reviewed by Sanotskii (1976), was likely, and the
17 likelihood that the study would have discovered a real increase in the rate of spontaneous abortions
18 was remote, as embryos with chromosomal abnormalities are spontaneously aborted early in
19 pregnancy. Many pregnancies are lost spontaneously often before a woman recognizes that she is
20 pregnant, and the clinical signs of miscarriage are often mistaken for a heavy or late menses (Griebel et
21 al., 2005). Thus, U.S. EPA (1985) concluded that it was not reasonable to draw conclusions on the
22 possible effect of chloroprene on early fetal losses based on the Sanotskii (1976) review. In addition,
23 the EPA suggested that the low participation of male volunteers available for sperm analysis (9.5%
24 participation, 15/143 workers) indicated that a large degree of selection bias may have been present. If
25 males with reproductive deficits self-selected themselves for participation, the meaningful
26 interpretation of the study results may be limited.
27 The final conclusion of the EPA analysis was that it is not possible to interpret the results in the
28 Sanotskii (1976) review with any degree of reliability (U.S. EPA, 1985). Savitz et al. (1994) and
29 Schrag and Dixon (1985) separately reviewed the study and also concluded that insufficient
30 methodological details were available to critically evaluate the observation reported by Sanotskii
31 (1976).
32 Sanotskii (1976) also reported a study of chromosome aberrations in leukocyte culture cells of
33 chloroprene production employees. The occurrence of chromosomal aberrations were significantly
34 higher (p <0.001) in the exposed group compared to the control group, as well as elevated compared to
35 reported levels among healthy persons. Similar results were reported for a different study of two sets
September 2009 4-20 DRAFT - DO NOT CITE OR QUOTE
-------
1 of female employees: (1) 20 women aged 19-23 and exposed to 3-7 mg/m3 (0.83-1.93 ppm)
2 chloroprene for 1-4 years; and (2) 8 women aged 19-50 and exposed to 1-4 mg/m3 (0.28-1.1 ppm) for
3 1-20 years. The results of these two studies are shown in Table 4-12.
Table 4-12. Frequency of chromosomal aberrations in lymphocyte culture cells
from chloroprene production workers
CHLOROPRENE
EXPOSURE
Chloroprene
Workers
Control
1-4 mg/m3
3-7 mg/m3
Population
Control
# EXAMINED
18
9
8
20
181
YEARS
EXPOSED
1-20
1-4
—
AGE
RANGE
19-50
19-23
—
# CELLS
ANALYZED
1,666
572
648
1,748
28,386
PERCENT
ABERRANT (+/-)
4.77 (0.57)a
0.65 (0.56)
2.5 (0.49)b
3.49(0.51)a
1.19(0.06)
PERCENT TYPE
ABERRANT
Chromatid
74.4
100
—
—
50.3
Chromosome
25.6
0
—
—
49.7
a p < 0.001. All values means ± SE
bp<0.05
Source: Sanotskii (1976).
4.2. SUBCHRONIC AND CHRONIC STUDIES AND CANCER BIOASSAYS IN ANIMALS-
ORAL AND INHALATION
4.2.1. Oral Exposure
4 The only available long-term animal study using the oral route of administration was part of a
5 developmental/reproductive study. Ponomarkov and Tomatis (1980) administered chloroprene
6 dissolved in olive oil by stomach tube to 17 female BD IV rats at a single dose (100 mg/kg body
7 weight) on gestational day (GD) 17. Progeny from treated females (81 males and 64 females) were
8 treated weekly with 50 mg/kg body weight by stomach tube from the time of weaning for life (120
9 weeks). A control group of 14 female rats was treated with 0.3 mL olive oil. The purity of the
10 chloroprene was reported as 99% with 0.8% 1-chlorobutadiene; storage conditions were not reported.
11 All survivors were sacrificed at 120 weeks or when moribund and autopsied. Major organs, as well as
12 those that showed gross abnormalities, were examined histologically.
13 Litter sizes and preweaning mortality, survival rates, and body weights did not differ between
14 chloroprene-treated animals and controls. Animals treated with chloroprene that died within the first
15 23-35 weeks of treatment showed severe congestion of the lungs and kidneys. Some animals (number
16 not specified) autopsied 80-90 weeks after the onset of treatment showed multiple liver necroses.
17 Animals that died after 90 weeks and some survivors euthanized at 120 weeks showed degenerative
18 lesions of the liver parenchymal cells (large cells with clear cytoplasm in nodular arrangements).
September 2009
4-21
DRAFT - DO NOT CITE OR QUOTE
-------
1 Tumor incidences and distribution reported in this study are summarized in Tables 4-13 and 4-
2 14. No statistically significant differences were reported between treated and control rats. However,
3 several tumors observed in males (intestinal leiomyosarcoma, osteoma, kidney mesenchymal tumor,
4 bone hemangioma, neurinoma of the optic nerve, transition-cell carcinoma of urinary bladder, and
5 forestomach papilloma.) and females (ovarian and mammary tumors) treated weekly with chloroprene
6 were not seen in the vehicle control group. Subcutaneous fibromas were more numerous in
7 chloroprene-treated male rats than in controls.
Table 4-13. Tumor incidence in female BD IV rats treated orally with chloroprene
(100 mg/kg) on GDI? and in their progeny treated (50 mg/kg) weekly for life
(120 weeks)
GROUP
Treated females
Treated progeny
Males
Females
Control females
Control progeny
Males
Females
NUMBER3
16
54
62
14
49
47
TUMOR BEARING
RATS
n
9
15
33
5
16
24
%
56.2
27.8
53.2
35.7
32.7
51.1
NUMBER OF
TUMORS
Total
14
18
37
7
16
29
Per rat
0.9
0.3
0.6
0.5
0.3
0.6
ANIMALS WITH
MORE THAN ONE
TUMOR
n
5
3
4
2
—
5
%
31.3
5.6
6.5
14.3
—
10.6
aSurvivors at the time the first tumors were observed.
Source: Ponomarkov and Tomatis (1980).
September 2009
4-22
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-14. Distribution of tumors in female BD IV rats treated orally with chloroprene
(100 mg/kg) on GDI? and their progeny treated (50mg/kg) weekly for life (120 weeks)
GROUP
Treated
Treated
Males
Females
Control
Control
Males
Females
ORAL
CAVITY
n
1
—
—
1
2
1
%
6.3
—
—
7.1
4.1
2.1
MAMMARY
n
6
—
25
4
—
22
%
37.5
—
40.3
28.6
—
46.8
OVARY
n
2
—
9
—
o
6
%
12.5
—
14.5
—
6.4
THYROID
n
1
1
—
—
%
1.9
1.6
—
—
SOFT
TISSUE
n
7
—
1
4
—
%
13.0
—
7.1
8.2
—
PITUITARY
n
1
2
2
—
2
1
%
6.3
3.7
3.2
—
4.1
—
OTHER
n
4a
8b
—
1°
8d
3e
%
25.0
14.8
—
7.1
16.3
6.4
a 1 each: uterine squamous cell carcinoma; lung reticulosarcoma; forestomach papilloma; sebaceous basal cell carcinoma.
b leach: intestinal leiomyosarcoma; osteoma; kidney mesenchymal tumor; bone hemangioma; neurinoma of the optic nerve;
adrenal cortical adenoma; transition-cell carcinoma of urinary bladder; forestomach papilloma.
0 Adrenal cortical adenoma.
d2 lymphomas; 1 each: lung epidermoid carcinoma; spleen hemangioma; osteosarcoma; mediastinal sarcoma; meningioma;
adrenal cortical adenoma.
e 1 each: stomach fibrosarcoma; lymphoma; uterine adenoma.
Source: Ponomarkov and Tomatis (1980).
4.2.2. Inhalation Exposure
1 The NTP conducted 16-day, 13-week, and 2-year inhalation exposure studies with chloroprene
2 in F344/N rats and B6C3F1 mice (NTP, 1998). Results of the 13-week study were reported by
3 Melnick et al. (1996), while the cancer results of the 2-year study were discussed separately by
4 Melnick et al. (1999) in relation to observations noted with 1,3-butadiene in mice. All experimental
5 regimes consisted of 6 hours per day, 5 days per week whole-body exposures. Group sizes were 10
6 animals/sex/group in the 16-day and 13-week studies and 50 animals/sex/group in the 2-year study.
7 Overall purity of the bulk chloroprene was determined to be approximately 96% by gas
8 chromatography. Vapor was generated in the 13-week and 2-year studies from chloroprene in an
9 evaporation flask kept at 66°C (72°C in the 16-day studies) followed by a temperature-controlled
10 condenser column (to remove less volatile impurities such as chloroprene dimers); the chloroprene
11 reservoir was kept at dry ice temperature (16-day study) or under nitrogen (13-week and 2-year
12 studies). The actual concentrations generated from the evaporator flask were within 99% of target
13 concentrations at the beginning of the exposures and were 95% pure at the end of the exposure period.
14 Chloroprene was dragged from the evaporator by a metered flow of nitrogen before being injected
15 into the mixer column, where it was diluted with HEPA- and charcoal-filtered air. Impurities more
16 volatile than chloroprene, such as chlorobutene, never exceeded more than 0.6% of the desired
17 chloroprene concentration when sampled from the distribution line, the last sampling point upstream
18 from the actual exposure chambers. Histopathology was performed by a study pathologist and
19 reviewed by a quality assurance pathologist and the Pathology Working Group.
September 2009
4-23
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
In the 16-day study, rats were exposed to 0, 32, 80, 200, or 500 ppm chloroprene (NTP, 1998).
On day 4, rats were placed in metabolism cages for 16-hour urine collection. A necropsy was
performed on all animals, and histopathological examinations were performed on controls, 80 ppm
female rats, and 200 and 500 ppm male and female rats. Tissues and organs examined included brain,
liver, kidney, lung, bone marrow, thymus, spleen, and testes. Sperm morphology and vaginal cytology
were not evaluated.
Survival and body weights of rats are given in Table 4-15. Only one male in the high-exposure
group (500 ppm) survived. Females in the high-exposure group had a higher survival (7/10) with a
significantly decreased body weight (-6% compared with controls). Significantly decreased body
weight gain was also observed in males and females at 200 ppm, and in females at 500 ppm.
Table 4-15. Survival and body weights of rats in the 16-day inhalation study of
chloroprene
SEX
Male
Female
EXPOSURE (ppm)
0
32
80
200
500
0
32
80
200
500
SURVIVAL
7/10
10/10
10/10
9/10
1/10
9/10
9/10
9/10
3/10
7/10
MEAN BODY WEIGHT (g)
Initial
115±4
113±4
118±5
114 ±4
114 ±4
100 ±2
100 ±2
103 ±2
101 ±2
102 ±2
Final
139 ±5
134 ±6
136 ±5
127 ±5
104
110 ±3
109 ±3
112 ±2
101 ±4
103 ±3
Change
+ 20 ±2
+ 20 ±2
+ 18+.1
+ 11 ±2**
4a
+ 9±1
+ 8± 1
+ 9±1
+ 4± 1**
-1±1**
11
12
13
14
15
16
17
18
a No standard error calculated due to high mortality
" Significantly different (p ^0.01) from the chamber control group by Williams' or Dunnett's test
Source: NTP (1998)
The incidences of minimal to mild olfactory epithelial degeneration in all exposed groups of
males and females were significantly greater than those in the chamber control groups (Table 4-16).
Mild to moderate centrilobular hepatocellular necrosis was observed in male and female rats exposed
to 200 or 500 ppm. Hematological and clinical chemistry parameters indicated increased serum
alanine aminotransaminase (ALT), glutamine dehydrogenase (GDH), and sorbitol dehydrogenase
(SDH) activities, as well as anemia and thrombocytopenia (decreased platelet count) in the 200 and
500 ppm groups, on day 4 only. In females, significant increases in kidney weights (right kidney only)
were seen at 80 and 500 ppm, and significantly increased liver weights were seen at 200 and 500 ppm.
September 2009
4-24
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-16. Incidences of selected nonneoplastic lesions in rats in the 16-day
inhalation study of chloroprene
CONTROL
32ppm
80 ppm
200 ppm
500 ppm
Male
Nose3
Degeneration, olfactory epithelium
Metaplasia, squamous, olfactory
epithelium
Metaplasia, respiratory, olfactory
epithelium
Metaplasia, squamous, olfactory
epithelium
Liver3
Necrosis, centrilobular
Inflammation, chronic
10/10
1/10
0/10
0/10
1/10
10/10
0/10
0/10
10/10
10/10C
0/10
2/10
1/10
1/10
0/10
0/10
10/10
10/10 c
0/10
5/10b
0/10
10/10
0/10
0/10
10/10
10/10 c
1/10
6/10 b
0/10
10/10
1/10
0/10
10/10
10/10 c
4/10 b
1/10
7/10
10/10
9/10 c
1/10
Female
Nose3
Degeneration, olfactory epithelium
Metaplasia, squamous, olfactory
epithelium
Metaplasia, respiratory, olfactory
epithelium
Metaplasia, squamous, olfactory
epithelium
Liver3
Necrosis, centrilobular
Inflammation, chronic
10/10
0/10
0/10
0/10
1/10
10/10
0/10
0/10
10/10
9/10 c
1/10
7/10 c
1/10
3/10
0/10
0/10
10/10
10/10 c
1/10
8/10 c
0/10
10/10
0/10
0/10
10/10
10/10 c
4/10 b
3/10
0/10
10/10
7/10 c
2/10
10/10
10/10 c
0/10
7/10 c
4/10
10/10
3/10
5/10 b
3 Number of animals with tissue examined microscopically.
bp<0.05.
0 Significantly different (p <_0.01) from the chamber control group by the Fisher's exact test.
Source: NTP(1998)
1 In the mouse portion of the 16-day NTP (1998) study, exposure levels were 0, 12, 32, 80, and
2 200 ppm. Additional groups of 10 male and 10 female mice designated for day 5 hematology and
3 clinical chemistry analyses were exposed to the same chloroprene concentrations. Histopathology
4 examinations were performed on chamber controls and 80 and 200 male and female mice as well as on
5 selected target organs in other groups. Tissues and organs examined were identical to those described
6 for the rat. Survival and body weights for mice are given in Table 4-17. All male and female animals
7 in the high-concentration group died, exhibiting signs of narcosis, hepatocellular and thymic necrosis,
8 and hypertrophy of the myocardium. Significantly decreased body weight gain (compared with
9 controls) was seen in males at 32 and 80 ppm. There were no other clinical findings related to
10 chloroprene exposure in the mouse.
September 2009
4-25
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-17. Survival and body weights of mice in the 16-day inhalation study of
chloroprene
EXPOSURE (ppm)
SURVIVAL
MEAN BODY WEIGHT (g)
Initial
Final
Change
Male
0
12
32
80
200
10/10
10/10
10/10
10/10
0/10
24.7 ±0.5
24.8 ±0.5
25.3 ±0.3
24.8 ±0.5
24.2 ±0.4
27.0 ±0.5
27.1 ±0.6
26.5 ±0.3
26.1 ±0.6
—
+ 2.3 ±0.1
+ 2.3 ±0.3
+ 1.2±0.3a
+ 1.3+0.23
—
Female
0
12
32
80
200
10/10
10/10
10/10
10/10
0/10
19.5 ±0.7
20.4 ±0.8
19.9 ±1.0
20.1 ±0.8
20.0 ±0.6
22.6 ±0.5
23.1 ±0.4
22.1 ±0.2
22.5 ±0.3
—
+ 2.3±0.3
+ 2.6 + 0.3
+ 1.8 + 0.3
+ 2.7 + 0.3
—
"Significantly different (p < 0.01) from the chamber control group by Williams' or Dunnett's test.
Source: NTP(1998)
1 A range-finding 13-week inhalation study was conducted by NTP (1998) (reported by Melnick
2 et al., 1996), using both mice and rats. In the rat, exposure groups were 0, 5, 12, 32, 80, and 200 ppm.
3 Separate groups of 10 male and 10 female rats designated for coagulation studies were exposed to
4 these concentrations for 2 days. Rats designated for hematology and clinical chemistry tests were first
5 placed in metabolism cages for 16-hour urine collections. Sperm samples were collected from male
6 rats at the end of the studies. Samples of vaginal fluid and cells were collected for up to 7 consecutive
7 days prior to the end of the studies for cytology evaluations. Five male and five female rats were
8 exposed to 0, 5, 32 or 200 ppm for glutathione evaluations. At week 11, all male and female core
9 study rats were administered neurobehavioral tests measuring the following parameters:
10 forelimb/hind-limb grip strength, horizontal activity, rearing activity, total activity, tail-flick latency,
11 startle response latency, and startle response amplitude. Survival and body weights of rats are given in
12 Table 4-18. No effects on final mean body weights were seen.
Table 4-18. Survival and body weights of rats in the 13-week inhalation study of
chloroprene
EXPOSURE (ppm)
SURVIVAL
MEAN BODY WEIGHT (g)
Initial
Final
Change
Male
0
5
12
32
80
200
10/10
10/10
10/10
10/10
10/10
9/10
109 + 4
119 + 2*
116+1
117 + 2
116+1
116 + 3
311 + 9
323 ± 11
306 + 9
327+11
301 + 8
304 + 8
+ 202 + 8
+ 204 ± 10
+ 190 + 8
+ 209 ± 10
+ 184 + 7
+ 185 + 7
September 2009
4-26
DRAFT - DO NOT CITE OR QUOTE
-------
EXPOSURE (ppm)
SURVIVAL
MEAN BODY WEIGHT (g)
Initial
Final
Change
Female
0
5
12
32
80
200
10/10
10/10
10/10
10/10
10/10
10/10
102 ±2
101 ±1
102 ±2
101 ±2
103 ±1
102 ±1
191 ±4
193 ±4
199 ±5
195 ±4
192 ±3
183 ±3
+ 89 ±3
+ 92 ±3
+ 97 ±4
+ 94 ±4
+ 90 ±3
+ 81 ±3
Significantly different (p < 0.05) from the chamber control group by Williams' or Dunnett's test.
Source: NTP(1998)
1 On day 2, minimal increases in hematocrit values, hemoglobin concentrations, and erythrocyte
2 counts occurred in males exposed to > 32 ppm and in females exposed to 200 ppm. At week 13, male
3 and female rats in the 200 ppm groups demonstrated decreased hematocrit values, hemoglobin
4 concentrations, and erythrocyte counts characterized as normocytic, normochromic anemia.
5 Thrombocytopenia, evidenced by a reduction in circulating platelet numbers, occurred in the male and
6 female rats in the 200 ppm groups on day 2 and in the females at 80 and 200 ppm on day 22. Platelet
7 numbers rebounded at study termination in the highest exposure groups for both male and female rats.
8 Activities of serum ALT, GDH, and SDH were elevated on day 22 in both sexes of the 200 ppm group.
9 However, these increases were transient, and serum activities of the enzyme levels returned to control
10 levels by the end of the exposure period. At week 13, an alkaline phosphatase (ALP) enzymeuria
11 occurred in males exposed to > 32 ppm and in females exposed to 200 ppm. In male rats in the 200
12 ppm group, proteinuria was seen at week 13. Significant reductions in nonprotein sulfhydryl (NPSH)
13 concentrations were observed in the livers from male rats exposed to 200 ppm for 1 day or 12 weeks,
14 as well as in female rats exposed to 200 ppm for 12 weeks. Nonprotein sulfhydryl concentrations were
15 reduced in the lung of 200 ppm female rats after 1 day but not after 12 weeks of exposure to 200 ppm.
16 Significant increases in kidney weights were seen in both male and female rats at 200 ppm and in
17 females at 80 ppm. In male rats exposed to 200 ppm, sperm motility was significantly less than that of
18 the chamber control group. Of the neurobehavioral parameters, horizontal activity was increased in
19 male rats exposed to > 32 ppm compared with chamber control animals. Total activity was increased
20 in male rats in the 32 and 200 ppm groups. There were no exposure-related effects on motor activity,
21 forelimb/hind-limb grip strength, or startle response.
22 Increased incidences of minimal to mild olfactory epithelial degeneration and respiratory
23 metaplasia occurred in male and female rats exposed to 80 or 200 ppm (Table 4-19). The incidence of
24 olfactory epithelial degeneration in females exposed to 32 ppm was significantly greater than in the
25 chamber control group. In female rats exposed to 200 ppm, the incidence of hepatocellular necrosis
26 was significantly greater than in the chamber control group. Variably sized aggregates of yellow or
27 brown material consistent with hemosiderin appeared in small vessels or lymphatics in or near portal
28 triads or in Kupffer cells of male and female rats exposed to 200 ppm and were significantly increased
29 compared with chamber controls.
September 2009 4-27 DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-19. Incidences of selected nonneoplastic lesions in rats in the 13-week
inhalation study of chloroprene
CONTROL
5 ppm
12 ppm
32 ppm
80 ppm
200 ppm
Male
Nose3
Degeneration, olfactory
epithelium
Metaplasia, respiratory,
olfactory epithelium
Liver3
Necrosis, centrilobular
Inflammation, chronic
Hemosiderin pigmentation
10/10
0/10
0/10
10/10
0/10
0/10
0/10
0/10
—
—
2/10
0/10
1/10
0/10
10/10
0/10
0/10
1/10
0/10
0/10
0/10
10/10
3/10
0/10
1/10
01/10
0/10
0/10
10/10
10/10 c
4/10 b
10/10
0/10
1/10
0/10
10/10
10/10 c
4/10 b
10/10
3/10
2/10
5/10 b
Female
Nose3
Degeneration, olfactory
epithelium
Metaplasia, respiratory,
olfactory epithelium
Liver3
Necrosis, centrilobular
Inflammation, chronic
Hemosiderin pigmentation
10/10
0/10
0/10
10/10
0/10
2/10
3/10
0/10
—
—
2/10
0/10
0/10
0/10
10/10
0/10
0/10
5/10
0/10
1/10
1/10
10/10
4/10 b
0/10
3/10
0/10
0/10
0/10
10/10
9/10 c
8/10 c
10/10
0/10
1/10
0/10
10/10
10/10 c
9/10 c
10/10
5/10 b
8/10 b
9/10 c
3 Number of animals with tissue examined microscopically.
b Significantly different ( p < 0.05) the chamber control group by Fisher's exact test.
0 Significantly different (p < 0.01) from the chamber control group by Fisher's exact test.
Source: NTP(1998)
1 In the mouse portion of the NTP 13-week inhalation study, exposure groups were 0, 5, 12, 32,
2 and 80 ppm. Survival and body weights are given in Table 4-20. There was no increased mortality in
3 any exposure group. Final mean body weights in 80 ppm males were significantly decreased
4 compared with controls.
Table 4-20. Survival and body weights of mice in the 13-week inhalation study of
chloroprene
SEX
-------
1
2
3
4
5
6
7
Hematology variables were similar to, although more mild than, in the 13-week rat study. A
minimal anemia, including decreased hematocrit values and erythrocyte counts, occurred in female
mice exposed to 32 and 80 ppm. Platelet counts were minimally increased in female mice exposed to
32 and 80 ppm, suggesting an increase in platelet production. No significant organ weight effects were
observed. Sperm morphology and vaginal cytology parameters were similar to those of the chamber
controls. Significantly increased incidences of squamous epithelial hyperplasia of the forestomach
were observed in male and female mice exposed to 80 ppm (Table 4-21).
Table 4-21. Incidences of forestomach lesions in mice in the 13-week inhalation
study of chloroprene
CONTROL
5 ppm
12 ppm
32 ppm
80 ppm
Male
Number examined microscopically
Squamous epithelial hyperplasia
10/10
0/10
3/10
0/10
0/10
—
10/10
0/10
10/10
4/10 a
Female
Number examined microscopically
Squamous epithelial hyperplasia
10/10
0/10
0/10
—
0/10
—
10/10
0/10
10/10
9/10 b
9
10
11
a Significantly different (p < 0.05) from the chamber control group by Fisher's exact test.
bp<0.01.
Source: NTP(1998)
In the 2-year NTP (1998) inhalation study of chloroprene in male and female rats, groups were
exposed to 0, 12.8, 32, and 80 ppm chloroprene. Estimates of 2-year survival probabilities are shown
in Table 4-22. Survival of males exposed to 32 or 80 ppm was significantly less than that of the
chamber control group.
September 2009
4-29
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-22. 2-Year survival probability estimates for F344/N rats chronically
exposed (2 years) to chloroprene by inhalation
SEX
-------
80 ppm chloroprene, and was characterized by focal thickening and folding of the squamous
epithelium.
Table 4-23. Incidence and severity of non-neoplastic lesions in F344/N rats
chronically exposed (2 years) to chloroprene by inhalation
TISSUE SITE/LESION TYPE
Oral cavity
Sqamous Cell Hyperplasia
Thyroid gland
Follicular Cell Hyperplasia
Lung
Alveolar Hyperplasia
Kidney (renal tubules)
Hyperplasia
Olfactory
Atrophyd
Basal Cell Hyperplasia
Metaplasia
Necrosis6
Chronic Inflammation
LESION INCIDENCE (SEVERITY)
Males (ppm)
0
0/50
0/50
5/50
(1.4)
14/50
(2.0)
3/50
(1.7)
0/50
6/50
1.7
0/50
0/50
12.8
0/50
2/50
(2.0)
16/50C
(1.4)
20/50
(2.6)
12/50b
(1.8)
0/50
5/50
(1.0)
ll/50b
(2.0)
5/50c
(1.0)
32
0/50
4/49b
(1.8)
14/49b
(1.9)
28/50c
(2.1)
46/49c
(2.2)
38/49c
(1.6)
45/49c
(1.8)
26/49c
(2.0)
9/49-
(1.6)
80
3/50
(2.7)a
1/50
(1.0)
25/50c
(1.4)
34/50c
(2.9)
48/49c
(3.6)
46/49c
(2.2)
48/49c
(3.1)
19/49C
(2.2)
49/49°
(2.7)
Females (ppm)
0
~
0/49
6/49
(1.8)
6/49
(1.3)
0/49
0/49
0/49
0/49
0/49
12.8
~
0/50
22/50c
1.4)
6/50
(1.8)
1/50
(1.0)
0/50
1/50
(1.0)
0/50
0/50
32
~
0/50
22/50c
(1.5)
11/50
(2.1)
40/50C
(1.3)
17/50C
(1.1)
35/50c
(1.0)
8/50c
(2.0)
2/50
(1.0)
80
-
2/50
(2.5)
34/50c
(1.3)
21/50C
(2.0)
50/50C
(2.9)
49/50c
(2.3)
50/50C
(2.7)
12/50C
(1.3)
33/50c
(2.0)
3
4
5
6
7
a Severity of lesions graded as: 1= minimal, 2 = mild, 3 = moderate, 4 = marked, average severity reported in parenthesis
bp < 0.05, ps correspond to the pairwise comparisons between the chamber controls and that exposed group. The
logistic regression test regards lesions in animals dying prior to terminal kill as nonfatal
cp<0.01
d Severity of atrophic lesions: Males: control -1 minimal, 2 mild; 12.8 ppm-6 minimal, 3 mild, 3 moderate; 32 ppm-
10 minimal, 19 mild, 15 moderate, 2 marked; 80 ppm- 1 mild, 18 moderate, 29 marked. Females: control -all 0; 12.8
ppm - 1 minimal; 32 ppm - 31 minimal, 7 mild, 2 moderate; 80 ppm - 10 mild, 33 moderate, 7 marked.
e Severity of necrotic lesions: Males: control -all 0, 12.8 ppm- 5 minimal, 1 mild, 5 moderate; 32 ppm- 8 minimal, 10
mild, 8 moderate; 80 ppm - 6 minimal, 4 mild, 8 moderate, 1 marked. Females: control -all 0; 12.8 ppm - all 0; 32
ppm - 3 minimal, 2 mild, 3 moderate; 80 ppm - 8 minimal, 4 mild.
Source: NTP(1998)
The incidences of thyroid gland follicular cell adenoma or carcinoma (combined) in male rats
exposed to 32 or 80 ppm were significantly greater than those in the chamber control group and
exceeded historical control ranges. The incidences of follicular cell adenoma and follicular cell
adenoma or carcinoma combined in female rats exposed to 80 ppm were increased but not significantly
greater than those in the chamber controls, although they did exceed the historical control range.
September 2009
4-31
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
Follicular cell carcinomas destroyed the thyroid gland and occasionally invaded the capsule or
adjacent structures. The incidence of follicular cell hyperplasia was significantly increased in male
rats exposed to 32 ppm. Hyperplasia was characterized by one or a few enlarged follicles with several
much smaller follicles inside and to one side.
Table 4-24. Incidence of neoplasms in F344/N rats chronically exposed (2 years) to
chloroprene by inhalation
TISSUE SITE/TUMOR TYPE
Oral cavity
Papillomas or carcinomas
Thyroid gland
Adenomas or carcinomas
Lung
Adenomas or carcinomas0
Kidney (renal tubules)
Adenomas or carcinomas (extended and
standard evaluations combined)
Mammary gland
Fibroadenomas
TUMOR INCIDENCE
Males (ppm)
0
0/50
0/50
2/50
1/50
12.8
2/50
2/50
2/50
8/50a
32
5/50a
4/49a
4/49
6/50b
80
12/50b
5/50a
6/50
8/50b
Females (ppm)
0
1/49
1/49
1/49
0/49
24/49
12.8
3/50
1/50
0/50
0/50
32/50
32
5/50
1/50
0/50
0/50
36/50a
80
ll/50b
5/50
3/50
4/50
36/50b
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
ap < 0.05, ps correspond to the pairwise comparisons between the chamber controls and that exposed group. The
logistic regression test regards lesions in animals dying prior to terminal kill as nonfatal
bp<0.01
0 Adenomas only in females
Source: NTP(1998)
The incidences of alveolar/bronchiolar carcinoma and alveolar/bronchiolar adenoma or
carcinoma (combined) in males exposed to 80 ppm were slightly greater than those in the chamber
control group. Although the increase in neoplasms was not statistically significantly increased relative
to control, the incidences exceeded the historical control range. The incidence of alveolar/bronchiolar
adenoma only was increased, though not significantly, in female rats exposed to 80 ppm chloroprene.
Alveolar/bronchiolar carcinomas were solid or papillary, obliterated normal pulmonary structure, and
sometimes invaded the pleura and other adjacent areas. The incidences of alveolar epithelial
hyperplasia (AEH) were significantly greater in all exposed groups of males and females compared
with the chamber control groups.
Renal tubule adenoma and hyperplasia were observed in male and female rats. Because renal
tubule neoplasms are rare in chamber control F344/N rats, additional kidney sections from male and
female control and exposed groups were examined to provide a clearer indication of the potential
effects of chloroprene on the kidney. The combined single- and step-section incidences of renal tubule
hyperplasia in males exposed to 32 and 80 ppm and in females exposed to 80 ppm and the incidences
of adenoma and adenoma or carcinoma combined in all exposed males were significantly greater than
those in the chamber controls.
September 2009
4-32
DRAFT - DO NOT CITE OR QUOTE
-------
1 The incidences of multiple fibroadenoma of the mammary gland in all exposed groups of
2 female rats were greater than in the chamber control group. The incidences of fibroadenoma in
3 females exposed to 32 and 80 ppm were significantly greater than in the chamber control group.
4 However, the incidences of fibroadenomas in all exposed females and the chamber control exceeded
5 the historical control range.
6 A slight increase in the incidence of transitional epithelium carcinoma of the urinary bladder
7 was observed in females exposed at 80 ppm. In addition, one male exposed at 32 ppm had a
8 transitional epithelium carcinoma and one male exposed at 80 ppm had a transitional cell papilloma.
9 No urinary bladder neoplasms have been observed historically in chamber control male or female
10 F344/N rats.
11 The incidences of atrophy, basal cell hyperplasia, metaplasia, and necrosis of the olfactory
12 epithelium in males and females exposed to 32 and 80 ppm and of atrophy and necrosis in males
13 exposed to 12.8 ppm were significantly greater than those in the chamber control groups. The
14 incidences of chronic inflammation were significantly increased in males exposed to 12.8 or 32 ppm
15 and in females exposed to 80 ppm. The incidences of fibrosis and adenomatous hyperplasia in males
16 and females exposed to 80 ppm were significantly greater than those in the chamber controls. Lesions
17 of the nasal cavity were generally minimal to moderate in average severity. Necrosis of the olfactory
18 epithelium was characterized by areas of karyorrhexis and sloughing of olfactory epithelium with cell
19 debris in the lumen of the dorsal meatus. Atrophy of the olfactory epithelium was characterized by
20 decreased numbers of layers of olfactory epithelium and included loss of Bowman's glands and
21 olfactory axons in more severe cases. Metaplasia was characterized by replacement of olfactory
22 epithelium with ciliated, columnar, respiratory-like epithelium. Basal cell hyperplasia was
23 characterized by proliferation or increased thickness of the basal cell layer in the turbinate and septum.
24 In the NTP 2-year mouse study, exposure concentrations were 0, 12.8, 32, and 80 ppm. All
25 animals were observed twice daily and body weights were recorded initially, weekly through week 12,
26 approximately every 4 weeks from week 15 through week 91, and every 2 weeks until the end of the
27 study. Clinical findings were recorded initially, at weeks 4, 5, 8, 12, every 4 weeks through week 91,
28 and every 2 weeks until the end of the study. A complete necropsy and microscopic examination were
29 performed on all mice as described for the rat portion of the 2-year study. Estimates of 2-year survival
30 probabilities are shown in Table 4-25.
September 2009 4-33 DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-25. 2-Year survival probabilities for B6C3F1 mice chronically exposed (2
years) to chloroprene by inhalation
-------
Table 4-26. Incidence and severity of non-neoplastic lesions in B6C3F1 mice
chronically exposed (2 years) to chloroprene by inhalation
TISSUE SITE/LESION TYPE
Lung
Bronchiolar Hyperplasia
Histiocytic Cell Infiltration
Kidney (renal tubule)
Hyperplasia
Mammary Gland
Hyperplasia
Forestomach
Epithelial Hyperplasia
Olfactory
Suppurative Inflammation
Atrophy
Metaplasia
Spleen
Hematopoietic Proliferation
LESION INCIDENCE (SEVERITY)
Males (ppm)
0
0/50
7/50
(1.6)
0/50
-
4/50
(3.0)
2/50
(2.0)
7/50
(1.1)
6/50
(1.0)
26/50
12.8
10/50C
(2.0)
8/50
(3.3)
4/49
(1.3)
-
6/48
(1.8)
1/48
(1.0)
8/48
(1.4)
5/50
(1.4)
22/49
32
18/50C
(1.7)
11/50
(2.5)
5/50b
(1.2)
-
7/49
(2.3)
4/50
(1.0)
7/50
(1.1)
5/50
(1.0)
35/50d
80
23/50c
(2.2)
22/50c
(2.9)
5/50b
(1.4)
-
29/50c
(2.2)
6/50
(1.5)
49/50c
(2.5)
49/50c
(2.5)
31/50d
Females (ppm)
0
0/50
1/50
(3.0)
-
0/49
4/50
(2.0)
0/50
6/50
(1.2)
2/50
(1.0)
13/50
12.8
15/49C
(2.0)
14/49C
(2.0)
-
1/49
(1.0)
3/49
(3.7)
1/49
(1.0)
5/49
(1.2)
3/49
91.0)
25/49d
32
12/50C
(2.2)
18/50C
(2.3)
-
1/50
(1.0)
8/49
(1.6)
3/49b
(1.7)
4/49
(1.3)
1/49
92.0)
42/49d
80
30/50C
(2.2)
23/23c
(2.4)
-
3/50
(2.0)
27/50c
(2.7)
4/50c
(1.5)
47/50c
(2.0)
44/50c
(2.0)
39/50d
a Severity of lesions graded as: 1= minimal, 2 = mild, 3 = moderate, 4 = marked, average severity reported in
parenthesis, average severity not reported for splenic hematopoietic proliferation
bp < 0.05, ps correspond to the pairwise comparisons between the chamber controls and that exposed group. The
logistic regression test regards lesions in animals dying prior to terminal kill as nonfatal
cp<0.01
d Significantly increased relative to controls, level of significance not reported
Source: NTP(1998)
1
2 The incidences of olfactory epithelial atrophy, adenomatous hyperplasia, and metaplasia in
3 males and females exposed to 80 ppm were significantly greater than those in the chamber controls.
4 The incidence of suppurative inflammation in females exposed to 32 and 80 ppm was significantly
5 greater than controls. Atrophy and metaplasia of the olfactory epithelium was similar to lesions
6 observed in rats exposed to chloroprene. Adenomas of the respiratory epithelium were present in one
7 female exposed to 32 ppm and one male exposed to 80 ppm.
8 In male mice, a pattern of nonneoplastic liver lesions along with silver-staining helical
9 organisms within the liver was observed, consistent with Helicobacter hepaticus infection.
10 Polymerase chain reaction-restriction fragment length polymorphism based assay confirmed an
September 2009 4-35 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
organism compatible with H. hepaticus. Historically, NTP studies with H. hepaticus associated
hepatitis showed increased incidences of hemangiosarcoma in male mice. Therefore,
hemangiosarcomas of the liver were excluded from the analyses of circulatory neoplasms in the males
in the chloroprene 2-year study. However, even with this exclusion, the combined occurrence of
hemangioma or hemangiosarcoma at other sites was significantly increased in all males exposed to
chloroprene and in females exposed to 32 ppm. The incidences of neoplasms at other sites were not
considered to have been significantly impacted by the infection with H. hepaticus or its associated
hepatitis. The incidences of hepatocellular carcinoma in all exposed female mice and hepatocellular
adenoma or carcinoma combined in females exposed to 32 and 80 ppm were significantly greater than
in the chamber control
Table 4-27. Incidence of neoplasms in B6C3F1 mice chronically exposed (2 years)
to chloroprene by inhalation
TISSUE SITE/TUMOR TYPE
Lung
Adenomas or carcinomas
All Organs
Hemangiomas or hemangiosarcomas
Harderian gland
Adenomas or carcinomas
Kidney (renal tubules)
Adenomas or carcinomas (extended and
standard evaluations combined)
Mammary gland
Carcinomas
Forestomach
Papillomas or carcinomas
Liver
Adenomas or carcinomas
Skin
Sarcoma
Mesentery
Sarcomas
Zymbal's gland
Carcinomas
TUMOR INCIDENCE
Males (ppm)
0
13/50
3/50
2/50
0/50
1/50
12.8
28/50c
14/50b
5/50
2/49
0/50
32
36/50c
23/50c
10/503
3/50a
2/50
80
43/50c
21/50C
12/50b
9/50b
4/50
Females (ppm)
0
4/50
4/50
2/50
3/50
1/50
20/50
0/50
0/50
0/50
12.8
28/49c
6/50
5/50
4/50
0/50
26/49
ll/50b
4/50
0/50
32
34/50c
18/50b
3/50
7/50
0/50
20/503
11/50C
8/50b
0/50
80
42/50c
8/50
9/50a
12/503
4/50
30/50C
18/50C
3/50
3/50
11
12
13
14
ap < 0.05, ps correspond to the pairwise comparisons between the chamber controls and that exposed group. The logistic
regression test regards lesions in animals dying prior to terminal kill as nonfatal
bp<0.01
cp< 0.001
Source: NTP (1998)
The incidences of Harderian gland adenoma and Harderian gland adenoma or carcinoma
combined in males exposed to 32 or 80 ppm and females exposed to 80 ppm were significantly greater
than in the chamber controls. The incidences of Harderian gland adenoma or carcinoma combined in
these groups exceeded the historical control range.
September 2009
4-36
DRAFT - DO NOT CITE OR QUOTE
-------
1 Although not significantly increased, the incidence of renal tubule adenoma in males exposed
2 to 80 ppm was greater than in the chamber control group. The incidence of this rare neoplasm
3 exceeded the historical control range. The incidences of renal tubule hyperplasia in males exposed to
4 32 or 80 ppm were significantly greater than in the chamber controls. The combined single- and step-
5 section incidence of renal tubule adenoma in males exposed to 80 ppm and the combined incidences of
6 renal tubule hyperplasia in all groups of exposed male mice were greater than in the chamber controls.
7 The incidences of mammary gland carcinoma in females exposed to 80 ppm were significantly
8 greater than in the chamber control group. The incidences of mammary gland carcinoma in females
9 exposed to 32 and 80 ppm exceeded the historical control range. Mammary gland hyperplasia was
10 present in a few females exposed to chloroprene, but was not significantly increased relative to
11 chamb er control s.
12 The incidence of forestomach squamous cell papilloma in females exposed to 80 ppm was
13 greater than in the chamber controls but statistically not significant. The incidence observed exceeded
14 the historical control range. In male and female mice exposed to 80 ppm, the incidences of hyperplasia
15 of the forestomach epithelium were significantly greater than in chamber controls, and the lesions were
16 similar to those seen in the 13-week study. Hyperplasia was a focal to multifocal change characterized
17 by an increase in the number of cell layers in the epithelium.
18 The incidences of sarcoma of the skin were significantly greater in all exposed female mice
19 compared with chamber controls. The incidences of sarcomas of the mesentery were increased in all
20 exposed female mice, with only the mice in the 32 ppm exposure group exhibiting a significant
21 increase.
22 Carcinomas of Zymbal's gland were observed in three females exposed to 80 ppm, and two
23 carcinomas had metastasized to the lung. Zymbal's gland carcinomas have not been reported in the
24 NTP historical database for control female mice.
25 Single papillary adenomas were detected in the trachea of one male exposed to 12.8 ppm and in
26 one male exposed to 32 ppm. These adenomas have not been documented in the NTP historical
27 database.
28 The incidences of splenic hematopoietic proliferation in males exposed to 32 and 80 ppm and
29 in all exposed groups of females were significantly greater than in the chamber controls.
30 Because of a large number of early deaths of mice exposed to chloroprene for 2-years, survival -
31 adjusted neoplasm rates were estimated by NTP by using the poly-3 survival-adjusted quantal response
32 method of Portier and Bailer (1989). This adjustment accounts for the impact of early mortality on the
33 expression of late-developing neoplasms and provides a clearer indication of exposure-response
34 relationships for neoplasms induced by chloroprene (Table 4-28). The neoplasm incidence values
35 provided represent the ratio of the number of animals in an exposure group bearing the specific
36 neoplasm relative to the adjusted number of animals at risk.
September 2009 4-37 DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-28. Survival-adjusted neoplasm rates for mice in the 2-year inhalation
study of chloroprene
Lung
Adenoma or carcinoma
Alveolar/bronchiolar adenoma or carcinoma
All Organs
Hemangioma or hemangiosarcoma
Harderian gland
Adenoma or carcinoma
Kidney (renal tubules)
Adenoma
(single section)
(single + step section)
Mammary gland
Adenoacanthoma or carcinoma
Forestomach
Squamous cell papilloma or carcinoma
Liver
Carcinoma
Adenoma or carcinoma
Skin
Sarcoma
Mesentery
Sarcoma
MALES (%)
0
14.1b
29.8b
2.4b
4.7b
Oa
ob
2.4b
—
12.8
28.3
63. 7b
28.2b
12.0
2.4
4.8
0
—
32
56.9b
79.2b
45.2b
26.3b
2.8
8.3
5.6
—
80
66.4b
92.9b
43. 6b
32.0b
8.2
23.9a
13.3
—
FEMALES (%)
0
4.6b
9.1b
9.0a
4.5b
—
6.7b
2.3b
9.0b
44.8
Oa
0
12.8
35.6b
68.3b
16.0
13.5
—
12.9
0
28.4a
62.9
27.5b
10.7a
32
53.8b
85.8b
53. lb
11.7
—
33.7b
0
47.5b
63.3
39.0b
28.9b
80
76.0b
96. lb
27.7a
31.2b
—
42.5b
14.6
58.2b
79.7b
52.6b
11.0
In the chamber control column,a indicates a significant trend (p< 0.05) across all exposure groups by the Poly-3 quantal
response test; b indicates a significant tend at p < 0.01
In the exposed group columns,a indicates a significant difference (p < 0.05) from the chamber control group by pairwise
comparison; b indicates a significant difference at p < 0.01.
Source: NTP(1998)
1 In another chronic inhalation study, Trochimowicz et al. (1998) exposed three groups of 100
2 Wistar rats and Syrian hamsters of each sex to chloroprene at 0, 10, or 50 ppm for 6 hours/day, 5
3 days/week for up to 18 months (hamsters) or 24 months (rats). Chemical purity of the bulk
4 chloroprene was reported to be 99.6%, with less than 50 ppm of dimers as determined by gas
5 chromatography. Bottles of test material were received weekly and were stored under nitrogen at -
6 20°C. Phenothiazine (0.01%) was added to prevent oxidation. A fresh sample of chloroprene from
7 cold storage was used for each day's exposure. To generate the test atmospheres, sufficient quantities
8 of the bulk material were vaporized with dried and filtered nitrogen at 0°C; vaporization at this
9 temperature was performed to inhibit the formation of degradation products. The saturated
10 chloroprene/nitrogen mixture was then directed into the inhalation chamber inlet, where it was mixed
11 with the main air flow to generate the desired exposure concentration. All animals were observed
12 daily and clinical signs and mortality were recorded. Rats and hamsters were weighed immediately
13 before the first exposure, weekly for the first 8 weeks, and at 4-week intervals thereafter. During the
14 last 6 months of each study, all animals were palpated monthly for the presence of tumors. Time of
15 tumor appearance, size, location, and progression were recorded. At study termination, both hamsters
September 2009
4-38
DRAFT - DO NOT CITE OR QUOTE
-------
1 and rats were sacrificed by exsanguination of the abdominal aorta. A postmortem examination was
2 conducted during which all major organs/tissues were examined for gross abnormalities. Gross
3 pathological examinations were conducted on all animals, including those that died intercurrently or
4 were killed in extremis, unless advanced autolysis or cannibalism prevented this. The following
5 organs were weighed: adrenals, brain (hamster), heart, kidneys, liver, lungs with trachea and larynx,
6 ovaries, pituitary, spleen, testes, and thyroid (rat). The following organs/tissues were preserved and
7 examined microscopically: all gross lesions, adipose tissue, aorta (rat), epididymides, external auditory
8 canal with Zymbal's glands, eyes, exorbital lachrymal glands, femur (with knee joint), gastrointestinal
9 tract (esophagus, stomach, duodenum, jejunum, ileum, cecum, and colon), lungs, lymph nodes
10 (auxiliary, cervical, and mesenteric), mammary glands, nasal cavity (four transverse sections),
11 pancreas, parotid salivary glands, preputial glands, prostate, sciatic nerve, seminal vesicles, skeletal
12 muscle, skin, spinal cord, sternum (bone marrow), sublingual and submaxillary salivary glands,
13 thymus, thyroid with parathyroid (hamster), urinary bladder, and uterus. Microscopic examinations
14 were performed on all organs from all control and high-exposure animals, and on the liver, spleen,
15 pituitary gland, thyroid glands, adrenals, and all grossly visible tumors and tumor-like lesions from the
16 low-exposure animals.
17 Mortality rate for rats was relatively low in all groups up to week 72, ranging from 1-3%.
18 During week 72, however, 87 males and 73 females of the 10 ppm exposure group died overnight from
19 suffocation resulting from accidental failure of the exposure chamber ventilation system. For
20 hamsters, the mortality rate was negatively correlated with the concentration of chloroprene exposure.
21 After 18 months of exposure, survival rates in the 0, 10, and 50 ppm groups were 88, 92, and 93% in
22 males and 63, 75, and 72% in females, respectively.
23 A slight but consistent growth retardation was found in male rats (-10%) and female rats (-5%)
24 in the 50 ppm exposure group. Both male and female hamsters showed a slight growth depression in
25 the 50 ppm group throughout the study. Appearance and behavior of the rats were not affected by
26 exposure to chloroprene, except that alopecia occurred more frequently in the 50 ppm group than in the
27 10 ppm group or in the controls. The alopecia varied from small, focal, mostly bilateral bald areas to
28 severe, diffuse, generalized hair loss. Alopecia was first observed after an exposure period of about 10
29 weeks, but by 25 weeks the incidence and degree of alopecia gradually decreased and in many animals
30 complete re-growth of hair was observed. No abnormalities were observed in hamsters; alopecia was
31 occasionally seen in each group during the first 64 weeks of study, regardless of exposure.
32 Body weights are given in Table 4-29. In both male and female rats, mean relative lung
33 weights were significantly lower in both exposure groups than in controls. In females exposed to 50
34 ppm, the mean relative spleen and thyroid weights were significantly lower. The kidney and pituitary
35 weights in males exposed to 10 ppm were significantly increased compared with controls, although
36 this was not observed in the 50 ppm exposure group. In hamsters, both male and female animals
September 2009 4-39 DRAFT - DO NOT CITE OR QUOTE
-------
exposed to 50 ppm had significantly higher brain weights compared with controls. Relative lung
weight was significantly higher in males exposed to 50 ppm than in controls.
Table 4-29. Selected mean relative organ weights of rats exposed for 24 months
and hamsters exposed for 18 months to chloroprene vapor
GROUP
(ppm)
NUMBER1
BW(g)
ADRENALS
BRAIN
KIDNEYS
LIVER
LUNGS
SPLEEN
THYROID
Rats
Males
0
10
50
77
9
76
494
500
496
...
—
—
—
—
—
0.61
0.68a
0.64
3.09
3.31
3.15
0.45
0.37b
0.38b
0.154
0.172
0.146
0.0056
0.0056
0.0056
Females
0
10
50
81
19
75
308
309
307
...
...
—
—
—
—
0.64
0.65
NR2
3.00
3.23a
3.13a
0.53
0.45a
0.45a
0.180
0.176
0.164b
0.0080
0.0073
0.0070b
Hamsters
Males
0
10
50
86
92
92
101
101
93
0.0311
0.0279a
0.0294
.10
.11
.19C
1.25
1.17b
1.22
5.11
4.75a
4.91
0.85
0.84
0.90b
0.197
0.190
0.174b
—
—
—
Females
0
10
50
60
74
72
99
98
90
0.0340
0.0356
0.0383
.13
.16
.24C
1.48
1.50
1.50
6.73
6.54
6.37
1.01
0.97
1.01
0.253
0.269
0.286
—
—
—
J
4
5
6
7
8
9
10
11
12
13
14
15
16
1 Number at sacrifice
2 Not recorded
a Significant, 0.1 < p < 0.005
b Significant, 0.001 < p < 0.01
c Significant, p < 0.001
Source : Trochimowicz et al. (1998)
Gross pathology revealed that lungs from rats exposed at 10 and 50 ppm had markedly lower
incidences of nodular pleural surfaces, consolidation, and gross changes consistent with, and
characterized as chronic respiratory disease, than did controls. Morphologic indicators of chronic
respiratory disease were seen in 28 of 196 controls, 0 of 37 in the 10 ppm group, and 4 of 200 in the 50
ppm group. The incidence of tumors or tumor-like lesions of the mammary glands was slightly higher
in the exposed animals terminated at the end of the study (10/24 and 34/100 in 10 and 50 ppm,
respectively) compared with controls (23/99); however, these differences were not statistically
significant unless animals that were moribund or dead before the terminal sacrifice were included in
the analysis. No other remarkable differences in gross pathology were seen in rats. Macroscopic
examination of hamsters revealed a slight, concentration-related decrease in the incidence of pale
adrenal glands in males.
The only nonneoplastic lesions in rats were observed in liver and lungs (only the livers of
animals that died accidentally due to a failure in the ventilation system were available for microscopic
September 2009 4-40 DRAFT - DO NOT CITE OR QUOTE
-------
1 examination). The number of female and male rats with one or more small foci of cellular alteration in
2 the liver was significantly increased in the 50 ppm group than in controls. Mild changes, such as
3 lymphoid aggregates around bronchi, bronchiole, and blood vessels, were observed in males and
4 females exposed to 50 ppm. Acute inflammatory processes in the lungs were found in the 50 ppm
5 exposure group and control animals to a similar extent.
6 The only nonneoplastic change seen in hamsters was a generalized amyloidosis (in the liver,
7 kidneys, spleen, and adrenals) that was lower in incidence in the 50 ppm exposed group compared with
8 controls.
9 Tumor incidences for rats and hamsters are shown in Tables 4-30 and 4-31, respectively. With
10 the exception of mammary gland tumors and squamous cell carcinomas, no individual organ or tissue
11 in rats exposed to chloroprene showed a statistically significant excess of tumors compared with
12 controls. The number of females bearing mammary tumors in the 50 ppm group was significantly
13 increased (p < 0.05). The observed increase in mammary tumors in the high dose animals was due to
14 the inclusion in the analysis of animals that were moribund or dead before the terminal sacrifice. No
15 difference was observed between control and test group animals that were sacrificed at the end of the
16 study. The number of mammary tumors per rat was not different between the 50 ppm group and the
17 control group. The increased incidence of mammary tumors was almost entirely due to the relatively
18 high number of animals of the test groups bearing benign fibroadenomas. Squamous cell carcinomas
19 involving the nasal cavity, sinus maxillaries, subcutis, and skin were found in 3 of 100 males of the 50
20 ppm group and in 1 of 99 females of the control group. Neither macroscopic nor microscopic
21 examination could clarify the exact origin of these tumors. If they originated as skin tumors, the total
22 number of squamous-cell carcinomas of the skin would have been 5/100 in the 50 ppm group, which
23 would be a statistically significant (p < .05) increase over controls (1/97).
24 In the hamster, the incidences of cystadenomatous polyps of the gallbladder and
25 pheochromocytoma were slightly, but significantly, elevated in the males exposed to 10 ppm. All other
26 tumors observed were about equally distributed among test and control groups or occurred in only one
27 or two hamsters.
28 Sanotskii (1976) provided a review of numerous Russian subchronic inhalation studies of
29 chloroprene (chemical purity and exposure regimen not specified) in rats and mice. According to
30 Sanotskii (1976), the studies evaluated the systemic effects of chloroprene exposure in white rats
31 exposed for 4.5 months to 0.051, 0.15, and 1.69 mg/m3 (0.014, 0.041, and 0.47 ppm) or C57BL/6 mice
32 exposed for 2 months to concentrations as high as 35 mg/m3 (9.7 ppm). Several "signs of systemic
33 effect" in male rats were reported at 1.69 ± 0.087 mg/m3, including an increase in a "summation
34 threshold index" (not defined) after 2.5 and 4.5 months, a decrease in the synthesis of hippuric acid
35 from sodium benzoate (Quick's test) at 4.5 months, and an inhibition of gas exchange after 4.5 months.
36 Chloroprene was reported to have had no effect on "the indicators used in the tests" (i.e., summation
September 2009 4-41 DRAFT - DO NOT CITE OR QUOTE
-------
1 threshold index, hippuric acid synthesis, and inhibition of gas exchange) in mice at concentrations as
2 high as 35 ± 0.7 mg/m3 (9.7 ppm).
Table 4-30. Incidence, site and type of tumor in selected organs and tissues of rats
exposed to chloroprene for 24 months
Site and type of tumor3
Initial number of rats
Number examined
Number tumor-bearingb
Total number primary tumors'3
Hematopoietic system
Lymphoid leukemia
Monocytic leukemia
Kidneys
Lipoma
Adenocarcinoma
Liver
Unidentified
Lungs
Anaplastic carcinoma
Mammary glands
Adenoma
Fibroadenoma
Adenocarcinoma
Papillary carcinoma
Unidentified tumor
Skin
Squamous cell carcinoma
Skin, nasal cavity, maxillary sinus,
Squamous cell carcinoma
Spleen
Hemangiosarcoma
Subcutis, nasal cavity, or maxillary sinus
Reticulum cell sarcoma
Testes
Leydig cell tumor
Testes/epididymides
Mesothelioma
Thyroid gland
Parafollicular cell adenoma
Small
Medium/large
Parafollicular cell carcinoma
Small
Large
Follicular adenoma
Small
Large
Papillary carcinoma
MALES
0 ppm
100
97
51
73/51
1
0
0
0
0
0
—
—
—
—
—
0
0
0
0
2
1
6
3
1
1
2
2
0
10 ppm
100
13
6
6/6
0
0
0
0
0
0
—
—
—
—
—
0
0
0
0
2
0
0
1
0
0
0
0
0
50 ppm
100
100
57
77/57
2
1
1
1
0
0
—
—
—
—
—
2
3
1
0
4
0
8
3
0
0
2
1
0
FEMALES
0 ppm
100
99
66
100/66
0
0
1
0
1
1
3
24
5
1
1
0
1
0
0
11
3
0
0
0
0
0
10 ppm
100
24
12
13/12
0
0
0
0
0
0
1
6
0
0
2
0
0
0
0
0
1
0
0
0
0
0
50 ppm
100
100
74
96/71
1
0
1
0
0
0
7
36
3
0
0
0
0
0
1
14
4
0
0
3
0
2
September 2009
4-42
DRAFT - DO NOT CITE OR QUOTE
-------
Site and type of tumor3
Urinary bladder
Transitional cell carcinoma (metastasizing)
Zymbal's gland
Adenoma
MALES
0 ppm
0
0
10 ppm
0
0
50 ppm
1
0
FEMALES
0 ppm
0
0
10 ppm
0
0
50 ppm
0
1
a Multiple tumors at one site were counted as one tumor
b Some animals had more than one tumor
Source: Trochimowicz et al. (1998)
Table 4-31. Incidence, site and type of tumor in selected organs and tissues of
hamsters exposed to chloroprene for 18 months
Initial number of hamsters
Number examined
Number tumor bearing3
Total number primary tumors3
Kidney
Cortical adenocarcinoma
Liver
Neoplastic (hepatocellular) nodule
Unidentified tumor-like lesion
Lung tumors
Gallbladder
Cystadenomatous polyp
Pancreas
Islet-cell adenoma
Islet-cell adenocarcinoma
Stomach
Papilloma
Unidentified papilloma-like lesion
Testes
Leydig-cell tumor
Colon
Adenomatous polyp
Pituitary
Adenoma
Thyroid gland
Parafollicular cell adenoma
Cystadenoma
Papillary adenoma
Follicular adenoma
Parathyroid
Adenoma
Adrenals
Cortical adenoma
Cortical carcinoma
Pheochromocytoma
Malignant pheochromocytoma
Ovaries
Granulosa-theca-cell tumor
MALES
0 ppm
100
100
14
15/14
2
0
0
0
1
1
0
0
1
1
0
0
2
1
0
2
0
4
0
0
0
—
10 ppm
100
97
17
18/17
0
1
1
0
6a
0
0
0
1
0
0
0
0
0
1
1
0
1
1
4b
0
—
50 ppm
100
97
20
23/20
0
0
0
0
1
2
0
2
1
0
0
1
0
0
1
0
0
10
0
2
2
FEMALES
0 ppm
100
94
10
11/11
0
0
0
0
1
0
1
0
1
—
2
2
0
0
1
1
0
0
1
0
0
0
10 ppm
100
93
11
11/11
0
0
0
0
2
0
0
0
2
—
0
0
2
0
0
2
1
0
0
0
0
2
50 ppm
100
97
15
18/15
0
0
1
0
3
0
1
0
0
—
0
0
1
0
2
2
0
3
1
0
0
1
September 2009
4-43
DRAFT - DO NOT CITE OR QUOTE
-------
Parotid salivary glands
Adenoma
Skin
Unidentified tumor-like lesion
Zymbal's gland
Sebaceous adenoma
Depot fat
Lipoma
Nose
Adenoma of Bowman's glands
Adenocarcinoma of Bowman's glands
Bone (ribs)
Osteosarcoma
Abdominal cavity
Reticulum cell sarcoma
MALES
0 ppm
0
0
0
0
0
0
0
1
10 ppm
0
1
0
0
0
0
0
0
50 ppm
0
0
1
0
0
0
0
0
FEMALES
0 ppm
0
0
0
0
1
0
1
0
10 ppm
0
0
0
0
0
0
0
0
50 ppm
1
0
0
1
0
1
0
0
a Some animals had more than one tumor
b Significant, p < 0.05 by chi-squared test
Source: Trochimowicz et al. (1998)
4.3. REPRODUCTIVE/DEVELOPMENTAL STUDIES—ORAL AND INHALATION
1 Ponomarkov and Tomatis (1980) administered chloroprene dissolved in olive oil by stomach
2 tube to 17 female BD IV rats at a single dose (100 mg/kg body weight) on gestational day (GD) 17.
3 Progeny from treated females (81 males and 64 females) were treated weekly with 50 mg/kg body
4 weight by stomach tube from the time of weaning for life (120 weeks). A control group of 14 female
5 rats was treated with 0.3 mL olive oil. Litter sizes and preweaning mortality, survival rates, and body
6 weights did not differ between chloroprene-treated animals and controls (see Section 4.2.1 for further
7 study details).
8 NTP (1998) evaluated sperm morphology and vaginal cytology in rats exposed to 0, 5, 32, or
9 200 ppm and mice exposed to 0, 12, 32, 80 ppm chloroprene for 13 weeks. Methods used were those
10 described in the NTP's sperm morphology and vaginal cytology evaluations protocol (NTP, 1985).
11 Table 4-32 is a summary of measured epididymal spermatozoal and estrous cycle parameters from
12 these 13-week studies. The sperm motility of male rats exposed to 200 ppm was significantly less than
13 that of controls. This was the only reproductive tissue or estrous cycle parameter affected, compared
14 with controls, in rats or mice at any exposure level.
September 2009
4-44
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-32. Summary of epididymal spermatozoal and estrous cycle parameters
for rats and mice in the 13-week study of chloroprene
n
RATS
0 ppm
10
5 ppm
10
32 ppm
9
200 ppm
9
MICE
0 ppm
7
12 ppm
8
32 ppm
10
80 ppm
10
Epididymal spermatozoa - males'1
Motility (%)
Abnormal sperm (%)
Sperm concentration
(106/g cauda epididymidis)
86.73
±1.04
0.70
±0.05
698
±40
83.62
±1.93
0.78
±0.11
722
±62
82.16
±1.84
0.73
±0.11
689
±46
Estrous cycle -
Length (days)
Diestrus stage (% of cycle)
Proestrus stage (% of cycle)
Estrus stage (% of cycle)
Metestrus stage (% of cycle)
Uncertain diagnosis stage (%
of cycle)
5.00
±0.15
42.9
15.7
18.6
22.9
0.0
4.67
±0.17b
35.7
18.6
22.9
22.9
0.0
5.00
± 0.27C
44.3
11.4
20.0
24.3
0.0
80.04
±1.99d
1.02
±0.14
683
± 25
79.09
±1.20
1.49
±0.42
1,632
±138
81.07
± 1.13
1.30
±0.22
1,447
±122
80.08
±1.19
0.98
±0.10
1,575
±104
80.04
±1.47
1.36
±0.22
1,672
±134
females'1
5.33
±0.17b
45.7
17.1
15.7
20.0
1.4
4.00
±0.00
31.4
20.0
24.3
24.3
4.30
±0.21
31.4
20.0
24.3
24.3
4.22
±0.15b
30.0
22.9
25.7
21.4
4.13
±.13C
35.7
25.7
20.0
18.6
a Epididymal spermatozoal parameters, and estrous cycle lengths are presented as mean ± standard error. Differences from
the control group are not significant by Dunn's test (epididymal spermatozoal abnormality and concentration, estrous cycle
length). By multivariate analysis of variance, exposed females do not differ significantly from the chamber control females
in relative length of time spent in the estrous stages.
b Estrous cycle was longer than 12 days or unclear in 1 of 10 animals.
0 Estrous cycle was longer than 12 days or unclear in 2 of 10 animals.
d Significantly different (p < 0.01) from the control group by Shirley's test.
Source: NTP, (1998).
1
2 Sanotskii (1976) reviewed several Russian studies that exposed white rats (strain unknown) to
3 various concentrations of chloroprene in order to determine the effect on reproductive and
4 developmental parameters. In male rats exposed for 4.5 months to 1.7 mg/m3 (0.5 ppm) of
5 chloroprene, reductions in the number of normal spermatogonia, increases in the percentage of dead
6 spermatozoa, and decreases in spermatozoal motility were reported. These effects were not observed
7 by NTP (1998) in F344 rats at much higher concentrations (Table 4-32). Sanotskii (1976) also
8 reported an increase in the number of seminiferous tubules with desquamating epithelium in male
9 C57BL/6 mice exposed to 0.32 mg/m3 (0.09 ppm) for 2 months and increased dominant lethal
10 mutations in germ cells of male and female C57BL/6 mice exposed to 3.5 mg/m3 (1 ppm) for 2
11 months.
12 Sanotskii (1976) also reported on an embryotoxicity study in which pregnant white rats were
13 exposed during their "whole period of pregnancy." Exposure to 4 mg/m3 (1.1 ppm) chloroprene was
14 reported to have resulted in an increase of embryonic mortality, a decrease in fetal weight, and a
15 disturbance in vascular permeability as evidenced by hemorrhaging into body cavities. Exposure to
16 0.13 mg/m3 (~ 0.04 ppm) chloroprene was reported to have resulted in increased postnatal mortality.
September 2009
4-45
DRAFT - DO NOT CITE OR QUOTE
-------
1 Exposure to 4 mg/m3 (1.1 ppm) chloroprene at various times during pregnancy was reported to have
2 resulted in cerebral hernia and hydrocephalus.
3 Culik et al. (1978) evaluated the embryotoxic, teratogenic, and reproductive toxicity of
4 chloroprene in rats. Culik et al. (1978) exposed pregnant CD rats to chloroprene by inhalation at 0, 1,
5 10, or 25 ppm (0.28, 2.8, or 6.9 mg/m3) for 4 hours daily, either on GDs 1-12 (embryotoxicity study)
6 or GDs 3-20 (teratology study). Pregnant rats in these embryotoxicity and teratology studies were
7 sacrificed and their litters examined on GDs 17 and 21, respectively. Male rats in a separate
8 reproduction study were exposed to 0 or 25 ppm (0 or 6.9 mg/m3) 4 hours daily for 22 days and bred
9 with untreated females for 8 consecutive weeks. The embryotoxicity study included 200 female rats
10 (50 per exposure group), the teratology study included 100 primigravida rats (25 per exposure group),
11 and the male reproduction study involved 10 male rats (5 per exposure group) and 3 virgin females per
12 male. The test material was reported to be > 99.9% pure and was stored under nitrogen at -20°C in
13 small glass bottles holding one day's supply for generating atmospheres. No chemical decomposition
14 was observed during the experiment.
15 In both the embryotoxicity and teratogenicity studies, litter size, average numbers of
16 implantation sites per litter, and preimplantation losses among exposed females did not differ
17 significantly from those of the controls (Table 4-33). In the teratology study, there was an increase in
18 the percentage of litters with resorptions that was statistically significant (p < 0.05, Fisher's exact test)
19 only in the 10 ppm exposure group (62% compared to 29% in the control group). The percentage of
20 litters with resorptions was also elevated in the 25 ppm group (59%), although this increase in effect
21 failed to achieve statistical significance. There was no effect on percentage of litters with resorptions
22 in any exposure group in the larger embryotoxicity study; all groups had approximately 50% of their
23 litters exhibiting resorption. The number of resorptions per litters with resorptions was not affected in
24 either study. The more frequently investigated endpoint of number of resorptions per litter (total) was
25 not reported by the study, but was calculated from the reported data and included in Table 4-33 for
26 reference. There was a slight, but statistically significant (p < 0.05), increase in the average body
27 weight of fetuses from dams exposed to chloroprene at 25 ppm in the teratology study. Fetuses from
28 dams in the teratology study exposed to 10 and 25 ppm chloroprene were significantly (p < 0.05)
29 longer than the control fetuses. The incidence of minor anomalies (minute subcutaneous hematomas
30 and petechial hemorrhages) in fetuses from exposed dams was similar to that found in control fetuses
31 (Table 4-34). No major compound-induced or concentration-related skeletal or soft tissue anomalies
32 were found. The number of unossified sternebrae and unossified thoracic vertebral centers were
33 similar in all groups regardless of treatment. The combined results of weekly matings for the 8-week
34 reproduction test indicated that there were no significant effects on reproduction due to chloroprene
35 exposure: the mating index, average number of pups per litter, viability index, and lactation index
36 were similar for exposed and control animals.
September 2009 4-46 DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-33. Results of teratology and embryotoxicity studies in rats exposed to
chloroprene by inhalation
PARAMETER
CONCENTRATION OF CHLOROPRENE (ppm)
0
1
10
25
Teratology Study
Number of litters
Pregnancy rate, %
Corpora lutea/dam
Implantation sites/dam
Median preimplantation loss, %
Live fetuses/litter
Litters with resorption, %
Litters totally resorbed
Median postimplantation loss
in litters with resorption, %
Resorptions/litters with
Resorptions
Resorptions/litters total
Fetal body weight, g
Fetal crown-rump length, mm
21
84 (21/25)
13 ±3
10 ±2
14.7
9±2
29 (6/21)
0
11.8
1.3 (8/6)
0.38 (8/21)
3.76 ±0.28
32.9 ±1.4
24
96 (24/25)
12 ±2
9±3
29.5
8±3
29 (7/24)
0
16.7
2.0 (14/7)
0.58 (14/24)
3.94 ±0.46
33.7 ± 1.6
21
84 (21/25)
12 ±2
9±2
20.0
8±3
62 (13/21) a
0
22.0
1.9 (25/13)
1.19(25/21)
3.96 ±0.26
33.8±0.7b
19
76 (19/25)
13 ±2
11 ±1
10.0
10 ±1
59(11/19)
0
16.7
1.6(17/11)
0.89 (17/19)
4.04 ± 0.27 b
34.1 ± 1.2 b
Embryotoxicity Study
Number of litters
Pregnancy rate, %
Corpora lutea/dam
Implantation sites/dam
Median preimplantation loss, %
Live fetuses/litter
Litters with resorption, %
Litters totally resorbed
Median postimplantation loss
in litters with resorption, %
Resorptions/litters with
resorptions
Resorptions/litters total
45
90 (45/50)
15 ±3
11 ±3
20.0
10 ±3
51 (23/45)
0
9.1
1.7 (39/23)
0.87 (39/45)
43
86 (43/50)
14 ±3
11±4
16.2
9±4
51 (22/43)
1
12.9
2.1 (47/22)
1.09 (47/43)
43
88 (43/49)
14 ±2
10 ±4
17.7
10 ±3
53 (23/43)
0
8.3
1.6 (37/23)
0.86 (37/43)
48
94(48/51)
13 ±3
10 ±3
16.0
10 ±3
50 (24/48)
0
9.1
1.4 (34/24)
0.71 (34/48)
a Significantly different (p < 0.05)
b Significantly different (p < 0.05)
test
from the control group by Fisher's exact test.
from the control group by an analysis of variance and least significant difference (LSD)
Source: Culiketal. (1978)
1 Culik et al. (1978) concluded that the statistically significant increase in litters with resorptions
2 observed in the teratology study at 10 ppm was not biologically significant because the increase at 25
3 ppm was not statistically significant and the effect was not observed in the embryotoxicity study,
4 which had larger numbers of animals per exposure group and was specifically designed to observe
5 such an effect. Further, the control group for the teratology study is the only group in either study
6 (embryotoxicity or teratology) that is far outside of the historical control range for number of
7 resorptions per litter (0.83 ± 0.34) for this strain of rat (Charles River Laboratories, 1996); the
8 corresponding control group in the embryotoxicity study had a response rate equivalent to historical
9 controls. Therefore, if the control group response in the teratology study is abnormally low, this may
10 indicate that the statistically significant increase seen in the 10 ppm group may be a spurious
September 2009
4-47
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
observation. Chloroprene exerts an effect on fetal weight and size, as evidenced by increases in both at
higher exposure levels. However, in the absence of other definitive markers of developmental toxicity,
the importance or adversity of this finding remains unclear. Given the lack of a defined dose-response
for litters with resorptions in either the embryotoxicity or teratology study, and that the control group
in the teratology study may be a statistical outlier compared to historic control data, there is no
compelling evidence that chloroprene displays developmental effects in CD rats at exposure levels up
to 25 ppm. Therefore, 25 ppm is identified as the NOAEL for this study.
Table 4-34. Incidence of anomalies in litters of rats exposed to chloroprene by
inhalation
Gross anomalies
Soft tissue anomalies
Skeletal anomalies
Gross anomalies
Runts3
Small subcutaneous hematomas
Petechial hemorrhages
Soft tissue anomalies
Hydronephrosis
Subcutaneous edema
Skeletal anomalies
Delayed ossification of one or more
sternebrae
14th rudimentary ribs(s) or spur(s)
Wavy ribs
Bipartite thoracic centra
CONCENTRATION OF CHLOROPRENE (ppm)
0
1
10
25
Number of litters (fetuses) examined
21 (192)
21 (66)
21 (126)
24 (191)
24 (69)
24 (122)
21 (172)
21 (60)
21(112)
19 (184)
19 (62)
19 (122)
Number of litters (fetuses) affected
1(1)
5(5)
5(5)
8(9)
0
17 (58)
20 (91)
4(4)
2(2)
0
9(9)
2(6)
4(6)
1(1)
15 (39)
22 (76)
4(5)
2(3)
1(1)
4(4)
3(3)
1(1)
0
13 (33)
20 (67)
2(3)
2(2)
1(1)
6(10)
2(2)
5(7)
0
14 (45)
19 (77)
3(4)
4(8)
9
10
11
12
13
14
15
16
17
18
19
a Body weight less than control mean weight minus 3 standard deviations
Source: Culiketal. (1978)
Mast et al. (1994) exposed groups of 15-16 pregnant New Zealand white rabbits by inhalation
to 10, 40, or 175 ppm chloroprene (36.2 144.8, or 633.5 mg/m3) for 6 hours/day on gestational days 6-
28. Maternal body weights were measured on days 0, 6, 15, 22, and 29 and animals were observed
twice daily (7 days/week) during the exposure period for signs of illness or mortality. On GD29, dams
were sacrificed and examined for gross tissue abnormalities. Maternal kidneys and liver were removed
and weighed. The uterus was removed and weighed, and the number, position, and status (live,
resorbed, or dead) of implants were recorded. Live fetuses were weighed and examined from gross,
visceral, and skeletal defects. Bulk chemical analysis was performed using infrared spectroscopy to
confirm test material identity. Purity and dimer determinations were conducted by gas chromatography.
Exposure atmospheres were generated by immersing an evaporation flask containing bulk material in
a 150° F water bath and passing a metered flow of nitrogen through the flask to a condenser. The
condenser's temperature was maintained at -2° C in order to control the chloroprene vapor
September 2009 4-48 DRAFT - DO NOT CITE OR QUOTE
-------
1 concentration, and to remove low volatility impurities from the vapor. From the condenser, the
2 chloroprene vapor was mixed with an appropriate amount of compressed air in order to achieve the
3 desired exposure concentration. The normal exposure concentrations in the study were between 98-
4 100% target concentrations, and there was no evidence of degradation products greater than 0.1%
5 target concentration.
6 There were no signs of maternal toxicity due to exposure to chloroprene. A few dams in each
7 group exhibited nasal discharge, vaginal bleeding, and loose stools at various times during the
8 exposure period. The overall pregnancy rate was 89%, with a range of 80-94% for each exposure
9 group. The incidence of clinical signs of toxicity was low during the exposure, and dams appeared to
10 be in excellent health at termination. No exposure-related effects on maternal weight change were
11 noted. Exposure to chloroprene had no effect on the number of implantations, live pups, or
12 resorptions. Fetal body, liver, and kidney weights were not affected by exposure. The incidence of
13 fetal malformations was not affected by exposure to chloroprene. The results of this study indicate that
14 exposure to chloroprene on GD6-28 in rabbits results in no observable developmental toxicity,
15 therefore the high exposure group, 175 ppm, was identified as the NOAEL for this study.
16 In an unpublished report, Appelman and Dreef van der Meulen (1979) exposed two successive
17 generations (Fo or Fi) of Wistar rats to 0, 10, 33, or 100 ppm (0, 36.2, 119.5, or362mg/m3)
18 chloroprene. In the F0-generation, groups of 25 males and females were exposed to chloroprene for 6
19 hours/day, 5 days/week for 13 weeks. After the termination of the exposure, the treated animals were
20 caged and mated with untreated stock animals for 20 days (1 male per 1 female). After the mating
21 period, the animals were separated: males were sacrificed and their testes were collected and
22 examined whereas females were caged individually and allowed to birth and rear their litters. After
23 their litters were weaned, the females were sacrificed and their uteri were collected and examined for
24 implantation sites. The number of pups in each litter was recorded at birth, as well as the total number
25 of survivors and total litter weight at days 1, 3, 14, and 28. Litters containing more than 8 siblings
26 were randomly culled to that number at day 4. From the Fi-litters, 20 males and females were selected
27 randomly from each exposure group one week after weaning and exposed to the same concentrations
28 of chloroprene from 10 weeks (6 hours/day, 5 days/week). In both the FO and FI rats, the general
29 condition, behavior, and signs of possible intoxication were checked daily and all signs of illness or
30 reaction to exposure were recorded. Individual body weights were recorded weekly during exposure.
31 In the FI rats, blood samples were collected from 15 rats/sex/exposure group at an age of 4 weeks and
32 analyzed for hemoglobin concentration. At the end of the exposure period, 10 FI rats/sex/exposure
33 group were sacrificed and their liver, lungs, and gonads were weighed and examined. Test
34 atmospheres were generated by evaporating bulk material with a metered flow of filtered and dried
35 nitrogen at 0° C. The chloroprene-saturated chloroprene was then mixed with air to achieve the
36 desired test concentration. Nominal concentrations were within 98-100% target concentrations.
September 2009 4-49 DRAFT - DO NOT CITE OR QUOTE
-------
1 The general condition and behavior of F0 rats did not differ between exposure groups. At 100
2 ppm, slight (less than 10% decrease relative to control), but significant, growth retardation was
3 observed in males in weeks 3, 6, 7, 8, and 10 and in females from week 2 to termination of exposure (p
4 < 0.05). There were statistically significant decreases in body weights in both sexes at various time
5 points in the low and mid-exposure groups compared to controls, but no consistent exposure-related
6 pattern was observed. No data on food consumption were provided, but the authors note that decreases
7 in body weight were most likely attributable to occasional shortages in food availability. The
8 percentage of females (exposed and non-exposed) that successfully mated was not affected by
9 chloroprene exposure. Sex ratios, mortality during lactation, and resorption quotients were not
10 significantly altered in any exposure group. The body weight of offspring descended from treated
11 females and untreated males was statistically reduced in the high exposure group. Body weights of
12 offspring descended from treated males and untreated females were not affected.
13 The general condition and behavior of FI rats did not differ between exposure groups.
14 Statistically significant decreases in body weight (greater than 10% reduction compared to control)
15 were observed in females descended from treated females during week 1 of exposure (p < 0.01), in
16 males descended from treated males during weeks 4, 6, 7, and 10 (p < 0.01), and in females descended
17 from treated males during weeks 5 and 6 (p < 0.01). Again, no food consumption data were provided,
18 precluding a determination of whether these decreases in body weight were related to exposure.
19 Hemoglobin levels were similar in all groups. The relative weights of testes from FI males were
20 statistically increased in all exposure groups in males descended from treated females (p < 0.05 at 10
21 and 33 ppm, p < 0.01 at 100 ppm) and at 33 and 100 ppm in males descended from treated males (p <
22 0.05). FI females descended from treated males and exposed to 100 ppm chloroprene had significantly
23 increased liver (p < 0.01), ovary (p < 0.001), and lung (p < 0.05) weights. Gross and microscopic
24 histopathological examinations revealed no treatment-related abnormalities in these organ systems.
25 Given the lack of histopathological findings in any examined organ system, the significant increases in
26 lung, liver, and gonad weights in FI males and females are not considered to be adverse.
27 The NOAEL for this study was identified as 33 ppm based on decreases in body weight during
28 lactation in pups descended from treated females and untreated males.
4.4. OTHER DURATION- OR ENDPOINT-SPECIFIC STUDIES
4.4.1. Acute and Subchronic Studies
29 Clary et al. (1978) conducted a study to investigate chloroprene's acute and subchronic toxicity
30 and to determine the dose range for a 2 year chronic inhalation study (chronic study by Trochimowicz
31 et al., 1998) in rats and hamsters. Groups of six male albino rats (from Charles River laboratories)
32 were exposed to chloroprene by the dermal (200 mg/kg), oral (50 mg/kg), or inhalation (2 mg/L [-550
33 ppm]) routes for 4 hours and sacrificed for histological examinations 14 days after exposure. This
34 exposure protocol was referred to as a "modified Class B poison test" (extension of sacrifice from 2-
September 2009 4-50 DRAFT - DO NOT CITE OR QUOTE
-------
1 14 days after exposure). A lethal concentration test was also conducted by exposing male rats to 0,
2 530, 1,690, 2,280, 3,535, or 3,610 ppm (0, 146, 467, 630, 976, or 997 mg/m3). The approximate lethal
3 concentration by inhalation (4 hours) in rats was determined to be 2,280 ppm (Table 4-35). In the
4 4-week range-finding inhalation study, Wistar rats were exposed to chloroprene at 0, 50, 200, or 800
5 ppm (actual mean concentrations were 0, 39, 161, or 625 ppm [0, 11, 44, or 173 mg/m3], respectively).
6 A similar study was conducted (after completion of the 4-week rat study) with Syrian golden hamsters
7 exposed to 0, 40, 160, or 625 ppm (actual mean concentrations were 0, 39, 162, or 630 ppm [0, 11, 45,
8 or 174 mg/m3], respectively). The purity of chloroprene used in this study was 99.9% with 0.01%
9 phenothiazine added as a polymerization inhibitor. Test atmospheres were generated by low
10 temperature (0°C) vaporization in nitrogen.
Table 4-35. Chloroprene-induced mortality in male rats
CONCENTRATION (ppm)
530
1,690
2,280
3,535
3,610
MORTALITY (DEAD/TOTAL)
0/6
0/6
1/6
2/6
2/6
Source: Clary etal. (1978)
11 Clary et al. (1978) reported no deaths from dermal, oral, or inhalation administration in the
12 standard Class B poison test (sacrifice 2 days after the 4-hour exposure period). There were mild to
13 moderate skin irritation and erythema after the dermal exposure. Irregular respiration, mild
14 lacrimation, and slight initial weight loss were reported after the inhalation exposure. For the modified
15 Class B poison test (sacrifice 14 days after the 4-hour exposure period), 2/6 and 3/6 animals died on
16 the sixth and seventh days, respectively.
17 In the 4-week range-finding study, exposure to 625 ppm chloroprene was associated with eye
18 irritation, restlessness, lethargy, nasal discharge, and orange-colored urine in rats and hamsters. Hair
19 loss was observed in female rats exposed to the two highest exposure groups (161 and 625 ppm).
20 Increased mortality in rats was observed at the two highest concentrations starting in week 1 (5/10
21 males and 3/10 females died at 625 ppm; 3/10 males died at 161 ppm at the end of the exposure period,
22 4 weeks). Mortality was 100% for male and female hamsters in the highest dose group (630 ppm) by
23 the end of week 1, and 1/10 males and 3/10 females at the mid-exposure (162 ppm) by the end of week
24 4. 1 male hamster died in the low exposure (39 ppm) group by week 4. Decreases in body weight
25 were observed at all concentrations in rats and at 162 ppm in hamsters. There were changes in the
26 relative weights of all organs except for the heart. The relative organ weights for kidneys were
27 increased at the 162 ppm exposure level for both male and female hamsters, the 625 ppm level for
28 male rats, and the 161 and 625 ppm level for female rats. Liver weights were increased in the high
29 exposure group in both species except for female hamsters. Male rats exhibited decreased liver
September 2009 4-51 DRAFT - DO NOT CITE OR QUOTE
-------
1 weights at 39 and 161 ppm. Relative lung weights were increased at 625 ppm for male and female
2 rats. Clary et al. (1978) noted that these increases in the relative weight of the kidneys, liver, and lungs
3 may have indicated a direct effect of chloroprene exposure, whereas weight changes in other organs
4 (spleen, brain, thyroid, and adrenal glands) may have been secondary to decreases in body weight.
5 In rats, gross pathological examination of the animals that died during exposure revealed dark,
6 swollen livers and grayish lungs with hemorrhagic areas. Dark swollen livers were also observed in
7 several animals exposed to the highest concentration when they were sacrificed at the end of the study.
8 Microscopic examination revealed slight to severe centrilobular liver degeneration in all male rats and
9 in 8/10 of the females at the high concentration. This change was also observed in 2/3 male rats
10 exposed to 161 ppm that died during the study. The kidneys of male and female rats exposed to 625
11 ppm had enlarged tubular epithelial cells. In addition, one male and one female rat exposed to 625
12 ppm showed foci of necrotic tubules in the intramedullary area of the kidneys.
13 In hamsters, the lungs of most of the animals that died within the first 24 hours of exposure (all
14 animals died after a single exposure to 630 ppm and 1/10 males and 1/10 females at 162 ppm) showed
15 gray-reddish edematous areas. Fecal and urinary incontinence were observed in 1/10 male and 3/10
16 females at 630 ppm. The heart of 1/2 females that died on the second day of exposure was pale with
17 severe myocarditis, and the thoracic cavity contained a considerable amount of fluid. The other female
18 had a small spleen and a pale liver with a pronounced lobular pattern. Significant body weight
19 decreases were observed only in the 162 ppm group. Histopathology examinations revealed necrosis
20 and midzonal degeneration of hepatocytes in most of the survivors of the 162 ppm group. Several
21 males and females (number not specified) exposed to either 39 or 162 ppm showed irritation of the
22 mucous membranes of the nasal cavity. This irritation was described as a slight flattening and thinning
23 of the layer of the olfactory epithelium in the dorsomedial part of the cavity.
4.4.2. Immunotoxicity
24 There are some laboratory animal data suggesting potential immunomodulatory effects in of
25 chloroprene; however the data are from standard toxicological studies and no targeted
26 immunotoxicological studies of chloroprene were identified. The studies discussed below were
27 described in detail previously in the assessment and only the relevant immune data are presented here.
28 NTP (1998) observed that thymus weights in adult male and female B6C3F1 mice exposed to 80 ppm
29 chloroprene for 16 days were significantly decreased compared to controls (p < 0.01) and thymic
30 necrosis, characterized by karyorrhexis of thymic lymphocytes, was observed in both sexes at 200
31 ppm. No changes in thymus weight or histopathology were reported in mice after chloroprene
32 exposure for a longer period (i.e., 13-week exposure) as part of the same NTP (1998) study.
33 Alterations in differential white blood cell counts (i.e., increased leukocyte, neutrophil, and monocyte
34 numbers) were observed at 500 ppm in male rats after 16 days exposure and segment neutrophils were
35 decreased in male rats at 200 ppm after 13 weeks of exposure. In the 2-year chronic portion of the
36 NTP study, splenic hematopoietic cell proliferation was significantly increased over controls in male
September 2009 4-52 DRAFT - DO NOT CITE OR QUOTE
-------
1 mice at 32 and 80 ppm, and in all exposed females (level of significance not reported). Hyperplasia of
2 the mediastinal lymph node was observed in females exposed to 32 or 80 ppm (significance not
3 stated).
4 Trochimowicz et al. (1998) observed that mean relative spleen and thymus weights were
5 significantly (p < 0.01) lower in female Wistar rats exposed to 50 ppm chloroprene for 2 years, but did
6 not report any accompanying histopathological changes in either organ. Clary et al. (1978) also
7 observed small spleens in hamsters (qualitative description) and decreased spleen weights (possibly
8 secondary to decreased body weights) in rats exposed to 625-630 ppm chloroprene for 4 weeks.
9 Sanotskii (1976) reported that chromosomal aberrations were observed in the bone marrow of mice
10 exposed to chloroprene and in leukocyte cultures of exposed chloroprene production workers.
11 These findings provide some evidence of immunomodulatory effects of chloroprene in
12 laboratory animals. The immune-related data for chloroprene include altered lymphoid organ weights
13 and histopathology, and chromosomal aberrations in bone marrow. However, it has been shown that
14 changes in lymphoid organ weights and genotoxicity observed in lymphoid organs are both poor
15 predictors of compound-related changes in immune function (Luster et al., 1992). The changes in
16 thymic histopathology reported after 16 days of exposure were not observed with longer exposure,
17 suggesting no chronic effects. The remaining data on increased hematopoietic cell proliferation and
18 lymph node hyperplasia are nonspecific effects that are difficult to interpret as potential immuntoxicity
19 of chloroprene. They may be related to general hematopoietic effects of chloroprene rather than an
20 effect on the immune system or immune function. In general, measures such as these (i.e.,
21 morphological disturbances) are not clear measures of a chemical's potential to cause changes in
22 immune function (Putman et al., 2003). Direct measures of immune function, such as antibody
23 production to a T-cell dependent antigen, are usually preferred to delineate a chemical's immunotoxic
24 potential (Luster et al., 1992; Putman et al., 2003).
4.5. MECHANISTIC DATA AND OTHER STUDIES IN SUPPORT OF MODE OF ACTION
4.5.1. Mode-of-Action Studies
25 Many of the available studies addressing the mode of action (MO A) of chloroprene have
26 focused on investigating the metabolic profile for chloroprene including identifying epoxide
27 metabolites, their reactivity with DNA, and adduct formation in vitro (Munter et al., 2002; Hurst and
28 Ali, 2007 ). Other studies have used molecular analysis to study alterations in ras proto- oncogenes
29 from lung and Harderian gland tumors identified in the NTP (1998) chronic bioassay that may indicate
30 events in chloroprene-induced neoplasia (Ton et al., 2007; Sills et al., 1999).
31 The metabolism of chloroprene into reactive epoxides has been primarily evaluated in vitro
32 with liver and lung tissue fractions from rat, mouse, hamster, and humans. Only a limited number of
33 studies have investigated the in vivo metabolism of chloroprene. In studies using mouse and human
September 2009 4-53 DRAFT - DO NOT CITE OR QUOTE
-------
1 liver microsomes, Bartsch et al. (1979) showed that 2-chloro-2-ethynyloxirane and/or (1-
2 chloroethenyl)oxirane could be intermediates in the biotransformation of chloroprene. Metabolism of
3 chloroprene into (l-chloroethenyl)oxirane was confirmed by Himmelstein et al. (2001b); oxidation of
4 chloroprene to (l-chloroethenyl)oxirane was evident in rodent and human liver microsomes and most
5 likely involved CYP2E1, as evidenced by the near complete in vitro inhibition with 4-methylpryazole.
6 A comparison across species suggested that a greater amount of (l-chloroethenyl)oxirane was present
7 in B6C3F1 mice and F344 rat liver microsomes, followed by the Wistar rat, then humans and
8 hamsters. A maximum concentration of (l-chloroethenyl)oxirane of 0.01-0.02 jiM was detected in
9 mouse liver microsomes between 5-10 minutes after initiation of exposure with 0.05 jiM chloroprene.
10 Preliminary data also showed that hydrolysis of (l-chloroethenyl)oxirane was slowest in the liver
11 microsomes of B6C3F1 mice. Further comparing metabolism between species, Cottrell et al. (2001)
12 observed that qualitative profiles of metabolites from liver microsomes obtained from B6C3F1 mice,
13 Sprague-Dawley or F344 rats, and humans were similar, with (l-chloroethenyl)oxirane being the
14 major metabolite in all species and genders. Himmelstein et al. (2004a) developed a two-compartment
15 closed vial model to describe both chloroprene and (l-chloroethenyl)oxirane metabolism in liver and
16 lung fractions from rat (two strains, F344 and Wistar), mouse, hamster, and humans. Estimates for
17 Vmax and Km for oxidation of chloroprene (into (l-chloroethenyl)oxirane) in liver microsomes ranged
18 from 0.068-0.29 |imol/hour/mg protein and 0.53-1.33 jiM, respectively. Oxidation (Vmax/Km) of
19 chloroprene in the liver was slightly faster in the mouse and hamster than in rats or humans. In lung
20 microsomes, Vmax/Km was much greater for mice compared with the other species. Conversely,
21 hydrolysis (Vmax/Km) of (l-chloroethenyl)oxirane in liver and lung microsomes was faster for the
22 human and hamster, than for rat or mouse. The observation that mice generally metabolized
23 chloroprene into its epoxide metabolite at equal or faster rates than other species and hydrolyzed the
24 epoxide more slowly may, in part, explain why mice were observed to be the most sensitive species in
25 regards to chloroprene's observed carcinogenicity.
26 The in vivo rodent studies support the postulated metabolic pathway for chloroprene. For
27 example, male Wistar rats administered 100 or 200 mg/kg chloroprene by gavage demonstrated a rapid
28 depletion of hepatic GSH and a dose-dependent increase in excreted urinary thioethers (presumably
29 GSH-conjugates), which is consistent with in vitro studies using isolated liver hepatocytes (Summer
30 and Greim, 1980). Pretreatment of rats or hepatocytes with phenobarbital or a poly chlorinated
31 biphenyl (PCB) mixture (Clophen A50) to induce the mixed-function oxidase enzymes enhanced the
32 GSH depletion effect.
33 Munter et al. (2002) investigated the reactivity of the chloroprene metabolite
34 (l-chloroethenyl)oxirane towards DNA nucleosides and calf thymus DNA. The adducts were isolated
35 by reverse-phase chromatography and characterized by their mass spectrometric features. The reaction
36 of (l-chloroethenyl)oxirane with the nucleoside 2'-deoxyguanosine yielded one major adduct derived
37 by nucleophilic attack of N-7 guanine on C-3' of the epoxide. In addition, another chloroprene
September 2009 4-54 DRAFT - DO NOT CITE OR QUOTE
-------
1 metabolite 2-chlorobut-2-en-l-al (See Figure 1, metabolite labeled as number 15) described as an
2 unsaturated aldehyde, yielded 2 major adducts. The reaction of (l-chloroethenyl)oxirane with double
3 stranded calf thymus DNAyielded N7-(3-chloro-2-hydroxy-3-buten-l-yl)-guanine (dGI) as the major
4 adduct, the same adduct seen when the chloroprene metabolite was incubated with 2'-deoxyguanosine
5 individually. N3-(3-chloro-2-hydroxy-b-buten-l-yl)-2'-deoxyuridine (dCI) was also detected. The
6 reaction of (l-chloroethenyl)oxirane with deoxycytidine in DNA may be significant because such
7 adducts are difficult to repair and may therefore be implicated in mutagenesis (Koskinen et al., 2000).
8 The in vitro reactivity of (l-chloroethenyl)oxirane with hemoglobin (adduct formation) and
9 enantiomer detoxification (i.e., disappearance of R- vs. S-enantiomer from the test system) in vitro
10 have been investigated by Hurst and Ali (2007). Mouse (C57BL/6) erythrocytes (RBCs) were
11 incubated with the R- and S-enantiomers of (l-chloroethenyl)oxirane in vitro. The authors reported a
12 greater persistence of the R- over the S-enantiomer upon incubation with RBCs in the in vitro system
13 tested. The authors also reported a greater amount of globin adducts formed with the R- than with the
14 S-enantiomer.
15 As part of the 2-year bioassay of chloroprene, NTP (1998) evaluated possible oncogene-
16 activating mechanisms for lung and Harderian gland neoplasms in the B6C3F1 mouse at 0, 12.8, 32,
17 and 80 ppm. The results were published by Sills et al. (1999). After isolation and amplification of
18 DNA from the neoplasms, H-ras and K-ras mutations were identified. A higher frequency (80%) of K-
19 ras mutations was detected in chloroprene-induced lung neoplasms than in spontaneous neoplasms of
20 control mice (30%). The predominant mutation was an A->T transversion (CAA->CTA) at K-ras
21 codon 61: 80% (8/10) of low dose, 71% (10/14) of mid dose, and 18% (4/22) of high dose lung tumors
22 were observed to have this mutation). This specific mutation was not observed in spontaneously
23 occurring lung neoplasms. A similar pattern of ras mutations was observed also with isoprene-induced
24 lung neoplasms but not in those induced by butadiene. Rare point mutations, not seen in spontaneous
25 lung neoplasms, were detected at codon 12. No consistent morphological pattern (papillary, solid, or
26 mixed) or type (benign or malignant) of neoplasm was co-observed with specific K-ras mutations.
27 Although definitve evidence is currently unavailable, there are a number of factors that may explain
28 the observation of the lower frequency of codon 61 CTA transversions in lung tumors of high dose
29 animals. In the lung, the lower frequencies in CTA transversions at high doses may be due to non-ras
30 mutation mechanisms of genotoxicity or carcinogenicity. Alternatively, differences in DNA-adduct
31 formation or induction of repair or removal mechanisms may explain the pattern observed.
32 A high incidence (100%) of both K-ras and H-ras mutations was detected in chloroprene-
33 induced Harderian gland neoplasms, compared with 56% in spontaneous Harderian gland tumors in
34 control mice, 100% in neoplasms from isoprene-exposed mice, or 69% in neoplasms from butadiene-
35 exposed mice. The predominant mutation was also a CAA->CTA transversion at K-ras codon 61
36 (93%), which only occurred in 7% (2/27) spontaneously occurring Harderian gland neoplasms. The
37 concentration-response was similar across exposure groups. It was suggested that the large number of
September 2009 4-55 DRAFT - DO NOT CITE OR QUOTE
-------
1 ras mutations at A:T base pairs after exposure to chloroprene, isoprene, or butadiene indicated an
2 interaction with DNA to form adenine adducts that may be important for tumor induction. Sills et al.
3 (2001) reported higher frequencies of K- and H-ras mutations (57%) in chloroprene-induced
4 forestomach tumors in B6C3F1 mice compared to spontaneous tumors (36%). The A->T transversion
5 (CAA->CTA) in H-ras codon 61 was identified in 29% of the chemically induced forestomach
6 neoplasms, but was not observed in spontaneous control tumors. Mutations at K-ras codon 61 were
7 not observed in chloroprene-induced forestomach tumors.
8 Ton et al. (2007) evaluated mutations in the K-ras oncogenes and loss of heterozygosity in the
9 region of K-ras on distal chromosome 6 in lung tumor samples collected from mice exposed to
10 chloroprene in the NTP 2-year inhalation study. DNA analysis included isolation from formalin fixed
11 tissue sections, and amplification, cycle sequencing of ras gene and analysis for loss of heterozygosity
12 (LOH). Chloroprene-induced mouse lung tumors had a high frequency of LOH on chromosome 6 in
13 the region of K-ras. The correlation between K-ras mutation and loss of the wildtype allele was high in
14 the tumors examined: of the 19 lung tumors with LOH from B6C3F1 mice exposed to chloroprene, 16
15 (84%) of them also had K-ras mutations.
4.5.2. Genotoxicity Studies
16 This section presents the findings of several genotoxicity studies that are summarized in Table
17 4-36.
Table 4-36. Genotoxicity assays of chloroprene
TEST SYSTEM
CELLS/STRAIN
TESTED
CONCENTRATIONS
RESULTS3
REFERENCE
Bacterial assays
Salmonella
typhimurium
TA100
TA100, TA1535
TA98
TA100, TA1535
TA98, TA1537, TA1538
TA100, TA1535, TA1537,
TA98
TA100
TA100
TA100, TA1535, TA97A,
TA98
0.5 to 8% (volume/
volume ) in air
10,000-40,000 ppm
10,000-40,000 ppm
up to 3,333 ug/plate
0-5 umol/plate
0-5 umol/plateb
0 to 69 mMc
+
+
-
+
-
—
+
+
Bartschetal. (1979)
Willems (1980)
Willems (1980)
Willems (1978)
Willems (1978)
NTP (1998)
Westphal et al. (1994)
Westphal et al. (1994)
Himmelstein et al. (200 la)
Mammalian cell assays
Micronucleus
Micronucleus
Chinese hamster V79
Chinese hamster V79
10% (v/v)
0.175mMa
-
-
Drevon and Kuroki (1979)
Himmelstein et al. (200 la)
September 2009
4-56
DRAFT - DO NOT CITE OR QUOTE
-------
TEST SYSTEM
CELLS/STRAIN
TESTED
CONCENTRATIONS
RESULTS3
REFERENCE
In vivo bioassays
Sex-linked recessive
lethal mutation
Sex-linked recessive
lethal mutation
Sister chromatid
exchange: bone
marrow
Chromosomal
aberration: bone
marrow
Chromosomal
aberration: bone
marrow
Micronucleus:
peripheral blood
Micronucleus: bone
marrow
Drosophila (Canton-S)
Drosophila (Berlin-K)
B6C3F1 mice
B6C3F1 mice
C57BL/6 mice
B6C3F1 mice
B6C3F1 mice
12.8, 32, 80 ppm
12.8, 32, 80 ppm
up to 1 ppm
12.8, 32, 80 ppm
—
+
+
—
—
Fouremanetal. (1994)
Vogel (1979)
NTP (1998); Shelby (1990);
Tice (1988a, 1988b)
NTP (1998)
Sanotskii (1976)
NTP (1998)
Shelby and Witt (1995)
a For bacterial assays, tests were performed in the absence or presence of the exogenous S9 metabolism system. In all cases
of positive mutagenicity (except Westphal et al., 1994), addition of S9 mixture enhanced the observed mutagenicity
bAged chloroprene distillates tested (in the absence of the exogenous S9 metabolism system).
0 Epoxide metabolite (l-chloroethenyl)oxirane tested .
4.5.2.1. Bacterial Mutagenicity Assays
1 Both positive and nonpositive mutagenic responses have been observed in bacterial mutagenic
2 assays.
3 Bartsch et al. (1979) exposed Salmonella typhimurium strain TA100 to 0.5-8%
4 (volume/volume [v/v]) of chloroprene within sealed desiccators for 4 hours at 37°C in the absence or
5 presence of the exogenous S9 metabolism system. Batch solutions were freshly prepared before use
6 and kept at-20°C. Chloroprene purity was 99% and contained a negligible amount of dimers. A
7 positive mutagenic response that was concentration-dependent was observed without S9 fraction; this
8 response increased threefold when S9 fractions from either phenobarbital-pretreated or untreated mice
9 were used.
10 Willems (1978, 1980) found that chloroprene (purity not stated, but sample was "freshly
11 supplied") was mutagenic with S. typhimurium strains TA100 and TA1535 in the presence or absence
12 of S9 (mutagenicity was more pronounced in the presence of the S9 fraction), indicating base pair
13 substitution mutations. Chloroprene, however, was not mutagenic in S. typhimurium strains TA98,
14 TA1537, and TA1538 indicating a lack of frameshift mutations. Petri plates were incubated at 37°C in
15 desiccators for either 48 or 24 hours, removed, and then incubated for another 24 hours. Positive
16 controls were used. Four dimers (chemical characterization not stated) were also tested under the same
17 conditions. Three of the four were mutagenic against both salmonella base pair substitution strains
18 (TA100andTA1535).
19 Westphal et al. (1994) investigated the mutagenicity of chloroprene with respect to the
20 compound stability and reactivity with solvents used in the test system. The Ames test was performed
September 2009 4-57 DRAFT - DO NOT CITE OR QUOTE
-------
1 using the S. typhimurium (strain TA100) with or without S9, in gas-tight chambers to prevent
2 chloroprene volatilization. Chloroprene was freshly distilled from a 50% xylene solution. The
3 distillates were stored at -20°C and checked for purity immediately before testing. The authors noted
4 that 2-5% xylenes remained in the chloroprene distillates. Another set of distillates were prepared in
5 the same manner and stored either under air or under argon and kept at room temperature (referred to
6 as aging) for 1, 2, or 3 days. Chromatographic analysis of the aged chloroprene revealed the presence
7 of decomposition products reported to be cyclic dimers. The influence of solvents was also tested in
8 this study by using either ethanol or dimethyl sulfoxide (DMSO) as vehicles. Propylene oxide (a
9 volatile direct mutagen) and benzo(a)pyrene were used as positive controls.
10 Freshly distilled chloroprene dissolved in either DMSO or ethanol as vehicles, with or without
11 S9, was not mutagenic in TA100. Aged chloroprene had a mutagenic effect on TA100 that increased
12 linearly with increasing age of the chloroprene distillates. Westphal et al. (1994) confirmed these
13 findings by obtaining positive results with 10 additional distillates containing different proportions
14 (quantitative details not specified) of the decomposition products, without S9. The mutagenicity of the
15 distillates correlated with the proportion of the decomposition products (which increased over time in
16 the aged samples). The mutagenicity of aged chloroprene towards TA100 was the same whether
17 chloroprene was stored under air or under an inert gas. The authors speculated that the mutagenic
18 products in aged chloroprene were less volatile than those in the fresh distillates, thus remaining in the
19 test medium long enough to cause toxicity.
20 Addition of GSH, both with and without S9, reduced the mutagenicity of aged chloroprene but
21 was less effective as the amount of decomposition products increased. Westphal et al. (1994) stated
22 that chloroprene diluted in DMSO was markedly more toxic and more mutagenic than chloroprene
23 dissolved in ethanol, although no data were provided to support this statement.
24 Chloroprene did not show any evidence of mutagenicity in any of four strains of S.
25 typhimurium (TA98, TA100, TA1535, or TA1537) tested at concentrations up to 3,333 jig/plate, in the
26 presence or absence of aroclor-induced rat or hamster liver S9 fraction (NTP, 1998).
27 Chloroprene monoepoxide, (l-chloroethenyl)oxirane (> 98% purity), was found to be
28 mutagenic in salmonella strains TA100 and TA1535. Some activity was also observed in strains
29 TA97A and TA98 without Aroclor-induced S9 activation (Himmelstein et al., 2001a); inclusion of S9
30 had no effect in the mutagenic response in all tester strains. Test concentrations were
31 0-69 mM in DMSO. Toxicity was noted at > 14 mM in plates without S9 and at > 34 mM in plates
32 with S9.
4.5.2.2. Mammalian Cell Assays
33 Chloroprene (99% pure) was evaluated for mutagenic potential in V79 Chinese hamster cells in
34 the presence of a liver supernatant (SI5 fraction) from phenobarbitone-pretreated rats and mice
35 (Drevon and Kuroki, 1979). Cells were incubated at 37°C for 5 hours or longer in 2.5 mL of reaction
September 2009 4-58 DRAFT - DO NOT CITE OR QUOTE
-------
1 mixture with or without S15 fraction from mice pretreated with phenobarbitone, plus cofactors, either
2 in liquid suspension or in 0.3 % agar. The petri dishes were placed in a desiccator and exposed to 0,
3 0.2, 1, 2, and 10% (v/v) chloroprene vapors for 5 hours. Toxicity was evaluated as a measure of
4 plating efficiency. Mutations were evaluated in terms of resistance to a purine analogue (8-
5 azaguanine) and ouabain (inhibitor of adenosine triphosphatase in cell membranes). Chloroprene
6 toxicity was observed at concentrations above 1%; this effect was enhanced with addition of the S15
7 fraction. The authors noted that this suggested the formation of atoxic metabolite. No mutations were
8 observed in the absence or presence of SI5.
9 Himmelstein et al. (2001a) evaluated the clastogenic potential of the (l-chloroethenyl)oxirane
10 (> 98% purity) using the cytochalasin-B blocked micronucleus test in Chinese hamster V79 cells
11 without metabolic activation. The V79 cells plated on tissue culture slides were placed inside sterile
12 bottles filled with culture medium followed by injection of
13 0-0.943 mM (l-chloroethenyl)oxirane dissolved in DMSO into the bottles and incubation for 3 hours.
14 Cells were then transferred to fresh medium containing cytochalasin-B and incubated for an additional
15 16 hours. A minimum of 500 binucleated cells were scored for micronuclei. Cytotoxicity, reported as
16 a reduction in the number of binucleated cells, and altered cell morphology were observed starting at
17 0.175 mM. No clastogenic response was noted at concentrations up to 0.175 mM.
4.5.2.3. In Vivo Bioassays
18 Vogel (1979) evaluated chloroprene (99% pure with negligible dimer content) dissolved in
19 DMSO (final DMSO concentration = 1%) in an experiment for induction of recessive lethal mutations
20 on the X chromosome of male Drosophila melanogaster (Berlin-K). Storage conditions and the
21 elapsed time between receipt and use were not reported. After mating, the F3 generation was evaluated
22 for recessive lethality. The increase in the percentage of observed recessive-lethal mutations was
23 marginal in several experiments and was not concentration dependent. However, when the data from
24 pooled samples from several experiments (53 lethals in 15,941 X chromosomes) were compared with
25 seven control experiments, the difference was statistically significant at p < 0.01. The authors noted
26 that the possible variation among samples could be related to the instability of chloroprene.
27 In a study by Foureman et al. (1994), chloroprene (purity not reported) dissolved in ethanol
28 was nonpositive (p > 0.01) for sex-linked recessive lethal mutations in postmeiotic and meiotic germ
29 cells of adult male D. melanogaster (Canton-S) when exposed by either the injection or feeding route.
30 The investigators suggested that the discrepancy between their nonpositive findings and those of Vogel
31 (1979) may be due to (1) differences in purity of the chloroprene sample, (2) differences between the
32 Berlin-K and Canton-S strains, (3) differences in sample sizes, and (4) possible genetic drift within the
33 female populations used by the two groups of investigators. Another possibility for the conflicting
34 results could be that chloroprene in ethanol is less genotoxic than if dissolved in DMSO (Westphal et
35 al., 1994; Gahlmann, 1993).
September 2009 4-59 DRAFT - DO NOT CITE OR QUOTE
-------
1 Cytogenetic tests using chloroprene were nonpositive. In studies performed by Brookhaven
2 National Laboratories for the NTP (1998), sister chromatid exchanges and chromosomal aberrations
3 (bone marrow cells) and the frequency of micronuclei in peripheral blood erythrocytes were evaluated
4 in male mice exposed by inhalation to chloroprene in the NTP (1998) bioassay. Results were
5 published separately by Shelby (1990), Tice (1988), and Tice et al (1988). Mice were exposed by
6 inhalation to chloroprene at 0, 12.8, 32, 80, or 200 ppm (0, 3.5, 8.8, 22, or 55 mg/m3) 6 hours/day for
7 12 days. Mortality was 100% at 200 ppm. There were no exposure-related effects compared with
8 controls in numbers of sister chromatid exchanges, chromosomal aberrations, or micronucleus
9 frequency in polychromatic or normochromatic erythrocytes. Tice (1988) and Tice et al (1988) did
10 report that the mitotic index (frequency of cells in metaphase) in mouse bone marrow cells was
11 elevated in chloroprene-exposed animals, with the increase being significant in the 80 ppm group.
12 Tice (1988), and Tice et al (1988) suggested that the lack of chloroprene-induced genotoxicity in bone
13 marrow may imply that any carcinogenic activity attributable to chloroprene would likely be localized
14 to tissues directly exposed to chloroprene (e.g., lung) or to tissues with a high metabolic activity that
15 form reactive intermediates.
16 The frequency of micronucleated cells in peripheral blood erythrocytes was not affected when
17 mice were exposed to chloroprene for 13 weeks to 0, 12.8, 32, or 80 ppm (0, 3.5, 8.8, or 22 mg/m3)
18 (NTP, 1998; MacGregor et al., 1990).
19 Sanotskii (1976) reported on a study identifying an increase in chromosomal aberrations in
20 bone marrow cells of mice exposed for 2 months to chloroprene concentrations of 3.5 mg/m3 (1 ppm)
21 and below. The protocol details and information about the purity and storage of chloroprene were not
22 provided.
23 Shelby and Witt (1995) found nonpositive results in vivo in the mouse bone marrow
24 micronucleus test and in chromosomal aberration tests when male B6C3F1 mice were injected
25 intraperitoneally with chloroprene in corn oil, three times, at 24-hour intervals. Dose levels, protocol
26 details, and information about the purity and storage of chloroprene were not provided.
27 Chloroprene was also tested in a dominant lethal assay with male Swiss mice (Immels and
28 Willems, 1978). Groups of 12 males were exposed to 0, 10, or 100 ppm (0, 2.8, or 28 mg/m3)
29 chloroprene 6 hours/day, 5 days/week for 2 weeks. Immediately after exposure, each male was mated
30 with two virgin females for seven days. Females were replaced each week for 8 weeks. There was no
31 sign of dominant lethal mutations or effects on mating performance or fertility.
4.5.3. Structural Alerts
32 Chloroprene is the 2-chloro analog of 1,3-butadiene, a multiorgan, cross-species carcinogen,
33 and is structurally similar to isoprene (2-methyl-l,3-butadiene). Inhalation studies have demonstrated
34 that, similar to butadiene and isoprene, chloroprene is a multisite carcinogen in rats and mice.
35 Butadiene and isoprene are both metabolized to epoxides and diepoxides that are known mutagens and
September 2009 4-60 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
9
10
11
12
13
14
15
are believed to be responsible for their carcinogenicity. Chloroprene is also metabolized to an epoxide
intermediate that may mediate its carcinogenic effects; however, there is no evidence of diepoxide
formation in the metabolism of chloroprene. The similarities in the sites of tumor induction in rodents
(see Table 4-37) between butadiene, isoprene, and chloroprene provide further evidence for a similar
MO A for these epoxide-forming compounds. A comparative report of the carcinogenicity of these
compounds highlights the qualitative and quantitative concordance of their tumorigenic effects
(Melnick and Sills, 2001). The female mouse lung was the most sensitive site of carcinogenicity for
both chloroprene and butadiene.
Table 4-37. Sites of increased incidences of neoplasms in the 2 year inhalation
studies of 1,3-butadiene, isoprene, and chloroprene in rats and mice
Site
Lymphatic/
hematopoietic
Circulatory
Lung
Liver
Forestomach
Harderian
gland
Mammary
gland
Brain
Thyroid
Pancreas
Testis
Zymbal's
gland
Kidney
Oral Cavity
Mice
Butadiene
M,Fa
M, F
M. F
M,F
M,F
M, F
F
M
Isoprene
M
M
M
M
M
M, F
Chloroprene
M, F
M, F
F
M,F
M, F
F
F
M
Rats
Butadiene
F
M
F
M
M
F
Isoprene
M, F
M
M
Chloroprene
M
F
M,F
M, F
M,F
a M = males ; F = females
Source: NTP (1998); Melnick et al. (1994); Placke et al. (1996); U.S. EPA (2002b)
4.6. SYNTHESIS OF MAJOR NONCANCER EFFECTS
There is a limited body of information on the toxicological consequences to humans who are
exposed to chloroprene. In a summary by Nystrom (1948), chloroprene was reported to cause
respiratory, eye, and skin irritation, chest pains, temporary hair loss, dizziness, insomnia headache, and
fatigue in occupationally exposed workers. Chest pains accompanied by tachycardia and dyspnea
were also reported. In a Russian review (Sanotskii, 1976) of the effects of chloroprene, medical
examinations of chloroprene production workers revealed changes in the nervous system (lengthening
of sensorimotor response to visual cues and increased olfactory thresholds), cardiovascular system
September 2009
4-61
DRAFT - DO NOT CITE OR QUOTE
-------
1 (muffled heart sounds, reduced arterial pressure, and tachycardia), and hematology (reduction in RBC
2 counts, decreased hemoglobin levels, erythrocytopenia, leucopenia, and thrombocytopenia). The
3 ambient concentration of chloroprene in work areas ranged from 1-7 mg/m3 (3.6-25 ppm).
4.6.1. Animal Studies
4.6.1.1. Oral Exposure
4 Chloroprene's toxic potential by the oral route has been assessed in only one study
5 (Ponomarkov and Tomatis, 1980). This was a reproductive study involving exposure of BDIV rats to a
6 single dose (100 mg/kg) of chloroprene on the 17th day of pregnancy and of their progeny to weekly
7 doses (50 mg/kg) for 120 weeks. Animals treated with chloroprene that died within the first 30 weeks
8 of treatment showed severe congestion of the lungs and kidneys.
4.6.1.2. Inhalation Exposure
9 The database for inhalation toxicity studies in animals on chloroprene includes two range-
10 finding studies for 16 days and 13 weeks (NTP, 1998 [also reported by Melnick et al., 1999]), two
11 chronic inhalation bioassays (NTP, 1998 [also reported by Melnick et al., 1999]; Trochimowicz et al.,
12 1998) and four reproductive developmental studies (Mast et al., 1994: Culik et al., 1978; Appelman
13 and Dreef van der Meulen, 1979; Sanotskii, 1976). These studies associate chloroprene inhalation
14 exposure with respiratory, kidney, liver, spleen, forestomach, reproductive, and developmental effects.
15 Inhalation exposure for 16 days (first range-finding study for the NTP [1998] chronic bioassay)
16 to chloroprene was associated with a range of effects. In rats, there were increased mortality at the
17 high exposure concentration (500 ppm) and decreased body weight starting at 200 ppm. Minimal to
18 mild olfactory epithelial degeneration was observed in all exposed groups (males and females). Mild
19 to moderate centrilobular hepatocellular necrosis was observed in male and female rats exposed to 200
20 or 500 ppm. Hematological and clinical chemistry parameters indicated increased serum enzyme
21 (ALT, GDH, and SDH) activities, as well as anemia and thrombocytopenia (decreased platelet count)
22 in the 200 and 500 ppm groups on day 4 only. In females, significant increases in kidney weights
23 (right kidney only) were seen at 80 and 500 ppm, and significantly increased liver weights were seen at
24 200 and 500 ppm.
25 In mice, all male and female animals in the high-exposure (200 ppm) group died, exhibiting
26 signs of narcosis, hepatocellular and thymic necrosis, and hypertrophy of the myocardium.
27 Significantly decreased body weight gains were seen in males at 32 and 80 ppm. There were no other
28 clinical findings related to chloroprene exposure in the mouse.
29 In the second (13-week) range-finding study, inhalation to chloroprene was associated with a
30 range of effects across several organ systems. In rats, on day 2, minimal increases in hematocrit values,
31 hemoglobin concentrations, and erythrocyte counts occurred in males exposed to 32 ppm or higher and
September 2009 4-62 DRAFT - DO NOT CITE OR QUOTE
-------
1 in females exposed to 200 ppm. At week 13, male and female rats in the 200 ppm group demonstrated
2 decreased hematocrit values, decreased hemoglobin concentrations, and decreased erythrocyte counts
3 characterized as normocytic, normochromic anemia. Thrombocytopenia, evidenced by a reduction in
4 circulating platelet numbers, occurred in the male and female rats in the 200 ppm group on day 2 and
5 in the females at 80 and 200 ppm on day 22. Transient increases in platelet numbers were observed at
6 80 and 200 ppm in treated males and females. Transient increases in activities of serum enzymes
7 (ALT, GDH, and SDH) were also observed on day 22 in both sexes at 200 ppm. Alkaline phosphatase
8 enzymeuria was observed in males at > 32 ppm and in females at 200 ppm. In male rats proteinuria
9 was observed at 200 ppm. Reductions in liver NPSH levels were observed in both sexes of rats at 200
10 ppm. Nonprotein sulfhydryl concentrations were also reduced in the lungs of female rats at 200 ppm.
11 Increases in kidney weights were seen in both male and female rats at 200 ppm and in females at 80
12 ppm. In male rats, sperm motility was decreased at 200 ppm. Of the neurobehavioral parameters
13 tested, horizontal activity was increased in male rats exposed to > 32 ppm. Total activity was
14 increased in male rats at 32 and 200 ppm. There were no exposure-related effects on motor activity,
15 forelimb/hind-limb grip strength, or startle response.
16 Increased incidences of minimal to mild olfactory epithelial degeneration and respiratory
17 metaplasia occurred in male and female rats at 80 or 200 ppm. Olfactory epithelial degeneration was
18 observed in females at 32 ppm. The incidence of hepatocellular necrosis was increased in female rats
19 at 200 ppm. Variably sized aggregates of yellow or brown material, consistent with hemosiderin
20 accumulation, appeared in small vessels or lymphatics in or near portal triads or in Kupffer cells of
21 male and female rats exposed to 200 ppm.
22 In mice, the hematological changes were similar to those observed in rats; however, they were
23 less severe. Minimal anemia, including decreased hematocrit values, erythrocyte counts, and platelet
24 counts were observed in females at 32 and 80 ppm. Sperm morphology and vaginal cytology
25 parameters were similar to those of the chamber controls. Significantly increased incidences of
26 squamous epithelial hyperplasia of the forestomach were observed in male and female mice at 80 ppm.
27 Exposure to chloroprene for 2 years (NTP, 1998) was associated with effects to the respiratory
28 tract (lung and nose) in rats and mice. The forestomach was also a target for chloroprene-induced
29 effects in mice.
30 In rats, the incidences of atrophy, basal cell hyperplasia, metaplasia, and necrosis of the
31 olfactory epithelium in males and females were increased at 32 and 80 ppm; atrophy and necrosis were
32 also increased in males at 12.8 ppm. The incidence of chronic inflammation was increased in males
33 exposed to 12.8 ppm and greater and in females exposed to 80 ppm. The incidences of fibrosis and
34 adenomatous hyperplasia of the olfactory epithelium were increased in males and females at 80 ppm.
35 The incidence of alveolar/bronchiolar hyperplasia was statistically significantly increased in males and
36 females in every exposure group.
September 2009 4-63 DRAFT - DO NOT CITE OR QUOTE
-------
1 In mice, there was increased mortality in males at 32 or 80 ppm and in females at all
2 concentrations tested. A decrease in mean body weights was observed in females at 80 ppm. Increases
3 in the incidences of olfactory epithelial atrophy, adenomatous hyperplasia, and metaplasia were
4 observed in males and females at 80 ppm. An increase in the incidence of forestomach epithelial
5 hyperplasia was observed at 80 ppm. Bronchiolar hyperplasia was increased in males and females at
6 doses 12.8 ppm and greater, whereas pulmonary histiocytic cellular infiltration was increased in every
7 dose group in females only. Hematopoietic cell proliferation was increased in males at 12.8 ppm and
8 greater. Renal tubule hyperplasia was observed in males at 32 and 80 ppm.
9 In the study by Trochimowicz et al. (1998), the only remarkable nonneoplastic lesions in rats
10 were observed in liver and lungs. The number of rats with one or more small foci of cellular alteration
11 in the liver was higher in the 50 ppm group than in controls. In males, there was an increased
12 incidence of hepatocellular lesions described as one or several small clear cell foci in the 50 ppm
13 group. Mild changes, such as lymphoid aggregates around bronchi, bronchioles, and blood vessels,
14 were observed in males and females exposed to 50 ppm. Acute inflammatory processes in the lungs
15 were found in the 50 ppm group and control animals to a similar extent. In hamsters, the only
16 exposure-related effect was a generalized amyloidosis (liver, kidney, spleen, and adrenal glands) that
17 was lower in incidence in the 50 ppm exposed group compared with controls.
18 Culik et al. (1978) exposed pregnant rats by inhalation to chloroprene at 0, 1, 10, or 25 ppm 4
19 hours/day on days 1-12 (embryotoxicity study) or days 3-20 (teratology study). In the teratology
20 study, an increase in the percentage of litters with resorptions was observed at 10 and 25 ppm, with
21 only the change in the 10 ppm group achieving statistically significance (p < 0.05) relative to controls.
22 An increase in the percentage of litters with resorptions was not observed by Culik et al. (1978) in the
23 larger embryotoxicity portion of the study, which was specifically designed to detect such effects. The
24 equally high numbers of litters with resorptions (-50%) in all experimental groups, including controls,
25 in the embryotoxicity study correspond well to the level of response observed at 10 ppm and 25 ppm in
26 the teratology study (62% and 59%, respectively). The observation that the control rates of litters with
27 resorptions differ so much between the teratology and embryotoxicity portions of the Culik et al.
28 (1978) warrants further consideration. When the potential increase in resorptions is expressed in
29 numbers of resorbed fetuses per litter, the control group for the teratology study is the only exposure
30 group which falls far outside of the historical control range for this strain of rat (Charles Rivers
31 Laboratories, 1976). This suggests that the control group response in the teratology study may be a
32 statistical outlier and that the finding of a statistically significant increase of litters with resorptions at
33 10 ppm is spurious. Chloroprene exposure did result in statistically significant increases in average
34 body weight of fetuses in the 25 ppm group (p < 0.05) and in the length of fetuses from dams in the 10
35 and 25 ppm groups (p < 0.05). No major compound-induced or dose-related skeletal or soft tissue
36 anomalies were found.
September 2009 4-64 DRAFT - DO NOT CITE OR QUOTE
-------
1 Mast et al. (1994) exposed groups of 15-16 pregnant New Zealand white rabbits by inhalation
2 to 10, 40, or 175 ppm chloroprene (36.2 144.8, or 633.5 mg/m3) for 6 hours/day on gestational days
3 6-28. There were no signs of maternal toxicity due to exposure to chloroprene. A few dams in each
4 group exhibited nasal discharge, vaginal bleeding, and loose stools at various times during the
5 exposure period. The overall pregnancy rate was 89%, with a range of 80-94% for each exposure
6 group. The incidence of clinical signs of toxicity was low during the exposure, and dams appeared to
7 be in excellent health at termination. No exposure-related effects on maternal weight change were
8 noted. Exposure to chloroprene had no effect on the number of implantations, live pups, or
9 resorptions. Fetal body, liver, and kidney weights were not affected by exposure. The incidence of
10 fetal malformations was not affected by exposure to chloroprene.
11 Appelman and Dreef van der Meulen (1979, unpublished report) exposed two successive
12 generations (Fo or FI) of Wistar rats to 0, 10, 33, or 100 ppm (0, 36.2, 119.5, or 362 mg/m3)
13 chloroprene via inhalation. The FO generation was exposed for 13 weeks and then allowed to mate
14 with untreated animals. After weaning, the FI was exposed for an additional 10 weeks. There were
15 statistically significant decreases in body weight reported in the F0 animals, however, concurrent food
16 consumption data was not reported and the authors report that the observed decreases were most likely
17 related to inadequacies in food availability. There were no reported effects on resorptions, sex ratio,
18 or mortality during lactation, although FI pups from treated females had reduced body weights at birth
19 and weaning compared to controls. The exposed FI rats also had decreased body weights relative to
20 controls, but as with the FO animals, no concurrent food consumption data were available to assess
21 whether those decreases were treatment-related. Exposed FI males had significantly smaller testes and
22 females had larger ovaries, livers, and lungs compared to controls, but no histopathological changes
23 were observed in those organs.
24 The general lack of effects in the Mast et al. (1994), Culik et al. (1978) and Appelman and
25 Dreef van der Meulen (1979) studies are not consistent the many positive effects seen in previous
26 Russian studies reviewed in Sanotskii (1976). Similar to other conflicting results in the chloroprene
27 toxicity database, the differences in the results in the reproductive and developmental studies may be
28 attributed to the purity of the test substance, differences in the species and strains used, and
29 experimental design and parameters evaluated in the individual studies.
4.7. EVALUATION OF CARCINOGENICITY
30 Under the Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005a), there is evidence that
31 chloroprene is "likely to be carcinogenic to humans" based on (1) statistically significant and dose-
32 related information from an NTP (1998) chronic inhalation bioassay demonstrating the early
33 appearance of tumors, development of malignant tumors, and the occurrence of multiple tumors within
34 and across animal species; (2) evidence of an association between liver cancer risk and occupational
35 exposure to chloroprene; (3) some evidence of an association between lung cancer risk and
September 2009 4-65 DRAFT - DO NOT CITE OR QUOTE
-------
1 occupational exposure; (4) the proposed mutagenic mode of action; and (5) structural similarities
2 between chloroprene and known human carcinogens, butadiene and vinyl chloride (see Table 4-38).
3 U.S. EPA's Guidelines for Carcinogen Risk Assessment (2005a) indicate that for tumors
4 occurring at a site other than the initial point of contact, the weight of evidence for carcinogenic
5 potential may apply to all routes of exposure that have not been adequately tested at sufficient doses.
6 An exception occurs when there is convincing toxicokinetic data that absorption does not occur by
7 other routes. Information available on the carcinogenic effects of chloroprene via the inhalation route
8 demonstrates that tumors occur in tissues remote from the site of absorption. Information on the
9 carcinogenic effects of chloroprene via the oral and dermal routes in humans or animals is limited or
10 absent. Data regarding the absorption via any route of exposure are unavailable. However, based on
11 the observance of systemic tumors following inhalation exposure, and in the absence of information to
12 indicate otherwise, it is assumed that an internal dose will be achieved regardless of the route of
13 exposure. Therefore, chloroprene is considered "likely to be carcinogenic to humans" by all routes of
14 exposure.
4.7.1. Summary of Overall Weight of Evidence
15 According to NTP (1998), there is clear evidence of carcinogenicity in the F344/N rat and
16 B6C3F1 mouse due to lifetime inhalation exposure to chloroprene. In rats, increased incidences of
17 neoplastic lesions primarily occurred in the oral cavity and lung (males only), kidney, and mammary
18 gland (females). In mice, increased incidences in neoplasms occurred in the lungs, circulatory system
19 (all organs), Harderian gland, forestomach, liver, skin and mesentery (females only), and kidney
20 (males only).
21 In the current document, a total of nine studies covering 11 cohorts of human subjects exposed to
22 chloroprene were reviewed to assess the occurrence of cancer. The most consistent findings across the
23 database were excess cancers of the liver (Bulbulyan et al., 1999, 1998; Li et al., 1989; Leet and
24 Selevan, 1982) and lung (Marsh et al., 2007b; Colonna and Laydevant, 2001; Bulbulyan et al., 1999,
25 1998; Leet and Selevan, 1982; Pell, 1978). The epidemiologic evidence for increased lung cancer
26 mortality due to chloroprene exposures is limited. The few studies that reported increased risk were
27 not statistically significant. In addition to lack of a consistent association and the small increased risks
28 that were detected, other study limitations, such as lack of smoking data, limit the ability to determine
29 possible causal associations between lung cancer and humans exposed occupationally to chloroprene
30 based on the available data.
September 2009 4-66 DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-38. Summary of animal and human tumor data and weight of evidence
descriptor for chloroprene
Statistically Significant Tumor
Types
In male F344/n rats, increased incidence of kidney (renal tubule)
adenoma or carcinoma in all dose groups, and oral papilloma or
carcinoma and thyroid adenoma or carcinoma at the two highest
dose groups
In female F344/n rats, increased incidence of mammary
fibroadenoma at the two highest dose groups and oral papilloma or
carcinoma at the highest dose
In male B6C3F1 mice, increased incidence of lung adenoma or
carcinoma and hemangioma/hemangiosarcoma in all organs in all
dose groups, and harderian gland adenoma or carcinoma and
kidney (renal tubule) adenoma or carcinoma at the two highest dose
groups
In female male B6C3F1 mice, increased incidence of lung adenoma
or carcinoma and skin sarcoma in all dose groups, liver adenoma or
carcinoma at the two highest dose groups, and harderian gland
adenoma or carcinoma and mammary gland fibroadenomas at the
highest dose. Hemangiomas/hemangiosarcomas in all organs and
mesentery sarcomas were observed in the middle dose.
In humans, significant increases in liver cancer mortality were
observed in 4 occupational epidemiology studies (out of 9 total
studies). Relative risk estimates for liver cancer (while not
statistically significant) increased with increasing exposure,
indicating a dose-response trend.
Rare Tumors
Statistically significant increase in rare kidney (renal tubule)
adenoma in male rats and mice. Non-statistical increase in females
at the high dose.
Statistically significant increases in primary (assumed) liver cancer
in four cohort studies and lung cancer mortality in two studies in
workers occupationally exposed to chloroprene
Multiple Studies
• Animals-NTP( 1998)
• Humans - Leet and Sullivan (1982), Li et al. (1989), Bulbulyan et
al. (1989), and Bulbulyan et al. (1999)
Conclusions
Tumors in both sexes of rats and mice
Decreased time to tumor in both sexes of rats and mice
Tumors in occupationally exposed workers
Methodological limitations of the occupational epidemiology
studies (e.g., no available data for some potential confounders
which precluded adjustment, limited statistical power due to small
sample sizes, and lack of precise quantitative exposure
ascertainment) make it difficult to draw firm conclusions regarding
the human cancer data
Rare tumors (kidney renal tubule adenomas in animals, primary
liver cancer in humans)
Metabolites include DNA-reactive epoxides and a mutagenic mode
of action is proposed.
Weight of Evidence
characterization
Likely to be carcinogenic to humans
September 2009
4-67
DRAFT - DO NOT CITE OR QUOTE
-------
1 There was a statistically significant excess of liver cancers in four of the cohorts reviewed
2 (Bulbulyan et al., 1999, 1998; Li et al., 1989; Leet and Selevan, 1982), with a two- to more than five-
3 fold increased risk in the SMR seen among these studies. Although no statistically significant increase
4 in risk of liver cancer was detected in the most recent and comprehensive cohort study involving
5 workers at four plants (Marsh et al., 2007b), the observed RR increased with increasing cumulative
6 exposure in the plant with the highest exposure levels, indicating a dose-response trend. Similar to the
7 Colonna and Laydevant (2001) study, there was only one case of liver cancer in the other three plants
8 included in this study (Marsh et al., 2007a, 2007b). Limitations in the existing epidemiological
9 database included the lack of information on individual workers' habits (i.e., alcohol consumption)
10 needed to control for potential confounding, incomplete enumeration of incidence and mortality cases,
11 and potential for biases that may lead to an underestimation of the risk (e.g., the healthy worker effect).
12 These limitations are further discussed in Section 4.7.2.1.
4.7.2. Synthesis of Human, Animal, and Other Supporting Evidence
4.7.2.1. Human
13 A number of occupational cohort studies have examined cancer mortality and incidence among
14 workers exposed to chloroprene monomer and/or polychloroprene latex in the United States, Russia
15 (Moscow), Armenia, France, China, and Ireland (Marsh et al., 2007a, 2007b; Colonna and Laydevant,
16 2001; Bulbulyan et al., 1999, 1998; Romazini et al., 1992; Li et al., 1989; Leet and Selevan, 1982; and
17 Pell, 1978). Concern that exposure to chloroprene may result in liver cancer derives principally from
18 its structural similarity to vinyl chloride, a chemical known to cause liver angiosarcoma in humans.
19 Exposed workers have included those involved in chloroprene monomer production using both the
20 acetylene process in which exposure to vinyl chloride was possible and the more recent butadiene
21 process which does not involve vinyl chloride exposure. Other workers were involved with
22 handling/sampling of partially finished products such as polychloroprene latex which contains various
23 amounts of dissolved monomer. Some studies span eras in which little or no worker safety protection
24 measures were likely used in contrast with years in which process improvements and concern for
25 worker safety were gradually instituted. Therefore, it is difficult to compare results across studies
26 given a wide range of exposure variability within and between these cohorts.
27 Despite these differences in occupational exposure to chloroprene and other chemicals, four of
28 the cohorts with observed liver/biliary passage cancer cases showed statistically significant
29 associations (i.e., two- to five-fold increased risk) with chloroprene exposure. Four mortality studies
30 reported SMRs of 571, 482, 240, and 339 when compared to external populations (Bulbulyan et al.,
31 1999, 1998; Li et al, 1989; and Leet and Selevan, 1982). Although sample size and statistical power
32 were limited (thus limiting the precision of risk estimates), Bulbulyan et al. (1999, 1998) observed
September 2009 4-68 DRAFT - DO NOT CITE OR QUOTE
-------
1 significantly elevated relative risk estimates for liver cancer incidence and mortality among
2 intermediate and highly exposed workers. The study involving four plants by Marsh et al. (2007b),
3 which had the largest sample size and most extensive exposure assessment, also observed increased
4 relative risk estimates for liver cancer in relation to cumulative exposure in the plant with the highest
5 exposure levels (trend p-value = 0.09, RRs 1.0, 1.90, 5.10, and 3.33 across quartiles of exposure, based
6 on 17 total cases). Athough not statistically significant, these findings were comparable to results (RR
7 range: 2.9-7.1) detected in two other studies for high and intermediate cumulative exposures
8 (Bulbulyan et al., 1999, 1998). Though several studies noted higher SMRs for lung cancer among
9 workers exposed to chloroprene, the evidence was not considered as strong as liver cancer. This was
10 mostly due to the inability to adequately control for confounding by smoking status, a strong risk
11 factor for lung cancer. There was also no evidence of exposure-response relationship across various
12 chloroprene exposure categories.
13 One of the strengths of several of the more recent epidemiologic studies was improved
14 exposure assessment data. These studies utilized industrial hygiene information to determine which
15 areas or jobs were most likely to have received higher chloroprene exposures. This allowed for
16 examination of various exposure contrasts and helped reduce the potential for exposure
17 misclassification. As such, valid internal analyses were conducted which were less impacted by bias
18 due to the healthy worker effect. Despite these improvements, several study limitations added to the
19 uncertainty in addressing the weight of evidence of the epidemiologic data.
20 A key limitation of most of the chloroprene studies (and other occupational studies) is the
21 potential for bias due to the healthy worker effect. Although this may be less of a concern for cancer
22 mortality outcomes, SMR analyses are based on external comparisons to the general population and
23 will often result in reduced SMR values for the occupational cohort. Two studies with more advanced
24 chloroprene exposure assessment conducted internal analyses to reduce this source of bias (Bulbulyan
25 et al., 1999; Marsh et al., 2007b). Among these studies, only Bulbulyan et al. (1999) observed a
26 statistically significant association between chloroprene exposure and liver cancer mortality. As with
27 most epidemiological research, the potential for bias due to residual confounding is another limitation
28 that exists in these studies. With respect to liver cancer, the lack of data on alcohol consumption is a
29 key limitation which precludes its examination, although there is no direct evidence that alcohol is
30 related to the exposure of interest (i.e., chloroprene). Given the nature of the work environment for
31 most of the study participants in these occupational studies, there is also a high likelihood of co-
32 exposures which may be confounders. Despite this potential, there is little evidence of substantial
33 exposure to liver carcinogens in these populations. One study with data on a co-exposure (vinyl
34 chloride) reported evidence of negative confounding (Marsh et al., 2007b). This would result in an
35 underestimate of the reported association between chloroprene and liver cancer if adjusted for vinyl
36 chloride which suggests that this co-exposure was unlikely to explain the association observed
37 between chloroprene and liver cancer in that population.
September 2009 4-69 DRAFT - DO NOT CITE OR QUOTE
-------
1 An additional limitation in several studies was incomplete enumeration of both incident cases
2 and deaths. In some studies, there were many workers who were exposed during time periods when
3 chloroprene levels were relatively high who could not be identified or located for inclusion in the
4 studies. This raises the possibility that the actual number of liver cancer cases might have been higher
5 than indicated from the data on the subset of individuals that were included in the studies. Another
6 concern in these occupational studies is the reliance on death certificates for outcome ascertainment in
7 the mortality analyses. Although misclassification of cause of death can be minimized by the review of
8 medical records or by histological confirmation, this was not done in any of the studies. The lack of
9 histological review of the liver cancer cases is an important limitation of the available studies using
10 internal controls.
11 These epidemiologic study results, when examined in the context of different plant operating
12 and worker exposure conditions over different time periods and a low number of incident liver cancers,
13 offer evidence of an association for exposure to chloroprene with an increase of liver cancer in
14 humans. Despite various limitations (e.g., healthy worker bias, potential co-exposure, and incomplete
15 enumeration of cases), internal and external comparisons showed consistent evidence of an association
16 between chloroprene exposures and liver cancer. The associations detected in some studies add
17 support to the cancer weight of evidence determination.
4.7.2.1.1. Evidence for Causality
18 The evidence for causality for cancer from the human studies is summarized in the paragraphs
19 that follow and is based on recommendations from the EPA (2005a) guidelines for carcinogen risk
20 assessment. It should be noted that there exists a number of methodological limitations of the
21 epidemiologic studies that may preclude drawing firm conclusions regarding the following criteria.
22 These limitations include lack of control of personal confounders and risk factors associated with the
23 outcomes in question, imprecise exposure ascertainment resulting in crude exposure categories,
24 incorrect enumeration of cases leading to misclassification errors, limited sample sizes, and the healthy
25 worker effect.
26 Temporality - exposure must precede the effect for causal inference. Furthermore, and
27 particularly with cancers, exposure must precede the effect with a sufficient latency to be considered
28 causal. In all the occupational studies reviewed the chloroprene exposure has preceded effect (either
29 incidence of or mortality due to liver cancer) with sufficient latency to be considered causally
30 associated. Several of the studies have specifically evaluated latencies of 15 to 20 years (Marsh et al.,
31 2007a, 2007b; Colonna and Laydevant, 2001; Bulbulyan et al., 1998; and Pell, 1978).
32 Strength of Association - refers to the magnitude of measures of association such as the ratio
33 of incidence or mortality (e.g., SMRs, SIRs, RRs or odds ratios) irrespective of statistical significance.
34 Studies reporting large, precise risks are less likely to be doing so due to chance, bias, or confounding.
35 Reports of modest risk, however, do not preclude a causal association and may reflect lower levels of
September 2009 4-70 DRAFT - DO NOT CITE OR QUOTE
-------
1 exposure or an agent of lower potency. When compared to external populations, there was a
2 stasticially significant two- to five-fold increased risk of liver cancer in four cohort studies in China (Li
3 et al, 1989), United States (Leet and Selevan, 1982), Russia (Bulbulyan et al., 1998) and Armenia
4 (Bulbulyan et al., 1999) despite evidence of healthy worker effect bias. Despite relatively small
5 numbers, there were also suggestive data from the Louisville cohort (Marsh et al., 2007b) which found
6 RRs ranging from 1.9-5.1 (not statistically significant) for cumulative exposures to chloroprene and
7 liver cancer mortality. These data were consistent in magnitude to two other studies (Bulbulyan et al.
8 1999, 1998) examining intermediate and high cumulative exposures to chloroprene and liver cancer
9 incidence (RRs = 2.9-4.9, statistically signficant) and mortality (RRs = 4.4-7.1, not statistically
10 significant), respectively.
11 Consistency - the observation of the same site-specific effect across several independent study
12 populations strengthens an inference of causality. Four different studies have shown an association
13 between chloroprene exposure and liver cancer incidence and mortality (Bulbulyan et al. 1998, 1999;
14 Li et al., 1989; Leet and Selevan, 1982), while a fifth study showed evidence suggesting an association
15 when examined in relation to detailed exposure data (Marsh et al, 2007b). Larger effect estimates for
16 liver cancer risk have been observed in diverse populations working in chloroprene monomer and
17 polymer production, neoprene manufacturing, and manufacturing utilizing polychloroprene products in
18 the U.S., China, Armenia, and Russia. The studies with internal comparisons showed consistently
19 elevated liver cancer relative risk estimates for intermediate (RR range: 2.9-7.1) and high cumulative
20 risk exposures (Range: 3.3-4.9) as noted above.
21 Specificity - as originally intended, this refers to increased inference of causation if a single
22 site effect as opposed to multiple effects is observed and associated with exposure. Based on current
23 understanding, this is now considered one of the weaker guidelines for causality (for example, many
24 agents cause respiratory disease and respiratory disease has multiple causes). However, when
25 specificity of effect is found, it strengthens causal inference. Chloroprene exposure has been found to
26 be associated specifically with increased risk of liver cancer in four cohorts (Bulbulyan et al. 1998,
27 1999; Li et al., 1989; Leet and Selevan, 1982).
28 Biological Gradient - refers to the presence of a dose-response and/or duration-response
29 between a health outcome and exposure of interest. The aforementioned internal analyses for
30 chloroprene (Bulbulyan et al., 1999; Marsh et al., 2007b) show a biological gradient by comparing
31 highly exposed workers to low or unexposed workers. This has also been observed in comparisons
32 between long-term employees and short-term employees
33 Biological Plausibility - refers to the observed effect having some biological link to the
34 exposure. Chloroprene has been found to be metabolized by humans and other species to epoxides,
35 which are known genotoxic metabolites, and has been shown to be a potent (early appearance,
36 multiplicity, malignancy of observed tumors) carcinogen in mice and rats. In addition, the structurally
September 2009 4-71 DRAFT - DO NOT CITE OR QUOTE
-------
1 related carcinogen, butadiene, is also metabolized to epoxides and produces a tumor profile resembling
2 that observed with chloroprene.
3 In summary, the temporality of exposure prior to occurrence of liver cancer, strength of
4 association, consistency, biological gradient, and biological plausibility provide some evidence for
5 chloroprene's carcinogenicity in humans.
4.7.2.2. Laboratory Animal
6 According to the NTP (1998), there is clear evidence of carcinogenicity in the F344/N rat and
7 B6C3F1 mouse due to lifetime inhalation exposure to chloroprene. The mouse is regarded as the most
8 sensitive species because tumor incidence and multisite distribution were greater than with the rat.
9 There was decreased survival in chloroprene-exposed rats and mice, and survival in mice was
10 significantly associated with the burden of neoplastic lesions. Mortality in rats was likely due to overt
11 toxicity across many organ systems. In rats, increased incidences of neoplastic lesions primarily
12 occurred in the oral cavity and lung (males only), kidney, and mammary gland (females). In mice,
13 increased incidences in neoplasms occurred in the lungs, circulatory system (all organ, Harderian
14 gland, forestomach, liver, skin and mesentery (females only), and kidney (males only). In contrast to
15 the neoplastic findings in the F334/N rat, only small numbers of neoplastic lesions were observed in
16 Wistar rats or Syrian golden hamsters (Trochimowicz et al., 1998). There is no unequivocal
17 explanation for why the results for the rat differ between these two studies. The stability of the bulk
18 material in the NTP (1998) study was monitored by gas chromatography, and the material was
19 analyzed for peroxide content. In addition, stabilizer concentrations were in an acceptable range and
20 no dimer peaks were found in the distribution lines leading to the exposure chamber. Concentrations
21 of volatile degradation products (e.g., 1-chlorobutadiene) never exceeded 0.6% of the atmospheric
22 concentration of chloroprene when sampled from either the distribution line or exposure chamber. In
23 the study in the Wistar rat by Trochimowicz et al. (1998), there was no evidence of degradation of the
24 freshly distilled chloroprene, and dimer concentrations were stated to be less than the limit of
25 detection. Thus, it is unlikely that the bulk materials or generated atmospheres differed to an extent
26 that would have caused the differences in results. The discrepancy between the carcinogenicity of
27 chloroprene observed in the two studies may be due to species and/or strain differences. Himmelstein
28 et al. (2001b) observed that liver microsomes from B6C3F1 mice and the F344 rats, the two species
29 used in the NTP (1998) study, produced more (l-chloroethenyl)oxirane than those from hamsters or
30 Wistar rats, the two species used in the Trochimowicz et al. (1998) study. These differences in
31 production of (l-chloroethenyl)oxirane were as great as 15-fold greater (F344 rats vs. hamsters).
32 The inhalation study by Dong et al. (1989) found that a 7-month exposure of the Kunming
33 strain of albino mice, a strain reported to have a low spontaneous rate of lung tumor formation,
34 resulted in a chloroprene-associated increase in lung tumors. Although quality assurance procedures
September 2009 4-72 DRAFT - DO NOT CITE OR QUOTE
-------
1 regarding histopathology were not reported, these study results are considered to support the findings
2 in the B6C3F1 mice in the NTP (1998) chronic bioassay.
3 In the only long-term oral cancer study (an Fl generation of inbred BD IV rats given weekly
4 doses of 50 mg/kg chloroprene by gavage), no significant neoplastic effects were reported
5 (Ponomarkov and Tomatis, 1980). The number of tumor-bearing animals was similar to controls.
4.7.3. Mode-of-Action Information
4.7.3.1. Hypothesized Mode of Action
6 The proposed hypothesis is that chloroprene acts via a mutagenic mode of action involving
7 reactive epoxide metabolites formed at target sites. DNA-epoxide adduct formation is an effect
8 observed for a number of carcinogens structurally related to chloroprene, including those with a known
9 mutagenic mode of action (i.e., vinyl chloride; EPA, 2005b, 2000f) and those for which a
10 preponderance of evidence strongly suggests a mutagenic mode of action (i.e., isoprene and 1,3-
11 butadiene) (Begemann et al., 2004; EPA, 2002b; Sills et al., 1999). This hypothesized mode of action
12 is presumed to apply to all tumor types. Mutagenicity is a well-established cause of carcinogenicity.
4.7.3.2. Experimental Support for the Hypothesized Mode of Action
13 Compelling evidence for the hypothesized mutagenic mode of action for chloroprene includes:
14 1) chloroprene, like butadiene and isoprene, is metabolized to epoxide intermediates and both
15 compounds are carcinogens; 2) chloroprene forms DNA adducts via its epoxide metabolite, and is a
16 point mutagen in vitro (in some but not all bacterial assays) and in vivo (in carcinogenicity bioassays,
17 with mutations occurring in proto-oncogenes); 3) observation of the genetic alterations (base-pair
18 transversions) in proto-oncogenes in chloroprene-induced lung, Harderian gland, and forestomach
19 neoplasms in mice and positive results in Salmonella typhimurium strains that test for base-pair
20 substitution mutations ; and 4) similarities in tumor sites and sensitive species between chloroprene
21 and butadiene in chronic rodent bioassays (NTP 1998 and 1999, respectively). These lines of evidence
22 are elaborated on below.
23 Evidence for the formation of reactive epoxide metabolites following exposure to chloroprene
24 has been observed in both genders of multiple species. Currently, in vivo data are unavailable for
25 blood or tissue-specific epoxide metabolism rates or concentrations. However, in studies using mouse
26 and human liver microsomes, Bartsch et al. (1979) showed that 2-chloro-2-ethynyloxirane and/or
27 (l-chloroethenyl)oxirane could be intermediates in the biotransformation of chloroprene. Himmelstein
28 et al. (2001b) confirmed the identity of the volatile metabolite reported by Bartsch et al. (1979) as the
29 epoxide (l-chloroethenyl)oxirane. Himmelstein et al. (2001b) reported that the oxidation of
30 chloroprene to (l-chloroethenyl)oxirane was evident in rodent and human liver microsomes and most
31 likely involved CYP 2E1. The oxidation of chloroprene to (l-chloroethynyl)oxirane is more prevalent
September 2009 4-73 DRAFT - DO NOT CITE OR QUOTE
-------
1 in B6C3F1 mice and F344 rat liver microsomes than in Wistar rats, humans, or hamsters. Comparing
2 metabolism between species, Cottrell et al. (2001) confirmed the results of Himmelstein et al. (2001b),
3 and further showed that the quantitative profiles of metabolites from liver microsomes obtained from
4 mice, rats, and humans were similar. In all species and either gender, (l-chloroethynyl)oxirane was the
5 major metabolite detected. One distinct difference between species was the stereospecificity of
6 epoxide metabolites formed. In 2 strains of rats (Sprague-Dawley and F344), the R-enantiomer was
7 preferentially formed, whereas this enantioselectivity was not observed in mice or humans. Hurst and
8 Ali (2007) reported that the S-(l-chloroethynyl)oxirane enantiomer was more quickly detoxified in
9 mouse erythrocytes than the R-enantiomer, suggesting that the R-enantiomer may be more toxic due to
10 its slower elimination. 1,3-butadiene exhibits similar biotransformation to reactive epoxide
11 metabolites. Oxidation of 1,3-butadiene to l,2-epoxy-3-butene has been observed in hepatic, lung, and
12 kidney microsomes, as well as lung tissue and bone marrow, in rats, mice, and humans (EPA, 2002b).
13 Further oxidation of l,2-epoxy-3-butene to 1,2,3,4-diepoxybutane has been observed rat, mouse, and
14 human liver microsomes, as well as in blood and tissues of mice and rats exposed by inhalation to 1,3-
15 butadiene (EPA, 2002b). Vinyl chloride and isoprene are also readily converted into their reactive
16 epoxide metabolites; vinyl chloride is converted to chloroethylene epoxide in rats and isoprene to
17 (2,2')-2-methylbioxirane in rats and mice (Watson et al., 2001; EPA, 2002f).
18 Chloroprene's metabolites have been shown to form DNA adducts when reacted with
19 nucleosides and double stranded DNA in vitro. Reaction of (l-chloroethenyl)oxirane with the
20 nucleoside 2'-deoxyguanosine yielded one major adduct derived by nucleophilic attack of N-7 guanine
21 on C-3' of the epoxide, whereas another metabolite, 2-chlorobut-2-en-l-al, yielded 2 major adducts
22 (Munter et al., 2002). The reaction of (l-chloroethenyl)oxirane with double stranded calf thymus DNA
23 yield the same adduct observed when the chloroprene metabolite was incubated with
24 2'-deoxyguanosine individually, (l-chloroethenyl)oxirane also reacted with deoxycytidine in double
25 stranded DNA to yield an adduct which may be significant as such adducts are difficult to repair and
26 may therefore be implicated in mutagenesis (Koskinen et al., 2000).
27 Evidence for the mutagenic potential of chloroprene has been shown in molecular analysis of
28 the genetic alteration of cancer genes including the ras proto-oncogenes (Sills et al., 1999, 2001; Ton et
29 al., 2007), which are alterations commonly observed in human cancers. Tissues from lung,
30 forestomach, and Harderian gland tumors from mice exposed to chloroprene in the NTP chronic
31 bioassay (1998) were shown to have a higher frequency of mutations in K- and H-ras proto-oncogenes
32 than in spontaneous occurring tumors (Sills et al., 1999, 2001). Further, there was a high correlation
33 between K-ras mutations and loss of heterozygosity in the same chromosome in chloroprene-induced
34 lung neoplasms in mice (Ton et al., 2007). Similar increases in the frequencies of K-ras mutations in
35 rodents were observed in isoprene-induced lung neoplasms and vinyl chloride-induced heptatocellular
36 carcinomas (NTP, 1998; U.S. EPA, 2002f). Activated K-ras oncogenes were observed in lung tumors,
37 hepatocellular carcinomas, and lymphomas in B6C3F1 mice exposed to 1,3-butadiene (EPA, 2002b).
September 2009 4-74 DRAFT - DO NOT CITE OR QUOTE
-------
1 Activated K-ras oncogenes have not been found in spontaneously occurring liver tumors or
2 lymphomas, and are found in only 1/10 spontaneous forming lymphomas in B6C3F1 mice (EPA,
3 2002b).
4 Although the genetic toxicity database for chloroprene includes numerous studies covering a
5 range of standard test batteries, their results have been conflicting. In general, bacterial base pair
6 substitution mutation (Salmonella typhimurium strains TA100 and TA 1535) assays have been positive
7 (Willems 1980; Bartsch et al., 1979) while the bacterial frame shift (S. typhimurium strains TA 97 and
8 TA98) assays have been nonpositive (NTP, 1998; Willems 1980; Willems 1978). The observation of
9 positive results in bacterial base pair substitution assays is in concordance with the finding that
10 mutations in H- and K-ras oncogenes in select neoplasms of exposed mice manifest in base pair
11 tranversions (Sills et al., 1999, 2001). In contrast, other studies (NTP, 1998) have reported nonpositive
12 results for all bacterial strains. Westphal et al. (1994) suggested that decomposition products of
13 chloroprene may be responsible for the mutagenicity seen in positive tests. Westphal et al. (1994)
14 exposed bacteria directly to liquid chloroprene in solution and observed no increase in mutagenicity,
15 whereas positive tests (Willem 1978, 1980; Bartsch et al., 1979) were conducted by exposure of
16 bacteria to chloroprene in the air. Atmospheric exposures of chloroprene may result in more
17 degradation products being formed, thereby increasing the mutagenicity of the parent compound. A
18 positive result with all bacterial strains was observed when exposed to chloroprene's epoxide
19 metabolite (l-chloroethenyl)oxirane in solution (Himmelstein et al., 2001).
20 Conflicting results (positive in Vogel, 1979; nonpositive in Foureman et al., 1994) have also
21 been reported for the in vivo Drosophila melanogaster sex-linked lethal mutation assay. Differences
22 observed may be due to differences in purity, strain susceptibilities, and sample size. Chloroprene has
23 been primarily nonpositive in the in vitro micronucleus assay (Himmelstein et al., 2001; Drevon and
24 Kuroki, 1979), in vivo chromosomal damage (NTP 1998) assay, and bone marrow micronucleus assays
25 (NTP 1998; Shelby and Witt, 1995). The lack of genotoxic damage induced in bone marrow or blood
26 by chloroprene suggests that the carcinogenic activity of this chemical may be site specific. The in
27 vivo uptake of chloroprene involves a balance of reactive epoxide formation and glutathione- or
28 epoxide hydrolase-dependent detoxification pathways. These pathways may be enhanced or more
29 active in some tissues, thus limiting DNA damage in those tissues. Bone marrow was not a target for
30 cancer in the chronic carcinogenicity bioassays (NTP, 1998), and the endpoints for chromosomal
31 damage in this tissue were nonpositive. Evidence for target organ-dependent mutagenicity is further
32 supported by the findings of K- and H-ras oncogene mutations in lung, forestomach, and Harderian
33 gland neoplasms in B6C3F1 mice (Sills et al., 1999, 2001). However, a positive result with all
34 bacterial strains was observed with chloroprene's epoxide intermediate (l-chloroethenyl)oxirane
35 (Himmelstein et al., 2001).
36 A comparative analysis by Melnick and Sills (2001) has shown that chloroprene, isoprene, and
37 butadiene share several tumor sites in rats (mammary gland, thyroid, and kidney) and mice
September 2009 4-75 DRAFT - DO NOT CITE OR QUOTE
-------
1 (hemangiomas and hemangiosarcomas [all organs], lung, liver, forestomach, Harderian gland, and
2 mammary gland). Similar to butadiene, the female mouse lung was the most sensitive site of
3 chloroprene carcinogenicity (see Section 4.5.3 and Tables 4-24 and 4-27). There are also remarkable
4 similarities in the potency and shape of the dose response between both compounds. Detailed
5 quantitative analysis (Melnick and Sills, 2001) has rated butadiene as being of slightly greater or equal
6 in potency at some of the common sites of tumor induction (mammary gland and Harderian gland),
7 and more importantly, of equal potency in the induction of the most sensitive tumor, lung neoplasms in
8 female mice.
9 In summary, the evidence supports the hypothesized mutagenic mode of action for chloroprene
10 . A mutagenic mode of carcinogenic action of chloroprene is supported by chloroprene's epoxide
11 metabolite formation, DNA-adduct formation, observation of in vivo and in vitro mutagenicity, and the
12 well known structure-activity relationship of similar epoxide-forming carcinogens. Chloroprene has
13 been found to be metabolized to epoxides by humans and rodents. The hypothesized mutagenic mode
14 of action is supported by evidence of base pair substitution mutations seen in H- and K-ras proto-
15 oncogenes in chloroprene-induced lung, forestomach, and Harderian gland neoplasms observed in the
16 NTP( 1998) study.
4.7.3.3 Conclusions about the Hypothesized Mode of Action
17 As noted above, the hypothesis is that chloroprene carcinogenicity has a mutagenic mode of
18 action. This hypothesized mode of action is presumed to apply to all of the tumor types. The key
19 events in the hypothesized mutagenic mode of action are metabolism to reactive epoxide intermediates
20 followed by binding to DNA, which leading to mutation. Epoxide-forming agents are generally
21 capable of forming DNA adducts which in turn have the potential to cause genetic damage, including
22 mutations; mutagenicity, in turn, is a well-established cause of carcinogenicity. This chain of key
23 events is consistent with current understanding of the biology of cancer. Further, the mutagenic mode
24 of action hypothesis is strongly supported by analogy with another epoxide-forming compound,
25 1,3-butadiene. In addition, alternative or additional modes of action for chloroprene carcinogenicity
26 have not been hypothesized or have supporting evidence.
27 Strength, Consistency, Specificity of Association - Data from NTP (1998) and Sills et al.
28 (1999, 2001) show codon-specific (codons 12, 13, and 61) mutations in the H- and K-ras proto-
29 oncogenes in chloroprene-induced lung, forestomach, and Harderian gland neoplasms. The high
30 incidence of ras proto-oncogene activation (37/46 lung, 27/27 Harderian gland, 4/7 forestomach) in
31 tumors in treated animals, in contrast with the lower incidence of oncogene activation in spontaneously
32 occurring tumors (25/82 lung, 15/27 Harderian gland, 4/11 forestomach), provides support for the role
33 of mutation in the ras oncogene as a precursor to tumor formation in animals treated with chloroprene.
34 Similar findings of ras oncogene activation for isoprene (11/11 lung, 30/30 Harderian gland, 7/10
35 forestomach) and 1,3-butadiene (6/9 lung, 20/29 Harderian gland, 20/24 forestomach) were observed
September 2009 4-76 DRAFT - DO NOT CITE OR QUOTE
-------
1 in tumors from animals treated with these structurally-related compounds (Sills et al., 2001, 1999).
2 These findings provide additional support for the importance of ras proto-oncogene activation via
3 mutation in the carcinogenesis of chloroprene and related compounds.
4 Dose-Response Concordance - High frequencies of K-ras codon 61 CTA mutations were
5 observed in lung tumors from animals exposed to the low- and mid-dose of chloroprene, but not the
6 high dose. Similarly high frequencies of K-ras mutations were observed at all doses in Harderian
7 gland tumors. There are a number of factors that might explain such observations. The higher
8 frequency of mutations at lower doses in lung neoplasms may indicate the saturation of one or more
9 metabolic pathways at higher doses or may suggest that non-ras mechanisms of genotoxicity are
10 operating at those doses. Dose-dependent differences in the mutation profile in the lung and Harderian
11 gland may be explained by differences in DNA-adduct formation or repair in low doses vs high doses.
12 Temporal Relationships - In mice exposed to chloroprene, tumors were observed in a
13 significant fraction of the exposed animals after 2 years of exposure. DNA-adduct formation and
14 subsequent ras mutations were most likely early mutagenic events in the development of lung,
15 Harderian gland, and forestomach neoplasms. The observation that ras mutations occured in benign
16 neoplasms in these organ systems (lung and Harderian gland adenomas and forestomach papillomas) is
17 supportive evidence of this. Additionally, in mice exposed to isoprene for 6 months and then allowed a
18 6 month recovery period, forestomach neoplasm with ras mutations did not regress (Melnick et al.,
19 1994). This suggests that ras mutations may have transformed forestomach epithelial cells at an early
20 time point and that the transformed cells progressed to neoplasia even after chemical exposure had
21 been terminated.
22 Biological Plausibility and Coherence - The biological plausibility of a mutagenic mode of
23 action for chloroprene is supported by evidence of mutations leading to ras proto-oncogene activation
24 in tumors from mice treated with chloroprene (Sills et al., 2001, 1999; NTP, 1998). These studies
25 provide the critical link between the in vitro evidence of mutagenicity (positive results in
26 S. typhiimurium strains 100 and 1535 that test for point mutations) and tumor formation in a specific
27 species. Similar findings with the structurally related chemicals 1,3-butadiene and isoprene and the
28 lower incidence of spontaneously occurring tumors displaying ras mutations in untreated animals (Sills
29 et al., 2001, 1999) enhance the database supporting this particular mode of action for chloroprene.
30 Additional evidence for the association between mutagenesis and tumor formation is the
31 observation that chloroprene exposure caused tumors in a wide variety of mouse tissues, including
32 lung, kidney, Harderian gland, mammary gland, forestomach, liver, skin, mesentery, and Zymbal's
33 gland (NTP, 1998). Tumors were also observed in a number of rat tissues, including oral cavity,
34 thyroid, lung, kidney, and mammary gland. Induction of tumors at multiple sites and in different
35 species is characteristic of carcinogens acting via mutagenesis (US EPA, 2005a).
36 Early-Life Susceptibility - According to the Supplemental Guidance for Assessing
37 Susceptibility from Early-Life Exposures to Carcinogens (U.S. EPA, 2005b) those exposed to
September 2009 4-77 DRAFT - DO NOT CITE OR QUOTE
-------
1 carcinogens with a mutagenic mode of action are assumed to have increased early-life susceptibility.
2 Data on chloroprene are not sufficient to develop separate risk estimates for childhood exposure.
3 There are no data comparing the carcinogen!city of chloroprene after exposure during early life with
4 the carcinogenicity after exposure during adulthood. Exposure to chloroprene commenced at about
5 6 weeks of age in mice and rats, and continued through adulthood in the 2-year chronic assay
6 (NTP, 1998).
7 Therefore, because the weight of evidence supports a mutagenic mode of action for
8 chloroprene carcinogenicity (see Section 4.7.3.2), and in the absence of chemical-specific data to
9 evaluate differences in susceptibility, early-life susceptibility should be assumed and the age-
10 dependent adjustment factors (ADAFs) should be applied, in accordance with the Supplemental
11 Guidance.
12 In conclusion, the weight of evidence supports a mutagenic mode of action for chloroprene
13 carcinogenicity and application of ADAFs to address assumed early-life susceptibility.
4.8. SUSCEPTIBLE POPULATIONS AND LIFE STAGES
4.8.1. Possible Childhood Susceptibility
14 No direct evidence has been found that indicates children are more susceptible to the toxic
15 effects of chloroprene exposure than adults: exposures of children have not been reported and the
16 metabolic fate of chloroprene in humans has not been sufficiently characterized. However, there are a
17 number of issues that, when considered together, suggest that childhood may represent a lifestage with
18 increased susceptibility to chloroprene effects.
19 There are indications of reduced metabolic capacity and elimination in children relative to
20 adults that may be a source of susceptibility. Glutathione levels are rapidly depleted in response to in
21 vitro (rat hepatocytes) and in vivo (Wistar rats) chloroprene exposures, suggesting a GSH-dependent
22 detoxification pathway (Summer and Greim, 1980). Additionally, chloroprene's major metabolite, (1-
23 chloroethenyl)oxirane, is rapidly detoxified via epoxide hydrolase-mediated hydrolysis in mouse liver
24 microsomes (Himmelstein et al., 2001b). The levels of both epoxide hydrolase and glutathione
25 transferase (GST) have been shown to be lower in infants than adults (Ginsberg et al., 2004). Epoxide
26 hydrolase is active at birth, but only at 50% of adult function for as long as 2 years. Evidence,
27 although limited, suggests that two forms of GST (GSTmu and oiB2) may be deficient (40-60% of adult
28 levels) in early life. This decrement in GST activity is especially relevant as GSTmu is critical to
29 epoxide conjugation to glutathione. Therefore, as both epoxide hydrolase and certain forms of GST
30 exhibit decreased activity in early life, newborns and young infants may experience higher and more
31 persistent blood concentrations of chloroprene and/or its metabolite than adults at similar dose levels.
32 Compensating mechanisms (i.e., other GST isozymes such as GSTpi) may be active in early life.
33 Reduced renal clearance in children may be another important source of potential susceptibility.
September 2009 4-78 DRAFT - DO NOT CITE OR QUOTE
-------
1 Excretion of chloroprene in exposed rats occurs through the elimination of urinary thioesters
2 (presumably glutathione conjugates) (Summer and Greim, 1980). Data indicating reduced renal
3 clearance for infants up to 2 months of age may suggest a potential to affect chloroprene excretion,
4 thus prolonging its toxic effects.
5 Further, a mutagenic mode of action is proposed for chloroprene's observed carcinogenicity
6 (See Section 4.7.3). In the absence of chemical-specific data to evaluate the differences between
7 adults and children, chemicals with such a mode of action are assumed to have increased early-life
8 susceptibility and age-dependent adjustment factors (ADAFs) should be applied, in accordance with
9 EPA's Supplemental Guidance for Assessing Susceptibility From Early-Life Exposure to Carcinogens
10 (U.S. EPA, 2005b).
4.8.2. Possible Gender Differences
11 In lifetime studies conducted in the rat, mouse, and hamster, chloroprene was not shown to
12 exhibit any remarkable gender-related differences in effects with the exception of a more pronounced
13 neoplastic response in B6C3F1 female mice compared to males.
September 2009 4-79 DRAFT - DO NOT CITE OR QUOTE
-------
5. DOSE-RESPONSE ASSESSMENTS
5.1. ORAL REFERENCE DOSE (RFD)
1 The available data are inadequate to derive an oral RfD for chloroprene. There are no human
2 data involving oral exposure. The only lifetime oral study exposed rats to chloroprene at one dose (50
3 mg/kg/day) and only qualitatively reported non-cancer effects (Ponomarkov and Tomatis, 1980).
4 In summary, this study identifies the liver (multiple liver necroses and degenerative lesions of
5 parenchymal cells), lung (severe congestion), and kidney (severe congestion) as potential target organs
6 for the oral toxicity of chloroprene; although, the available information is insufficient to characterize
7 toxicity outcomes or dose-response relationships. An RfD was not derived due to the significant
8 uncertainty associated with the oral database for chloroprene.
5.2. INHALATION REFERENCE CONCENTRATION (RFC)
9 RfCs are derived for exposures via the inhalation route. In general, the RfC is an estimate of a
10 daily exposure to the human population (including susceptible subgroups) that is likely to be without
11 an appreciable risk of adverse health effects over a lifetime. It is derived from a statistical lower
12 confidence limit on the benchmark dose (BMDL), a no-observed-adverse-effect level (NOAEL), a
13 lowest-observed-adverse-effect level (LOAEL), or another suitable point of departure (POD), with
14 uncertainty/variability factors applied to reflect limitations of the data used. The inhalation RfC is
15 analogous to the oral RfD but provides a continuous inhalation exposure estimate. The inhalation RfC
16 considers toxic effects for both the respiratory system (portal-of-entry) effects and systems peripheral
17 to the respiratory system (extra-respiratory or systemic effects). It is generally expressed in mg/m3.
5.2.1. Choice of Principal Study and Critical Effect(s)
18 While literature exists on the carcinogenic potential of chloroprene exposure in humans, no
19 human studies are available that would allow for the quantification of sub-chronic or chronic non-
20 cancer effects. Two inhalation studies investigating portal-of-entry (nasal and pulmonary) and
21 systemic effects were identified in the literature and considered for the principal study for derivation of
22 an RfC: a 2-year chronic study in B6C3F1 mice and F344 rats (NTP, 1998), and a 2-year chronic study
23 in Wistar rats and Syrian gold hamsters (Trochimowicz et al., 1998).
24 The chronic NTP inhalation bioassay (1998) exposed groups of 50 mice and rats of each sex to
25 0, 12.8, 32 or 80 ppm chloroprene for 6 hours/day, 5 days/week for 2 years. This study observed a
26 range of chloroprene-induced nonneoplastic effects across several organ systems including the
27 respiratory tract (from the nose to the alveolar region) in both mice and rats, the kidneys of rats and
28 male mice, the forestomach of male and female mice and the spleen of male and female mice (NTP,
29 1998). In addition, many histopathological lesions were significantly increased compared to controls
30 at the lowest level tested (12.8 ppm), including alveolar epithelial hyperplasia in male and female rats,
September 2009 5-1 DRAFT - DO NOT CITE OR QUOTE
-------
1 bronchiolar hyperplasia in male and female mice, lung histiocytic cell infiltration in female mice,
2 hematopoietic cell proliferation in the spleen in female mice, and atrophy, necrosis, and chronic
3 inflammation of the nasal olfactory epithelium in male rats.
4 Trochimowicz et al. (1998) exposed three groups of 100 Wistar rats and Syrian hamsters of
5 each sex to chloroprene at 0, 10, or 50 ppm for 6 hours/day, 5 days/week for up to 18 months
6 (hamsters) or 24 months (rats). Unlike the NTP (1998) study, this study did not observe a wide range
7 of nonneoplastic effects in multiple organ systems. Gross pathology revealed that the lungs from rats
8 exposed at 10 and 50 ppm had markedly lower incidences of pathological changes consistent with, and
9 characterized as, chronic respiratory disease than did controls. Male hamsters exhibited a
10 concentration-related decrease in the incidence of pale adrenal glands. The only remarkable
11 nonneoplastic lesions statistically increased in male and female rats were observed in the liver and
12 lungs at 50 ppm: an increase in foci of cellular alteration in the liver and mild changes, such as
13 lymphoid aggregates around the bronchi, bronchiole, and blood vessels, in the lungs. Accidental
14 failure of the exposure chamber ventilation system suffocated 87 males and 73 females in the low
15 exposure (10 ppm) group during week 72 of exposure, and limited the histopathological examinations
16 performed in this study. Only the livers of rats that died accidentally were processed for microscopic
17 examination. No morphological disturbances were noted in the liver of low exposure group animals.
18 The only nonneoplastic change seen in hamsters was a generalized amyloidosis (in the liver, kidneys,
19 spleen, and adrenals) that was lower in incidence in the 50 ppm exposed group compared with
20 controls.
21 The chronic NTP (1998) study was chosen as the principal study for the derivation of the RfC.
22 Based on the non-cancer database for chloroprene, this study demonstrated exposure concentration-
23 related effects more extensively than any other study. It was a well conducted study that utilized 50
24 animals per sex, per exposure group, a range of exposure concentrations based on the results of
25 preliminary, shorter-duration studies (16 day and 13 weeks), and thoroughly examined chloroprene's
26 observed toxicity in two species. Trochimowicz et al. (1998) was not chosen as the principal study
27 primarily due to the lack of observed effects at similar exposure levels as the NTP (1998) study
28 (Trochimowicz et al., 1998; see Section 4.7.2.2 for discussion of potential causes of differences in
29 observed toxicity between the NTP and Trochimowicz studies). Concerns regarding the abnormally
30 high mortality in the low dose animals in the Trochimowicz et al. (1998) also influenced the choice to
31 disregard it as the principal study.
32 From the NTP (1998) study, all portal-of-entry and systemic nonneoplastic lesions that were
33 statistically increased at the lowest exposure concentration (12.8 ppm), compared to chamber controls,
34 were considered candidates for the critical effect. The candidate endpoints included bronchiolar
35 hyperplasia, pulmonary histiocytic cell infiltration, and splenic hematopoietic cell proliferation in
36 mice, and alveolar epithelial hyperplasia, nasal chronic inflammation, olfactory necrosis, and olfactory
37 epithelium atrophy in rats (Table 5-1).
September 2009 5-2 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
The incidence data for atrophic and necrotic olfactory epithelial lesions in male rats were
mostly coincident, suggesting an interdependence between these two lesions. In the 12 ppm group, 10
of the 13 males affected with one of these lesions showed the other as well, and all male rats with
necrosis at higher exposures also showed atrophy. Because both are degenerative lesions and not
enough information was available to establish if one lesion was a precursor for the other (e.g., atrophy
progressing to necrosis), these lesions were combined and considered as one critical effect,
degenerative nasal lesions. In addition, with the exception of one high concentration male rat, all rats
with nasal chronic inflammation showed nasal atrophy.
Table 5-1. Incidences of nonneoplastic lesions statistically significantly increased
at lowest exposure concentration resulting from chronic exposure (ppm) to
chloroprene
SPECIES
Mice
Rats
TISSUE
Lung
Spleen
Nose
Lung
ENDPOINT
Bronchiolar
hyperplasia
Histiocytic cell
infiltration
Hematopoietic
cell
proliferation
Inflammation,
chronic
Atrophy
Necrosis
Atrophy or
Necrosisd
Alveolar
epithelial
hyperplasia
MALE
0
0/50
-
-
0/50
3/50
0/50
3/50
5/50
12.8
10/50b
-
-
5/50a
12/503
ll/50b
13/50
16/50b
32
18/50b
-
-
9/49b
46/49b
26/49b
47/49
14/493
80
23/50b
-
-
49/49b
48/49b
19/49b
48/49
25/50b
FEMALE
0
0/50
1/50
13/50
-
-
—
-
6/49
12.8
15/49b
14/49b
25/49c
-
-
—
-
22/50b
32
12/50b
18/50b
42/49c
-
-
~
-
22/50b
80
30/50b
23/50b
39/50c
-
-
~
-
34/50b
9
10
11
12
13
14
ap<0.05
bp<0.01
0 Reported as significantly greater than controls, but level of significance not reported
d Combination of two separately reported endpoints
~ Endpoint not observed
Source: NTP(1998)
5.2.2. Methods of Analysis
This assessment uses benchmark dose (BMD) methodology to estimate a POD for the
derivation of an RfC for chloroprene. The use of the BMD approach improves the assessment by
including consideration of the shape of the dose-response curve, providing independence from
experimental doses, and providing estimation of the experimental variability associated with the
calculated dose-response relationship. Use of BMD methods involves fitting mathematical models to
dose-response data and provides a BMD and its 95% lower confidence limit (BMDL) associated with a
September 2009 5-3 DRAFT - DO NOT CITE OR QUOTE
-------
1 predetermined benchmark response (BMR). The BMDL is then used as the POD for deriving the RfC.
2 The suitability of these methods to determine a POD is dependent on the nature of the toxicity
3 database for a specific chemical. Alternatively, the NOAEL/LOAEL approach was used when the data
4 could not be appropriately modeled.
5 A BMR of 10% extra risk was chosen for these endpoints under the assumption that it
6 represents a minimal biologically significant change. In any case, a 10% increase in incidence relative
7 to controls is recommended for the BMR when using dichotomous models, to facilitate a consistent
8 basis of comparison across assessments and endpoints (U.S. EPA, 2000b). All available dichotomous
9 models in the EPA BMD software (BMDS) (version 2.0) were fit to the incidence data for lung, spleen,
10 and nasal effects in rats and mice (Table 5-1). Details of the BMD modeling analysis are included in
11 Appendix Bl.
12 The models selected for a particular endpoint were chosen based on global and local goodness-
13 of-fit criteria (Akaike Information Criterion [AIC] and chi-square [%2] residual values, respectively)
14 and visual inspection. The BMDs and BMDLs associated with an extra risk of 10% for the best-fitting
15 models for each endpoint are shown in Table 5-2. NOAELs and LOAELs were used as potential
16 PODs for the bronchiolar hyperplasia and splenic hematopoietic cell proliferation endpoints in female
17 mice that could not be modeled adequately.
5.2.3. Exposure Duration and Dosimetric Adjustments
18 Because an RfC is a measure that assumes continuous human exposure over a lifetime, data
19 derived from animal studies need to be adjusted to account for the noncontinuous exposure protocols
20 used in animal studies. In the NTP (1998) study, rats were exposed to chloroprene for 6 hours/day, 5
21 days/week for 2 years. Therefore, the duration-adjusted PODs for lung, nasal, and spleen lesions in
22 rats and mice are calculated as follows:
23
24 PODAoj (ppm) = POD (ppm) x hours exposed per day724 hours x days exposed per week/7days
25
26 RfCs are typically expressed in units of mg/m3; the above ppm value needs to be converted
27 using the chemical specific conversion factor of 1 ppm = 3.62 mg/m3 (see Table 2-1) for chloroprene.
28 Therefore, the final PODADJ values are calculated as follows:
29
30 PODADJ (mg/m3) = PODADJ (ppm) x 3.62 mg/m3/1 ppm
31
32 Finally, this PODADJ value must be converted to a human equivalent concentration (HEC).
33 Chloroprene is a relatively water-insoluble, non-reactive gas that primarily induces nasal (i.e.,
34 olfactory atrophy) and thoracic (i.e., bronchiolar/alveolar hyperplasia) non-cancer effects. Water-
35 insoluble, non-reactive chemicals typically do not partition greatly into the aqueous mucus coating of
36 the upper respiratory system. Rather, they tend to distribute to the lower portions of the respiratory
September 2009 5-4 DRAFT - DO NOT CITE OR QUOTE
-------
1 tract where larger surface areas and the thin alveolar-capillary barrier facilitate uptake (Medinsky and
2 Bond, 2001). However, the pattern of respiratory effects seen following chloroprene exposure is
3 consistent with what is known about its metabolism and the expression of cytochrome P450 enzymes
4 in the olfactory and respiratory mucosa in rats. The proposed mode of action of chloroprene involves
5 the conversion of the parent compound into its reactive epoxide metabolite by P450 isoform CYP2E1.
6 The olfactory and respiratory mucosa of rats have been shown to specifically express CYP2E1
7 (Thornton-Manning and Dahl, 1997). Further, the olfactory mucosa of rats contains levels of P450s
8 that are more similar to hepatic levels than any other non-hepatic tissue examined and respiratory
9 mucosa P450 concentrations are approximately 25% that of the olfactory mucosa. Conversion of
10 chloroprene into its reactive epoxide metabolite in the olfactory and respiratory epithelia may facilitate
11 its uptake in those regions.
12 Therefore, in accordance with the U.S. EPA (1994b) RfC methods, chloroprene is characterized
13 as a Category 1 gas for portal-of-entry effects, and as such the HECs for lung and nasal lesions in rats
14 and mice were calculated by the application of a dosimetric adjustment factor (DAF). DAFs are ratios
15 of animal and human physiologic parameters, and are dependent on the nature of the contaminant
16 (particle or gas) and the target site (e.g., respiratory tract or remote to the portal-of-entry) (U.S. EPA,
17 1994b). For gases with extrathoracic portal-of-entry effects (i.e. nasal), the DAF is the regional gas
18 dose ratio (RGDRET) and is expressed (in this particular case, for male rats) as follows:
19 RGDRET = (MVr/SET_r)/ (MVh/SET_h)
20 where:
21 MVr = F344/N male rat minute volume (0.294 L/min)2
22 MVh = human minute volume, (13.8 L/min)
23 SET r = surface area of the extrathoracic region in rats (15 cm2)
24 SET h = surface area of the extrathoracic region in humans (200 cm2),
25 RGDRET= (.294/15)7(13.8/200)
26 RGDRET=0.28
27 Therefore, the HEC for portal-of-entry effects in male rats is calculated as follows:
28 HEC (mg/m3) = PODADJ (mg/m3) x DAF
29 = PODADJ (mg/m3) x 0.28
30 The calculated HEC values for all considered endpoints are presented in the last column of
31 Table 5-2.
2 Calculated according to U.S. EPA (1994b): ln(MVr) = b0 + bi x ln(BW). Default minute volume is in L/min; b0 and bi =
species-specific (rat) intercept and coefficient used; body weight in kg. Time-weighted average body weight was 0.456, kg
for male rats
September 2009 5-5 DRAFT - DO NOT CITE OR QUOTE
-------
Table 5-2. Human equivalent concentration estimates for best fitting models of the
BMD from chronic exposure to chloroprene
Endpoint
Species/
Sex
NOAEL
(ppm)
LOAEL
(ppm)
Model3
BMD10b
(ppm)
BMDL10b
(ppm)
PODadic
(mg/m3)
DAFde
PODHEC
(mg/m3)
Lung
Alveolar
epithelial
hyperplasia
Bronchiolar
hyperplasia
Histiocytic cell
infiltration
Rat/male
Rat/female
Mouse/ male
Mouse/
female
Mouse/
female
n/a
n/a
n/a
n/a
n/a
12.8
12.8
12.8
12.8
12.8
Log-logistic
Log-logistic
Log-logistic
f
f
11.4
4.9
7.5
~
~
7.1
3.3
5.6
~
~
4.6
2.1
3.6
8.3
8.3
3.4
2.3
4.1
4.1
4.1
15.6
4.9
14.8
33.9
33.9
Nasal
Chronic
Inflammation
Atrophy
Necrosis
Atrophy or
Necrosis
Rat/male
Rat/male
Rat/male
Rat/male
n/a
n/a
n/a
n/a
12.8
12.8
12.8
n/a
Log-logistic8
Logistic8
Log-probit8
Logistic8
14.6
7.7
7.9
7.4
9.3
6.0
6.4
5.7
6.0
3.9
4.1
3.7
0.28
0.28
0.28
0.28
1.7
1.1
1.2
1.0
Systemic Effects
Hematopoietic
proliferation
Mouse/
female
n/a
12.8
Probit8
4.0
3.3
2.1
1.0
2.1
"Best fitting model as determined by goodness-of-fit statistics. Bold numbers indicate which value (NOAEL, LOAEL, or
BMDL) is used in determination of POD^c-
bAt BMR = 10% extra risk.
"Duration adjusted POD [mg/nf] (PODadj) = POD [ppm] x (3.62 mg/m3/ppm) x (5 days/7 days) x (6 hours/24 hours), in
accordance with EPA policy (2002a)
dDAF = dosimetric adjustment factor.
e For portal-of-entry effects (lung and nasal effects) the DAF = the regional gas dose ratio (RGDR) and is expressed as:
(MV/SA)animais/( MV/SA)hmnan, where MV = minute volume and SA = respiratory surface area. F344/N male rat MV =
0.294 L/min, F344/N female rat MV = 0.203 L/min, B6C3F1 male mouse MV = 0.052L/min, B6C3F1 female mouse MV
= 0.053 L/min, human minute volume = 13.8 L/min. Surface area of the pulmonary region in F344/N rats =0.34 m2,
surface area of the pulmonary region in B6C3F1 mice = 0.05 m2, surface area of the pulmonary region in humans = 54 m2,
surface area of the extrathoracic region in rats =15 cm2, surface area of the extrathoracic region in humans = 200 cm2.
Minute volumes were calculated according to U.S. EPA 1994b using the following time-weighted average body weights:
0.456, 0.290, 0.0437, and 0.0443 kg for male rats, female rats, male mice, and female mice, respectively. For systemic
effects (splenic effects) the DAF is expressed as the ratio of animal to human blood:air partition coefficients. As given in
Table 3-1, the blood:air partition coefficients for rats are in the range of 7.3-8.0, while that for humans is 4.5. A default
value of 1 is substituted when the laboratory animal value exceeds the human.
f No model fits appropriately according fit statistics or visual inspection. Therefore, the NOAEL/LOAEL approach is
recommended to determine a POD
8 High dose group was dropped in order to obtain adequate model fit.
Source: NTP(1998)
September 2009
5-6
DRAFT - DO NOT CITE OR QUOTE
-------
1 The BMD modeling indicated that degenerative nasal lesions were the most sensitive endpoint,
2 when comparing the BMDioS and BMDLioS. Selection of the most relevant BMR for this endpoint to
3 use for developing the RfC involved evaluating the relative biological significance of the degenerative
4 nasal lesions. The nasal lesion data for chloroprene over the range of exposure concentrations tested
5 (NTP, 1998) indicate a progression in both incidence and severity from no necrosis and minimal
6 atrophy in the controls to manifestations of cellular injury and diminished tissue function at the highest
7 exposures (see Table 4-23 for severity grade details). Considering the mild to moderate degenerative
8 nasal lesions observed in 26% of male rats exposed to 12.8 ppm (average severity 1.8 for atrophy and
9 2.0 for necrosis), a BMR of 10% extra risk was selected for the POD based on the assumption that a
10 10% increase in incidence of this effect (with presumably less than moderate severity) is minimally
11 biologically significant. Therefore, 1.0 mg/m3 was selected to serve as the human equivalent POD for
12 the derivation of the RfC.
5.2.4. RfC Derivation—Including Application of Uncertainty Factors
13 A PODHEc value of 1.0 mg/m3 for increased incidence of degenerative nasal lesions in male
14 F344/N rats (NTP, 1998) was used as the POD to derive the chronic RfC for chloroprene because it
15 was the lowest PODHEc calculated after duration and dosimetric adjustments. A total UF of 100 was
16 applied to this POD as described below:
17 • A 3-fold UFA was used to account for uncertainty in extrapolating from laboratory animals to
18 humans (i.e., interspecies variability). This uncertainty factor is comprised of two separate
19 and equal areas of uncertainty to account for differences in the toxicokinetics and
20 toxicodynamics of animals and humans. In this assessment, toxicokinetic uncertainty was
21 accounted for by the calculation of a human equivalent concentration by the application of a
22 dosimetric adjustment factor as outlined in the RfC methodology (U.S. EPA, 1994b). As the
23 toxicokinetic differences are thus accounted for, only the toxicodynamic uncertainties
24 remain, and a UF of 3 is retained to account for this residual uncertainty.
25 • A default 10-fold UFH was used to account for variation in susceptibility among members of
26 the human population (i.e., interindividual variability). Limited information is available to
27 predict potential variability in human susceptibility.
28 • An UFs was not needed to account for subchronic-to-chronic extrapolation because a chronic
29 inhalation study is being used to derive the chronic RfC.
30 • An UF for LOAEL-to-NOAEL extrapolation was not applied because the current approach is
31 to address this factor as one of the considerations in selecting a BMR for benchmark dose
32 modeling. In this case, a BMR of 10% change in degenerative nasal lesions was selected
33 under an assumption that is represents a minimal biologically significant change.
September 2009 5-7 DRAFT - DO NOT CITE OR QUOTE
-------
1 • A 3-fold UF was used to account for deficiencies in the database. The major strength of the
2 database is the observation of exposure-response effects in multiple organ systems in a well-
3 designed chronic inhalation study that utilized 50 animals per sex per dose group, a range of
4 doses based on the results of preliminary, shorter-duration studies (16 day and 13 weeks),
5 and thoroughly examined chloroprene's observed toxicity in two species (rat and mouse).
6 The database further contains another chronic inhalation bioassay investigating outcomes in
7 another species (hamster), and well-designed embryotoxicity, teratological, and reproductive
8 toxicity studies. The database also contains subchronic studies and chronic studies observing
9 potential neurotoxic and immunotoxic effects. A limitation in the database is the lack of a
10 two-generation reproductive toxicity study.
11 Application of this 100-fold composite uncertainty factor yields the calculation of the chronic
12 RfC for chloroprene as follows:
13 RfC = PODHEc - UF = 1.0 mg/m3 - 100 = 1.0 x 10'2 mg/m3
5.2.5. Previous RfC Assessment
14 The IRIS Program has not previously evaluated the noncancer inhalation toxicity of
15 chloroprene.
5.2.6. RfC Comparison Information
16 Figure 5-1 presents PODs, applied UFs, and derived sample RfCs for all of the endpoints from
17 the chronic inhalation NTP (1998) study that were considered for the critical effect for determination
18 of an RfC. Of the considered studies, the NTP (1998) study was considered the most suitable to derive
19 an RfC. The endpoints considered for the critical effects from the NTP (1998) study included any
20 histopathological lesion that was significantly increased in the lowest dose group relative to controls.
21 The PODs are either based on the best fit models from BMD models or the LOAEL of 12.8 ppm and
22 were adjusted for duration and dosimetry before applications of uncertainty factors.
September 2009 5-8 DRAFT - DO NOT CITE OR QUOTE
-------
Olfactory Effects
Pulmonary Effects
Systemic
Effects
I
Atrophy, male Necrosis, male Atrophy and/or
rats(BMDLIO) rats (BMDL10) Necrosis, male
(NTP, 1998) (NTP, 1998) rats (BMDL10)
Alveolar
Epithelial
Hyperplasia,
male rats
(BMDL10)
(NTP, 1998)
Alveolar
Epithelial
Hyperplasia,
female rats
(BMDL10)
(NTP, 1998)
Bronchiolar Bronchiolar Histiocytic Cell
Hyperplasia, Hyperplasia, Infiltration,
male mice female mice female mice
(BMDL10) (NTP, (LOAEL) (NTP, (LOAEL) (NTP,
1998) 1998) 1998)
Splenic
Hematopoietic
Proliferation,
female mice
(BMDL10)
(NTP, 1998)
Human-to-Animal
Human Variation
LOAEL to NOAEL I _ I Database
Figure 5-1. Points of departure (in mg/m ) for selected endpoints with corresponding
applied uncertainty factors and derived sample RfCs (chosen RfC value is circled).
5.3. UNCERTAINTIES IN THE INHALATION REFERENCE CONCENTRATION
1 As presented earlier in the previous section, the UF approach, following EPA practices and RfC
2 guidance (U.S. EPA, 1994b), was applied to the PODHEc in order to derive the chronic RfC. Factors
3 accounting for uncertainties associated with a number of steps in the analyses were adopted to account
4 for extrapolating from an animal bioassay to human exposure, a diverse population of varying
5 susceptibilities, POD determination methodologies, and to account for database deficiencies. The
6 following is a more extensive discussion of the uncertainties associated with the RfC for chloroprene
7 beyond which is described quantitatively in Section 5.2.4. A summary is provided in Table 5-3.
8 Choice ofendpoint. Sample RfCs considered from the NTP (1998) chronic inhalation study
9 ranged from 1.0 x 10"2 to 3.4 x 10"1 mg/m3. Sample RfC values primarily depended on whether
10 olfactory or pulmonary effects were considered. The chosen critical effect, increased incidence of
11 degenerative nasal lesions in male rats, is considered to be the most sensitive endpoint because it
12 returned the lowest PODnEc value compared to all other considered endpoints. Its portal-of-entry
13 nature is consistent with the other portal-of-entry effects (e.g., nasal chronic inflammation, alveolar
14 epithelial hyperplasia, etc.) observed in the NTP (1998) study. The location of the effect is also
15 consistent with what is currently proposed for chloroprene's metabolism: conversion of chloroprene
16 into its reactive metabolite (l-chloroethenyl)oxirane by CYP2E1. The olfactory mucosa of rats has
17 been shown to specifically express CYP2E1 (Thornton-Manning and Dahl, 1996) and levels of total
18 P450 enzymes most similar to hepatic levels than any other non-hepatic tissue examined. Choice of
September 2009
5-9
DRAFT - DO NOT CITE OR QUOTE
-------
1 other olfactory effects (chronic inflammation, necrosis or atrophy considered separately) as the critical
2 effect would not appreciably alter the derived RfC (1.7 x 10"2 for chronic inflammation, 1.2 x 10"2 for
3 olfactory necrosis, or 1.1 x 10"2 for olfactory atrophy) relative to degenerative nasal lesions (i.e.,
4 atrophy and necrosis considered together). Choice of degenerative nasal lesions as the critical effect is
5 supported as the RfC derived is marginally lower than RfCs for other olfactory effects, and thus this
6 value is presumed to be protective for any individual lesion type. Choice of pulmonary effects (e.g.,
7 alveolar epithelial hyperplasia) as the critical effect would result in RfC values up to 30-fold higher
8 than the RfC for degenerative nasal lesions. Choice of the only systemic effect considered from the
9 NTP (1998) study, splenic hematopoietic proliferation, would result in an RfC approximately 2-fold
10 higher than the RfC for degenerative nasal lesions.
11 Choice of model for BMDL derivation. The logistic model fit the data for degenerative nasal
12 lesions in male rats adequately (global goodness of fit p-value = 0.231). Data points are well-predicted
13 near the BMD (%2 residual = 0.905). Use of sample models would either increase the RfC by
14 approximately 50% or decrease the RfC by approximately 3-fold. However, the logistic model was
15 chosen over these models based on current BMD technical guidance (U.S. EPA, 2000b).
16 Choice ofBMR. There is uncertainty in the selection of the benchmark response (BMR) level.
17 For increased incidence of degenerative nasal lesions in male rats, definitive data do not exist to
18 further inform the selection of what the appropriate BMR should be; therefore a BMR of 10% extra
19 risk was chosen based on the assumption that a 10% increase in incidence of this effect (with
20 presumably less than moderate severity) is minimally biologically significant.
21 Statistical uncertainty at POD. For the logistic model applied to degenerative nasal lesions in
22 male rats, there is a reasonably small degree of statistical uncertainty at the 10% extra risk level (the
23 point of departure for derivation of the RfC), with the BMDL being about 25% below the BMD.
24 Choice ofbioassay. The NTP (1998) chronic inhalation study was used for development of the
25 RfC because it was a well designed study that was conducted in 2 relevant species, used 50 animals per
26 sex per exposure group, and thoroughly examined a wide-range of appropriate toxicological endpoints.
27 The other bioassays were discounted for use as the principal study due either a general lack of effects
28 at similar exposure levels (i.e., Trochimowicz et al., 1998) or interpretational difficulties in respect to
29 the observed effects (i.e., Culik et al., 1978).
30 Choice of species. The RfC was based on increased incidence of degenerative nasal lesions in
31 male rats exposed to chloroprene via inhalation for 2 years. Use of other effects that occurred in
32 another species, B6C3F1 mice, would result in RfCs approximately 2-30 times greater than the current
33 RfC.
34 Human population variability. The extent of inter-individual variation of chloroprene
35 metabolism in humans has not been well characterized. However, a number of issues, including lower
36 enzyme levels and renal clearance in children, potential distribution of chloroprene to breast milk, and
37 chloroprene's proposed mutagenic mode of action suggest that childhood may represent a potentially
September 2009 5-10 DRAFT - DO NOT CITE OR QUOTE
-------
1 susceptible lifestage to chloroprene's toxicity. The 10-fold default uncertainty value is applied to the
2 PODHEc primarily due to the limited data on human variability or potential susceptible subpopulations.
Table 5-3. Summary of Uncertainties in the Chloroprene noncancer risk
assessment
CONSIDERATION
POTENTIAL
IMPACT3
DECISION
JUSTIFICATION
Choice of endpoint
Use of other
endpoints could t
RfC by up to 30-fold
RfC is based on the
endpoint with the lowest
PGDnEc, increased
incidence of
degenerative nasal
lesions in male rats,
Chosen endpoint is considered to be the most
sensitive (based on P()DHEC values). Its
portal-of-entry nature is consistent with other
portal-of-entry effects observed in the NTP
(1998) study.
Choice of model for
BMDL derivation
Other models would
t or|RfC
Logistic model used
U.S. EPA (2000b) BMD technical guidance
used to choose model based on global and
local measures of model fit
Choice of BMR
Other BMRs would
or | RfC
BMR of 10% extra risk
chosen
BMR of 10% extra risk was chosen based on
the assumption that a 10% increase in
incidence of this effect (with presumably less
than moderate severity) is minimally
biologically significant
Statistical uncertainty
at POD
RfC would be-25%
higher if BMD (vs.
BMDL) were used
BMDL used as POD per
U.S. EPA guidance
(2000b)
Size of bioassay results in sampling
variability; lower bound is 95% confidence
interval on administered exposure
Choice of bioassay
Other bioassays
could t or | RfC
NTP (1998) used as
critical study
Other bioassays were available but were
discounted as principal study due to lack of
effects or interpretational difficulties. The
chosen bioassay was well-conducted and
reported and resulted in the lowest BMDL
for derivation of RfC
Choice of species
RfC would t if based
on another species
Rats chosen
RfC is based on the most sensitive endpoint
(incidence of degenerative nasal lesions) in
the most sensitive species (rat), based on
PODHEC
Human population
variability
RfC could for | if a
non-default value of
UF was used
10-fold uncertainty
factor applied to derived
the RfC
10-fold UF, the default value, is applied
principally because of limited data on human
variability or potential susceptible
subpopulations
a t = increase, J, = decrease
September 2009
5-11
DRAFT - DO NOT CITE OR QUOTE
-------
5.4. CANCER ASSESSMENT
5.4.1. Choice of Study/Data—with Rationale and Justification
1 The NTP (1998) study was used for development of an inhalation unit risk. This was a well-
2 designed study, conducted in both sexes of two species with 50 animals per sex per dose group and
3 with examination of appropriate toxicological endpoints in both sexes of rats and mice. Tumor
4 incidences were elevated with increasing exposure level at numerous sites across all sex/species
5 combinations, involving point of contact in the respiratory system and more distant locations. The
6 Trochimowicz et al. (1998) was not considered for quantification purposes, primarily due to the
7 general lack of observed neoplastic effects at similar exposure levels as the NTP (1998) study (see
8 Section 4.2.2. for study details).
5.4.2. Dose-Response Data
9 In the NTP (1998) study, groups of 50 male and female F344 rats and B6C3F1 mice were
10 exposed via inhalation to 0, 12.8, 32, or 80 ppm chloroprene for 6 hours/day, 5 days/week for 2 years.
11 These data are summarized in Tables 5-4 (mice) and 5-5 (rats).
12 Mice were the more sensitive species, with statistically significant increases in tumor incidence
13 observed at multiple sites: all organs (hemangiomas, hemangiosarcomas), lung (bronchiolar/alveolar
14 adenomas and carcinomas), forestomach, Harderian gland (adenomas and carcinomas), kidney
15 (adenomas [males]), skin and mesentery (sarcomas), liver, and mammary gland (females). These
16 tumors generally appeared earlier with increasing exposure levels and showed statistically significantly
17 increasing trends with increasing exposure level (by life table test or logistic regression, p < 0.001).
18 Etiologically similar tumor types, benign and malignant tumors of the same cell type, were combined
19 for these tabulations because of the possibility that the benign tumors could progress to the malignant
20 form (U.S. EPA, 2005a). Survival for all chloroprene-exposed female mice and for male mice in the
21 two higher exposed groups was statistically significantly lower than for the corresponding control
22 mice.
September 2009 5-12 DRAFT - DO NOT CITE OR QUOTE
-------
Table 5-4. Tumor incidence in female and male B6C3F1 mice exposed to
chloroprene via inhalation
TISSUE
CUMULATIVE
INCIDENCE; TIME
OF FIRST
OCCURRENCE
ADMINISTERED CHLOROPRENE
CONCENTRATION (ppm)
Control
12.8
32
80
Females
All organs: hemangioma or
hemangiosarcoma
Lung: alveolar/bronchiolar adenoma or
carcinoma
Liver: hepatocellular adenoma or carcinoma
Skin or mesentery: sarcoma
Mammary gland: adenocarcinoma,
carcinoma or adenoacanthoma
Forestomach: squamous cell papilloma or
carcinoma
Harderian gland: adenoma or carcinoma
Zymbal's gland: carcinoma
Unadjusted
KM (%)
First incidence (days)
Unadjusted
KMa (%)
First incidence (days)
Unadjusted
KM (%)
First incidence (days)
Unadjusted
KM (%)
First incidence (days)
Unadjusted
KM (%)
First incidence (days)
Unadjusted
KM (%)
First incidence (days)
Unadjusted
KM (%)
First incidence (days)
Unadjusted
KM (%)
First incidence (days)
4/50
11%
541
4/50
11%
706
20/50
52%
493
0/50
0%
3/50
7.4%
527
1/50
2.9%
734
2/50
5%
527
0/50
0%
6/49
26%
482
28/49
84%
447
26/49
82%
440
15/49
63%
285
6/49
23%
440
0/49
0%
5/49
24%
621
0/50
0%
18/50
100%
216
34/50
100%
346
20/50
100%
503
18/50
100%
463
11/50
42%
394
0/50
0%
3/50
12%
524
0/50
0%
8/50
50.8%
523
42/50
100%
324
30/50
100%
384
19/50
72%
443
14/50
61%
336
4/50
19%
576
9/50
77%
427
3/50
11%
565
Males
All organs: hemangioma or
hemangiosarcoma
Lung: alveolar/bronchiolar adenoma or
carcinoma
Forestomach: squamous cell papilloma or
carcinoma
Harderian gland: adenoma or carcinoma
Kidney: renal tubule adenomas or
carcinomas (extended and standard
evaluations combined)
Unadjusted
KM (%)
First incidence (days)
Unadjusted
KM (%)
First incidence (days)
Unadjusted
KM (%)
First incidence (days)
Unadjusted
KM (%)
First incidence (days)
Unadjusted
KM (%)
First incidence (days)
3/50
11%
733
13/50
39%
635
1/50
4%
733
2/50
5.8%
596
0/50
0%
14/50
42%
659
28/50
79%
530
0/50
0%
5/49
18%
701
2/49
7.1%
111
23/50
71%
495
36/50
89%
382
2/50
14%
733
10/50
42%
596
3/50
20%
715
21/50
74%
454
43/50
100%
523
4/50
15%
587
12/50
58%
589
9/50
41%
567
a Kaplan-Meier estimated neoplasm incidence rate at the end of the study,
and under the assumption that the observed tumors were fatal.
Source: NTP(1998).
involving adjustment for intercurrent mortality
September 2009
5-13
DRAFT - DO NOT CITE OR QUOTE
-------
Table 5-5. Tumor incidence in female and male F344 rats exposed to chloroprene
via inhalation
TISSUE
CUMULATIVE
INCIDENCE; TIME OF
FIRST OCCURRENCE
ADMINISTERED CHLOROPRENE
CONCENTRATION (ppm)
Control
12.8
32
80
Females
Oral cavity: papillomas or
carcinomas
Thyroid gland: follicular
cell adenomas or
carcinomas
Mammary gland:
fibroadenomas
Kidney: renal tubule
adenomas or carcinomas
(extended and standard
evaluations combined)
Unadjusted
KMa (%)
First incidence (days)
Unadjusted
KM (%)a
First incidence (days)
Unadjusted
KM (%)
First incidence (days)
Unadjusted
KM (%)
First incidence (days)
1/49
3.0%
687
1/49
3.4%
733
24/49
65%
366
0/49
0%
3/50
9.2%
681
1/50
3.2%
721
32/50
86%
302
0/50
0%
5/50
17%
588
1/50
3.8%
733
36/50
85%
470
0/50
0%
11/50
41%
660
5/50
17%
617
36/50
90%
433
4/50
10%
609
Males
Oral cavity: papillomas or
carcinomas
Thyroid gland: follicular
cell adenomas or
carcinomas
Lung: alveolar/bronchiolar
adenoma or carcinoma
Kidney: renal tubule
adenomas or carcinomas
(extended and standard
evaluations combined)
Unadjusted
KM (%)
First incidence (days)
Unadjusted
KM (%)
First incidence (days)
Unadjusted
KM (%)
First incidence (days)
Unadjusted
KM (%)
First incidence (days)
0/50
0%
0/50
0%
2/50
6.9%
616
1/50
7.7%
733
2/50
14%
701
2/50
14%
597
2/50
18%
702
8/50
43%
600
5/50
28%
609
4/49
30%
569
4/49
20%
505
6/50
53%
679
12/50
75%
539
5/50
36%
307
6/50
59%
540
8/50
71%
625
a Kaplan-Meier estimated neoplasm incidence rate at the
and under the assumption that the observed tumors were
Source: NTP(1998).
end of the study, involving adjustment for intercurrent mortality
fatal.
5.4.3. Dose Adjustments and Extrapolation Methods
1 The current EPA Guidelines for Carcinogen Risk Assessment (EPA, 2005a) emphasize that the
2 method used to characterize and quantify cancer risk from a chemical is determined by what is known
3 about the MOA of the carcinogen and the shape of the cancer dose-response curve. The dose response
4 is assumed to be linear in the low dose range when evidence supports a mutagenic MOA because of
5 DNA reactivity or if another MOA that is anticipated to be linear is applicable A mutagenic mode of
6 carcinogenic action of chloroprene is supported by chloroprene's epoxide metabolite formation, DNA-
7 adduct formation, observation of in vivo and in vitro mutagenicity, and the well known structure-
8 activity relationship of similar epoxide-forming carcinogens. The determination of a mutagenic mode
9 of action is also supported by evidence of base pair substitution mutations seen in H- and K-ras proto-
September 2009
5-14
DRAFT - DO NOT CITE OR QUOTE
-------
1 oncogenes in chloroprene-induced lung, forestomach, and Harderian gland neoplasms observed in the
2 NTP( 1998) study.
3 . For these reasons, a linear low-dose extrapolation approach was used to estimate human
4 carcinogenic risk associated with chloroprene exposure. Because the weight of evidence supports a
5 mutagenic mode of action for chloroprene carcinogenicity, and in the absence of chemical-specific
6 data on early-life susceptibility, increased early-life susceptibility should be assumed and, if there is
7 early-life exposure, the age-dependent adjustment factors (ADAFs) should be applied, as appropriate,
8 in accordance with EPA's Supplemental Guidance for Assessing Susceptibility From Early-Life
9 Exposure to Carcinogens (U.S. EPA, 2005b).
10 Due to the occurrence of multiple tumor types, earlier occurrence with increasing exposure, and
11 increased mortality with increasing exposure level, methods that can reflect the influence of competing
12 risks and intercurrent mortality on site-specific tumor incidence rates are preferred. EPA has generally
13 used the multistage Weibull model, because it incorporates the time at which death-with-tumor
14 occurred. The multistage Weibull model has the following form:
15 P(d) = 1 - exp[-(q0 + qid + q2d2 + ... + q^) x ft - t0)z]
16 where P(d) represents the lifetime risk (probability) of cancer at dose d (i.e., human equivalent
17 exposure in this case); parameters qt > 0, for i = 0, 1, ..., k; t is the time at which the tumor was
18 observed; and z is a parameter estimated in fitting the model, which characterizes the change in
19 response with age. The parameter to represents the time between when a potentially fatal tumor
20 becomes observable and when it causes death and is generally set to 0 because of a lack of data to
21 estimate the time reliably. The dose-response analyses were conducted using the computer software
22 program TOX_RISK, version 5.3 (ICF, Fairfax, VA), which is based on Weibull models drawn from
23 Krewski et al. (1983). Parameters were estimated using the method of maximum likelihood estimate
24 (MLE).
25 Other characteristics of the observed tumor types were considered prior to modeling, including
26 allowance for different, although possibly unidentified, MO As and for relative severity of tumor types.
27 First, etiologically different tumor types were not combined across sites prior to modeling in order to
28 allow for the possibility that different tumor types can have different dose-response relationships
29 because of varying time courses or other underlying mechanisms or factors. Consequently, all of the
30 tumor types listed separately in Table 5-4 were modeled separately. A further consideration allowed by
31 the software program is the distinction between tumor types as being either fatal or incidental in order
32 to adjust for competing risks. Incidental tumors are those tumors thought not to have caused the death
33 of an animal, while fatal tumors are thought to have resulted in animal death.
34 Specific multistage Weibull models were selected for the individual tumor types for each sex,
35 based on the values of the log-likelihoods according to the strategy used by EPA (U.S. EPA, 2002b). If
36 twice the difference in log-likelihoods was less than a %2 with degrees of freedom equal to the
September 2009 5-15 DRAFT - DO NOT CITE OR QUOTE
-------
1 difference in the number of stages included in the models being compared, the models were considered
2 comparable, and the most parsimonious model (i.e., the lowest-stage model) was selected contingent
3 on visual fits of the data as follows. For incidental tumors, plots of model fits compared with Hoel-
4 Walburg estimates of cumulative incidence were also examined for goodness of fit in the lower
5 exposure region of the observed data (Gart et al., 1986). For fatal tumors, plots of model fits were
6 compared with Kaplan-Meier estimates of cumulative incidence. If a model with one more stage fitted
7 the low-dose data better than the most parsimonious model, then the model with one higher stage was
8 selected.
9 Tumor types were categorized by tumor context as either fatal or incidental tumors. Incidental
10 tumors are those tumors thought not to have caused the death of an animal, and fatal tumors are
11 thought to have resulted in animal death. Hemangiosarcomas were treated as fatal tumors, unless
12 observed at an interim or terminal sacrifice, in which case they were considered incidental.
13 Furthermore, these fatal tumors were deemed rapidly fatal, and t0 was set equal to 0 (the data were
14 considered insufficient to reliably estimate to in any event, without any interim sacrifices for example).
15 Tumors at all other sites were treated as incidental. This is consistent with the tumor context
16 determinations in the IRIS assessment of 1,3-butadiene (US EPA, 2002b)
17 PODs for estimating low-dose risk were identified at doses at the lower end of the observed
18 data, generally corresponding to 10% extra risk, defined as the extra risk over the background tumor
19 rate, [P(d) - P(0)]/[l - P(0)]. PODs were converted to continuous human-equivalent exposure levels
20 by multiplying by (6 hours)/(24 hours) x (5 days)/(7 days), or 0.178. Additionally, in accordance with
21 the U.S. EPA (1994b) RfC methodology, the HECs for the various tumors were calculated by the
22 application of DAFs (see section 5.2.3 and Table 5-2 footnote f for explanations of derivation). With
23 the exception of the lung tumors, all tumors were treated as systemic effects. For these sites a DAF of
24 1.0 was applied as the value for the rat blood:air partition coefficient exceeded the human value . For
25 alveolar/bronchiolar tumors, the HEC was calculated treating the neoplasms alternatively as portal-of-
26 entry effects or systemic effects. As there is evidence that chloroprene and/or its metabolite are
27 distributed systemically (i.e., the observation of tumors in multiple organ systems), there is the
28 potential that chloroprene is redistributed to the lungs. In this manner, chloroprene may induce lung
29 tumors as a systemically delivered carcinogen in addition to inducing tumors via inhalation. However,
30 the contribution of either route of delivery (i.e., inhalation vs. bloodstream) to the induction of lung
31 tumors is currently unknown. Therefore, chloroprene-induced lung tumors were treated as either
32 point-of-entry lesions using a DAF of 4.1, or as systemic lesions using a DAF of 1.0 due to a lack of
33 data clarifying whether one or both modes were more likely to be operating.
34 The lifetime continuous inhalation unit risk for humans is defined as the slope of the line from
35 the lower 95% bound on the exposure at the POD. This 95% upper confidence limit represents a
36 plausible upper bound on the true risk. Unit risks for each tumor site were calculated by dividing the
September 2009 5-16 DRAFT - DO NOT CITE OR QUOTE
-------
1 BMR level (usually 10%) by its corresponding lower bound on the benchmark concentration
2 (BMDLio).
5.4.4. Oral Slope Factor and Inhalation Unit Risk
3 In the absence of any data on the carcinogenicity of chloroprene via the oral route, or a suitable
4 PBPK model allowing route-to-route extrapolation, no oral slope factor was derived. An inhalation
5 unit risk was derived based on the carcinogenic effects of chloroprene via the inhalation route. The
6 results of applying the multistage Weibull models to the male and female mouse tumor incidence data
7 are provided in Tables 5-6 and 5-7, respectively. Human equivalent unit risks estimated from the
8 mouse tumor sites with statistically significant increases ranged from 7.2 x 1CT6 to 1.7 x 1CT4 per
9 ug/m3, approximately a 25-fold range. The highest unit risk (1.7 x 1CT4 per ug/m3) corresponded to
10 lung tumors (treated as systemic lesions) in female mice; the highest unit risk in male mice was
11 associated with lung tumors (treated as systemic lesions) at 8.5 x 1CT5 per ug/m3. Lung tumors were
12 the most sensitive response in mice for 1,3-butadiene as well. Alternatively, if the lung tumors resulted
13 strictly from portal-of-entry processes, the estimated risks associated with this site would be four-fold
14 lower, and the highest unit risk estimates come from hemangiomas and hemangiosarcomas, at 8.3 x
15 1CT5 per ug/m3 for female mice and 6.6 x 1CT5 per ug/m3 for male mice.
16 Given the multiplicity of tumor sites, however, basing the unit risk on one tumor site may
17 underestimate the carcinogenic potential of chloroprene. An approach suggested in the cancer
18 guidelines would be to estimate cancer risk from tumor-bearing animals. EPA traditionally used this
19 approach until the document Science and Judgment in Risk Assessment (National Research Council
20 [NRC], 1994) made a case that this approach would tend to underestimate overall risk when tumor
21 types occur in a statistically independent manner. In addition, application of one model to a composite
22 data set does not accommodate biologically relevant information that may vary across sites or may
23 only be available for a subset of sites. For instance, the time courses of the multiple tumor types
24 evaluated varied substantially, as is suggested by the variation in estimates of z (see Tables 5-6 and 5-
25 7) which shows an association of increasing incidence with time, from about 1.0-10 for female mice
26 and from about 2-10 for male mice. Fitting a model like the multisage-Weibull with mechanism-
27 related parameters to composite data would not characterize the evident range of variation. A simpler
28 empirical model could be used for the composite data, such as the multistage model, but available
29 biological information (time of tumor observation) would then be ignored.
September 2009 5-17 DRAFT - DO NOT CITE OR QUOTE
-------
Table 5-6. Dose-response modeling summary for male mouse tumor sites
associated with inhalation exposure to chloroprene
TUMOR TYPE*
Lung: alveolar/
bronchiolar
adenoma or
carcinomaf
All organs:
hemangiosarcomas,
hemangiomas
Forestomach:
squamous cell
papilloma or
carcinoma
Harderian gland:
adenoma or
carcinoma
Kidney: renal tubule
adenomas or
carcinomas
(extended and
standard evaluations
combined)
MLE
COEFFICIENTS3
q0 = 4.01 x 1Q-8
qi = 4.46 x 1(T9
z = 3.5
q0 = 8.38 x 1(T22
qi = 1.16 x 1(T22
z =10
q0= 3.03 x 10~6
q1 = 2.34 x 10~7
z= 1.8
q0 = 3.26 x 10~13
qj = 3.60 x 10~14
z = 5.6
ql = 2.03 x 10~15
z = 6.1
POINT OF DEPARTURE11 (jig/m3)
Modeled from
bioassay
BMDL10
6.64 x 103
8.50 x 103
1.68 x 105
3.86 x 104
6.07 x 104
BMD10
9.08 x 103
1.10 x 104
4.01 x 105
6.15 x 104
9.85 x 104
Continuous, human
equivalent0
BMDL10
1.18 x Iff
4.84 x 103
1.51 x 103
2.99 x 103
6.87 x 103
1.08 x 104
BMD10
1.62 x Iff
6.62 x 103
1.95 x 103
7.14 x 104
1.10 x 104
1.75 x 104
UNIT
RISKd
Oig/m3)-1
8.5 x Iff5
2.01 x 10~5
6.6 x 10~5
3.3 x 10~5
1.5 x 10~5
9.3 x KT6
OVERALL
UNIT
RISK6
Oig/m3)-1
1.7x Iff4
1.1 x IQ-4
a Model: multistage-Weibull: P(d) = 1 - exp[-(q0 + qid + q2dz + ... + qkd ) x (t - to)z], coefficients estimated in
terms of ppm as administered in bioassay; lower stage q; not listed were estimated to be zero.
b BMD10 = Concentration at 10% extra risk; BMDL10 = 95% lower bound on concentration at 10% extra risk.
0 Continuous equivalent estimated by multiplying exposures by (6 hr)/(24 hr) x (5 days)/(7 days).
d Unit risk estimated by dividing the BMR (0.1) by the BMDL10.
e Overall unit risk estimate, across all sites listed; see text for method.
f Values in italics indicate BMD/BMDL when lung tumors are treated as systemic lesions
* Tumor incidence data from NTP (1998)
September 2009
5-18
DRAFT - DO NOT CITE OR QUOTE
-------
Table 5-7. Dose-response modeling summary for female mouse tumors associated
with inhalation exposure to chloroprene
TUMOR TYPE*
Lung: alveolar/
bronchiolar
adenoma or
carcinomaf
Skin: sarcoma
All organs:
hemangiosarcomas,
hemangiomas
Mammary gland:
adenocarcinoma,
carcinoma or
adenoacanthoma
Liver:
hepatocellular
adenoma or
carcinoma
Forestomach:
squamous cell
papilloma or
carcinoma
Harderian gland:
adenoma or
carcinoma
Zymbal's gland:
carcinoma
MLE
COEFFICIENTS3
q0 = 5.95 x 1(T10
q2 = 5.76 x 1(T10
z =4.1
qi = 7.40 x 1(T7
z = 2.1
q0=1.35 x 1(T15
q2=1.97 x 10~17
z =6.8
q0 = 6.82 x 1(T4
qi = 4.97 x 1(T5
z= 1.0
q0=1.15 x ID"10
qj = 5.27 x 1(T12
z = 4.8
q0 = 6.98 x 1(T23
ql = 1.33 x 1(T23
z=10
q0 = 4.68 x 1(T14
ql = 6.62 x 1(T15
z = 5.9
q3 = 2.22 x 1(T27
z=10
POINT OF DEPARTURE11 (jig/m3)
Modeled from bioassay
BMDL10
3.38 x 103
2.54 x 104
1.97 x 104
4.88 x 104
9.74 x 103
6.07 x 104
4.81 x 104
7.86 x 104
BMD10
4.22 x 103
3.91 x 104
3.56 x 104
7.52 x 104
1.30 x 104
1.97 x 105
6.05 x 104
2.52 x 105
Continuous, Human
equivalent0
BMDL10
6.02 x Iff
2.47 x 103
4.53 x 103
3.51 x 103
8.69 x 103
1.73 x 103
1.08 x 104
8.56 x 103
1.40 x 104
BMD10
7.51 x Iff
3.08 x 103
6.95 x 103
6.34 x 103
1.34 x 104
2.31 x 103
3.51 x 104
1.08 x 104
4.49 x 104
UNIT
RISKd
/ftig/m3)
1.7* Iff4
4.1 x 10~5
2.2 x 10~5
8.3 x 10~5
1.2 x 10~5
5.78x 10~5
9.3 x 10^
1.2 x 10~5
7.2x 10^
OVERALL
UNIT
RISK6
/Oig/m3)
3.3 x Iff4
2.1 x 10~4
a Model: multistage-Weibull: P(d) = 1 - exp[-(q0 + qid + q2d2 + ... + qkdk) x (t - t0)z], coefficients estimated in terms
of ppm as administered in bioassay; lower stage q; not listed were estimated to be zero.
b BMD10 = Concentration at 10% extra risk; BMDL10= 95% lower bound on concentration at 10% extra risk.
0 Continuous equivalent estimated by multiplying exposures by (6 hours)/(24 hours) x (5 days)/(7 days).
d Unit risk estimated by dividing the BMR (0.1) by the BMDL10.
e Overall unit risk estimate, across all sites listed; see text for method.
f Values in italics indicate BMD/BMDL when lung tumors are treated as systemic lesions
* Tumor incidence data from NTP (1998)
1 Following the recommendations of the NRC (1994) and the current Guidelines for Carcinogen
1 Risk Assessment (U.S. EPA, 2005a) to consider total risk, a statistically-based upper bound on total risk
3 was estimated in order to gain some understanding of the total risk from multiple tumor sites in female
4 and male B6C3F1 mice. Note that this upper bound estimate of overall risk describes the risk of
5 developing any combination of the tumor types considered, not just the risk of developing all
6 simultaneously. Statistical methods which can accommodate the underlying distribution of slope
7 factors are optimal, such as through maximum likelihood estimation or through bootstrapping or
8 Bayesian analysis. However, these methods have not yet been extended to models such as the
9 multistage-Weibull model. Consequently, this analysis used the same method as in several previous
September 2009 5-19 DRAFT - DO NOT CITE OR QUOTE
-------
1 assessments (e.g., 1,3-butadiene (US EPA, 2002b), 1,2-dibromoethane (US EPA, 2005)) which
2 involves assuming that variability in slope factors can be characterized by a normal distribution. Each
3 overall risk estimate involved the following steps (detailed in Appendix C):
4 • It was assumed that the tumor types associated with chloroprene exposure were statistically
5 independent - that is, that the occurrence of a hemangiosarcoma, say, was not dependent on
6 whether there was a forestomach tumor. This assumption cannot currently be verified and if
7 not correct could lead to an overestimate of risk from summing across tumor sites. However,
8 NRC (1994) argued that a general assumption of statistical independence of tumor-type
9 occurrences within animals was not likely to introduce substantial error in assessing
10 carcinogenic potency from rodent bioassay data.
11 • The models previously fitted to estimate the BMDs and BMDLs were used to extrapolate to a
12 lower level of risk (R) where the BMDs and BMDLs were in a linear range. For these data a
13 10~2 risk was generally the lowest risk necessary. Although this step appears to differ from the
14 explicit recommendation of the cancer guidelines (U.S. EPA, 2005a) to estimate cancer risk
15 from a POD "near the lower end of the observed range, without significant extrapolation to
16 lower doses," this method is recommended in the cancer guidelines as a method for combining
17 multiple extrapolations. A sensitivity analysis considering risks nearer the lower end of the
18 observed ranges for each tumor type was also considered and is described below with the
19 results. The unit risk for each site was then estimated by R/BMDLR, as for the estimates for
20 each tumor site above.
21 • The central tendency estimates of unit potency (that is, risk per unit of exposure) at each
22 BMDR, estimated by R/BMDR, were summed across the sites listed in Table 5-6 for male mice
23 and similarly across the sites for female mice listed in Table 5-7.
24 • An estimate of the 95% upper bound on the overall unit risk was calculated by assuming a
25 normal distribution for the individual risk estimates and deriving the variance of the risk
26 estimate for each tumor site from its 95% upper confidence limit (UCL) according to the
27 following formula:
28 95% UCL = MLE + 1.645 x SD (1)
29 rearranged to:
30 SD = (UCL - MLE)/1.645 (2)
31 where 1.645 is the t-statistic corresponding to a one-sided 95% confidence interval and > 120 degrees
32 of freedom, and the standard deviation (SD) is the square root of the variance of the MLE. The
33 variances (variance = SD2) for each site-specific estimate were summed across tumor sites to obtain
September 2009 5-20 DRAFT - DO NOT CITE OR QUOTE
-------
1 the variance of the sum of the MLEs. The 95% UCL on the sum of the individual MLEs was
2 calculated from expression (1) using the variance of the MLE to obtain the relevant SD (SD =
3 variance172).
4 The resulting combined unit risk for all tumor types for female mice was 3.3 x 10"4 per ug/m3
5 (with lung tumors treated as a systemic effect). Overall, the consideration of the other tumor sites
6 increased the unit risk by two-fold from the highest unit risk for any individual tumor type, 1.7 x 10"4
7 per ug/m3 for lung tumors treated as a systemic lesion. The increase was due largely to the
8 hemangiosarcomas and liver tumors, with little contribution from the other tumor sites. A sensitivity
9 analysis (not included in this document) showed that the overall risk was essentially the same (to 2
10 significant digits) whether or not the individual risks were estimated in the region of 10"2 risk or near
11 thePODs.
12 For male mice the combined unit risk for all tumor types was 1.7 x 10"4 per ug/m3 (with lung
13 tumors treated as a systemic lesion), a 2-fold increase compared to the highest unit risk for any
14 individual tumor type, 8.5 x 10"5 per ug/m3 for lung tumors treated as a systemic lesion. The increase
15 was due almost entirely to the risk associated with the hemangiosarcomas. As with the overall risk for
16 female mice, there was a trivial difference whether or not the individual risks were estimated in the
17 region of 10"2 risk or near the PODs.
18 For estimates in both species, if the lung tumors are primarily site of contact lesions, the
19 estimated overall risk decreases to 2.1 x 10"4 per ug/m3 (females) and 1.1 x 10"4 per ug/m3 (males).
20 Based on the relatively high fatair partition coefficients (see Section 3.2.) in rodents and humans,
21 chloroprene is likely to be absorbed rapidly (U.S. EPA, 1994), consistent with the possibility that the
22 lung tumors are both portal-of-entry and systemic lesions.
23 Based on the analyses discussed above, the recommended upper bound estimate on human
24 extra cancer risk from continuous lifetime exposure to chloroprene is 3 x 10"4 per ug/m3, rounding the
25 overall risk for female mice above to one significant digit. This unit risk should not be used with
26 continuous lifetime exposures greater than 600 ug/m3 (0.6 mg/m3), the human equivalent POD for the
27 female lung tumors, because the observed dose-response relationships do not continue linearly above
28 this level and the fitted dose-response models better characterize what is known about the
29 carcinogenicity of chloroprene. The recommended unit risk estimate reflects the time-to-tumor
30 dimension of the responses as well as the exposure-response relationships for the multiple tumor sites
31 in both sexes of mice.
5.4.5 Application of Age-Dependent Adjustment Factors
32 Because a mutagenic mode of action for chloroprene carcinogenicity is sufficiently supported
33 by in vivo and in vitro data and relevant to humans (see Section 4.7.3.1), and in the absence of
34 chemical-specific data to evaluate the differences in susceptibility, increased early-life susceptibility is
35 assumed and the age-dependent adjustment factors (ADAFs) should be applied, as appropriate, along
36 with specific exposure data in accordance with EPA's Supplemental Guidance for Assessing
September 2009 5-21 DRAFT - DO NOT CITE OR QUOTE
-------
1 Susceptibility From Early-Life Exposure to Carcinogens (U.S. EPA, 2005b). The inhalation unit risk
2 of 3 x icr4 per ug/m3, calculated from data for adult exposures, does not reflect presumed early-life
3 susceptibility for this chemical. Example evaluations of cancer risks based on age at exposure are
4 given in Section 6 of the Supplemental Guidance.
5 The Supplemental Guidance establishes ADAFs for three specific age groups. The current
6 ADAFs and their age groupings are 10 for <2 years, 3 for 2 to <16 years, and 1 for 16 years and above
7 (U.S. EPA, 2005b). The 10-fold and 3-fold adjustments in slope factor are to be combined with age
8 specific exposure estimates when estimating cancer risks from early life (<16 years age) exposure to
9 chloroprene.
10 To illustrate the use of the ADAFs established in the Supplemental Guidance (U.S. EPA,
11 2005b), sample calculations are presented for a lifetime risk estimate for continuous exposure from
12 birth with a life expectancy of 70 years. The ADAFs are first applied to obtain risk estimates for
13 continuous exposure over the three age groups:
14 • Risk for birth through < 2 yr = 3 x 1CT4 per ug/m3 x 10 x 2yr/70yr = 8.6 x 10"5 per ug/m3
15 • Risk for ages 2 through < 16 = 3 x ID"* per ug/m3 x 3 x 14yr/70yr = 1.8 x 10"4per ug/m3
16 • Risk for ages 16 until 70 = 3 x 1CT4 per ug/m3 x 1 x 54yr/70yr = 2.3 x 10"4per ug/m3
17 To calculate the lifetime risk estimate for continuous exposure from birth for a population with
18 default life expectancy of 70 years, the risk associated with each of the three relevant time periods is
19 summed:
20 • Risk =8.6 x 10'5+1.8 x 10'4+2.3 x 10'4=5.0 x 10'4 per ug/m3
21 Using the above full lifetime unit risk estimate of 5 x 10"4 per ug/m3 for continuous exposure
22 from birth to 70 years, the lifetime chronic exposure level of chloroprene corresponding to an extra
23 risk of 1 x 10"6 can be estimated as follows:
24 1 x lO'6 4- 5 x 10'4 per ug/m3 = 0.002 ug/m3
5.4.6. Previous Cancer Assessment
25 The carcinogenicity of chloroprene has not been evaluated previously for the IRIS program.
5.4.7. Uncertainties in Cancer Risk Values
26 A number of uncertainties underlie the cancer unit risk for chloroprene. These are discussed in
27 the following paragraphs. Specifically addressed is the impact on the assessment of issues such as the
28 use of models and extrapolation approaches, the use of other bioassay data, and the choices made and
29 the data gaps identified. In addition, the use of assumptions, particularly those underlying the
30 Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005a) is explained and the decision
September 2009 5-22 DRAFT - DO NOT CITE OR QUOTE
-------
1 concerning the preferred approach is given and justified. Principal uncertainties are discussed below
2 and summarized in Table 5-8.
Table 5-8. Summary of uncertainties in chloroprene cancer unit risk estimate
Consideration
Human
population
variability in
metabolism and
response/
sensitive
subpopulations
Low-dose
extrapolation
procedure
Dose metric
Bioassay
Species /gender
combination
Cross-species
extrapolation
Potential Impact8
Low-dose risk could
t or J, to an
unknown extent
Unknown; not clear
what departure from
Cancer Guidelines
would be plausible
Alternatives could t
or J, low-dose risk
per unit
concentration by an
unknown extent
Others unavailable
Human risk could J,
or t, depending on
relative sensitivity
Alternatives for
lung tumors differ
by 4-fold: human
risk for any site
could J, or |. Low-
dose risk would J,
approximately 40%
if lung tumors were
treated as portal-of-
entry effects
Decision
Considered qualitatively
Multistage-Weibull
model to determine
POD, linear low-dose
extrapolation from POD
Used administered
concentration
NTP study
Multiple sites in female
mice
RfC methodology:
Equal risk per unit of air
concentration for all
sites; for lung also
considered relative
surface areas of affected
region. Treat lung
tumors as systemic
effects.
Justification
No data to support range of human
variability /sensitivity. Mutagenic MO A
indicates potentially increased early -life
susceptibility.
Multistage-Weibull model addresses
competing risks from other tumors and
intercurrent mortality. Mutagenic MOA
supports linear low-dose extrapolation.
Experimental evidence supports a role for
metabolism in toxicity, but actual
responsible metabolites are neither clearly
identified nor quantifiable. Use of
administered concentration provides an
unbiased estimate if proportional to the
actural carcinogen(s).
Standard design, well conducted,
extensively peer reviewed; carcinogenic
response consistently observed across all 4
species/sex combinations.
Unit risk is based on the most sensitive
endpoint (risk of any tumor type) in the
most sensitive species and gender (female
mouse), based on PODHEc It was assumed
that humans are as sensitive as the most
sensitive rodent gender/species tested; true
correspondence is unknown. Site
concordance for liver tumors for humans
and female mice was observed, but human
data not sufficient to rule out other types
seen in mice or rats.
There are no data to support other
alternatives. There is evidence that
chloroprene is distributed systematically
(observation of tumors at multiple sites),
and correspondingly the possibility that
chloroprene is redistributed to the lungs.
The contribution of one route of delivery
(i.e., inhalation vs. bloodstream) to the
induction of lung tumors is currently
unknown, therefore the derivation approach
that returns the highest unit risk was used
September 2009
5-23
DRAFT - DO NOT CITE OR QUOTE
-------
Consideration
Statistical
uncertainty at
POD
Potential Impact"
I risk per unit
concentration 1.2-
fold if BMD10 used
rather than BMDL10
Decision
BMCL (default approach
for calculating plausible
upper bound)
Justification
Limited size of bioassay results in sampling
variability; lower bound is 95% confidence
interval on concentration.
a t = increase, J, = decrease
1 Human population variability. The extent of inter-individual variability in chloroprene
2 metabolism has not been characterized. A separate issue is that the human variability in response to
3 chloroprene is also poorly understood. The effect of metabolic variation, including potential
4 implications for differential toxicity, has not been well studied. Although a mutagenic MO A indicates
5 increased early-life susceptibility, there are no data exploring whether there is differential sensitivity to
6 chloroprene carcinogenicity across human life stages. This lack of understanding about potential
7 differences in metabolism and susceptibility across exposed human populations thus represents a
8 source of uncertainty.
Choice of low-dose extrapolation approach. The MOA is a key consideration in clarifying how
risks should be estimated for low-dose exposure. A multistage Weibull time-to-tumor model was the
preferred model because it can account for differences in mortality and other competing risks between
the exposure groups in the mouse bioassay; however, it is unknown how well this model predicts low-
dose extrapolated risks for chloroprene.
Dose metric. Chloroprene is metabolized to intermediates with carcinogenic potential, most
likely an epoxide. However, data sufficient to estimate quantities were not available. Under the
assumption that the carcinogenic form(s) of chloroprene are produced in proportion to low exposures
of chloroprene, the derived unit risk is an unbiased estimate.
Choice ofbioassay/species/gender. The NTP inhalation bioassay followed an accepted
protocol, was well conducted, and extensively peer reviewed. The carcinogenic response occurs in
both species and sexes of rodents as well as in humans. The calculated combined unit risk is based on
the most sensitive endpoint (risk of any tumor type) in the most sensitive species and gender (female
mouse). There is no information on chloroprene to indicate that the observed rodent tumors are not
relevant to humans. Further, no data exist to guide quantitative adjustment for differences in
sensitivity among rodents and humans. While site concordance generally is not assumed across
species, e.g., due to potential differences in pharmacokinetics, DNA repair, other protective systems
across species and tissues (U.S. EPA, 2005a), it is notable that human-mouse site concordance was
observed for liver tumors. In addition, rat and mouse tumor types overlapped but included different
tumor types observed for each species/sex combination. Human data were insufficient to rule out the
occurrence of these additional tumor types in humans.
9 Cross-species scaling. Another source of uncertainty comes from the interspecies extrapolation
10 of risk from mouse to human. The two rodent species for which bioassay data were available— mouse
September 2009
5-24
DRAFT - DO NOT CITE OR QUOTE
-------
1 and rat—vary in their carcinogenic responses to chloroprene, in terms of both site specificity and
2 magnitude of response (see Chapter 4). Ideally, a PBPK model for the internal dose(s) of the reactive
3 metabolite(s) would decrease some of the quantitative uncertainty in interspecies extrapolation;
4 however, current PBPK models are inadequate for this purpose (Chapter 3). Existing pharmacokinetic
5 models cannot yet adequately explain the species differences in carcinogenic response, and it is
6 possible that there are pharmacodynamic as well as pharmacokinetic differences between the mouse
7 and rat with respect to their sensitivities to chloroprene.
8 While concordance of specific sites between rodents and humans (e.g., liver tumors) tends to
9 support the relevance of rodent species to humans, lack of specific site concordance (other tumors)
10 does not diminish concern for human carcinogenic potential. The mouse was the more sensitive
11 species to the carcinogenic effects of chloroprene exposure. Although the derivation took into account
12 some known differences between mice and humans in tissue dosimetry (US EPA, 1994), differences in
13 anatomy of the upper respiratory tract and resulting differences in absorption or in local respiratory
14 system effects are sources of uncertainty.
Statistical uncertainty at the Point of Departure (POD). Parameter uncertainty within the
chosen model reflects the limited sample size of the cancer bioassay. For the multistage-Weibull
model applied to this data set, there is a reasonably small degree of uncertainty at the 10% extra risk
level (the POD for linear low-dose extrapolation). Central estimates of risk differed from their upper
bounds by about 1.2-fold for lung tumors and for the overall risk estimates.
HEC derivation. A source of uncertainty in the derivation of the HEC comes from whether or
nor chloroprene induces lung tumors due to portal-of-entry or systemic effects. Systemic distribution
of chloroprene is evidenced by the induction of tumors in multiple organs and suggests that
chloroprene may be redistributed back to the lungs and may potentially act as a systemically delivered
carcinogen rather than, or in addition to, a portal-of-entry toxicant. However, the contribution of either
route of delivery (i.e., inhalation vs. bloodstream) to the induction of lung tumors is currently
unknown. Treating lung tumors as systemic effects returns the highest combined unit risk
(approximately 60% greater than if lung tumors are treated as portal-of-entry effects).
September 2009 5-25 DRAFT - DO NOT CITE OR QUOTE
-------
6. MAJOR CONCLUSIONS IN CHARACTERIZATION OF HAZARD AND DOSE RESPONSE
6.1. HUMAN HAZARD POTENTIAL
1 Chloroprene (C4H5C1, 2-chloro-1,3-butadiene, CASRN 126-99-8) is a volatile and flammable
2 liquid monomer that can be produced by dimerization of acetylene and addition of hydrogen chloride
3 or by chlorination of 1,3-butadiene. Chloroprene is polymerized to form elastomers for use in the
4 manufacture of belts, hoses, gloves, wire coatings, tubing, solvents, and adhesives. Chloroprene is
5 also a structural analogue of isoprene (2-methyl-1,3-butadiene) and resembles vinyl chloride as far as
6 having a chlorine bound to a double-bonded carbon (alkene) backbone.
7 Toxicokinetic information on the absorption, distribution, and in vivo metabolism and excretion
8 of chloroprene and/or its metabolites is nonexistent for humans and limited for animals. Several in
9 vitro studies have focused on chloroprene metabolism in lung and liver tissue fractions from rat,
10 mouse, hamster, and humans (Hurst and Ali, 2007; Munter et al., 2007a, b, 2003; Himmelstein et al.,
11 2004a, 2001a, 2001b; Cottrell et al., 2001; Summer and Greim, 1980). These studies suggest that
12 chloroprene is metabolized via the CYP450 enzyme system to monoepoxides [(1-
13 chloroethenyl)oxirane and 2-chloro-2-ethynyloxirane], further metabolized to aldehydes and ketone
14 intermediates and subsequent mercapturic acid derivatives, and cleared via hydrolysis and/or
15 glutathione conjugation reactions. Similar to 1,3-butadiene, the epoxide metabolite is considered to be
16 the toxic moiety. The metabolic profile for chloroprene is qualitatively similar across species.
17 However, in vitro kinetic studies using tissues from rodents and humans suggest quantitative species
18 and tissue-specific differences that, if operative in vivo, could contribute to the species, strain, and
19 gender differences observed in chloroprene-induced effects.
20 Limited information exists on the noncancer effects of chloroprene due to oral ingestion. In
21 rats, oral exposures from weaning until death (at 120 weeks) resulted in indices of liver toxicity (liver
22 necroses and degenerative lesions of th eparenchymal cells). No information is available on
23 chloroprene's oral toxicity in humans.
24 Limited information exists on the noncancer effects of chloroprene via the inhalation route in
25 humans. Chloroprene was reported to cause respiratory, ocular, and dermal irritation, chest pains,
26 temporary hair loss, dizziness, insomnia, headache, and fatigue. Chest pains accompanied by
27 tachycardia and dyspnea were also reported. In a Russian review of the effects of chloroprene,
28 Sanotskii (1976) reported that medical examinations of chloroprene production workers revealed
29 changes in the nervous system (lengthening of sensorimotor response to visual cues and increased
30 olfactory thresholds), cardiovascular system (muffled heart sounds, reduced arterial pressure, and
31 tachycardia), and hematology (reduction in red blood cell (RBC) counts, decreased hemoglobin levels,
32 erythrocytopenia, leucopenia, and thrombocytopenia). The ambient concentration of chloroprene
33 associated with these effects ranged from 1-7 mg/m3.
September 2009 6-1 DRAFT - DO NOT CITE OR QUOTE
-------
1 Chloroprene's toxic and carcinogenic potential by the inhalation route has been assessed in
2 several laboratory animal studies, including a rat and mouse subchronic (16 days and 13 weeks) and
3 chronic inhalation bioassays conducted by NTP (1998), a subchronic range-finding and a chronic study
4 in rats and hamsters conducted by Trochimowicz et al. (1998), an embryotoxicity and a teratology
5 study by Culik et al. (1978), and a series of Russian reproductive and developmental toxicity studies
6 reviewed by Sanotskii (1976). These studies associate chloroprene inhalation exposure with
7 respiratory, kidney, liver, forestomach, reproductive, and developmental effects. The pulmonary
8 (alveolar hyperplasia) and nasal (olfactory epithelium) lesions were the most sensitive endpoints in
9 chronically exposed test animals, having been observed at all the doses tested (12.8-80 ppm) in the
10 NTP (1998) study of rats and mice. In the chronic study by Trochimowicz et al. (1998), lesions in
11 lungs (inflammation, lymphoid aggregates around the bronchi, bronchiole, and blood vessels) and
12 livers (small foci of cellular alteration) of rats were observed at 50 ppm. Embryotoxicity and fetal
13 resorptions were reported in the inhalation developmental toxicity study (Culik et al., 1978). However,
14 interpretational difficulties obscure whether this effect is an actual outcome or rather a statistical
15 artifact of an abnormally low background rate in control animals.
16 Chloroprene's carcinogenic potential in humans has been assessed in a number of occupational
17 epidemiologic studies among workers exposed to chloroprene monomer and/or polychloroprene latex
18 conducted in 11 cohorts from the United States, Russia, Armenia, France, China, and Ireland. Five
19 cohorts with sufficient numbers of liver/biliary passage cancer cases showed some evidence of
20 association with occupational chloroprene exposure. Four mortality studies reported elevated SMRs
21 when compared to external populations (Marsh et al., 2007a; Bulbulyan et al., 1999; Li et al., 1989;
22 Leet and Selevan, 1982). These measures of association were strong, especially in the presence of the
23 healthy worker effect bias. Several studies were able to use more advanced exposure assessments and
24 internal reference populations, which should reduce this bias. These studies showed relatively
25 consistent elevated relative risk estimates among intermediate and highly exposed workers, despite
26 limited sample size and statistical power (Marsh et al., 2007a; Bulbulyan et al., 1999, 1998). Several
27 studies also reported higher SMRs for lung cancer among workers exposed to chloroprene. These
28 associations are not considered as strong as those with liver cancer due to the inability to control for
29 confounding by smoking status, a strong indicator of lung cancer.
30 Chloroprene has been shown to induce multisite, malignant tumors in rats and mice in the 2-
31 year NTP (1998) bioassay. Dose-related increasing trends in tumors were noted in rats at the following
32 sites: oral cavity, thyroid gland, lung, kidney, mammary gland. Dose related increasing trends in
33 tumors were noted in mice at the following sites: lung, all organs (hemangiomas and
34 hemangiosarcomas), Harderian gland, forestomach, kidney, skin, liver, mammary gland, mesentery,
35 Zymbal's gland. All of these tumor sites showed statistically significantly positive trends with
36 increasing exposure level (Cochran-Armitage test for trend p < 0.05, most with p < 0.001). In
September 2009 6-2 DRAFT - DO NOT CITE OR QUOTE
-------
1 addition, many early deaths and moribund sacrifices were associated with chloroprene-induced
2 neoplasms.
3 The genetic toxicity database includes numerous studies covering a range of standard genotoxicity test
4 batteries; however, the results have been conflicting, making it difficult to ascertain the mutagenic
5 potential of chloroprene. In general, bacterial base pair substitution (S. typhimurium strains TA100
6 and TA 1535) mutation assays have been positive (Willems, 1980; Bartsch et al., 1979), while the
7 bacterial frame shift (S. typhimurium strains TA97 and TA98) mutation assays have been nonpositive
8 (NTP, 1998; Willems, 1980). In contrast, other studies (NTP, 1998) have reported nonpositive results
9 for all bacterial strains. A positive result with all bacterial strains was observed with chloroprene's
10 epoxide intermediate epoxide (l-chloroethenyl)oxirane (Himmelstein et al., 2001a). Chloroprene has
11 been primarily nonpositive in in vitro micronucleus assays (Himmelstein et al., 2001a; Drevon and
12 Kuroki, 1979), in vivo chromosomal damage assays (NTP, 1998), and bone marrow micronucleus
13 assays (NTP, 1998; Shelby and Witt, 1995). Conflicting results (positive in Vogel [1979]; nonpositive
14 in Foureman et al. [1994]) have been reported for the in vivo drosophila sex-linked lethal mutation
15 assay. Further in vivo evidence for chloroprene's mutagenicity is the observation that tissues from
16 lung, forestomach, and Harderian gland tumors from mice exposed to chloroprene in the NTP chronic
17 bioassay (1998) were shown to have a higher frequency of mutations in K- and H-ras proto-oncogenes
18 than in spontaneous occurring tumors (Stills et al., 1999, 2001). There was also a high correlation
19 between K-ras mutations and loss of heterozygosity in the same chromosome in chloroprene-induced
20 lung neoplasms in mice (Ton et al., 2007). Possible explanations for the conflicting mutagenic
21 responses of chloroprene in standard genotoxicity assays include methods of exposure that do not
22 control for the high volatility of chloroprene (i.e., chloroprene is not present in the test system), the
23 presence of more stable (perhaps more toxic) chloroprene dimers, the use of microsomal inducers that
24 did not elicit a broad range of metabolic enzymes (specifically, in bacterial assays), and the reactivity
25 (perhaps deactivation) of chloroprene with treatment vehicle (e.g., DMSO vs. ethanol).
26 The likely MOA for chloroprene is via mutagenicity involving epoxide metabolites formed at
27 the target sites. The MOA determination is supported by chloroprene's epoxide metabolite formation,
28 DNA-adduct formation, observation of in vivo and in vitro mutagenicity, and the well known
29 structure-activity relationship of similar epoxide-forming carcinogens. Chloroprene has been found to
30 be metabolized to epoxides by humans and rodents. The hypothesized mutagenic mode of action is
31 supported by evidence of base pair substitution mutations seen in H- and K-ras proto-oncogenes in
32 chloroprene-induced lung, forestomach, and Harderian gland neoplasms observed in the NTP (1998)
33 study.
34 In addition, chloroprene is the 2-chloro analog of 1,3-butadiene. Inhalation studies have
35 demonstrated that, similar to 1,3-butadiene and isoprene, chloroprene is a multisite carcinogen in rats
36 and mice. Butadiene and isoprene are metabolized to epoxides and diepoxides which are believed to
37 be responsible for their carcinogenicity. Chloroprene is also metabolized to epoxide intermediates
September 2009 6-3 DRAFT - DO NOT CITE OR QUOTE
-------
1 that, similarly to butadiene, may mediate its carcinogenic effects. The similarities in the sites of tumor
2 induction in rodents (mammary gland and thyroid gland in rats, lung, Harderian gland, forestomach,
3 kidney, and liver in mice) between butadiene and chloroprene provide further evidence for a similar
4 MOAfor these epoxide-forming compounds. In addition, the mouse lung was the most sensitive site
5 of carcinogen!city for both chloroprene and butadiene. Similar to butadiene, DNA reactivity and
6 adduct formation have been described for chloroprene. Areas of uncertainty exist in the data
7 supporting a mutagenic MOA for chloroprene carcinogenicity, more specifically in the genotoxicity
8 database. There is conflicting evidence in the bacterial genotoxocity assays and generally nonpositive
9 findings in mammalian in vivo tests, but these results are weighed against the base pair substitution
10 mutations seen in H- and K-ras proto-oncogenes in chloroprene-induced lung, forestomach, and
11 Harderian gland neoplasms observed in the NTP (1998) study.
6.2. DOSE RESPONSE
12 The chronic inhalation study conducted by NTP (1998) was considered as the principal study
13 for both the non-neoplastic and neoplastic effects of chloroprene exposure.
14 The respiratory system is the primary targets of chloroprene-induced non-neoplastic effects via
15 inhalation. A range of portal-of-entry non-neoplastic effects from the NTP study (1998), including
16 alveolar epithelial hyperplasia, bronchiolar hyperplasia, pulmonary histiocytic cell infiltration,
17 olfactory epithelial atrophy, chronic inflammation, and necrosis were considered as candidates for the
18 selection of the critical effect for derivation of the RfC. BMD modeling was used to determine
19 potential PODs for deriving the chronic RfC by estimating the effective dose at a specified level of
20 response (benchmark concentration [BMDio]) and it's BMDLio for each selected chloroprene-induced
21 respiratory and systemic effect (see Table 5-2). The HEC for each of the selected endpoints was then
22 estimated for the best fitting models of the BMD at a BMR of 10% extra risk. Degenerative nasal
23 lesions in male rats (characterized by atrophy or necrosis of the olfactory epithelium) resulted in the
24 lowest POD(HEC) value of approximately 1.0 mg/m3. This POD was then divided by a 100-fold UF (3
25 for uncertainty associated with animal to human differences, 10 for consideration of human variability,
26 and 3 for database deficiencies) to obtain a chronic RfC of 1 x 10~2 mg/m3.
27 Statistically significant increases in tumor incidence were observed at multiple sites in the
28 mouse (the most sensitive species) in the NTP study: all organs (hemangiomas and
29 hemangiosarcomas), lung (bronchiolar/alveolar adenomas and carcinomas), forestomach, Harderian
30 gland (adenomas and carcinomas), kidney (adenomas), skin and mesentery, liver, and mammary
31 glands. These tumors generally appeared earlier with increasing exposure level and showed
32 statistically significantly increasing trends with increasing exposure level (by life table test or logistic
33 regression, p < 0.001). Dose-response modeling was used to determine potential PODs for deriving
34 the inhalation unit risk by estimating the effective dose at a specified level of response (benchmark
35 concentration [BMDio]) and its BMDLio for each selected chloroprene-induced tumor (see Tables 5-6
September 2009 6-4 DRAFT - DO NOT CITE OR QUOTE
-------
1 and 5-7). Lung tumors, treated as a systemic lesion (see Section 5.4.3 and 5.4.7 for details), in female
2 mice resulted in the highest inhalation unit risk (1.7 x 10"4 per ug/m3) when modeled as an individual
3 lesion. When etiologically different tumors were considered together (given the multiplicity of the
4 tumor sites, basing unit risk on only one tumor site may underestimate the carcinogenic potential of
5 chloroprene), the resulting combined inhalational unit risk for female mice was 3.3 x 10"4per ug/m3
6 (when lung tumors were considered systemic lesions). Based on these modeling results, the upper
7 bound estimate on human extra lifetime cancer risk from continuous lifetime (adult) exposure to
8 chloroprene is 3 x 10"4per ug/m3. Application of the ADAFs to account for early-life susceptibility to
9 chloroprene's proposed mutagenic mode of action yields an adjusted human lifetime cancer risk of 5 x
10 10"4per ug/m3.
11 Confidence in the principal study (NTP, 1998) is judged to be high as it was a well-designed
12 study using two test species (rats and mice) with 50 animals per dose group. This study appropriately
13 characterizes a range of chloroprene-induced non-neoplastic and neoplastic lesions. In addition, the
14 key histopathological lesions observed are appropriately described, and suitable statistical analysis is
15 applied to all animal data.
16 Confidence in the critical non-cancer effect identified in the principal study is medium to high.
17 The critical non-cancer effect, increased incidence of degenerative nasal lesions in male rats, is
18 consistent with what is known about chloroprene's metabolism and the expression of cytochrome p450
19 enzymes in the olfactory and respiratory mucosa of rats, as well as the effects of structurally analogous
20 chemicals (i.e. 1,3-butadiene).
21 Confidence in the overall database specific to chloroprene is medium to high. The major
22 strength of the database is the observation of dose-response effects in multiple organ systems in a well-
23 designed chronic inhalation study that utilized 50 animals per sex per dose group, a range of doses
24 based on the results of preliminary, shorter-duration studies (16 day and 13 weeks), and thoroughly
25 examined chloroprene's observed toxicity in two species (rat and mouse). The database further
26 contains another chronic inhalation bioassay investigating outcomes in another species (hamster), and
27 well-designed embryotoxicity, teratological, and reproductive toxicity studies. The database also
28 contains subchronic studies and chronic studies observing potential neurotoxic and immunotoxic
29 effects. A major limitation in the database is the lack of a two-generation reproductive toxicity study.
30 Therefore, confidence in the RfC is judged to be medium to high.
September 2009 6-5 DRAFT - DO NOT CITE OR QUOTE
-------
7. REFERENCES
1 Appelman, LM and Dreef van der Meulen HC. (1979) Reproduction study with p-chloroprene vapor in rats. Report no. R
2 6225. Produced by the Central Institute for Nutrition and Food Research (CIVO), The Netherlands, submitted by DuPont
3 Chemical, Wilmington, DE, under TSCA Section 8D; EPA Document No. 878215005; NTIS No. OTS0206752.
4 Bartsch, H; Malaveille, C; Barbin, A; et al. (1979) Mutagenic and alkylating metabolites of halo-ethylenes,
5 chlorobutadienes and dichlorobutenes produced by rodent or human liver tissues. Evidence for oxirane formation by P450-
6 linked microsomal mono-oxygenases. Arch Toxicol 41(4):249-277.
7 Begemann, P; Christova-Georgieva, NI; Sangaiah, R; et al (2004) Synthesis, characterization, and identification of N7-
8 guanine adducts of isoprene monoepoxides in vitro. Chem Res Toxicol 17: 929-936.
9 Bulbulyan, MA; Changuina, OV; Zaridze, DG; et al. (1998) Cancer mortality among Moscow shoe workers exposed to
10 chloroprene (Russia). Cancer Causes Control 9(4):381-387.
11 Bulbulyan, MA; Margaryan, AG; Ilychova, SA; et al. (1999) Cancer incidence and mortality in a cohort of chloroprene
12 workers from Armenia. Int J Cancer 81(1):31—33.
13 CambridgeSoft Corp. (2007) Chloroprene. ChemFinder. CambridgeSoft Corporation, Cambridge, MA. Available online at
14 http://chemfmder.cambridgesoft.com.
15 Charles River Laboratories. (1996) Historical control data (1992-1994) for developmental and reproductive toxicity studies
16 using the Crl:CD (SD)BR rat. Compiled by the Middle Atlantic Reproduction and Teratology Association. Available on-
17 line at: http://www.criver.com/sitecollectiondocuments/rm rmrtox studies cried sdbrrat.pdf
18 Clary, JJ; Feron, VJ; Reuzel, PG. (1978) Toxicity of beta-chloroprene (2-chlorobutadiene-l,3): acute and subacute toxicity.
19 Toxicol Appl Pharmacol 46(2):375-384.
20 Colonna, M; Laydevant, G. (2001) A cohort study of workers exposed to chloroprene in the department of Isere, France.
21 ChemBiol Interact 135-136:505-514.
22 Corrao, G; Bagnardi, V; Zambon, A; La Vecchia, CA. (2004) A meta-analysis of alcohol consumption and the risk of 15
23 diseases. PrevMed, 38:613-619.
24 Cottrell, L; Golding, BT; Munter, T; et al. (2001) In vitro metabolism of chloroprene: species differences, epoxide
25 stereochemistry and a de-chlorination pathway. Chem Res Toxicol 14(11): 1552-1562.
26 Cox, DR. (1972) Regression models and life-tables. J R Stat Soc B34: 187-220.
27 Crump, KS. (1984) A new method for determining allowable daily intakes. Fundam Appl Toxicol 4:854-871.
28 Crump, KS. (1995) Calculation of benchmark doses from continuous data. Risk Anal 15:79-89.
29 Culik, R; Kelly, DP; Clary, JJ. (1978) Inhalation studies to evaluate the teratogenic and embryotoxic potential of beta-
30 chloroprene (2-chlorobutadiene-l,3). Toxicol Appl Pharmacol 44(l):81-88.
31 Dong, QA; Xiao, BL; Hu, YH; et al. (1989) Short-term test for the induction of lung tumor in mouse by chloroprene.
32 Biomed Environ Sci 2(2):150-153.
33 Drevon, C; Kuroki, T. (1979) Mutagenicity of vinyl chloride, vinylidene chloride and chloroprene in V79 Chinese hamster
34 cells. Mutat Res 67(2): 173-182.
35 Esmen, NA; Hall, TA; Phillips, ML; et al. (2007a) Chemical process based reconstruction of exposures for an
36 epidemiological study. I. Theoretical and methodological issues. ChemBiol Interact 166(l-3):254-263.
September 2009 7-1 DRAFT - DO NOT CITE OR QUOTE
-------
1 Esmen, NA; Hall, TA; Phillips, ML; et al. (2007b) Chemical process-based reconstruction of exposures for an
2 epidemiological study. Part II. Estimated exposures to chloroprene and vinyl chloride. Chem Biol Interact 166(l-3):264-
3 276.
4 Esmen, NA; Kennedy, KJ; Hall, TA; et al. (2007c) Classification of worker exposures. Chem Biol Interact 166(l-3):245-
5 253.
6 Foureman, P; Mason, JM; Valencia, R; et al. (1994) Chemical mutagenesis testing in drosophila. X. Results of 70 coded
7 chemicals tested for the National Toxicology Program. Environ Mol Mutagen 23(3):208-227.
8 Gahlmann, R. (1993) Pros and cons of transgenic mouse mutagenesis test systems. J Exp Anim Sci 35(5-6):232-243.
9 Gargas, ML; Burgess, RJ; Voisard, DE; et al. (1989) Partition coefficients of low-molecular-weight volatile chemicals in
10 various liquids and tissues. Toxicol Appl Pharmacol 98(l):87-99.
11 Gart, JJ; Krewski, D; Lee, PN; et al. (1986) Statistical methods in cancer research. Volume III. The design and analysis of
12 long-term animal experiments. IARC Sci Publ 79:1-219.
13 Ginsberg, G; Slikker Jr., W; Bruckner, J; Sonawane, B. (2004) Incorporating children's toxicokinetics into a risk
14 framework. Environ Health Perspect 112(2): 272-283.
15 Griebel, CP; Halvorsen, J; Golemon, TB; et al. (2005) Management of spontaneous abortion. Am Fam Physician
16 72(7): 1243-1250.
17 Grosjean, D. (1990) Atmospheric chemistry of toxic contaminants 1. Reaction rates and atmospheric persistence. J Air
18 Waste Manag Assoc 40:1397-1402.
19 Grosjean, D. (1991) Atmospheric chemistry of toxic contaminants 5. Unsaturated halogenated aliphatics: allyl chloride,
20 chloroprene, hexachlorocyclopentadiene, vinylidene chloride. J Air Waste Manag Assoc 41:182-189.
21 Haley, TJ. (1978) Chloroprene (2-chloro-l, 3-butadiene)~what is the evidence for its carcinogenicity? Clin Toxicol
22 13(2):153-170.
23 Hall, TA; Esmen, NA; Jones, EP; et al. (2007) Chemical process based reconstruction of exposures for an epidemiological
24 study. III. Analysis of industrial hygiene samples. Chem Biol Interact 166(l-3):277-284.
25 Hill, AB. (1965) The environment and disease: Association or causation? Proc R Soc Med 58:295—300.
26 Himmelstein, MW; Gladnick, NL; Dormer, EM; et al. (2001a) In vitro genotoxicity testing of (l-chloroethenyl)oxirane, a
27 metabolite of beta-chloroprene. Chem Biol Interact 135-136:703—713.
28 Himmelstein, MW; Carpenter, SC; Hinderliter, PM; et al. (200 Ib) The metabolism of beta-chloroprene: Preliminary in-vitro
29 studies using liver microsomes. Chem Biol Interact 135-136:267-284.
30 Himmelstein, MW; Carpenter, SC; Hinderliter, PM. (2004a) Kinetic modeling of beta-chloroprene metabolism: I. In vitro
31 rates in liver and lung tissue fractions from mice, rats, hamsters, and humans. Toxicol Sci 79(1): 18-27.
32 Himmelstein, MW; Carpenter, SC; Evans, MV; et al. (2004b) Kinetic modeling of beta-chloroprene metabolism: II. The
33 application of physiologically based modeling for cancer dose response analysis. Toxicol Sci 79(l):28-37.
34 Hurst, HE; Ali, MY. (2007) Analyses of (l-chloroethenyl)oxirane headspace and hemoglobin N-valine adducts in
35 erythrocytes indicate selective detoxification of (l-chloroethenyl)oxirane enantiomers. Chem Biol Interact 166(l-3):332-
36 340.
37 IARC (International Agency for Research on Cancer). (1999) IARC monographs on the evaluation of carcinogenic risks of
38 chemicals to humans. Vol. 71. Re-evaluation of some organic chemicals, hydrazine, and hydrogen peroxide. Lyon, France:
39 International Agency for Research on Cancer. Available online at
40 http://monographs.iarc.fr/ENG/Monographs/vol7 l/volume71 .pdf.
September 2009 7-2 DRAFT - DO NOT CITE OR QUOTE
-------
1 Immels, H; Willems, M. (1978) Initial submission: Dominant lethal assay with beta-chloroprene in male mice with cover
2 letter dated 09/01/92. Produced by the Central Institute for Nutrition and Food Research (CIVO), The Netherlands; Report
3 No. 5756; submitted by Dupont Chemical, Wilmington, DE, under TSCA Section 8ECP; EPA Document No. 88-
4 920008417; NTIS No. OTS0570699.
5 Jemal, A; Murray, T; Samuels, A; et al. (2003) Cancer statistics, 2003. CA Cancer J Clin 53:5-26.
6 Koskinen, M; Calebiro, Davide; Hemminki, K. (2000) Styrene oxide-induced 2'-deoxycytidine adducts: implications for
7 the mutagenicity of styrene oxide. ChemBiol Interact 126:201-213.
8 Krewski, D; Crump, KS; Farmer, J; et al. (1983) A comparison of statistical methods for low dose extrapolation utilizing
9 time-to-tumor data. Fundam Appl Toxicol 3(3):140-160.
10 Kroshwitz, JI; Howe-Grant, M; eds. (1993) Chloroprene. Kirk-Othmer encyclopedia of chemical technology. Vol. 6. 4th
11 edition. New York, NY: Wiley; pp. 70-78.
12 Leet, T; Selevan, S. (1982) Mortality analysis of workers exposed to chloroprene [final report]. E.I. du Pont de Nemours
13 and Co., Inc., Wilmington, DE. Available from the National Technical Information Service, Springfield, VA; PB83-193581.
14 Lefaux, R; ed. (1968) Chloroprene. In: Practical toxicology of plastics. London: Iliffe Books, Ltd.; pp. 69-71.
15 Li, SQ; Dong, QN; Liu, YQ; et al. (1989) Epidemiologic study of cancer mortality among chloroprene workers. Biomed
16 Environ Sci 2(2): 141-149.
17 Luster, MI; Portier, C; Pait, DG; White, Jr., KL; Gennings, G; Munson, AE; Rosenthal, GJ. (1992) Risk assessment in
18 immunotoxicity. I. Sensitivity and predictability of immune tests. Fundam Appl Toxicol 8:200-210.
19 Lynch, J. (200la) BD monomer and elastomer production processes. Chem Biol Interact 135—136:147—153.
20 Lynch, M. (2001b) Manufacture and use of chloroprene monomer. ChemBiol Interact 135-136:155-167.
21 MacGregor, JT; Wehr, CM; Henika, PR; et al. (1990) The in vivo erythrocyte micronucleus test: measurement at steady
22 state increases assay efficiency and permits integration with toxicity studies. Fundam Appl Toxicol 14(3):513-522.
23 Marsh, GM; Youk, AO; Buchanich, JM; et al. (2007a) Mortality patterns among industrial workers exposed to chloroprene
24 and other substances. I. General mortality patterns. Chem Biol Interact 166(l-3):285-300.
25 Marsh, GM; Youk, AO; Buchanich, JM; et al. (2007b) Mortality patterns among industrial workers exposed to chloroprene
26 and other substances. II. Mortality in relation to exposure. Chem Biol Interact 166(1-3): 301—316.
27 Mast, TJ; Evanhoff, JJ; Westerberg, RL; et al. (1994) Inhalation developmental toxicology studies: developmental toxicity
28 of chloroprene vapors in New Zealand white rabbits. Prepared by Battelle Pacific Northwest Laboratories, Richland, WA,
29 for the National Institute of Environmental Health Sciences, U.S. Department of Health and Human Services, Research
30 Triangle Park, NC; PNL-7458. Available from the National Technical Information Service, Springfield, VA; DE94012384.
31 Medinsky, MA and Bond, JA. (2001) Sites and mechanisms for uptake of gases and vapors in the repiratory tract.
32 Toxicology 160: 165-172.
33 Melnick, RL; Sills, RC. (2001) Comparative carcinogenicity of 1,3-butadiene, isoprene, and chloroprene in rats and mice.
34 ChemBiol Interact 135-136:27-42.
35 Melnick, RL; Elwell, MR; Roycroft, JH; et al. (1996) Toxicity of inhaled chloroprene (2-chloro-1,3-butadiene) inF344 rats
36 and B6C3F(1) mice. Toxicology 108(l-2):79-91.
37 Melnick, RL; Sills, RC; Portier, CJ; et al. (1999) Multiple organ carcinogenicity of inhaled chloroprene (2-chloro-1,3-
38 butadiene) in F344/N rats and B6C3F1 mice and comparison of dose-response with 1,3-butadiene in mice. Carcinogenesis
39 20(5):867-878.
September 2009 7-3 DRAFT - DO NOT CITE OR QUOTE
-------
1 Melnick, RL; Sills, RC; Roycroft, JH; Chou, BJ; Ragan, HA; Miller, RA (1994) Isoprene, an endogenous hydrocarbon and
2 industrial chemical, induces multiple organ neoplasia in rodents after 26 weeks of inhalation exposure. Can Res 54:5333-
3 5339.
4 Munter, T; Cottrell, L; Ghai, R; et al. (2007a) The metabolism and molecular toxicology of chloroprene. Chem Biol
5 Interact 166(l-3):323-331.
6 Munter, T; Cottrell, L; Ghai, R; et al. (2007b) Erratum to "The metabolism and molecular toxicology of chloroprene"
7 [Chem Biol Interact 2007; 166(l-3):323-331]. Chem Biol Interact 168(2): 169.
8 Munter, T; Cottrell, L; Golding, BT; et al. (2003) Detoxication pathways involving glutathione and epoxide hydrolase in the
9 in vitro metabolism of chloroprene. Chem Res Toxicol 16(10): 1287—1297.
10 Munter, T; Cottrell, L; Hill, S; et al. (2002) Identification of adducts derived from reactions of (l-chloroethenyl)oxirane
11 with nucleosides and calf thymusDNA. Chem Res Toxicol 15: 1549-1560.
12 NLM (National Library of Medicine). (2008) 2-Chloro-l,3-butadiene. HSDB (Hazardous Substances Data Bank). National
13 Institutes of Health, U.S. Department of Health and Human Services, Bethesda, MD. Available online at
14 http://toxnet.nlm.nih.gov.
15 NRC (National Research Council). (1983) Risk assessment in the federal government: Managing the process. Washington,
16 DC: National Academy Press.
17 NRC (National Research Council). (1994) Science and judgement in risk assessment. Washington, DC:National Academy
18 Press.
19 NTP (National Toxicology Program). (1985) Technical protocol for sperm morphology and vaginal cytology evaluations in
20 toxicity testing for rats and mice, 10/31/82 version (updated March 1985). Public Health Service, U.S. Department of
21 Health and Human Services. Available from the National Institute of Environmental Health Sciences, Research Triangle
22 Park, NC.
23 NTP (National Toxicology Program). (1996) Toxicology and carcinogenesis studies of chloroprene (CAS No. 126-99-8) in
24 F344/N rats and B6C3F1 mice (inhalation studies) [draft]. Public Health Service, U.S. Department of Health and Human
25 Services; NTP TR-467. Available from the National Institute of Environmental Health Sciences, Research Triangle Park,
26 NC.
27 NTP (National Toxicology Program). (1998) Toxicology and carcinogenesis studies of chloroprene (CAS No. 126-99-8) in
28 F344/N rats and B6C3F1 mice (inhalation studies). Public Health Service, U.S. Department of Health and Human Services;
29 NTP TR-467. Available from the National Institute of Environmental Health Sciences, Research Triangle Park, NC, and
30 online at http://ntp.niehs.nih.gov/ntp/htdocs/LT_rpts/tr467.pdf.
31 NTP (National Toxicology Program). (2000) 9th Report on carcinogens. Public Health Service, U.S. Department of Health
32 and Human Services, Research Triangle Park, NC.
33 NTP (National Toxicology Program). (2005) 11th Report on carcinogens. Public Health Service, U.S. Department of
34 Health and Human Services, Research Triangle Park, NC. Available online at http://ntp.niehs.nih.gov/?objectid=035E5806-
35 F735-FE81-FF769DFE5509AFOA.
36 Nystrom, A. (1948) Health hazards in the chloroprene rubber industry and their prevention. Acta Med Scand 132(Suppl.
37 219):1-125.
38 OECD (Organisation for Economic Co-operation and Development). (1998) Chloroprene. Screening Information DataSet
39 (SIDS) High Production Volume Chemicals. United Nations Environment Programme (UNEP) Publications, Nairobi,
40 Kenya. Available online at http://www.inchem.org/documents/sids/sids/Chloroprene.pdf.
41 Pell, S. (1978) Mortality of workers exposed to chloroprene. J Occup Med 20(l):21-29.
September 2009 7-4 DRAFT - DO NOT CITE OR QUOTE
-------
1 Placke, ME; Griffs, L; Bird, M; Bus, J; Parsing, RL; Cox Jr, AC (1996) Chronic inhalation oncogenicity study of isoprene
2 inB6C3Fl mice. Toxicology 110: 253-262.
3 Ponomarkov, V; Tomatis, L. (1980) Long-term testing of vinylidene chloride and chloroprene for carcinogenicity in rats.
4 Oncology 37(3): 136-141.
5 Portier, CJ; Bailer, AJ. (1989) Testing for increased carcinogenicity using a survival-adjusted quantal response test.
6 Fundam Appl Toxicol 12(4):731-737.
7 Putman, E; van der Laan, JW; van Loveren, H. (2003) Assessing immunotoxicity: guidelines. Fund Clinical Pharamcol
8 17(5): 615-626.
9 Rice, JM; Boffetta, P. (2001) 1,3-Butadiene, isoprene and chloroprene: Reviews by the IARC monographs programme,
10 outstanding issues, and research priorities in epidemiology. Chem Biol Interact 135-136:11-26.
11 Romazini, S; Laydevant, G; Lutz, J; et al. (1992) Mortality study in occupational exposure to chloroprene. Arch Mai Prof
12 8:721-725.
13 Rothman, K. (1986) Modern epidemiology. Boston, MA: Little, Brown and Co.
14 Sanotskii, IV. (1976) Aspects of the toxicology of chloroprene: Immediate and long-term effects. Environ Health Perspect
15 17:85-93.
16 Savitz, DA; Sonnenfeld, NL; Olshan, AF. (1994) Review of epidemiologic studies of paternal occupational exposure and
17 spontaneous abortion. Am JIndMed25(3):361-383.
18 Schrag, SD; Dixon, RL. (1985) Occupational exposures associated with male reproductive dysfunction. Annu Rev
19 Pharmacol Toxicol 25:567-592.
20 Shelby, MD. (1990) Results of NTP-sponsored mouse cytogenetic studies on 1,3-butadiene, isoprene, and chloroprene.
21 Environ Health Perspect 86:71-73.
22 Shelby, MD; Witt, KL. (1995) Comparison of results from mouse bone marrow chromosome aberration and micronucleus
23 tests. EnvironMolMutagen25(4):302-313.
24 Sills, RC; Hong, HL; Boorman, GA; et al. (2001) Point mutations of K-ras and H-ras genes in forestomach neoplasm from
25 control B6C3F1 mice and following exposure to 1,3-butadiene, isoprene or chloroprene for up to 2-years. Chem Bio Inter
26 135-136: 373-386.
27 Sills, RC; Hong, HL; Melnick, RL; et al. (1999) High frequency of codon 61 K-ras A~>T transversions in lung and
28 Harderian gland neoplasms of B6C3F1 mice exposed to chloroprene (2-chloro-1,3-butadiene) for 2 years, and comparisons
29 with the structurally related chemicals isoprene and 1,3-butadiene. Carcinogenesis 20(4):657-662.
30 Stewart, KJ. (1971) Dimerization of chloroprene and related dimers. J Am Chem Soc 93:4815-4821.
31 Summer, KH; Greim, H. (1980) Detoxification of chloroprene (2-chloro-l,3-butadiene) with glutathione in the rat.
32 BiochemBiophys Res Commun 96(2):566-573.
33 Tarone, RE. (1975) Tests for trend in lifre table analysis. Biometrika 62: 679-682.
34 Thornton-Manning, JR and Dahl, AR. (1997) Metabolic capacity of nasal tissue: interspecies comparisons of xenobiotic-
35 metabolizing enzymes. Mutat Res 380: 43-59.
36 Tice, RR. (1988) The cytogenetic evaluation of in vivo genotoxic and cytotoxic activity using rodent somatic cells. Cell
37 Biol Toxicol 4(4):475-486.
38 Tice, RR; Boucher, R; Luke, CA; et al. (1988) Chloroprene and isoprene: cytogenetic studies in mice. Mutagenesis
39 3(2):141-146.
September 2009 7-5 DRAFT - DO NOT CITE OR QUOTE
-------
1 Ton, TV; Hong, HH; Devereux, TR; et al. (2007) Evaluation of genetic alterations in cancer-related genes in lung and brain
2 tumors from B6C3F1 mice exposed to 1,3-butadiene or chloroprene. Chem Biol Interact 166(1-3): 112—120.
3 Trochimowicz, H; Loser, E; Feron, V; et al. (1998) Chronic inhalation toxicity and carcinogenicity studies onbeta-
4 chloroprene in rats and hamsters. Inhal Toxicol 10:443-472.
5 U.S. EPA, (Environmental Protection Agency). (1985) Summary overview of health effects associated with chloroprene.
6 Environmental Criteria and Assessment Office, Office of Health and Environmental Assessment
7 U.S. EPA (Environmental Protection Agency). (1986a) Guidelines for the health risk assessment of chemical mixtures.
8 Federal Register 51(185):34014-34025. Available online at http://www.epa.gov/ncea/raf/rafguid.htm.
9 U.S. EPA (Environmental Protection Agency). (1986b) Guidelines for mutagenicity risk assessment. Federal Register
10 51(185):34006-34012. Available online at http://www.epa.gov/ncea/raf/rafguid.htm.
11 U.S. EPA (Environmental Protection Agency). (1988) Recommendations for and documentation of biological values for use
12 in risk assessment. Environmental Criteria and Assessment Office, Office of Health and Environmental Assessment,
13 Cincinnati, OH; EPA/600/6-87/008. Available from the National Technical Information Service, Springfield, VA, PB88-
14 179874/AS, and online at http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=34855.
15 U.S. EPA (Environmental Protection Agency). (1991) Guidelines for developmental toxicity risk assessment. Federal
16 Register 56(234):63798-63826. Available online at http://www.epa.gov/ncea/raf/rafguid.htm.
17 U.S. EPA (Environmental Protection Agency). (1994a) Interim policy for particle size and limit concentration issues in
18 inhalation toxicity: Notice of availability. Federal Register 59(206):53799. Available online at http://www.epa.gov/EPA-
19 PEST/1994/October/Day-26/pr-11 .html.
20 U.S. EPA (Environmental Protection Agency). (1994b) Methods for derivation of inhalation reference concentrations and
21 application of inhalation dosimetry. Environmental Criteria and Assessment Office, Office of Health and Environmental
22 Assessment, Cincinnati, OH; EPA/600/8-90/066F. Available from the National Technical Information Service, Springfield,
23 VA, PB2000-500023, and online at http://cfpub.epa.gov/ncea/raf/recordisplay.cfm?deid=71993.
24 U.S. EPA (Environmental Protection Agency). (1995) Use of the benchmark dose approach in health risk assessment. Risk
25 Assessment Forum, Washington, DC; EPA/630/R-94/007. Available from the National Technical Information Service,
26 Springfield, VA, PB95-213765, and online at
27 http://cfpub.epa.gov/ncea/raf/raf_pubtitles.cfm?detype=document&excCol=archive.
28 U.S. EPA (Environmental Protection Agency). (1996) Guidelines for reproductive toxicity risk assessment. Federal
29 Register 61(212):56274-56322. Available online at http://www.epa.gov/ncea/raf/rafguid.htm.
30 U.S. EPA (Environmental Protection Agency). (1998) Guidelines for neurotoxicity risk assessment. Federal Register
31 63(93):26926-26954. Available online at http://www.epa.gov/ncea/raf/rafguid.htm.
32 U.S. EPA (Environmental Protection Agency). (2000a) Science policy council handbook: Risk characterization. Office of
33 Science Policy, Office of Research and Development, Washington, DC. EPA/100-B-00-002. Available online at
34 http://www.epa.gov/OSA/spc/pdfs/prhandbk.pdf.
35 U.S. EPA (Environmental Protection Agency). (2000b) Benchmark dose technical guidance document [external review
36 draft]. Risk Assessment Forum, Washington, DC; EPA/630/R-00/001. Available online at http://cfpub.epa.gov/ncea/cfm/
37 nceapublication.cfm?ActType=PublicationTopics&detype=DOCUMENT&subject=BENCHMARK+DOSE&subjtype=TIT
38 LE&excCol=Archive.
39 U.S. EPA (Environmental Protection Agency). (2000c) Supplementary guidance for conducting health risk assessment of
40 chemical mixtures. Risk Assessment Forum, Washington, DC; EPA/630/R-00/002. Available online at
41 http://cfpub.epa.gov/ncea/raf/chem_mix.cfm.
September 2009 7-6 DRAFT - DO NOT CITE OR QUOTE
-------
1 U.S. EPA (Environmental Protection Agency). (2000d) Chloroprene (2-chloro-l,3-butadiene). Hazard summary,
2 Technology Transfer Network Air Toxics Web Site. Office of Air and Radiation, Washington, DC. Available online at
3 http://www.epa.gov/ttn/atw/hlthef/chloropr.html.
4 U.S. EPA (Environmental Protection Agency). (2000e) Guidance for data quality assessment: Practical methods for data
5 analysis. Office of Environmental Information, Washington, DC; EPA/600/R-96/084.
6 U.S. EPA (Environmental Protection Agency). (2000f) Toxicological review of vinyl chloride. Integrated Risk Information
7 System. National Center for Environmental Assessment. Washington, DC: EPA/635R-00/004. Available online at:
8 http://www.epa.gov/ncea/iris/toxreviews/1001-tr.pdf
9 U.S. EPA (Environmental Protection Agency). (2002a) A review of the reference dose concentration and reference
10 concentration processes. Risk Assessment Forum, Washington, DC; EPA/630/P-02/002F. Available online at
11 http ://cfpub .epa. gov/ncea/raf/raf_pubtitles. cfm?detvpe=document&excCol=archive.
12 U.S. EPA (Environmental Protection Agency). (2002b) Health assessment of 1,3-butadiene. National Center for
13 Environmental Assessment. Washington, DC: EPA/600/P-98/00IF. Available online at
14 http://www.epa.gov.ncea/iris/supdocs/buta-sup.pdf
15 U.S. EPA (Environmental Protection Agency). (2005a) Guidelines for carcinogen risk assessment. Federal Register
16 70(66):17765-18717. Available online at http://www.epa.gov/cancerguidelines.
17 U.S. EPA (Environmental Protection Agency). (2005b) Supplemental guidance for assessing susceptibility from early-life
18 exposure to carcinogens. Risk Assessment Forum, Washington, DC; EPA/630/R-03/003F. Available online at
19 http://www.epa.gov/cancerguidelines.
20 U.S. EPA (Environmental Protection Agency). (2006a) Science policy council handbook: Peer review. 3rd edition. Office
21 of Science Policy, Office of Research and Development, Washington, DC; EPA/100/B-06/002. Available online at
22 http://www.epa.gov/OSA/spc/2peerrev.htm.
23 U.S. EPA (Environmental Protection Agency). (2006b) A framework for assessing health risk of environmental exposures
24 to children. National Center for Environmental Assessment, Washington, DC; EPA/600/R-05/093F. Available online at
25 http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=158363.
26 Vogel, E. (1979) Mutagenicity of chloroprene, l-chloro-l,3-trans-butadiene, l-4-dichlorobutene-2 and l,4-dichloro-2,3-
27 epoxybutane in Drosophila melanogaster. MutatRes 67(4):377—381.
28 Watson, WP; Cottrell, L; Zhang, D; Golding, BT (2001) Metabolism and molecular toxicology of isoprene. Chem-Biol
29 Inter 135-136: 223-238.
30 Westphal, GA; Blaszkewicz, M; Leutbecher, M; et al. (1994) Bacterial mutagenicity of 2-chloro-l,3-butadiene
31 (chloroprene) caused by decomposition products. Arch Toxicol 68(2):79-84.
32 Willems, M. (1978) Evaluation of beta-chloroprene and five dimmers in the Salmonella/microsome mutagenicity test.
33 Report no. R 6392. Produced by the Central Institute for Nutrition and Food Research (CIVO), The Netherlands; submitted
34 by Dupont Chemical, Wilmington, DE, under TSCA Section 8ECP
35 Willems, M. (1980) Evaluation of beta-chloroprene and 4-chloroprene dimers in the Ames test by atmospheric exposure of
36 the tester strains with cover letter dated 090192. Produced by the Central Institute for Nutrition and Food Research (CIVO),
37 The Netherlands; submitted by Dupont Chemical, Wilmington, DE, under TSCA Section 8ECP; EPA Document No. 88-
38 920008421; NTIS No. OTS0570703.
September 2009 7-7 DRAFT - DO NOT CITE OR QUOTE
-------
APPENDIX A. SUMMARY OF EXTERNAL PEER REVIEW AND PUBLIC COMMENTS
AND DISPOSITION
1
2
3
4 [page intentionally left blank]
September 2009 A-1 DRAFT - DO NOT CITE OR QUOTE
-------
APPENDIX B. BENCHMARK DOSE MODELING RESULTS FOR THE DERIVATION OF
THE RFC
1 Benchmark Dose (BMD) modeling was performed to identify the point of departure for the
2 derivation of the chronic RfC for chloroprene. The modeling was conducted in accordance with the
3 draft EPA guidelines (U. S. EPA, 2000b) using Benchmark Dose Software Version 2.0 (BMDS). The
4 BMDS model outputs for the derivation of the chronic RfC are attached.
5 The following critical effects were modeled using BMDS: alveolar epithelial hyperplasia,
6 bronchiolar hyperplasia, lung histiocytic cell infiltration, nasal epithelial chronic inflammation, nasal
7 epithelium atrophy, nasal epithelial necrosis, and splenic hematopoietic cell proliferation. The
8 endpoint being modeled specified which set of models, continuous (linear, polynomial, power, and
9 Hill) or dichotomous (gamma, logistic, multi-stage, probit, quantal-linear, quantal-quadratic, Weibull,
10 and dichotomous Hill), would be utilized. Model eligibility was determined by assessing the
11 goodness-of-fit using a value of a = 0.1 (when appropriate), visual fit, and ranking by Akaike
12 Information Criterion (AIC).
13 The critical endpoint selected for the derivation of the chronic RfC was increased incidence of
14 degenerative nasal lesions in male rats. The logistic model provided the best fit for this data set. The
15 following tables (B-l through B-9) are summaries of the modeling results for all considered endpoints.
16 The best fitting model for each endpoint is indicated in bold and the model plot and output are
17 included immediately after the table.
Table B-l. Modeling Results for Alvelolar Epithelial Hyperplasia in Male F344/N
Rats
Model:
Restricted:
BMR:
BMR Type:
BMD
BMDL
ChiA2
Goodness-
of-fit p-
value
AIC
Fitted Log-
Likelihood
Fitted
p-value
Gamma
Yes
0.1
Extra
Risk
14.8657
10.0883
4.05
0.1317
231.042
-113.521
0.1421
Logistic
0.1
Extra
Risk
24.4838
19.1571
5.12
0.0775
232.34
-114.17
0.0743
Log-
Logistic
Yes
0.1
Extra
Risk
11.4228
7.06934
3.48
0.1753
230.479
-113.24
0.1883
Log-
Probit
Yes
0.1
Extra
Risk
28.604
19.5927
6.63
0.0363
233.859
-114.93
0.03475
Probit
0.1
Extra
Risk
23.3986
18.2584
5.02
0.0813
232.209
-114.104
0.07933
Weibull
Yes
0.1
Extra
Risk
14.866
10.0883
4.05
0.1317
231.042
-113.521
0.1421
Quantal-
Linear
0.1
Extra
Risk
14.866
10.0883
4.05
0.1317
231.042
-113.521
0.1421
Hill
Yes
0.1
Extra
Risk
3.54712
9.05E-08
1.989357
NA
233.164
-112.582
NA
September 2009
B-1
DRAFT - DO NOT CITE OR QUOTE
-------
Scaled Residuals:
0
12.8
32
80
-0.928
1.698
-0.509
-0.224
-0.928
1.698
-0.509
-0.224
-0.649
1.566
-0.763
-0.16
-1.676
1.939
-0.087
-0.23
-1.427
1.711
-0.126
-0.198
-0.928
1.698
-0.509
-0.224
-0.928
1.698
-0.509
-0.224
-0.1573
0.6599
-1.122
0.5197
I
C
0
0.7
0.6
0.5
0.4
0.3
0.2
0.1
Log-Logistic Model with 0.95 Confidence Level
Log-Logistic
BMDL BMD
10 20
30
40
dose
50
60
70
80
09:46 02/14 2008
Figure B-l. Log-logistic model fit for alveolar epitheliala hyperplasia in male F344/N rats
1
2
o
J
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Logistic Model. (Version: 2.10; Date: 09/23/2007)
Input Data File: U:\Chloroprene\Male_F344_rat\alv_hyper_loglog.(d)
Gnuplot Plotting File: U:\Chloroprene\Male_F344_rat\alv_hyper_loglog.plt
ThuFeb 14 09:46:06 2008
BMDS Model Run
The form of the probability function is:
Pfresponse] = background+(l-background)/[l+EXP(-intercept-slope*Log(dose))]
Dependent variable = alveolar_hyper
Independent variable = DOSE
Slope parameter is restricted as slope >= 1
Total number of observations = 4
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
User has chosen the log transformed model
Default Initial Parameter Values
background = 0.1
September 2009
B-2
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
o
J
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
o o
JJ
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
intercept =
slope =
-4.4782
1
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameters) -slope
have been estimated at a boundary point, or have been specified by the user,
and do not appear in the correlation matrix )
background intercept
background
intercept
1
-0.66
-0.66
1
Parameter Estimates
Variable
background
intercept
slope
Estimate
0.130984
-4.63283
1
95.0% Wald Confidence Interval
S td. Err. Lower Conf. Limit Upper Conf. Limit
* - Indicates that this value is not calculated.
Analysis of Deviance Table
Model Log(likelihood) # Param's Deviance Test d.f. P-value
Full model -111.57 4
Fittedmodel -113.24 2 3.33902 2 0.1883
Reducedmodel -121.815 1 20.4898 3 0.0001343
AIC:
230.479
Goodness of Fit
Scaled
Dose Est._Prob. Expected Observed Size Residual
0.0000
12.8000
32.0000
80.0000
0.1310
0.2272
0.3373
0.5113
6.549
11.360
16.526
25.564
5
16
14
25
50
50
49
50
-0.649
1.566
-0.763
-0.160
ChiA2 = 3.48 d.f. = 2
P-value = 0.1753
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Extra risk
Confidence level = 0.95
BMD = 11.4228
BMDL = 7.06934
September 2009
B-3
DRAFT - DO NOT CITE OR QUOTE
-------
Table B-2. Modeling Results for Alveolar Epithelial Hyperplasia in Female
F344/N Rats
Model:
Restricted:
BMR:
BMR Type:
BMD
BMDL
ChiA2
Goodness-
of-fit p-
value
AIC
Fitted Log-
Likelihood
Fitted
p-value
Gamma
Yes
0.1
Extra
Risk
8.0322
5.89582
5.59
0.0612
245.78
-120.89
0.06481
Logistic
0.1
Extra
Risk
14.8564
11.9857
8.24
0.0163
248.949
-122.475
0.01328
Log-
Logistic
Yes
0.1
Extra
Risk
4.90719
3.27097
3.45
0.1779
243.677
-119.839
0.1854
Log-
Probit
Yes
0.1
Extra
Risk
15.342
10.7468
9.68
0.0079
249.954
-122.977
0.00804
Probit
0.1
Extra
Risk
14.4844
11.8082
8.14
0.0171
248.806
-122.403
0.01427
Weibull
Yes
0.1
Extra
Risk
8.03223
5.89582
5.59
0.0612
245.78
-120.89
0.06481
Quantal-
Linear
0.1
Extra
Risk
8.03223
5.89582
5.59
0.0612
245.78
-120.89
0.06481
Hill
Yes
0.1
Extra
Risk
1.42027
0.001489
1.853207
0.1734
244.161
-119.081
0.1733
Scaled Residuals:
0
12.8
32
80
-1.098
1.998
-0.328
-0.531
-2.002
1.99
0.271
-0.442
-0.453
1.536
-0.91
-0.246
-1.77
2.446
-0.201
-0.723
-1.96
2.006
0.275
-0.439
-1.098
1.998
-0.328
-0.531
-1.098
1.998
-0.328
-0.531
-0.05399
0.6151
-1.089
0.5347
September 2009
B-4
DRAFT - DO NOT CITE OR QUOTE
-------
T3
0
0
0
t
<
C
0
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
Log-Logistic Model with 0.95 Confidence Level
Log-Logistic
'- BMDL BMD
10
20
30
40
dose
50
60
70
80
13:10 02/14 2008
Figure B-2. Log-logistic model fit for alveolar epithelial hyperplasia in female F344/N
rats
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Logistic Model. (Version: 2.10; Date: 09/23/2007)
Input Data File: U:\Chloroprene\Female_F344_rat\alv_hyper_loglog.(d)
Gnuplot Plotting File: U:\Chloroprene\Female_F344_rat\alv_hyper_loglog.plt
ThuFebl413:10:352008
BMDS Model Run
The form of the probability function is:
P[response] = background+( 1 -background)/[ l+EXP(-intercept-slope*Log(dose))]
Dependent variable = alveolar_hyper
Independent variable = DOSE
Slope parameter is restricted as slope >= 1
Total number of observations = 4
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
User has chosen the log transformed model
Default Initial Parameter Values
background = 0.122449
intercept = -3.74532
slope = 1
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -slope
have been estimated at a boundary point, or have been specified by the user,
and do not appear in the correlation matrix )
September 2009
B-5
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
o o
JJ
34
35
36
37
38
39
40
41
42
43
44
45
background intercept
background 1 -0.62
ntercept -0.62 1
Parameter Estimates
Variable Estimate Std. Err.
background 0.145252 *
intercept -3.78793 *
slope 1 *
95.0% Wald Confidence Interval
Lower Conf. Limit Upper Conf. Limit
* *
* *
* *
* - Indicates that this value is not calculated.
Analysis of Deviance Table
Model Log(likelihood) #
Full model -118.153
Fitted model 119.839
Reduced model -135.512
AIC: 243.677
Goodness of Fit
Param's Deviance Test d.f. P-value
4
2 3.37005 2 0.1854
1 34.7167 3 <0001
Scaled
Dose Est. Prob. Expected Observed Size Residual
0.0000 0.1453 7.117
12.8000 0.3373 16.866
32.0000 0.5044 25.218
80.0000 0.6960 34.799
ChiA2 = 3.45 d.f = 2 P-value = 0.
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Extra risk
Confidence level = 0.95
BMD = 4.90719
BMDL = 3.27097
6 49 -0.453
22 50 1.536
22 50 -0.910
34 50 -0.246
1779
September 2009
B-6
DRAFT - DO NOT CITE OR QUOTE
-------
Table B-3. Modeling Results for Bronchiolar Hyperplasia in Male B6C3F1 Mice
Model:
Restricted:
BMR:
BMR Type:
BMD
BMDL
ChiA2
Goodness-
of-fit p-
value
AIC
Fitted Log-
Likelihood
Fitted
p-value
Gamma
Yes
0.1
Extra
Risk
9.962
7.95025
6.24
0.1003
192.219
-95.1094
0.1195
Logistic
0.1
Extra
Risk
25.582
20.9208
12.55
0.0019
206.147
-101.073
0.000138
Log-
Logistic
Yes
0.1
Extra
Risk
7.54241
5.60381
2.32
0.5085
188.645
-93.3224
0.5186
Log-
Probit
Yes
0.1
Extra
Risk
18.0076
12.7086
13.97
0.0009
203.779
-99.8893
0.000452
Probit
0.1
Extra
Risk
23.8731
19.6205
12.13
0.0023
205.312
-100.656
0.00021
Weibull
Yes
0.1
Extra
Risk
9.962
7.95025
6.24
0.1003
192.219
-95.1094
0.1195
Quantal-
Linear
0.1
Extra
Risk
9.962
7.95025
6.24
0.1003
192.219
-95.1094
0.1195
Hill
Yes
0.1
Extra
Risk
6.46897
0.13456
0
0.9997
190.376
-92.1882
0.9997
Scaled Residuals:
0
12.8
32
80
0
1.561
1.139
-1.585
-2.549
0.897
2.136
-0.826
0
0.8
0.6
-1.149
-1.669
2.278
1.259
-2.098
-2.442
0.992
2.094
-0.891
0
1.561
1.139
-1.585
0
1.561
1.139
-1.585
0
-0.0003
-0.00015
0.000113
September 2009
B-7
DRAFT - DO NOT CITE OR QUOTE
-------
T3
0
0
0
t
<
C
0
0.6
0.5
0.4
0.3
0.2
0.1
Log-Logistic Model with 0.95 Confidence Level
Log-Logistic
BMDL BMD
10 20
30 40
dose
50 60
70
80
07:46 02/20 2008
Figure B-3. Log-logistic model fit for bronchiolar hyperplasia in male B6C3F1 Mice
1
2
o
J
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
o o
JJ
34
35
Logistic Model. (Version: 2.10; Date: 09/23/2007)
Input Data File: U:\Chloroprene\Male_B6C3Fl_mouse\bronc_hyper_loglog.(d)
Gnuplot Plotting File: U:\Chloroprene\Male_B6C3Fl_mouse\bronc_hyper_loglog.plt
Wed Feb 20 07:46:23 2008
BMDS Model Run
The form of the probability function is:
Pfresponse] = background+(l-background)/[l+EXP(-intercept-slope*Log(dose))]
Dependent variable = bronchiolar_hyper
Independent variable = DOSE
Slope parameter is restricted as slope >= 1
Total number of observations = 4
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
User has chosen the log transformed model
Default Initial Parameter Values
background = 0
intercept = -4.24694
slope = 1
Asymptotic Correlation Matrix of Parameter Estimates
(*** The model parameter(s) -background -slope
have been estimated at a boundary point, or have been specified by the user,
and do not appear in the correlation matrix )
intercept
September 2009
B-8
DRAFT - DO NOT CITE OR QUOTE
-------
1 intercept 1
2
3 Parameter Estimates
4
5 95.0%Wald Confidence Interval
6 Variable Estimate Std. Err. Lower Conf. Limit Upper Conf. Limit
7 background 0 * * *
8 intercept -4.21777 * * *
9 slope 1 * * *
10
11 * - Indicates that this value is not calculated.
12
13 Analysis of Deviance Table
14
15 Model Log(likelihood) # Param's Deviance Test d.f. P-value
16 Full model -92.1882 4
17 Fitted model -93.3224 1 2.26827 3 0.5186
18 Reduced model -113.552 1 42.7283 3 <0001
19
20 AIC: 188.645
21
22
23 Goodness of Fit
24 Scaled
25 Dose Est._Prob. Expected Observed Size Residual
26
27
28
29
30
31
32 ChiA2 = 2.32 d.f. = 3 P-value = 0.5085
o o
JJ
34
35 Benchmark Dose Computation
36 Specified effect = 0.1
37 Risk Type = Extra risk
38 Confidence level = 0.95
39 BMD = 7.54241
40 BMDL = 5.60381
0.0000
12.8000
32.0000
80.0000
0.0000
0.1586
0.3204
0.5410
0.000
7.932
16.019
27.049
0
10
18
23
50
50
50
50
0.000
0.800
0.600
-1.149
September 2009 B-9 DRAFT - DO NOT CITE OR QUOTE
-------
Table B-4. Modeling Results for Bronchioloar Hyperplasia in Female B6C3F1
Mice
Model:
Restricted:
BMR:
BMR Type:
BMD
BMDL
ChiA2
Goodness-
of-fit p-
value
AIC
Fitted Log-
Likelihood
Fitted p-
value
Gamma
Yes
0.1
Extra
Risk
8.36521
6.74162
11.72
0.0084
194.774
-96.387
0.01856
Logistic
0.1
Extra
Risk
21.7233
17.9832
14.42
0.0007
205.342
-100.671
9.29E-05
Log-
Logistic
Yes
0.1
Extra
Risk
6.16449
4.6013
7.96
0.0469
192.407
-95.2036
0.05422
Log-
Probit
Yes
0.1
Extra
Risk
16.922
11.5061
18.88
0.0001
206.563
-101.282
5.04E-05
Probit
0.1
Extra
Risk
20.2623
16.905
14.34
0.0008
204.825
-100.412
0.00012
Weibull
Yes
0.1
Extra
Risk
8.36519
6.74162
11.72
0.0084
194.774
-96.387
0.01856
Quantal-
Linear
0.1
Extra
Risk
8.36519
6.74162
11.72
0.0084
194.774
-96.387
0.01856
Hill
Yes
0.1
Extra
Risk
2.97514
0.149667
5.982981
0.0502
192.953
-94.4763
0.04552
Scaled Residuals:
0
12.8
32
80
0
3.092
-1.377
-0.513
-2.5
2.835
-0.285
-0.235
0
2.128
-1.847
0.137
-2.04
3.615
-1.163
-0.549
-2.409
2.889
-0.359
-0.249
0
3.092
-1.377
-0.513
0
3.092
-1.377
-0.513
0
1.122
-1.956
0.9482
Note: No model fits appropriately. Therefore, recommend using the NOAEL/LOAEL approach to determining a
POD
September 2009
B-10
DRAFT - DO NOT CITE OR QUOTE
-------
Table B-5. Modeling Results for Histiocytic Cell Infiltration in Male B6C3F1 Mice
x-. T Log- Log- _ , .^ „, ., „ Quantal- „.„
Gamma Logistic _ . ,. _ !V Probit Weibull ^ • Hill
Model: Logistic Probit Linear
Restricted:
BMR:
BMR Type:
BMD
BMDL
ChiA2
Goodness-
of-fit p-
value
AIC
Fitted Log-
Likelihood
Fitted p-
value
Yes
0.1
Extra
Risk
11.1361
8.17864
8.97
0.0113
215.559
-105.78
0.01235
0.1
Extra
Risk
24.8436
19.8574
12.27
0.0022
222.023
-109.012
0.000488
Yes
0.1
Extra
Risk
7.44347
5.28235
5.38
0.0679
211.84
-103.92
0.07927
Yes
0.1
Extra
Risk
22.6862
15.5226
14.62
0.0007
223.144
-109.572
0.000278
0.1
Extra
Risk
23.4156
18.7645
12.04
0.0024
221.531
-108.765
0.000624
Yes
0.1
Extra
Risk
11.1363
8.17864
8.97
0.0113
215.559
-105.78
0.01235
0.1
Extra
Risk
11.1363
8.17864
8.97
0.0113
215.559
-105.78
0.01235
Yes
0.1
Extra
Risk
1615.32
-807.661
1.3 1E-
304
Scaled Residuals:
0
12.8
32
80
-1.224
2.188
0.816
-1.421
-2.606
1.653
1.488
-0.733
-0.49
1.762
0.222
-1.409
-2.356
2.379
1.188
-1.414
-2.524
1.714
1.458
-0.779
-1.224
2.188
0.817
-1.421
-1.224
2.188
0.817
-1.421
Note: No model fits appropriately. Therefore, recommend using the NOAEL/LOAEL approach to determining a
POD
Table B-6. Modeling Results for Olfactory Chronic Inflammation in Male F344/N
Rats
Gamma Logistic T Log" DL°g" Probit Weibull Q".antal- ffill
Model: Logistic Probit Linear
Restricted:
BMR:
BMR Type:
BMD
BMDL
ChiA2
Goodness-of-
fit p-value
AIC
Fitted Log-
Likelihood
Fitted p-
value
Yes
0.1
Extra
Risk
15.2489
10.1164
0.22
0.8964
81.4586
-39.7293
0.8993
0.1
Extra
Risk
23.8087
18.9473
2.83
0.0925
87.0594
-41.5297
0.05085
Yes
0.1
Extra
Risk
14.6428
9.27776
0.12
0.9398
81.3682
-39.4291
0.9409
Yes
0.1
Extra
Risk
17.7991
13.7362
3.08
0.2144
83.9766
-40.9883
0.2553
0.1
Extra
Risk
15.2489
10.1164
0.22
0.8964
81.4586
-41.3298
0.06467
Yes
0.1
Extra
Risk
15.2489
10.1164
0.22
0.8964
81.4586
-39.7293
0.8993
0.1
Extra
Risk
15.2489
10.1164
0.22
0.8964
81.4586
-39.7293
0.8993
Yes
0.1
Extra
Risk
~
~
-
-
~
~
~
September 2009
B-11
DRAFT - DO NOT CITE OR QUOTE
-------
Scaled Residuals:
0
12.8
32
0.000
0.390
-0.258
-1.120
1.221
-0.291
0.000
0.286
-0.207
0.000
1.458
-0.976
-1.027
1.195
-0.345
0.000
0.390
-0.258
0.000
0.390
-0.258
~
~
~
Note: High dose group was dropped in order to obtain adequate model fit
I
C
0
0.35
0.3
0.25
0.2
0.15
0.1
0.05
Log-Logistic Model with 0.95 Confidence Level
Log-Logistic
BMDL
BMD
10
15
dose
20
25
30
10:02 12/12 2008
Figure B-4. Log-logistic model fit for olfactory chronic inflammation in male F344/N rats
1
2
o
J
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Logistic Model. (Version: 2.12; Date: 05/16/2008)
Input Data File: M:\Chloroprene\Male_F344_rat\ntp_inflammation_hdd_loglog.(d)
Gnuplot Plotting File: M:\Chloroprene\Male_F344_rat\ntp_inflammation_hdd_loglog.plt
Fri Dec 12 08:22:05 2008
BMDS Model Run
The form of the probability function is:
Pfresponse] = background+(l-background)/[l+EXP(-intercept-slope*Log(dose))]
Dependent variable = Col3
Independent variable = Coll
Slope parameter is restricted as slope >= 1
Total number of observations = 3
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
User has chosen the log transformed model
September 2009
B-12
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
o
J
4
5
6
7
«
O
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
o o
JJ
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
Default Initial Parameter Values
background = 0
intercept = -4.79799
slope = 1
Asymptotic Correlation Matrix of Parameter Estimates
(*** The model parameter(s) -background -slope
have been estimated at a boundary point, or have been specified by the user,
and do not appear in the correlation matrix )
intercept
intercept 1
Parameter Estimates
95.0% Wald Confidence Interval
Variable Estimate Std. Err. Lower Conf. Limit Upper Conf. Limit
background 0 * * *
intercept -4.88117 * * *
slope 1 * * *
* - Indicates that this value is not calculated.
Analysis of Deviance Table
Model Log(likelihood) # Param's Deviance Test d.f.
Full model -39.6231 3
Fitted model 39.6841 1 0.121914 2
Reducedmodel -46.4291 1 13.6119 2
AIC: 81.3682
Goodness of Fit
Scaled
Dose Est. Prob. Expected Observed Size Residual
0.0000 0.0000 0.000 0.000 50 0.000
12.8000 0.0885 4.426 5.000 50 0.286
32.0000 0.1954 9.574 9.000 49 -0.207
ChiA2 = 0.12 d.f. = 2 P-value = 0.9398
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Extra risk
Confidence level = 0.95
BMD = 14.6428
BMDL = 9.27776
P-value
0.9409
0.001107
September 2009
B-13
DRAFT - DO NOT CITE OR QUOTE
-------
Table B-7. Modeling Results for Olfactory Atrophy in Male F344/N Rats
Model:
Restricted:
BMR:
BMR Type:
BMD
BMDL
ChiA2
Goodness-of-
flt p-value
AIC
Fitted Log-
Likelihood
Fitted p-
value
Gamma
Yes
0.1
Extra Risk
10.6003
7.99938
0.00
NA
106.376
-50.1882
NA
Logistic
0.1
Extra Risk
7.70048
5.97454
1.24
0.2655
105.53
-50.7651
0.2828
Log-
Logistic
Yes
0.1
Extra Risk
10.81
8.62799
0.00
NA
106.376
-50.1882
NA
Log-
Probit
Yes
0.1
Extra Risk
10.9386
8.79455
0.00
NA
106.376
-50.1882
NA
Probit
0.1
Extra Risk
6.91725
5.40111
2.02
0.1555
106.283
-51.1417
0.1673
Weibull
Yes
0.1
Extra Risk
9.95012
7.06875
0.00
NA
106.376
-50.1882
NA
Quantal-
Linear
0.1
Extra Risk
2.28431
1.80011
18.03
0.0000
125.166
-60.5831
5.12E-006
Hill
Yes
0.1
Extra Risk
-
-
-
-
-
-
-
Scaled Residuals:
0
12.8
32
-0.000
-0.000
0.000
0.847
-0.597
0.406
0.000
0.000
-0.000
0.000
-0.000
0.000
0.977
-0.901
0.502
0.000
0.000
-0.000
0.459
-3.280
2.658
-
-
-
Note: High dose group was dropped in order to obtain adequate model fit
T3
0
0
0
t
<
C
0
0.8
0.6
0.4
0.2
10:11 12/12 2008
Logistic Model with 0.95 Confidence Level
Logistic
BMDL BMD
10
15
dose
20
25
30
Figure B-5. Logistic model fit for olfactory atrophy in male F344/N rats
September 2009
B-14
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
o
J
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
o o
JJ
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
Logistic Model. (Version: 2.12; Date: 05/16/2008)
Input Data File: M:\Chloroprene\Male_F344_rat\ntp_atrophy_hdd_logistic.(d)
Gnuplot Plotting File: M:\Chloroprene\Male_F344_rat\ntp_atrophy_hdd_logistic.plt
Fri Dec 12 08:51:22 2008
BMDS Model Run
The form of the probability function is:
Pfresponse] = l/[l+EXP(-intercept-slope*dose)]
Dependent variable = Col3
Independent variable = Coll
Slope parameter is not restricted
Total number of observations = 3
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial Parameter Values
background = 0 Specified
intercept = -2.84277
slope = 0.164779
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -background
have been estimated at a boundary point, or have been specified by the user,
and do not appear in the correlation matrix )
intercept
slope
intercept
1
-0.85
slope
-0.85
1
Parameter Estimates
Variable
intercept
slope
Estimate
-3.25094
0.179356
95.0% Wald Confidence Interval
Std. Err. Lower Conf. Limit Upper Conf. Limit
0.484263 -4.20007 -2.3018
0.0262753
0.127857
0.230855
Analysis of Deviance Table
Model Log(likelihood) # Param's Deviance Test d.f P-value
Full model -50.1882 3
Fitted model -50.7651 2 1.15379 1 0.2828
Reducedmodel -100.819 1 101.262 2 <0001
AIC:
105.53
Goodness of Fit
Scaled
Dose Est._Prob. Expected Observed Size Residual
0.847
-0.597
0.0000
12.8000
0.0373
0.2778
1.865
13.892
3.000
12.000
50
50
September 2009
B-15
DRAFT - DO NOT CITE OR QUOTE
-------
1 32.0000 0.9233 45.243 46.000 49 0.406
2
3 ChiA2=1.24 d.f. = 1 P-value = 0.2655
4
5
6 Benchmark Dose Computation
7 Specified effect = 0.1
8 Risk Type = Extra risk
9 Confidence level = 0.95
10 BMD = 7.70048
11 BMDL= 5.97454
September 2009 B-16 DRAFT - DO NOT CITE OR QUOTE
-------
Table B-8. Modeling Results for Olfactory Necrosis in Male F344/N Rats
Model:
Restricted:
BMR:
BMR Type:
BMD
BMDL
ChiA2
Goodness-
of-fit p-
value
AIC
Fitted Log-
Likelihood
Fitted p-
value
Gamma
Yes
0.1
Extra
Risk
6.46561
3.70666
0.00
1.0000
124.435
-60.2177
1.00
Logistic
0.1
Extra
Risk
12.1684
9.77545
4.56
0.0328
130.942
-63.4712
0.01075
Log-
Logistic
Yes
0.1
Extra
Risk
6.92124
2.96263
0.00
1.0000
124.435
-60.2177
1.00
Log-
Probit
Yes
0.1
Extra
Risk
7.98173
6.41755
0.06
0.9686
122.499
-60.2494
0.9688
Probit
0.1
Extra
Risk
11.3581
9.13936
3.86
0.0494
129.762
-62.881
0.021
Weibull
Yes
0.1
Extra
Risk
6.31726
3.70666
0.00
1.0000
124.435
-60.2177
1.00
Quantal-
Linear
0.1
Extra
Risk
4.75407
3.65317
0.30
0.8622
122.737
-60.3685
0.86
Hill
Yes
0.1
Extra
Risk
~
~
~
~
~
-
~
Scaled Residuals:
0
12.8
32
0.000
-0.000
0.000
-1.510
1.450
-0.418
0.000
0.000
-0.000
0.000
0.188
-0.169
-1.313
-0.461
1.387
0.000
-0.000
0.000
0.000
-0.443
0.317
-
~
~
Note: High dose group was dropped in order to obtain adequate model fit
September 2009
B-17
DRAFT - DO NOT CITE OR QUOTE
-------
T3
0
0
0
t
<
C
0
LogProbit Model with 0.95 Confidence Level
0.7
0.6
0.5
0.4
0.3
0.2
0.1
LogProbit
BMDL BMD
10
15
dose
20
25
30
10:14 12/12 2008
Figure B-6. Logistic model fit for olfactory necrosis in male F344/N rats
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Probit Model. (Version: 3.1; Date: 05/16/2008)
Input Data File: M:\Chloroprene\Male_F344_rat\ntp_necrosis_hdd_logprobit.(d)
Gnuplot Plotting File: M:\Chloroprene\Male_F344_rat\ntp_necrosis_hdd_logprobit.plt
Fri Dec 12 09:00:57 2008
BMDS Model Run -
The form of the probability function is:
P[response] = Background + (1-Background) * CumNorm(Intercept+Slope*Log(Dose)),
where CurnNormQ is the cumulative normal distribution function
Dependent variable = Col3
Independent variable = Coll
Slope parameter is restricted as slope >= 1
Total number of observations = 3
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
User has chosen the log transformed model
Default Initial (and Specified) Parameter Values
background = 0
intercept = -3.33803
slope = 1
Asymptotic Correlation Matrix of Parameter Estimates
(*** The model parameter(s) -background -slope
have been estimated at a boundary point, or have been specified by the user,
and do not appear in the correlation matrix )
intercept
September 2009
B-18
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
o o
JJ
34
35
36
37
38
39
40
41
42
intercept
Parameter Estimates
Variable
background
intercept
slope
Estimate
0
-3.35871
1
Std. Err.
NA
0.133307
NA
95.0% Wald Confidence Interval
Lower Conf. Limit Upper Conf. Limit
-3.61998
-3.09743
NA - Indicates that this parameter has hit a bound
implied by some inequality constraint and thus
has no standard error.
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced model
Log(likelihood) # Param's Deviance Test d.f P-value
-60.2177 3
-60.2494 1 0.063351 2 0.9688
-83.5122 1 46.5889 2 <0001
AIC:
122.499
Goodness of Fit
Scaled
Dose Est._Prob. Expected Observed Size Residual
0.0000
12.8000
32.0000
0.0000
0.2092
0.5426
0.000
10.459
26.588
0.000
11.000
26.000
50
50
49
0.000
0.188
-0.169
ChiA2 = 0.06 d.f. = 2
P-value = 0.9686
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Extra risk
Confidence level = 0.95
BMD= 7.98173
BMDL = 6.41755
September 2009
B-19
DRAFT - DO NOT CITE OR QUOTE
-------
Table B-9. Modeling Results for Degenerative Nasal Lesions (olfactory atrophy or
necrosis) in Male F344/N Rats
Model:
Restricted:
BMR:
BMR Type:
BMD
BMDL
ChiA2
Goodness-
of-fit p-
value
AIC
Fitted Log-
Likelihood
Fitted p-
value
Gamma
Yes
0.1
Extra
Risk
10.3392
7.89967
0
n/a
102.417
-48.3572
n/a
Logistic
0.1
Extra
Risk
7.38383
5.71667
1.44
0.2309
102.053
-49.0267
0.2472
Log-
Logistic
Yes
0.1
Extra
Risk
10.6373
8.62242
0
n/a
102.714
-48.3572
n/a
Log-
Probit
0.1
Extra
Risk
10.7236
8.71969
0
n/a
102.714
-48.3572
n/a
Probit
0.1
Extra
Risk
6.61322
5.14757
2.11
0.1460
102.721
-49.3605
0.1566
Weibull
Yes
0.1
Extra
Risk
9.55894
6.8742
0
n/a
102.714
-48.3572
n/a
Quantal-
Linear
0.1
Extra
Risk
2.13213
1.68322
18.61
0.0
122.619
-59.3096
2.8E-006
Hill
Yes
0.1
Extra
Risk
~
~
~
-
~
-
~
Scaled Residuals:
0
12.8
32
-0.0
0.0
-0.0
0.905
-0.614
0.490
0.0
0.0
-0.0
0.0
-0.0
0.0
0.993
-0.903
0.558
-0.0
0.0
0.0
0.445
-3.304
2.738
~
~
~
Note: High dose group was dropped in order to obtain adequate model fit
September 2009
B-20
DRAFT - DO NOT CITE OR QUOTE
-------
T3
0
0
0
t
<
C
0
Logistic Model with 0.95 Confidence Level
0.8
0.6
0.4
0.2
Logistic
BMDL BMD
10
15
dose
20
25
30
10:31 01/08 2009
Figure B-7. Logistic model fit for degenerative nasal lesions (olfactory atrophy or
necrosis) in male F344/N rats
1
1
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Logistic Model. (Version: 2.12; Date: 05/16/2008)
Input Data File: M:\Chloroprene\Male_F344_rat\olfactory_logistic_hdd.(d)
Gnuplot Plotting File: M:\Chloroprene\Male_F344_rat\olfactory_logistic_hdd.plt
Tue Jan 06 12:25:14 2009
BMDS Model Run
The form of the probability function is:
P[response] = l/[l+EXP(-intercept-slope*dose)]
Dependent variable = Col3
Independent variable = Coll
Slope parameter is not restricted
Total number of observations = 3
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial Parameter Values
background = 0 Specified
intercept = -2.85849
slope = 0.176122
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameters) -background
have been estimated at a boundary point, or have been specified by the user,
and do not appear in the correlation matrix )
September 2009
B-21
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
o
J
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
o o
JJ
34
35
36
37
38
intercept slope
intercept 1 -0.85
slope -0.85 1
Parameter Estimates
95.0% Wald Confidence Interval
Variable Estimate Std. Err. Lower Conf. Limit Upper Conf. Limit
intercept -3.28386 0.49695 -4.25787 -2.30986
slope 0.190277 0.0288302 0.133771 0.246783
Analysis of Deviance Table
Model Log(likelihood) # Param's Deviance Test d.f
Full model -48.3572 3
Fitted model -49.0267 2 1.33895 1
Reduced model -101.497 1 106.279 2
AIC: 102.053
Goodness of Fit
Scaled
Dose Est. Prob. Expected Observed Size Residual
0.0000 0.0361 1.806 3.000 50 0.905
12.8000 0.2998 14.989 13.000 50 -0.614
32.0000 0.9429 46.204 47.000 49 0.490
ChiA2 = 1.44 d.f. = 1 P-value = 0.2309
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Extra risk
Confidence level = 0.95
BMD= 7.38383
BMDL = 5.71667
P-value
0.2472
<0001
September 2009
B-22
DRAFT - DO NOT CITE OR QUOTE
-------
Table B-10. Modeling Results for Splenic Hematopoietic Cell Proliferation in
Female B6C3F1 Mice
Model:
Restricted:
BMR:
BMR Type:
BMD
BMDL
ChiA2
Goodness-
of-fit p-
value
AIC
Fitted Log-
Likelihood
Fitted
p-value
Gamma
Yes
0.1
Extra
Risk
5.73584
1.90919
0.00
NA
171.405
-82.7026
NA
Logistic
0.1
Extra
Risk
4.06642
3.28512
0.02
0.8993
169.421
-82.7106
0.8993
Log-
Logistic
Yes
0.1
Extra
Risk
6.5828
2.43228
0.00
NA
171.405
-82.7026
NA
Log-
Probit
Yes
0.1
Extra
Risk
6.91076
3.48982
0.00
NA
171.405
-82.7026
NA
Probit
0.1
Extra
Risk
4.03306
3.33147
0.00
0.9466
169.41
-82.7048
0.9466
Weibull
Yes
0.1
Extra
Risk
5.17994
1.90919
0.00
NA
171.405
-82.7026
NA
Quantal-
Linear
0.1
Extra
Risk
2.34557
1.7616
1.35
0.2455
171.405
-82.7026
NA
Hill
Yes
0.1
Extra
Risk
~
~
-
-
~
~
~
Scaled Residuals:
0
12.8
32
0.000
-0.001
0.000
0.064
-0.095
0.054
0.000
0.000
-0.000
0.000
0.000
0.000
0.033
-0.052
0.026
-0.000
0.000
-0.000
-0.000
0.000
-0.000
~
~
~
Note High dose group was dropped in order to obtain adequate model fit
September 2009
B-23
DRAFT - DO NOT CITE OR QUOTE
-------
Probit Model with 0.95 Confidence Level
T3
0
0
0
t
<
C
0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
Probit
BMDL BMD
10
15
dose
20
25
30
10:17 12/12 2008
Figure B-8. Probit model fit for splenic hematopoietic cell proliferation in female
B6C3F1 mice
1
1
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
Probit Model. (Version: 3.1; Date: 05/16/2008)
Input Data File: M:\Chloroprene\Female_B6C3Fl_mouse\ntp_hemato _prolif_hdd_probit.(d)
Gnuplot Plotting File: M:\Chloroprene\Female_B6C3Fl_mouse\ntp_hemato_j>rolif_hdd_probit.plt
Fri Dec 12 09:20:03 2008
BMDS Model Run
The form of the probability function is:
P[response] = CumNorm(Intercept+Slope*Dose),
where CurnNormQ is the cumulative normal distribution function
Dependent variable = Col3
Independent variable = Coll
Slope parameter is not restricted
Total number of observations = 3
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial (and Specified) Parameter Values
background = 0 Specified
intercept = -0.643083
slope = 0.0528681
Asymptotic Correlation Matrix of Parameter Estimates
September 2009
B-24
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
( *** The model parameters) -background
have been estimated at a boundary point, or have been specified by the user,
and do not appear in the correlation matrix )
intercept
intercept 1
slope -0.73
Parameter Estimates
Variable Estimate
intercept -0.649733
slope 0.0534876
slope
-0.73
1
95.0% Wald Confidence Interval
Std. Err. Lower Conf. Limit Upper Conf. Limit
0.16595 -0.97499 -0.324476
0.00913534 0.0355826 0.0713925
Analysis of Deviance Table
Model Log(likelihood)
Full model -82.7026
Fitted model -82.7048
# Param's Deviance Test d.f. P- value
3
2 0.00449095 1 0.9466
Reduced model -102.099 1 38.7924 2 <0001
AIC: 169.41
Goodness of Fit
Scaled
Dose Est. Prob. Expected Observed Size Residual
0.0000 0.2579
12.8000 0.5139
32.0000 0.8559
ChiA2 = 0.00 d.f. = 1
12.897 13.000 50 0.033
25.182 25.000 49 -0.052
41.937 42.000 49 0.026
P- value = 0.9466
Benchmark Dose Computation
Specified effect = 0
Risk Type = Extra
.1
risk
Confidence level = 0.95
BMD = 4.03306
BMDL= 3.33147
September 2009
B-25
DRAFT - DO NOT CITE OR QUOTE
-------
APPENDIX C. CANCER DOSE-RESPONSE MODELING RESULTS
1 rwinrannR Fatal Graph
.o i u^/ lu/^uup_HEM3fatal.TTD - Chloroprene: female mice, circulatory system
Model: Two Stage Weib
U)
ir:
0.8
0.6
0.4
0.2
0
Dose (ppm)=12.8
Dose (ppm)=32
Kaplan Meier (12.8)
Kaplan Meier (32)
20
40
60
Time (wks)
80
100
Figure C-l. Female mice, hemangiomas and hemangiosarcomas in all organs; high dose
dropped,) hemangiosarcomas occurring before termination considered fatal
September 2009
C-1
DRAFT - DO NOT CITE OR QUOTE
185
-------
.cy n4/1 n/9nna Incidental Graph
.o/ u^/ lu/^uup_HEM3fatal.TTD - Chloroprene: female mice, circulatory system
Model: Three Stage Weib
0.8
0.6
0.4
0.2
0
Dose (ppm)=12.8
Dose (ppm)=32
HoelWalburg(12.8)
Hoel Walburg (32)
H
20
40
60
Time (wks)
80
100
120
Figure C-2. Female mice, hemangiomas and hemangiosarcomas in all organs; high dose
dropped, hemangiosarcomas occurring before termination considered fatal. Details
below.
Model: Two Stage Weib Dataset: M:\_ToxRiskData\Chloroprene\F_HEM3fatal.TTD
Functional form: 1 - EXP [ ( -QO - Ql * D - Q2 * D^2) * (T - TO)*Z]
Maximum Log-Likelihood = -1.431146e+002
Parameter Estimates :
Q 0 = 1.352245E-015
Q 1 = O.OOOOOOE+000
Q 2 = 1.970859E-017
Z = 6 .820986E+000
TO = O.OOOOOOE+000
Set by User
Avg. Doses
(ppm)
0
12 .80
32
Model: Two Stage Weib
Target Species: Human
Route: Air
of animals
50
50
50
J.>J ULLU^C i.
with fatal
tumors
1
4
16
with incidental
tumors
3
2
2
Exposure Pattern
Age Begins: 0 Age Ends: 70
Weeks/Year: 52 Days/Week: 7
Hours/Day : 24
Animal to human conversion method: PPM IN AIR
Human Equivalent Dose Estimates (ug/m^3)
September 2009
C-2
95.00 %
Upper Bound
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
DRAFT - DO NOT CITE OR QUOTE
185
-------
16:4504/10/2008
1
0.8
0.6
0.4
0.2
Incidental Graph
F_LUNG.TTD - Alv/Bronch tumors, females
Model: Multistage Weib
Dose (ppm)=12.8
Dose (ppm)=32
— - Dose (ppm)=80
' HoelWalburg(12.8)
---- Hoel Walburg (32)
— - Hoel Walburg (80)
20
40
60
Time (wks)
100
120
Figure C-3. Female mice, alveolar/bronchiolar tumors. Details below.
Model: Three Stage Weib Dataset: M:\_ToxRiskData\Chloroprene\F_LUNG.TTD
Functional form: 1 - EXP[( -QO - Ql * D - Q2 * D^2 - Q3 * D^3 )
* (T - TO)*Z]
Maximum Log-Likelihood = -8.624362e+001
Parameter Estimates :
Avg. Doses
(ppm)
0
12 .80
32
80
Q 0
Q 1
Q 2
Q 3
z
TO
= 5.
= 5.
= 0.
= 0.
= 4 .
= 0.
of animals
50
50
50
50
.948316E-010
.755951E-010
.OOOOOOE+000
.OOOOOOE+000
.067307E+000
.OOOOOOE+000 Set
with fatal
tumors
0
0
0
0
by User
with incidental
tumors
4
28
34
42
Model: Three Stage Weib
Target Species: Human
Route: Air
Exposure Pattern
Age Begins: 0 Age Ends: 70
Weeks/Year: 52 Days/Week: 7
Hours/Day : 24
Animal to human conversion method: PPM IN AIR
Human Equivalent Dose Estimates (ug/m^3)
Incid Extra Risk
l.OOOOE-006
l.OOOOE-005
0.0001
0.0010
0.01
0.10
Time (yr)
70.00
70.00
70.00
70.00
70.00
70.00
95.00 %
Lower Bound
3.2072E-002
3.2072E-001
3.2074E+000
3.2088E+001
3.2234E+002
3.3791E+003
MLE
4.0061E-002
4.0061E-001
4.0063E+000
4.0081E+001
4.0263E+002
4.2208E+003
95.00 %
Upper Bound
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
September 2009
C-3
DRAFT - DO NOT CITE OR QUOTE
185
-------
16:53 04/10/2008
Incidental Graph
FJJV.TTD - Chloroprene: female mice, hep. care, and aden.
Model: One Stage Weib
0.8
0.6
0.4
0.2
0
Dose (ppm)=12.8
Dose(ppm)=32
— - Dose(ppm)=80
' Hoel Walburg(12.8)
Hoel Walburg (32)
' Hoel Walburg (80)
20
40
60
Time (wks)
80
100
120
Figure C-4. Female mice, hepatocellular adenomas and carcinomas. Details below.
Model: One Stage Weib Dataset: M:\_ToxRiskData\Chloroprene\F_LIV.TTD
Functional form: 1 - EXP[( -QO - Ql * D ) * (T - TO)*Z]
Maximum Log-Likelihood = -1.233781e+002
Parameter Estimates :
Q 0 = 1.153457E-010
Q 1 = 5.270331E-012
Z = 4.834857E+000
TO = O.OOOOOOE+000
Set by User
Avg. Doses
(ppm)
0
12 .80
32
80
of animals
50
49
50
50
with fatal
tumors
0
0
0
0
with incidental
tumors
20
26
20
30
Exposure Pattern
Model: One Stage Weib Age Begins: 0 Age Ends: 70
Target Species: Human Weeks/Year: 52 Days/Week: 7
Route: Air Hours/Day : 24
Animal to human conversion method: PPM IN AIR
Human Equivalent Dose Estimates (ug/m^3)
Incid Extra Risk
l.OOOOE-006
l.OOOOE-005
0.0001
0.0010
0.01
0.10
Time (yr)
70.00
70.00
70.00
70.00
70.00
70.00
95.00 %
Lower Bound
9.2486E-002
9.2486E-001
9.2490E+000
9.2532E+001
9.2951E+002
9.7443E+003
MLE
1.2383E-001
1.2383E+000
1.2384E+001
1.2389E+002
1.2445E+003
1.3047E+004
95.00 %
Upper Bound
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
September 2009
C-4
DRAFT - DO NOT CITE OR QUOTE
185
-------
17:2304/10/2008
Incidental Graph
F_FORST.TTD - Chloroprene: female mice, forestomach
Model: One Stage Weib
0.8
0.6
0.4
0.2
0
Dose (ppm)=12.8
Dose (ppm)=32
Dose (ppm)=80
' ' Hoel Walburg (80)
20
40
60
Time (wks)
80
100
120
Figure C-5. Female mice, forestomach tumors. Details below.
Model: One Stage Weib Dataset: M:\_ToxRiskData\Chloroprene\F_FORST.TTD
Functional form: 1 - EXP[( -QO - Ql * D ) * (T - TO)*Z]
Maximum Log-Likelihood = -3.359600e+001
Parameter Estimates :
Avg. Doses
(ppm)
0
12 .80
32
80
Q 0 = 6.983602E-023
Q 1 = 1.330500E-023
Z = l.OOOOOOE+001
TO = O.OOOOOOE+000
Set by User
of animals
50
50
49
50
with fatal
tumors
0
0
0
0
with incidental
tumors
1
0
0
4
Model: One Stage Weib
Target Species: Human
Route: Air
Exposure Pattern
Age Begins: 0 Age Ends: 70
Weeks/Year: 52 Days/Week: 7
Hours/Day : 24
Animal to human conversion method: PPM IN AIR
Human Equivalent Dose Estimates (ug/m^3)
Incid Extra Risk
l.OOOOE-006
l.OOOOE-005
0.0001
0.0010
0.01
0.10
Time (yr)
70.00
70.00
70.00
70.00
70.00
70.00
95.00 %
Lower Bound
5.7604E-001
5.7604E+000
5.7607E+001
5.7632E+002
5.7894E+003
6.0692E+004
MLE
1.8723E+000
1.8723E+001
1.8724E+002
1.8733E+003
1.8818E+004
1.9727E+005
95.00 %
Upper Bound
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
September 2009
C-5
DRAFT - DO NOT CITE OR QUOTE
185
-------
17:2804/10/2008
Incidental Graph
F_HARD.TTD - Chloroprene: female mice, hard.gland tumors
Model: One Stage Weib
1
0.8
0.6
0.4
0.2
Dose (ppm)=12.8
Dose (ppm)=32
— - Dose (ppm)=80
HoelWalburg(12.8)
Hoel Walburg (32)
Hoel Walburg (80)
20
40
60
Time (wks)
80
100
120
Figure C-6. Female mice, Harderian gland tumors. Details below.
Model: One Stage Weib Dataset: M:\_ToxRiskData\Chloroprene\F_HARD.TTD
Functional form: 1 - EXP[( -QO - Ql * D ) * (T - TO)*Z]
Maximum Log-Likelihood = -6.640761e+001
Parameter Estimates :
Avg. Doses
(ppm)
0
12 .80
32
80
Model: One Stage Weib
Target Species: Human
Route: Air
Q 0 = 4.680639E-014
Q 1 = 6.624162E-015
Z = 5.942561E+000
TO = O.OOOOOOE+000
Set by User
of animals
50
50
50
50
J.>J ULLU^C i.
with fatal
tumors
0
0
0
0
with incidental
tumors
2
5
3
9
Age Begins: 0
Weeks/Year: 52
Exposure Pattern
Age Ends: 70
Days/Week: 7
Hours/Day : 24
Animal to human conversion method: PPM IN AIR
Human Equivalent Dose Estimates (ug/m^3)
Incid Extra Risk
l.OOOOE-006
l.OOOOE-005
0.0001
0.0010
0.01
0.10
Time (yr)
70.00
70.00
70.00
70.00
70.00
70.00
95.00 %
Lower Bound
4.5697E-001
4.5697E+000
4.5699E+001
4.5720E+002
4.5927E+003
4 .8146E+004
MLE
5.7446E-001
5.7446E+000
5.7449E+001
5.7474E+002
5.7735E+003
6.0525E+004
95.00 %
Upper Bound
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
September 2009
C-6
DRAFT - DO NOT CITE OR QUOTE
185
-------
1 7-33 04/1 0/20
.^ u-r/ '"'^"
Incidental Graph
- Chloroprene: female mice, mammary care, aden.
Model: One Stage Weib
1
0.8
0.6
0.4
0.2
Dose (ppm)=12.8
Dose (ppm)=32
— - Dose (ppm)=80
HoelWalburg(12.8)
Hoel Walburg (32)
Hoel Walburg (80)
20
40
60
Time (wks)
80
100
120
Figure C-7. Female mice, mammary gland tumors. Details below.
Model: One Stage Weib Dataset: M:\_ToxRiskData\Chloroprene\F_MAMM.TTD
Functional form: 1 - EXP[( -QO - Ql * D ) * (T - TO)*Z]
Maximum Log-Likelihood = -9.075190e+001
Parameter Estimates :
Avg. Doses
(ppm)
0
12 .80
32
80
Model: One Stage Weib
Target Species: Human
Route: Air
Q 0 = 6.820061E-004
Q 1 = 4.969301E-005
Z = l.OOOOOOE+000
TO = O.OOOOOOE+000
Set by User
of animals
50
50
50
50
with fatal
tumors
0
0
0
0
with incidental
tumors
3
6
11
14
Exposure Pattern
Age Begins: 0 Age Ends: 70
Weeks/Year: 52 Days/Week: 7
Hours/Day : 24
Animal to human conversion method: PPM IN AIR
Human Equivalent Dose Estimates (ug/m^3)
Incid Extra Risk
l.OOOOE-006
l.OOOOE-005
0.0001
0.0010
0.01
0.10
Time (yr)
70.00
70.00
70.00
70.00
70.00
70.00
95.00 %
Lower Bound
4.6346E-001
4.6347E+000
4.6349E+001
4.6369E+002
4.6580E+003
4 .8831E+004
MLE
7.1352E-001
7.1352E+000
7.1355E+001
7.1387E+002
7.1711E+003
7.5176E+004
95.00 %
Upper Bound
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
September 2009
C-7
DRAFT - DO NOT CITE OR QUOTE
185
-------
10:1004/14/2008
Incidental Graph
F_SKIN.TTD - Chloroprene: female mice, skin sarcomas
Model: One Stage Weib
1
0.8
0.6
in
ce
0.4
0.2
c
Dose
— - Dose
' Hoel
' Hoel
' Hoel
) 10
'' ' ' ' ' ' ' -
; (ppm)=32 •
; (ppm)=80
Walburg (12.8) :
Walburg (32) :
Walburg (80) ;
:
I , ~^— r '
_-— - " _ _ - - " " ^____ — — :
20 30 40 50 60 70 80 90
Time (wks)
Figure C-8. Female mice, skin sarcomas. Details below.
Model: One Stage Weib Dataset: M:\_ToxRiskData\Chloroprene\F_SKIN.TTD
Functional form: 1 - EXP[( -QO - Ql * D ) * (T - TO)*Z]
Maximum Log-Likelihood = -9.120449e+001
Parameter Estimates :
Avg. Doses
(ppm)
0
12 .80
32
80
Q 0 = O.OOOOOOE+000
Q 1 = 7.398518E-007
Z = 2.086144E+000
TO = O.OOOOOOE+000
Set by User
of animals
50
50
50
50
J.>J ULLU^C i.
with fatal
tumors
0
0
0
0
with incidental
tumors
0
11
11
18
Model: One Stage Weib
Target Species: Human
Route: Air
Exposure Pattern
Age Begins: 0 Age Ends: 70
Weeks/Year: 52 Days/Week: 7
Hours/Day : 24
Animal to human conversion method: MG/KG BODY WEIGHT(3/4)/DAY
Human Equivalent Dose Estimates (ug/m^3)
Incid Extra Risk
l.OOOOE-006
l.OOOOE-005
0.0001
0.0010
0.01
0.10
Time (yr)
70.00
70.00
70.00
70.00
70.00
70.00
95.00 %
Lower Bound
2.4134E-001
2.4134E+000
2.4135E+001
2.4146E+002
2.4255E+003
2.5428E+004
MLE
3 .7066E-001
3.7066E+000
3.7068E+001
3.7085E+002
3.7253E+003
3.9053E+004
95.00 %
Upper Bound
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
September 2009
C-8
DRAFT - DO NOT CITE OR QUOTE
185
-------
11:21 04/11/2008
Incidental Graph
F_Zymb.TTD - Chloroprene: female mice, Zymbal's gland
Model: Three Stage Weib
0.8
0.6
0.4
0.2
Dose (ppm)=12.8
Dose (ppm)=32
— - Dose (ppm)=80
' Hoel Walburg (80)
20
40
60
Time (wks)
80
100
120
Figure C-9. Female mice, Zymbal's gland tumors. Details below.
Model: Three Stage Weib
Functional form: 1 - EXP[( -QO - Ql '
* (T - TO)*Z]
Maximum Log-Likelihood = -2.897667e+001
Dataset: M:\_ToxRiskData\Chloroprene\F_Zymb.TTD
D - Q2 * D^2 - Q3 * D^3 )
Parameter Estimates :
Avg. Doses
(ppm)
0
12 .80
32
80
Q 0 = O.OOOOOOE+000
Q 1 = O.OOOOOOE+000
Q 2 = O.OOOOOOE+000
Q 3 = 2 .224776E-027
Z = l.OOOOOOE+001
TO = O.OOOOOOE+000
Set by User
of animals
50
50
49
50
J.>J ULLU^C i.
with fatal
tumors
0
0
0
0
with incidental
tumors
0
0
0
3
Model: Three Stage Weib
Target Species: Human
Route: Air
Exposure Pattern
Age Begins: 0 Age Ends: 70
Weeks/Year: 52 Days/Week: 7
Hours/Day : 24
Animal to human conversion method: PPM IN AIR
Human Equivalent Dose Estimates (ug/m^3)
Incid Extra Risk Time (yr)
l.OOOOE-006 70.00
l.OOOOE-005 70.00
0.0001 70.00
0.0010 70.00
0.01 70.00
0.10 70.00
95.00 %
Lower Bound
4.8038E-001
4.8039E+000
4.8041E+001
4.8062E+002
1.1261E+004
7.8634E+004
MLE
5.3398E+003
1.1504E+004
2.4786E+004
5.3407E+004
1.1524E+005
2.5220E+005
95.00 %
Upper Bound
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
September 2009
C-9
DRAFT - DO NOT CITE OR QUOTE
185
-------
17:4004/10/2008
Incidental Graph
MJHEM.TTD - Chloroprene: Male mice, circulatory system
Model: One Stage Weib
1
0.8
0.6
0.4
0.2
0
Dose (ppm)=12.8
Dose (ppm)=32
Dose (ppm)=80
HoelWalburg(12.8)
Hoel Walburg (32)
Hoel Walburg (80)
H
20
40
60
Time (wks)
80
100
120
17:4704/10/2008
Fatal Graph
M_HEM.TTD - Chloroprene: Male mice, circulatory system
Model: One Stage Weib
1
0.8
0.6
0.4
0.2
Dose (ppm)=12.8
Dose (ppm)=32
Dose (ppm)=80
Kaplan Meier (12.8)
Kaplan Meier (32)
Kaplan Meier (80)
20
40
60
Time (wks)
80
100
Figure C-10. Male mice, hemangiomas and hemangiosarcomas; hemangiosarcomas
occurring before termination considered fatal. Details below.
September 2009
C-10
DRAFT - DO NOT CITE OR QUOTE
185
-------
Model: One Stage Weib Dataset: M:\_ToxRiskData\Chloroprene\M_HEM.TTD
Functional form: 1 - EXP[( -QO - Ql * D ) * (T - TO)*Z]
Maximum Log-Likelihood = -2.022502e+002
Parameter Estimates :
Q 0 = 8.379828E-022
Q 1 = 1.163863E-022
Z = l.OOOOOOE+001
TO = O.OOOOOOE+000
Set by User
Avg. Doses
(ppm)
0
12 .80
32
80
of animals
50
50
50
50
J.>J ULLU^C i.
with fatal
tumors
0
6
16
13
with incidental
tumors
3
8
7
8
Model: One Stage Weib
Target Species: Human
Route: Air
Exposure Pattern
Age Begins: 0 Age Ends: 70
Weeks/Year: 52 Days/Week: 7
Hours/Day : 24
Animal to human conversion method: PPM IN AIR
Human Equivalent Dose Estimates (ug/m^3)
IncidExtra Risk Time
l.OOOOE-006
l.OOOOE-005
0.0001
0.0010
0.01
0.10
(yr)
70.00
70.00
70.00
70.00
70.00
70.00
95.00 %
Lower Bound
8.0694E-002
8.0694E-001
8.0698E+000
8.0734E+001
8.1100E+002
8.5019E+003
MLE
1.0423E-001
1.0423E+000
1.0424E+001
1.0428E+002
1.0475E+003
1.0982E+004
95.00 %
Upper Bound
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
September 2009
C-11
DRAFT - DO NOT CITE OR QUOTE
185
-------
17:51 04/10/2008
Incidental Graph
M_LUNG.TTD - Chloroprene: Alv/Bronch tumors, males
Model: One Stage Weib
1
0.8
0.6
0.4
0.2
0
C
_ ' , ,
Dose (ppm)=;
— - Dose (ppm)=?
• Hoel Walburg
• Hoel Walburg
2.8
32 _^J :
30 _ ^-^ H ,--"" ;
(12.8) I / I ,,-
(32) ' r------R //- :
<80> X r-,^-' /" ;
r-y- -^i _//! — I :
/ ."' ,- H :
) 20
40 60 80 100 120
Time (wks)
Figure C-ll. Male mice, alveolar/bronchiolar tumors. Details below.
Model: One Stage Weib Dataset
Functional form: 1 - EXP[( -QO - Ql * D ) * (T - TC
Maximum Log-Likelihood = -1.049275e+002
Parameter Estimates :
Q 0 = 4.009395E-008
Q 1 = 4 .460491E-009
Z = 3.461551E+000
TO = O.OOOOOOE+000
M: \_ToxRiskData\Chloroprene\M_LUNG.TTD
Set by User
of animals
50
50
50
50
with fatal
tumors
0
0
0
0
with incidental
tumors
13
28
36
43
Avg. Doses
(ppm)
0
12 .80
32
80
Generating Extrapolated Doses Table
TITLE: Chloroprene: Alv/Bronch tumors, males
Dataset: M:\_ToxRiskData\Chloroprene\M_LUNG.TTD
Exposure Pattern
Model: One Stage Weib Age Begins: 0 Age Ends: 70
Target Species: Human Weeks/Year: 52 Days/Week: 7
Route: Air Hours/Day : 24
Animal to human conversion method: PPM IN AIR
Human Equivalent Dose Estimates (ug/m^3)
Incid Extra Risk
l.OOOOE-006
l.OOOOE-005
0.0001
0.0010
0.01
0.10
Time (yr)
70.00
70.00
70.00
70.00
70.00
70.00
95.00 %
Lower Bound
6.2992E-002
6.2992E-001
6.2995E+000
6.3024E+001
6.3309E+002
6 .6369E+003
MLE
8 .6156E-002
8 .6157E-001
8 .6160E+000
8.6199E+001
8.6590E+002
9.0774E+003
95.00 %
Upper Bound
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
September 2009
C-12
DRAFT - DO NOT CITE OR QUOTE
185
-------
Incidental Graph
_ Ch|oroprene: Ma|e mice forestomach papillomas
Model: One Stage Weib
1
0.8
0.6
0.4
0.2
Dose (ppm)=12.8
Dose (ppm)=32
Dose (ppm)=80
Hoel Walburg (32)
Hoel Walburg (80)
20 40 60 80
Time (wks)
100
120
Figure C-12. Male mice, forestomach tumors. Details below.
Model: One Stage Weib Dataset: M:\_ToxRiskData\Chloroprene\M_FORST.TTD
Functional form: 1 - EXP[( -QO - Ql * D ) * (T - TO)*Z]
Maximum Log-Likelihood = -2.831876e+001
Parameter Estimates :
Avg. Doses
(ppm)
0
12 .80
32
80
Model: One Stage Weib
Target Species: Human
Route: Air
Q 0 = 3.027002E-006
Q 1 = 2.341463E-007
Z = 1.793090E+000
TO = O.OOOOOOE+000
Set by User
of animals
50
50
50
50
with fatal
tumors
0
0
0
0
with incidental
tumors
1
0
2
4
Exposure Pattern
Age Begins: 0 Age Ends: 70
Weeks/Year: 52 Days/Week: 7
Hours/Day : 24
Animal to human conversion method: PPM IN AIR
Human Equivalent Dose Estimates (ug/m^3)
Incid Extra Risk
l.OOOOE-006
l.OOOOE-005
0.0001
0.0010
0.01
0.10
Time (yr)
70.00
70.00
70.00
70.00
70.00
70.00
95.00 %
Lower Bound
1.5961E+000
1.5961E+001
1.5962E+002
1.5969E+003
1.6041E+004
1.6816E+005
MLE
3 .8065E+000
3 .8065E+001
3 .8067E+002
3 .8084E+003
3 .8256E+004
4.0105E+005
95.00 %
Upper Bound
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
September 2009
C-13
DRAFT - DO NOT CITE OR QUOTE
185
-------
17:56 04/10/:
1
0.8
0.6
i
0.4
0.2
Incidental Graph
- Chloroprene: Male mice, harderian gland tumors
Model: One Stage Weib
Dose (ppm)=12.8
Dose (ppm)=32
— - Dose (ppm)=80
HoelWalburg(12.8)
Hoel Walburg (32)
Hoel Walburg (80)
20
40
60
Time (wks)
80
100
120
Figure C-13. Male mice, Harderian gland tumors. Details below.
Generating Model Fit Table
TITLE: Chloroprene: Male mice, Harderian gland tumors
Model: One Stage Weib Dataset: M:\_ToxRiskData\Chloroprene\M_HARD.TTD
(T - TO)*Z]
000 Set by User
Number
with fatal with incidental
tumors tumors
0 2
0 5
0 10
0 12
Generating Extrapolated Doses Table
TITLE: Chloroprene: Male mice, Harderian gland tumors
Dataset: M:\_ToxRiskData\Chloroprene\M_HARD.TTD
Exposure Pattern
Model: One Stage Weib Age Begins: 0 Age Ends: 70
Target Species: Human Weeks/Year: 52 Days/Week: 7
Route: Air Hours/Day : 24
Animal to human conversion method: PPM IN AIR
Unit Potency [ per mg/kg/day ] (computed for Risk of l.OE-6)
Lower Bound = Not Reqstd MLE = 5.9938E-003 Upper Bound(ql*) = 9.5529E-003
Induction Time (TO) Set by User to 0
'unctional form:
Maximum
1 - EXP [ ( -QO - Ql
Log-Likelihood =
*
7.
D ) * ('
.3663946
Parameter Estimates :
(ppm)
0
12 .80
32
80
Q 0 =
Q 1 =
Z
TO =
3 .
3 .
5.
0.
.258829E
.597993E
.574590E
.OOOOOOE
of animals w
50
50
50
50
Human Equivalent Dose Estimates (ug/m^3)
Incid Extra Risk
l.OOOOE-006
l.OOOOE-005
0.0001
0.0010
0.01
0.10
Time (yr)
70.00
70.00
70.00
70.00
70.00
70.00
95.00 %
Lower Bound
3.6653E-001
3.6653E+000
3.6655E+001
3.6671E+002
3.6837E+003
3.8618E+004
MLE
5.8417E-001
5.8417E+000
5.8420E+001
5.8446E+002
5.8711E+003
6.1548E+004
95.00 %
Upper Bound
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
September 2009
C-14
DRAFT - DO NOT CITE OR QUOTE
185
-------
17:5804/10/2008
Incidental Graph
M_KIDN.TTD - Chloroprene: Male mice, kidney adenomas
Model: One Stage Weib
1
0.8
0.6
0.4
0.2
Dose (ppm)=12.8
Dose (ppm)=32
— - Dose (ppm)=80
HoelWalburg(12.8)
Hoel Walburg (32)
Hoel Walburg (80)
20
40
60
Time (wks)
80
100
120
Figure C-14. Male mice, renal tubule tumors. Details below.
Model: One Stage Weib Dataset: M:\ ToxRiskData\Chloroprene\M_KIDN.TTD
Functional form: 1 - EXP[( -QO - Ql * D ) * (T - TO)^Z]
Maximum Log-Likelihood = -4.100330e+001
Parameter Estimates
Avg. Doses
(ppm)
0
12 .80
32
80
Q 0 = O.OOOOOOE+000
Q 1 = 2.031241E-015
Z = 6 .092308E+000
TO = O.OOOOOOE+000
Set by User
of animals
50
50
50
50
Model: One Stage Weib
Target Species: Human
Route: Air
Number -
with fatal
tumors
0
0
0
0
Exposure Pattern
Age Begins: 0 Age Ends
Weeks/Year: 52 Days/Week
Hours/Day
with incidental
tumors
0
2
3
9
70
7
24
Animal to human conversion method: PPM IN AIR
Human Equivalent Dose Estimates (ug/m^3)
Incid Extra Risk
l.OOOOE-006
l.OOOOE-005
0.0001
0.0010
0.01
0.10
Time (yr)
70.00
70.00
70.00
70.00
70.00
70.00
95.00 %
Lower Bound
5.7583E-001
5.7583E+000
5.7586E+001
5.7612E+002
5.7873E+003
6.0670E+004
MLE
9.3451E-001
9.3451E+000
9.3455E+001
9.3498E+002
9.3921E+003
9.8460E+004
95.00 %
Upper Bound
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
Not Reqstd
September 2009
C-15
DRAFT - DO NOT CITE OR QUOTE
185
-------
Table C-l. Summary of human equivalent overall cancer risk values estimated
R/BMDR, based on male and female mouse tumor incidence (NTP, 1998)
by
Tumor site
Assumed
Dosimetry
Risk,
R
BMDR,
Hg/m3
BMDLR,
Hg/m3
Unit Risk
at
BMDRa,
per
Hg/m3
Unit
risk",
per
Hg/m3
SD
SD2
Proportion
of total
variance
Female Mice
Lung
Skin
All
hemangiomas,
hemangiosarco
mas
Mammary
adenomas,
carcinomas,
adenocanthom
as
Hepatocellular
tumors
Forestomach
Harderian
gland
Zymbal's gland
Systemic
Portal-of-
Entry
Systemic
Systemic
Systemic
Systemic
Systemic
Systemic
Systemic
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.001
3.22 x
IO2
1.32 x
IO3
2.43 x
IO3
6.43 x
IO2
4.66 x
IO3
9.30 x
IO2
5.79 x
IO3
4.59 x
IO3
4.81 x
IO2
4.03 x
IO2
1.65 x
IO3
3.73 x
IO3
8.63 x
IO2
7.17 x
IO3
1.24 x
IO3
1.88 x
IO4
5.77 x
IO3
5.34 x
IO4
Sum, MLE Cancer Risks':
3.10 x
io-5
7.57. x
io-6
4.12 x
io-6
1.55 x
io-5
2.15 x
1.08 x
io-5
1.73 x
io-6
2.18 x
io-6
2.08 x
10"6
5.08 x
io-5c
3.20 x
io-5
Upper Bound on Sum of MLE Risk Estimates6:
Continuous Human Equivalent Overall Unit Riskf:
2.48 x
io-5
6.06 x
io-6
2.68 x
io-6
1.16 x
io-5
1.39 x
io-6
8.04 x
10"6
5.31 x
io-7
1.73 x
io-6
1.87 x
10"8
5.92 x
io-5
3.79 x
io-5
3.32 x
io-4
2.13 x
io-4
3.76 x
io-6
9.17 x
io-7
8.74 x
io-7
2.41 x
io-6
4.57 x
io-7
1.66 x
10"6
7.27 x
io-7
2.71 x
io-7
1.25 x
10"6
Sum,
SD .
Overall
SDd:
1.41 x
io-11
8.41 x
io-13
7.65 x
io-13
5.80 x
io-12
2.09 x
io-13
2.74 x
io-12
5.29 x
io-13
7.33 x
io-14
1.57 x
io-12
2.58 x
io-11
1.25 x
io-11
5.08 x
10"6
3.54 x
io-6
0.07
0.06
0.46
0.02
0.22
0.04
0.01
0.13
0.55
0.03
0.22
0.01
0.11
0.02
0.00
0.06
September 2009
C-16
DRAFT - DO NOT CITE OR QUOTE
185
-------
Tumor site
Assumed
Dosimetry
Risk,
R
BMDR,
Hg/m3
BMDLR,
Hg/m3
Unit
risk
at
BMDRa,
per
Hg/m3
Unit
riskb,
per
Hg/m3
SD
SD2
Proportion
of total
variance
Male Mice
Lung
All
hemangiomas,
hemangiosarco
mas
Forestomach
Harderian
gland
Kidney
Systemic
Portal-of-
Entry
Systemic
Systemic
Systemic
Systemic
0.01
0.01
0.01
0.01
0.01
0.01
6.33 x
IO2
2.60 x
IO3
8.11 x
IO2
1.60 x
IO4
3.68 x
IO3
5.79 x
IO3
8.66 x
IO2
3.55 x
IO3
1.05 x
IO3
3.83 x
IO4
5.87 x
IO3
9.39 x
IO3
Sum, MLE Cancer Risks':
1.58 x
io-5
3.85 x
io-6
1.23 x
io-5
6.23 x
io-7
2.71 x
io-6
1.73 x
io-6
2.41 x
io-5
1.54 x
io-5
Upper Bound on Sum of Risk Estimates6:
Continuous Human Equivalent Overall Unit Riskf:
1.15 x
io-5
2.82 x
io-6
9.55 x
io-6
2.61 x
io-7
1.70 x
io-6
1.06 x
io-6
2.94 x
10 5
1.86 x
io-5
1.65 x
io-4
1.05 x
io-4
2.58 x
io-6
6.30 x
io-7
1.69 x
io-6
2.20 x
io-7
6.15 x
io-7
4.03 x
io-7
Sum
SD2:
Over
all
SDd:
6.67 x
io-12
3.96 x
io-13
2.86 x
io-12
4.84 x
io-14
3.78 x
io-13
1.63 x
io-13
1.01 x
io-11
3.85 x
io-12
3.18x
10 6
1.96 x
10"6
0.10
0.74
0.01
0.10
0.04
0.66
0.28
0.00
0.04
0.02
aR/BMDR
bR/BMDLR
0 Summary statistics in italics were calculated using the "systemic" entries. The other summary statistics were
calculated using the "portal-of-entry" estimate for lung tumors and all entries for the other tumor sites.
d Overall SD = (Sum, SD2)0 5
e Upper bound on the overall risk estimate = Sum of MLE cancer risks + 1.645 x Overall SD.
f Adjusted for continuous exposure by multiplying unit risk in previous line by 6/24 (hours) x 5/7 (days) = 5.6.
September 2009
C-17
DRAFT - DO NOT CITE OR QUOTE
185
------- |