f/EPA
EPA/63 5/R-14/312a
External Review Draft
www.epa.gov/iris
Toxicological Review of Benzo[a]pyrene
(CASRN 50-32-8)
In Support of Summary Information on the
Integrated Risk Information System (IRIS)
September 2014
NOTICE
This document is an External Review draft. This information is distributed solely for the purpose
of pre-dissemination peer review under applicable information quality guidelines. It has not been
formally disseminated by EPA. It does not represent and should not be construed to represent any
Agency determination or policy. It is being circulated for review of its technical accuracy and
science policy implications.
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC
-------
Toxicological Review ofBenzo[a]pyrene
DISCLAIMER
This document is a preliminary draft for review purposes only. This information is
distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should
not be construed to represent any Agency determination or policy. Mention of trade names or
commercial products does not constitute endorsement of recommendation for use.
This document is a draft for review purposes only and does not constitute Agency policy.
ii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
CONTENTS
AUTHORS | CONTRIBUTORS | REVIEWERS ix
PREFACE xii
PREAMBLE TO IRIS TOXICOLOGICAL REVIEWS xvi
EXECUTIVE SUMMARY xxxiv
LITERATURE SEARCH STRATEGY! STUDY SELECTION xli
1. HAZARD IDENTIFICATION 1-1
1.1. PRESENTATION AND SYNTHESIS OF EVIDENCE BY ORGAN/SYSTEM 1-1
1.1.1. Developmental Toxicity 1-1
1.1.2. Reproductive Toxicity 1-22
1.1.3. Immunotoxicity 1-37
1.1.4. Other Toxicity 1-44
1.1.5. Carcinogenicity 1-52
1.2. SUMMARY AND EVALUATION 1-81
1.2.1. Weight of Evidence for Effects Other than Cancer 1-81
1.2.2. Weight of Evidence for Carcinogenicity 1-82
2. DOSE-RESPONSE ANALYSIS 2-1
2.1. ORAL REFERENCE DOSE FOR EFFECTS OTHER THAN CANCER 2-1
2.1.1. Identification of Studies and Effects for Dose-Response Analysis 2-1
2.1.2. Methods of Analysis 2-5
2.1.3. Derivation of Candidate Values 2-8
2.1.4. Derivation of Organ/System-Specific Reference Doses 2-12
2.1.5. Selection of the Proposed Overall Reference Dose 2-14
2.1.6. Confidence Statement 2-14
2.1.7. Previous IRIS Assessment: Reference Dose 2-15
2.2. INHALATION REFERENCE CONCENTRATION FOR EFFECTS OTHER THAN CANCER 2-15
2.2.1. Identification of Studies and Effects for Dose-Response Analysis 2-15
2.2.2. Methods of Analysis 2-17
2.2.3. Derivation of Candidate Values 2-19
2.2.4. Derivation of Organ/System-Specific Reference Concentrations 2-22
This document is a draft for review purposes only and does not constitute Agency policy.
iii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
2.2.5. Selection of the Proposed Reference Concentration 2-23
2.2.6. Confidence Statement 2-23
2.2.7. Previous IRIS Assessment: Reference Concentration 2-24
2.2.8. Uncertainties in the Derivation of the RfD and RfC 2-24
2.3. ORAL SLOPE FACTOR FOR CANCER 2-25
2.3.1. Analysis of Carcinogenicity Data 2-25
2.3.2. Dose-Response Analysis—Adjustments and Extrapolation Methods 2-28
2.3.3. Derivation of the Oral Slope Factor 2-28
2.3.4. Uncertainties in the Derivation of the Oral Slope Factor 2-30
2.3.5. Previous IRIS Assessment: Oral Slope Factor 2-32
2.4. INHALATION UNIT RISK FOR CANCER 2-32
2.4.1. Analysis of Carcinogenicity Data 2-33
2.4.2. Dose-Response Analysis—Adjustments and Extrapolation Methods 2-34
2.4.3. Inhalation Unit Risk Derivation 2-35
2.4.4. Uncertainties in the Derivation of the Inhalation Unit Risk 2-36
2.4.5. Previous IRIS Assessment: Inhalation Unit Risk 2-38
2.5. DERMAL SLOPE FACTOR FOR CANCER 2-39
2.5.1. Analysis of Carcinogenicity Data 2-39
2.5.2. Dose-Response Analysis—Adjustments and Extrapolation Methods 2-40
2.5.3. Derivation of the Dermal Slope Factor 2-42
2.5.4. Dermal Slope Factor Cross-Species Scaling 2-44
2.5.5. Uncertainties in the Derivation of the Dermal Slope Factor 2-45
2.5.6. Previous IRIS Assessment: Dermal Slope Factor 2-47
2.6. APPLICATION OF AGE-DEPENDENT ADJUSTMENT FACTORS (ADAFS) 2-47
REFERENCES R-l
This document is a draft for review purposes only and does not constitute Agency policy.
iv DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
TABLES
Table ES-1. Organ/system-specific RfDs and proposed overall RfD for benzo[a]pyrene xxxv
Table ES-2. Organ/system-specific RfCs and proposed overall RfCfor benzo[a]pyrene xxxvii
Table LS-1. Summary of the search strategy employed for benzo[a]pyrene xli
Table 1-1. Evidence pertaining to developmental effects of benzo[a]pyrene in humans 1-4
Table 1-2. Evidence pertaining to developmental effects of benzo[a]pyrene in animals 1-5
Table 1-3. Evidence pertaining to the neurodevelopmental effects of benzo[a]pyrene from PAH
mixtures 1-13
Table 1-4. Evidence pertaining to the neurodevelopmental effects of benzo[a]pyrene in animals 1-15
Table 1-5. Evidence pertaining to the male reproductive toxicity of benzo[a]pyrene in adult
animals 1-25
Table 1-6. Evidence pertaining to the female reproductive effects of benzo[a]pyrene in humans 1-32
Table 1-7. Evidence pertaining to the female reproductive effects of benzo[a]pyrene in adult
animals 1-33
Table 1-8. Evidence pertaining to the immune effects of benzo[a]pyrene in animals 1-41
Table 1-9. Evidence pertaining to other toxicities of benzo[a]pyrene in animals 1-49
Table 1-10. Cancer sites for PAH-related agents reviewed by IARC 1-55
Table 1-11. Summary of epidemiologic studies of benzo[a]pyrene (direct measures) in relation
to lung cancer risk: Tier 1 studies 1-55
Table 1-12. Summary of epidemiologic studies of benzo[a]pyrene (direct measures) in relation
to lung cancer risk: Tier 2 studies 1-56
Table 1-13. Summary of epidemiologic studies of benzo[a]pyrene (direct measures) in relation
to bladder cancer risk 1-60
Table 1-14. Tumors observed in chronic oral animal bioassays 1-63
Table 1-15. Tumors observed in chronic inhalation animal bioassays 1-66
Table 1-16. Tumors observed in chronic dermal animal bioassays 1-68
Table 1-17. Experimental support for the postulated key events for mutagenic mode of action 1-75
Table 1-18. Supporting evidence for the carcinogenic to humans cancer descriptor for
benzo[a]pyrene 1-87
Table 2-1. Summary of derivation of PODs 2-7
Table 2-2. Effects and corresponding derivation of candidate values 2-10
Table 2-3. Organ/system-specific RfDs and proposed overall RfD for benzo[a]pyrene 2-13
Table 2-4. Summary of derivation of PODs 2-19
Table 2-5. Effects and corresponding derivation of candidate values 2-21
Table 2-6. Organ/system-specific RfCs and proposed overall RfCfor benzo[a]pyrene 2-22
Table 2-7. Summary of the oral slope factor derivations 2-29
Table 2-8. Summary of uncertainties in the derivation of cancer risk values for benzo[a]pyrene
oral slope factor 2-31
Table 2-9. Summary of the inhalation unit risk derivation 2-36
Table 2-10. Summary of uncertainties in the derivation of cancer risk values for benzo[a]pyrene
(inhalation unit risk) 2-38
Table 2-11. Summary of dermal slope factor derivations, unadjusted for interspecies differences 2-43
Table 2-12. Summary of uncertainties in the derivation of cancer risk values for benzo[a]pyrene
dermal slope factor 2-47
Table 2-13. Sample application of ADAFs for the estimation of benzo[a]pyrene cancer risk
following lifetime (70-year) oral exposure 2-48
This document is a draft for review purposes only and does not constitute Agency policy.
v DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Table 2-14. Sample application of ADAFs for the estimation of benzo[a]pyrene cancer risk
following lifetime (70-year) inhalation exposure 2-49
Table 2-15. Sample application of ADAFs for the estimation of benzo[a]pyrene cancer risk
following lifetime (70-year) dermal exposure 2-49
FIGURES
Figure LS-1. Study selection strategy xliii
Figure 1-1. Exposure-response array for developmental effects following oral exposure to
benzo[a]pyrene 1-8
Figure 1-2. Exposure-response array for neurodevelopmental effects following oral exposure 1-18
Figure 1-3. Exposure-response array for male reproductive effects following oral exposure in
adult animals 1-28
Figure 1-4. Exposure-response array for female reproductive effects following oral exposure in
adult animals 1-34
Figure 1-5. Exposure-response array for immune effects following oral exposure 1-42
Figure 1-6. Proposed metabolic activation pathways and key events in the carcinogenic mode of
action for benzo[a]pyrene 1-70
Figure 2-1. Candidate values with corresponding PODs and composite UFs 2-12
Figure 2-2. Candidate values with corresponding PODs and composite UFs 2-22
This document is a draft for review purposes only and does not constitute Agency policy.
vi DRAFT—DO NOT CITE OR QUOTE
-------
ABBREVIATIONS
Toxicological Review ofBenzo[a]pyrene
1-OH-Py 1-hydroxypyrene ETS
AchE acetylcholine esterase EU
ADAF age-dependent adjustment factor Fe203
Ah aryl hydrocarbon FSH
AHH aryl hydrocarbon hydroxylase GABA
AhR aryl hydrocarbon receptor GD
AIC Akaike's Information Criterion GI
AKR aldo-keto reductase GJIC
AMI acute myocardial infarction
ANOVA analysis of variance GSH
ARNT Ah receptor nuclear translocator GST
AST aspartate transaminase GSTM1
ATSDR Agency for Toxic Substances and hCG
Disease Registry HEC
BMC benchmark concentration HED
BMCL benchmark concentration lower HERO
confidence limit
BMD benchmark dose HFC
BMDL benchmark dose, 95% lower bound HPLC
BMDS Benchmark Dose Software
BMR benchmark response hprt
BPDE benzo[a]pyrene-7,8-diol-9,10-epoxide
BPQ benzo [ajpyrene semiquinone HR
BrdU bromodeoxyuridine Hsp90
BSM benzene-soluble matter i.p.
BUN blood urea nitrogen i.v.
BW body weight Ig
CA chromosomal aberration IHD
CAL/EPA California Environmental Protection IRIS
Agency LDH
CASRN Chemical Abstracts Service Registry LH
Number LOAEL
CERCLA Comprehensive Environmental MAP
Response, Compensation, and Liability MCL
Act MCLG
CHO Chinese hamster ovary MIAME
CI confidence interval
GYP cytochrome MLE
CYP450 cytochrome P450 MMAD
DAF dosimetric adjustment factor MN
dbcAMP dibutyl cyclic adenosine MPPD
monophosphate mRNA
DMSO dimethyl sulfoxide MS
DNA deoxyribonucleic acid NCE
EC European Commission NCEA
EH epoxide hydrolase
ELISA enzyme-linked immunosorbent assay NIOSH
EPA Environmental Protection Agency
EROD 7-ethoxyresorufin-O-deethylase NK
environmental tobacco smoke
European Union
ferrous oxide
follicle stimulating hormone
gamma-aminobutyric acid
gestational day
gastrointestinal
gap junctional intercellular
communication
reduced glutathione
glutathione-S-transferase
glutathione-S-transferase Ml
human chorionic gonadotropin
human equivalent concentration
human equivalent dose
Health and Environmental Research
Online
high-frequency cell
high-performance liquid
chromatography
hypoxanthine guanine phosphoribosyl
transferase
hazard ratio
heat shock protein 90
intraperitoneal
intravenous
immunoglobulin
ischemic heart disease
Integrated Risk Information System
lactate dehydrogenase
luteinizing hormone
lowest-observed-adverse-effect level
mitogen-activated protein
Maximum Contaminant Level
Maximum Contaminant Level Goal
Minimum Information About a
Microarray Experiment
maximum likelihood estimate
mass median aerodynamic diameter
micronucleus
Multi-Path Particle Deposition
messenger ribonucleic acid
mass spectrometry
normochromatic erythrocyte
National Center for Environmental
Assessment
National Institute for Occupational
Safety and Health
natural-killer
This document is a draft for review purposes only and does not constitute Agency policy.
vii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
NMDA N-methyl-D-aspartate SEM
NOAEL no-observed-adverse-effect level SHE
NPL National Priorities List SIR
NQO NADPH:quinone oxidoreductase SMR
NRC National Research Council SOAR
NTP National Toxicology Program SOD
OECD Organisation for Economic SRBC
Co-operation and Development SSB
OR odds ratio TCDD
ORD Office of Research and Development TK
PAH polycyclic aromatic hydrocarbon ToxR
PBMC peripheral blood mononuclear cell TPA
PBPK physiologically based pharmacokinetic TUNEL
PCA Principal Components Analysis
PCE polychromatic erythrocyte TWA
PCNA proliferating cell nuclear antigen UCL
PND postnatal day UDP-UGT
POD point of departure
PUVA psoralen plus ultraviolet-A UDS
RBC red blood cell UF
RDDRER regional deposited dose ratio for UFA
extrarespiratory effects UFn
RfC inhalation reference concentration UFH
RfD oral reference dose UFi
RNA ribonucleic acid UFS
ROS reactive oxygen species
RR relative risk UVA
s.c. subcutaneous UVB
SCC squamous cell carcinoma WBC
SCE sister chromatid exchange WESPOC
SCSA sperm chromatin structure assay WT
SD standard deviation WTC
SE standard error XPA
standard error of the mean
Syrian hamster embryo
standardized incidence ratio
standardized mortality ratio
Systematic Omics Analysis Review
superoxide dismutase
sheep red blood cells
single-strand break
2,3,7,8-tetrachlorodibenzo-p-dioxin
thymidine kin as e
Toxicological Reliability Assessment
12-0-tetradecanoylphorbol-13-acetate
terminal deoxynucleotidyl transferase
dUTP nick end labeling
time-weighted average
upper confidence limit
uridine diphosphate-
glucuronosyltransferase
unscheduled DNA synthesis
uncertainty factor
interspecies uncertainty factor
database deficiencies uncertainty factor
intraspecies uncertainty factor
LOAEL-to-NOAEL uncertainty factor
subchronic-to-chronic uncertainty
factor
ultraviolet-A
ultraviolet-B
white blood cell
water escape pole climbing
wild type
World Trade Center
xeroderma pigmentosum group A
This document is a draft for review purposes only and does not constitute Agency policy,
8 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
AUTHORS | CONTRIBUTORS | REVIEWERS
Assessment Team
Christine Cai, MS
Glinda Cooper, Ph.D.
Louis D'Amico, Ph.D.
Jason Fritz, Ph.D.
Martin Gehlhaus, MHS
Catherine Gibbons, Ph.D.
Karen Hogan, MS
Andrew Kraft, Ph.D.
Kathleen Newhouse, MS (Assessment Manager)
Linda Phillips, Ph.D.
Margaret Pratt, Ph.D.
Keith Salazar, Ph.D.
John Schaum, MS (retired)
Suryanarayana Vulimiri, DVM
Gene Ching-Hung Hsu, Ph.D. (formerly with
EPA)
Lyle Burgoon, Ph.D.
John Cowden, Ph.D.
Amanda Persad, Ph.D.
John Stanek, Ph.D.
Chris Brinkerhoff, Ph.D.
Emma McConnell, MS
Scott Glaberman, Ph.D.
U.S. EPA
Office of Research and Development
National Center for Environmental
Assessment
Washington, DC
U.S. EPA
Office of Research and Development
National Center for Environmental
Assessment
Research Triangle Park, NC
Oak Ridge Institute for Science and
Education Fellow
American Association for the
Advancement of Science Fellow
Scientific Support
Lynn Flowers, Ph.D.
John Fox, Ph.D.
Paul White, Ph.D.
Maria Spassova, Ph.D.
U.S. EPA
Office of Research and Development
National Center for Environmental
Assessment
Washington, DC
Production Team
Taukecha Cunningham
Maureen Johnson
Terri Konoza
Vicki Soto
U.S. EPA
Office of Research and Development
National Center for Environmental
Assessment
Washington, DC
Contractor Support
Heather Carlson-Lynch, S.M.
Peter McClure, Ph.D.
SRC, Inc., Syracuse, NY
This document is a draft for review purposes only and does not constitute Agency policy.
ix DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Megan Riccardi
Kelly Salinas
Joe Santodonato
Julie Stickney, Ph.D.
George Holdsworth, Ph.D.
Lutz W.Weber, Ph.D.
Janusz Z. Byczkowski, Ph.D., D.Sc.
Oak Ridge Institute for Science and
Education, Oak Ridge, TN
JZB Consulting, Fairborn, OH
Executive Direction
Kenneth Olden, Ph.D., Sc.D., L.H.D.
(Center Director)
Lynn Flowers, Ph.D.
(Associate Director for Health)
Vincent Cogliano, Ph.D.
(IRIS Program Director—acting)
Samantha Jones, Ph.D.
(IRIS Associate Director for Science)
Jamie B. Strong, Ph.D.
(Toxic Effects Branch Chief)
U.S. EPA/ORD/NCEA
Washington, DC
Internal Review Team
Stephen Nesnow, Ph.D. (retired)
Rita Schoeny, Ph.D.
U.S. EPA
National Health and Environmental
Effects Research Laboratory
Research Triangle Park, NC
U.S. EPA
Office of Water
Washington, DC
Reviewers
This assessment was provided for review to scientists in EPA's Program and Region Offices.
Comments were submitted by:
Office of Children's Health Protection, Washington DC
Office of Policy, Washington, DC
Office of Solid Waste and Emergency Response, Washington DC
Office of Water, Washington DC
Region 2, New York, NY
Region 3, Philadelphia, PA
Region 8, Denver, CO
This assessment was provided for review to other federal agencies and the Executive Office of the
President Comments were submitted by:
This document is a draft for review purposes only and does not constitute Agency policy.
x DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Agency for Toxic Substances and Disease Registry, Centers for Disease Control,
Department of Health and Human Services
Department of Defense
National Aeronautics and Space Administration
National Institute for Occupational Safety and Health, Centers for Disease Control,
Department of Health and Human Services
Office of Management and Budget, Executive Office of the President
White House Council on Environmental Quality, Executive Office of the President
This document is a draft for review purposes only and does not constitute Agency policy.
xi DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2 PREFACE
3 This Toxicological Review, prepared under the auspices of the U.S. Environmental
4 Protection Agency's (EPA's) Integrated Risk Information System (IRIS) program, critically reviews
5 the publicly available studies on benzo[a]pyrene in order to identify potential adverse health effects
6 and to characterize exposure-response relationships. Benzo[a]pyrene is found in the environment
7 and in food. Benzo[a]pyrene occurs in conjunction with other structurally related chemical
8 compounds known as polycyclic aromatic hydrocarbons (PAHs).1 Benzo[a]pyrene is universally
9 present in these mixtures and is routinely analyzed and detected in environmental media
10 contaminated with PAH mixtures: thus it is often used as an indicator chemical to measure
11 exposure to PAH mixtures [Bostrometal.. 2002). and as an index chemical for deriving potency
12 factors for PAH mixtures.
13 Benzo[a]pyrene is listed as a hazardous substance under the Comprehensive Environmental
14 Response, Compensation, and Liability Act of 1980 (CERCLA), is found at 524 hazardous waste sites
15 on the National Priorities List (NPL) and is ranked number 8 out of 275 chemicals on the Priority
16 List of Hazardous Substances for CERCLA [ATSDR. 2011). This ranking is based on a combination
17 of factors that include the frequency of occurrence at NPL sites, the potential for human exposure,
18 and the potential health hazard. Benzo[a]pyrene is also listed as a drinking water contaminant
19 under the Safe Drinking Water Act and a Maximum Contaminant Level Goal (MCLG) and
20 enforceable Maximum Contaminant Level (MCL) have been established2. It is also one of the
21 chemicals included in EPA's Persistent Bioaccumulative and Toxic Chemical Program
22 [http://www.epa.gov/pbt/pubs/benzo.htm]. In air, benzo[a]pyrene is regulated as a component in
23 a class of chemicals referred to as Polycyclic Organic Matter, defined as a Hazardous Air Pollutant
24 by the 1990 amendments to the Clean Air Act.
25 This assessment updates IRIS assessment of benzo[a]pyrene that was developed in 1987.
26 The previous assessment included a cancer descriptor and oral slope factor. New information has
27 become available, and this assessment reviews information on all health effects by all exposure
28 routes. Organ/system-specific reference values are calculated based on developmental,
29 reproductive, and immune system toxicity data. These reference values may be useful for
30 cumulative risk assessments that consider the combined effect of multiple agents acting on the
31 same biological system. In addition, in consideration of the Agency's need to estimate the potential
iPAHs are a large class of chemical compounds formed during the incomplete combustion of organic matter.
They consist of only carbon and hydrogen arranged in two or more fused rings.
2MCLG = 0, MCL = 0.0002 mg/L.
This document is a draft for review purposes only and does not constitute Agency policy.
xii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 for skin cancer from dermal exposure [U.S. EPA, 2004], especially in children exposed to
2 contaminated soil, this assessment includes the IRIS Program's first dermal slope factor.
3 This assessment was conducted in accordance with EPA guidance, which is cited and
4 summarized in the Preamble to Toxicological Reviews. Appendices for chemical and physical
5 properties, toxicokinetic information, and summaries of toxicity studies are provided as
6 Supplemental Information to this assessment.
7 For additional information about this assessment or for general questions regarding IRIS,
8 please contact EPA's IRIS Hotline at 202-566-1676 (phone), 202-566-1749 (fax), or
9 hotline.iris@epa.gov.
10 Chemical Properties
11 Benzo[a]pyrene is a five-ring PAH. It is a pale yellow crystalline solid with a faint aromatic
12 odor. It is relatively insoluble in water and has low volatility. Benzo[a]pyrene is released to the air
13 from both natural and anthropogenic sources and removed from the atmosphere by photochemical
14 oxidation; reaction with nitrogen oxides, hydroxy and hydroperoxy radicals, ozone, sulfur oxides,
15 and peroxyacetyl nitrate; and wet and dry deposition to land or water. In air, benzo[a]pyrene is
16 predominantly adsorbed to particulates but may also exist as a vapor at high temperatures (ATSDR,
17 19951
18 Uses and Pathways of Exposure
19 There is no known commercial use for benzo[a]pyrene; it is only produced as a research
20 chemical. Benzo[a]pyrene is ubiquitous in the environment primarily as a result of incomplete
21 combustion emissions. It is found in fossil fuels, crude oils, shale oils, and coal tars (HSDB, 2012). It
22 is released to the environment via both natural sources (such as forest fires) and anthropogenic
23 sources including stoves/furnaces burning fossil fuels (especially wood and coal), motor vehicle
24 exhaust, cigarettes, and various industrial combustion processes (ATSDR. 1995). Benzo[a]pyrene is
25 also found in soot and coal tars. Several studies have reported that urban run-off from asphalt-
26 paved car parks treated with coats of coal-tar emulsion seal could account for a large proportion of
27 PAHs in many watersheds (Rowe and O'Connor. 2011: Van Metre and Mahler. 2010: Mahler etal..
28 2005). Benzo[a]pyrene exposure can also occur to workers involved in the production of
29 aluminum, coke, graphite, and silicon carbide, and in coal tar distillation. The major sources of non-
30 occupational exposure are tobacco products, inhalation of polluted air, ingestion of contaminated
31 food and water, and through cooking processes that involve smoke (HSDB. 2012). Dermal exposure
32 can occur through contact with materials containing soot, tar, or crude petroleum, including
33 pharmaceutical products containing coal tar, such as coal tar-based shampoos and treatments for
34 eczema and psoriasis (Gal/EPA. 2010: IARC. 2010).
35 It persists for a long period of time in the atmosphere in the particulate phase and is thus
36 efficiently transported over long distances. It is lipophilic with low water solubility; therefore, once
37 deposited in water or sediments, it adsorbs strongly to sediments and particulate matter and
This document is a draft for review purposes only and does not constitute Agency policy.
xiii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 degrades slowly over several years [Gal/EPA, 2010: GLC, 2007]. Because of its presence in high
2 concentrations in the waters and sediments of the Great Lakes and St Lawrence river ecosystem, it
3 is 1 of the 12 level I substances identified and targeted for reduction in the Great Lakes Region
4 fGLC. 20071.
5 Most aquatic organisms metabolize benzo[a]pyrene, eliminating it in days, and thus, it is not
6 expected to bioconcentrate in these organisms; however, several aquatic organisms such as
7 plankton, oysters, and some fish cannot metabolize benzo[a]pyrene [U.S. EPA. 2010a]. Thus, the
8 data on benzo[a]pyrene bioconcentration in aquatic organisms varies from low to very high [HSDB.
9 2012]. Biomagnification of benzo[a]pyrene in the food chain has not been reported [ATSDR, 1995].
10 Additional information on benzo[a]pyrene exposure and chemical properties can be found in
11 Appendix A.
12 Implementation of the 2011 National Research Council Recommendations
13 On December 23, 2011, The Consolidated Appropriations Act, 2012, was signed into law
14 [U.S. Congress. 2011]. The report language included direction to EPA for the IRIS Program related
15 to recommendations provided by the National Research Council (NRC] in their review of EPA's
16 draft IRIS assessment of formaldehyde [NRC, 2011]. The report language included the following:
17 The Agency shall incorporate, as appropriate, based on chemical-specific datasets
18 and biological effects, the recommendations of Chapter 7 of the National Research
19 Council's Review of the Environmental Protection Agency's Draft IRIS Assessment of
20 Formaldehyde into the IRIS process...For draft assessments released in fiscal year
21 2012, the Agency shall include documentation describing how the Chapter 7
22 recommendations of the National Academy of Sciences (NAS] have been
23 implemented or addressed, including an explanation for why certain
24 recommendations were not incorporated.
25 The NRC's recommendations, provided in Chapter 7 of their review report, offered
26 suggestions to EPA for improving the development of IRIS assessments. Consistent with the
27 direction provided by Congress, documentation of how the recommendations from Chapter 7 of the
28 NRC report have been implemented in this assessment is provided in the table below. Where
29 necessary, the documentation includes an explanation for why certain recommendations were not
30 incorporated.
31 The IRIS Program's implementation of the NRC recommendations is following a phased
32 approach that is consistent with the NRC's "Roadmap for Revision" as described in Chapter 7 of the
33 formaldehyde review report. The NRC stated that "the committee recognizes that the changes
34 suggested would involve a multi-year process and extensive effort by the staff at the National
35 Center for Environmental Assessment and input and review by the EPA Science Advisory Board and
36 others."
37 Phase 1 of implementation has focused on a subset of the short-term recommendations,
38 such as editing and streamlining documents, increasing transparency and clarity, and using more
This document is a draft for review purposes only and does not constitute Agency policy.
xiv DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 tables, figures, and appendices to present information and data in assessments. Phase 1 also
2 focused on assessments near the end of the development process and close to final posting. The
3 IRIS benzo[a]pyrene assessment is in Phase 2 and represents a significant advancement in
4 implementing the NRC recommendations shown in Table F-l in Appendix F. The Program is
5 implementing all of these recommendations, but recognizes that achieving full and robust
6 implementation of certain recommendations will be an evolving process with input and feedback
7 from the public, stakeholders, and external peer review committees. Phase 3 of implementation
8 will incorporate the longer-term recommendations made by the NRC as outlined in Table F-2 in
9 Appendix F, including the development of a standardized approach to describe the strength of
10 evidence for noncancer effects. In May 2014, the NRC released their report reviewing the IRIS
11 assessment development process. As part of this review, the NRC reviewed current methods for
12 evidence-based reviews and made several recommendations with respect to integrating scientific
13 evidence for chemical hazard and dose-response assessments. In their report, the NRC states that
14 EPA should continue to improve its evidence-integration process incrementally and enhance the
15 transparency of its process. The committee did not offer a preference but suggests that EPA
16 consider which approach best fits its plans for the IRIS process. The NRC recommendations will
17 inform the IRIS Program's efforts in this area going forward. This effort is included in Phase 3 of
18 EPA's implementation plan.
19 Assessments by Other National and International Health Agencies
20 Toxicity information on benzo[a]pyrene has been evaluated by the World Health
21 Organization, Health Canada, the International Agency for Research on Cancer, and the European
22 Union. The results of these assessments are presented in Appendix B. It is important to recognize
23 that these assessments were prepared at different times, for different purposes, using different
24 guidelines and methods, and that newer studies have been included in the IRIS assessment.
This document is a draft for review purposes only and does not constitute Agency policy,
xv DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
PREAMBLE TO IRIS TOXICOLOGICAL REVIEWS
3 1. Scope of the IRIS Program
4 Soon after the EPA was established in
5 1970, it was at the forefront of developing
6 risk assessment as a science and applying it
7 in decisions to protect human health and the
8 environment. The Clean Air Act, for example,
9 mandates that the EPA provide "an ample
10 margin of safety to protect public health;"
11 the Safe Drinking Water Act, that "no
12 adverse effects on the health of persons may
13 reasonably be anticipated to occur, allowing
14 an adequate margin of safety." Accordingly,
15 the EPA uses information on the adverse
16 effects of chemicals and on exposure levels
17 below which these effects are not
18 anticipated to occur.
19 IRIS assessments critically review the
20 publicly available studies to identify adverse
21 health effects from exposure to chemicals
22 and to characterize exposure-response
23 relationships. In terms set forth by the
24 National Research Council [NRC. 1983). IRIS
25 assessments cover the hazard identification
26 and dose-response assessment steps of risk
27 assessment, not the exposure assessment or
28 risk characterization steps that are
29 conducted by the EPA's program and
30 regional offices and by other federal, state,
31 and local health agencies that evaluate risk
32 in specific populations and exposure
33 scenarios. IRIS assessments are distinct from
34 and do not address political, economic, and
35 technical considerations that influence the
36 design and selection of risk management
37 alternatives.
38 An IRIS assessment may cover a single
39 chemical, a group of structurally or
40 toxicologically related chemicals, or a
41 complex mixture. These agents may be found
42 in air, water, soil, or sediment Exceptions
43 are chemicals currently used exclusively as
44 pesticides, ionizing and non-ionizing
45 radiation, and criteria air pollutants listed
46 under Section 108 of the Clean Air Act
47 (carbon monoxide, lead, nitrogen oxides,
48 ozone, particulate matter, and sulfur oxides).
49 Periodically, the IRIS Program asks other
50 EPA programs and regions, other federal
51 agencies, state health agencies, and the
52 general public to nominate chemicals and
53 mixtures for future assessment or
54 reassessment Agents may be considered for
55 reassessment as significant new studies are
56 published. Selection is based on program
57 and regional office priorities and on
58 availability of adequate information to
59 evaluate the potential for adverse effects.
60 Other agents may also be assessed in
61 response to an urgent public health need.
62 2. Process for developing and peer
63 reviewing IRIS assessments
64 The process for developing IRIS
65 assessments (revised in May 2009 and
66 enhanced in July 2013) involves critical
67 analysis of the pertinent studies,
68 opportunities for public input, and multiple
69 levels of scientific review. The EPA revises
70 draft assessments after each review, and
71 external drafts and comments become part
72 of the public record (U.S. EPA. 20141
73 Before beginning an assessment, the IRIS
74 Program discusses the scope with other EPA
75 programs and regions to ensure that the
76 assessment will meet their needs. Then a
77 public meeting on problem formulation
78 invites discussion of the key issues and the
79 studies and analytical approaches that might
80 contribute to their resolution.
81 Step 1. Development of a draft
82 Toxicological Review. The draft
83 assessment considers all pertinent
84 publicly available studies and applies
85 consistent criteria to evaluate study
86 quality, identify health effects, identify
This document is a draft for review purposes only and does not constitute Agency policy.
xvi DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 mechanistic events and pathways,
2 integrate the evidence of causation for
3 each effect, and derive toxicity values. A
4 public meeting prior to the integration of
5 evidence and derivation of toxicity
6 values promotes public discussion of the
7 literature search, evidence, and key
8 issues.
9 Step 2. Internal review by scientists in
10 EPA programs and regions. The draft
11 assessment is revised to address the
12 comments from within the EPA.
13 Step 3. Interagency science consultation
14 with other federal agencies and the
15 Executive Offices of the President. The
16 draft assessment is revised to address
17 the interagency comments. The science
18 consultation draft, interagency
19 comments, and the EPA's response to
20 major comments become part of the
21 public record.
22 Step 4. Public review and comment,
23 followed by external peer review. The
24 EPA releases the draft assessment for
25 public review and comment A public
26 meeting provides an opportunity to
27 discuss the assessment prior to peer
28 review. Then the EPA releases a draft for
29 external peer review. The peer reviewers
30 also receive written and oral public
31 comments and the peer review meeting
32 is open to the public. The peer reviewers
33 assess whether the evidence has been
34 assembled and evaluated according to
35 guidelines and whether the conclusions
36 are justified by the evidence. The peer
37 review draft, written public comments,
38 and peer review report become part of
39 the public record.
40 Step 5. Revision of draft Toxicological
41 Review and development of draft IRIS
42 summary. The draft assessment is
43 revised to reflect the peer review
44 comments, public comments, and newly
45 published studies that are critical to the
46 conclusions of the assessment The
47 disposition of peer review comments
48 and public comments becomes part of
49 the public record.
50 Step 6. Final EPA review and interagency
51 science discussion with other federal
52 agencies and the Executive Offices of
53 the President. The draft assessment and
54 summary are revised to address the EPA
55 and interagency comments. The science
56 discussion draft, written interagency
57 comments, and the EPA's response to
58 major comments become part of the
59 public record.
60 Step 7. Completion and posting. The
61 Toxicological Review and IRIS summary
62 are posted on the IRIS website [http://
63 www.epa.gov/iris/].
64 The remainder of this Preamble
65 addresses step 1, the development of a draft
66 Toxicological Review. IRIS assessments
67 follow standard practices of evidence
68 evaluation and peer review, many of which
69 are discussed in EPA guidelines [U.S. EPA.
70 2005a. b, 2000b. 1998. 1996. 1991c. 1986a.
71 b) and other methods [U.S. EPA. 2012a. c,
72 2011. 2006a. b, 2002. 19941 Transparent
73 application of scientific judgment is of
74 paramount importance. To provide a
75 harmonized approach across IRIS
76 assessments, this Preamble summarizes
77 concepts from these guidelines and
78 emphasizes principles of general
79 applicability.
SO 3. Identifying and selecting
81 pertinent studies
82 3.1. Identifying studies
83 Before beginning an assessment, the EPA
84 conducts a comprehensive search of the
85 primary scientific literature. The literature
86 search follows standard practices and
87 includes the PubMed and ToxNet databases
88 of the National Library of Medicine, Web of
89 Science, and other databases listed in the
90 EPA's HERO system (Health and
91 Environmental Research Online, http://
92 hero.epa.gov/]. Searches for information on
This document is a draft for review purposes only and does not constitute Agency policy.
xvii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 mechanisms of toxicity are inherently
2 specialized and may include studies on other
3 agents that act through related mechanisms.
4 Each assessment specifies the search
5 strategies, keywords, and cut-off dates of its
6 literature searches. The EPA posts the
7 results of the literature search on the IRIS
8 website and requests information from the
9 public on additional studies and ongoing
10 research.
11 The EPA also considers studies received
12 through the IRIS Submission Desk and
13 studies (typically unpublished) submitted
14 under the Toxic Substances Control Act or
15 the Federal Insecticide, Fungicide, and
16 Rodenticide Act Material submitted as
17 Confidential Business Information is
18 considered only if it includes health and
19 safety data that can be publicly released. If a
20 study that may be critical to the conclusions
21 of the assessment has not been peer-
22 reviewed, the EPA will have it peer-
23 reviewed.
24 The EPA also examines the toxicokinetics
25 of the agent to identify other chemicals (for
26 example, major metabolites of the agent) to
27 include in the assessment if adequate
28 information is available, in order to more
29 fully explain the toxicity of the agent and to
30 suggest dose metrics for subsequent
31 modeling.
32 In assessments of chemical mixtures,
33 mixture studies are preferred for their
34 ability to reflect interactions among
35 components. The literature search seeks, in
36 decreasing order of preference (U.S. EPA.
37 2000b. 32.2: 1986c. 32.11:
38 - Studies of the mixture being assessed.
39 - Studies of a sufficiently similar mixture.
40 In evaluating similarity, the assessment
41 considers the alteration of mixtures in
42 the environment through partitioning
43 and transformation.
44 - Studies of individual chemical
45 components of the mixture, if there are
46 not adequate studies of sufficiently
47 similar mixtures.
48 3.2. Selecting pertinent epidemiologic
49 studies
50 Study design is the key consideration for
51 selecting pertinent epidemiologic studies
52 from the results of the literature search.
53 - Cohort studies, case-control studies, and
54 some population-based surveys (for
55 example, NHANES) provide the strongest
56 epidemiologic evidence, especially if they
57 collect information about individual
58 exposures and effects.
59 - Ecological studies (geographic
60 correlation studies) relate exposures and
61 effects by geographic area. They can
62 provide strong evidence if there are
63 large exposure contrasts between
64 geographic areas, relatively little
65 exposure variation within study areas,
66 and population migration is limited.
67 - Case reports of high or accidental
68 exposure lack definition of the
69 population at risk and the expected
70 number of cases. They can provide
71 information about a rare effect or about
72 the relevance of analogous results in
73 animals.
74 The assessment briefly reviews
75 ecological studies and case reports but
76 reports details only if they suggest effects
77 not identified by other studies.
78 3.3. Selecting pertinent experimental
79 studies
80 Exposure route is a key design
81 consideration for selecting pertinent
82 experimental animal studies or human
83 clinical studies.
84 - Studies of oral, inhalation, or dermal
85 exposure involve passage through an
86 absorption barrier and are considered
87 most pertinent to human environmental
88 exposure.
89 - Injection or implantation studies are
90 often considered less pertinent but may
91 provide valuable toxicokinetic or
92 mechanistic information. They also may
This document is a draft for review purposes only and does not constitute Agency policy.
xviii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 be useful for identifying effects in
2 animals if deposition or absorption is
3 problematic (for example, for particles
4 and fibers).
5 Exposure duration is also a key design
6 consideration for selecting pertinent
7 experimental animal studies.
8 - Studies of effects from chronic exposure
9 are most pertinent to lifetime human
10 exposure.
11 - Studies of effects from less-than-chronic
12 exposure are pertinent but less
13 preferred for identifying effects from
14 lifetime human exposure. Such studies
15 may be indicative of effects from less-
16 than-lifetime human exposure.
17 Short-duration studies involving animals
18 or humans may provide toxicokinetic or
19 mechanistic information.
20 For developmental toxicity and
21 reproductive toxicity, irreversible effects
22 may result from a brief exposure during a
23 critical period of development. Accordingly,
24 specialized study designs are used for these
25 effects fU.S. EPA. 2006b. 1998.1996.1991cl
26 4. Evaluating the quality of
27 individual studies
28 After the subsets of pertinent
29 epidemiologic and experimental studies
30 have been selected from the literature
31 searches, the assessment evaluates the
32 quality of each individual study. This
33 evaluation considers the design, methods,
34 conduct, and documentation of each study,
35 but not whether the results are positive,
36 negative, or null. The objective is to identify
37 the stronger, more informative studies based
38 on a uniform evaluation of quality
39 characteristics across studies of similar
40 design.
41 4.1. Evaluating the quality of
42 epidemiologic studies
43 The assessment evaluates design and
44 methodological aspects that can increase or
45 decrease the weight given to each
46 epidemiologic study in the overall evaluation
47 [U.S. EPA. 2005a. 1998.1996.1994.1991c):
48 - Documentation of study design,
49 methods, population characteristics, and
50 results.
51
52
53
54
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
Definition and selection of the study
group and comparison group.
Ascertainment of exposure
chemical or mixture.
to the
55 - Ascertainment of disease or health effect
Duration of exposure and follow-up and
adequacy for assessing the occurrence of
effects.
Characterization
critical periods.
of exposure during
Sample size and statistical power to
detect anticipated effects.
- Participation rates and potential for
selection bias as a result of the achieved
participation rates.
- Measurement error (can lead to
misclassification of exposure, health
outcomes, and other factors) and other
types of information bias.
- Potential confounding and other sources
of bias addressed in the study design or
in the analysis of results. The basis for
consideration of confounding is a
reasonable expectation that the
confounder is related to both exposure
and outcome and is sufficiently prevalent
to result in bias.
For developmental toxicity, reproductive
toxicity, neurotoxicity, and cancer there is
further guidance on the nuances of
evaluating epidemiologic studies of these
effects (U.S. EPA. 2005a. 1998.1996.1991c).
4.2. Evaluating the quality of
experimental studies
85 The assessment evaluates design and
86 methodological aspects that can increase or
87 decrease the weight given to each
This document is a draft for review purposes only and does not constitute Agency policy.
xix DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 experimental animal study, in vitro study, or
2 human clinical study (U.S. EPA. 2005a. 1998.
3 1996. 1991c). Research involving human
4 subjects is considered only if conducted
5 according to ethical principles.
6 - Documentation of study design, animals
7 or study population, methods, basic data,
8 and results.
9 - Nature of the assay and validity for its
10 intended purpose.
11 - Characterization of the nature and extent
12 of impurities and contaminants of the
13 administered chemical or mixture.
14 - Characterization of dose and dosing
15 regimen (including age at exposure) and
16 their adequacy to elicit adverse effects,
17 including latent effects.
18 - Sample sizes and statistical power to
19 detect dose-related differences or trends.
20 - Ascertainment of survival, vital signs,
21 disease or effects, and cause of death.
22 - Control of other variables that could
23 influence the occurrence of effects.
24 The assessment uses statistical tests to
25 evaluate whether the observations may be
26 due to chance. The standard for determining
27 statistical significance of a response is a
28 trend test or comparison of outcomes in the
29 exposed groups against those of concurrent
30 controls. In some situations, examination of
31 historical control data from the same
32 laboratory within a few years of the study
33 may improve the analysis. For an uncommon
34 effect that is not statistically significant
35 compared with concurrent controls,
36 historical controls may show that the effect
37 is unlikely to be due to chance. For a
38 response that appears significant against a
39 concurrent control response that is unusual,
40 historical controls may offer a different
41 interpretation [U.S. EPA. 2005a. §2.2.2.1.3).
42 For developmental toxicity, reproductive
43 toxicity, neurotoxicity, and cancer there is
44 further guidance on the nuances of
45 evaluating experimental studies of these
46 effects [U.S. EPA. 2005a. 1998. 1996. 1991c).
47 In multigeneration studies, agents that
48 produce developmental effects at doses that
49 are not toxic to the maternal animal are of
50 special concern. Effects that occur at doses
51 associated with mild maternal toxicity are
52 not assumed to result only from maternal
53 toxicity. Moreover, maternal effects may be
54 reversible, while effects on the offspring may
55 be permanent fU.S. EPA. 1998. §3.1.2.4.5.4:
56 1991c. 33.1.1.41.
57 4.3. Reporting study results
58 The assessment uses evidence tables to
59 present the design and key results of
60 pertinent studies. There may be separate
61 tables for each site of toxicity or type of
62 study.
63 If a large number of studies observe the
64 same effect, the assessment considers the
65 study quality characteristics in this section
66 to identify the strongest studies or types of
67 study. The tables present details from these
68 studies and the assessment explains the
69 reasons for not reporting details of other
70 studies or groups of studies that do not add
71 new information. Supplemental information
72 provides references to all studies
73 considered, including those not summarized
74 in the tables.
75 The assessment discusses strengths and
76 limitations that affect the interpretation of
77 each study. If the interpretation of a study in
78 the assessment differs from that of the study
79 authors, the assessment discusses the basis
80 for the difference.
81 As a check on the selection and
82 evaluation of pertinent studies, the EPA asks
83 peer reviewers to identify studies that were
84 not adequately considered.
85 5. Evaluating the overall evidence
86 of each effect
87 5.1. Concepts of causal inference
88 For each health effect, the assessment
89 evaluates the evidence as a whole to
90 determine whether it is reasonable to infer a
91 causal association between exposure to the
This document is a draft for review purposes only and does not constitute Agency policy.
xx DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 agent and the occurrence of the effect This
2 inference is based on information from
3 pertinent human studies, animal studies, and
4 mechanistic studies of adequate quality.
5 Positive, negative, and null results are given
6 weight according to study quality.
7 Causal inference involves scientific
8 judgment, and the considerations are
9 nuanced and complex. Several health
10 agencies have developed frameworks for
11 causal inference, among them the U.S.
12 Surgeon General [CDC. 2004: HEW. 19641.
13 the International Agency for Research on
14 Cancer [IARC. 2006] , the Institute of
15 Medicine [IOM. 2008). and the U.S. EPA.
16 f2010b.31.6: 2005a. 32.51. Although
17 developed for different purposes, the
18 frameworks are similar in nature and
19 provide an established structure and
20 language for causal inference. Each
21 considers aspects of an association that
22 suggest causation, discussed by Hill [1965]
23 and elaborated by Rothman and Greenland
24 f!998j. and U.S. EPA f2005a. §2.2.1.7: 1994.
25 §2.2.1.71
26 Strength of association: The finding of a
27 large relative risk with narrow
28 confidence intervals strongly suggests
29 that an association is not due to chance,
30 bias, or other factors. Modest relative
31 risks, however, may reflect a small range
32 of exposures, an agent of low potency, an
33 increase in an effect that is common,
34 exposure misclassification, or other
35 sources of bias.
36 Consistency of association: An inference of
37 causation is strengthened if elevated
38 risks are observed in independent
39 studies of different populations and
40 exposure scenarios. Reproducibility of
41 findings constitutes one of the strongest
42 arguments for causation. Discordant
43 results sometimes reflect differences in
44 study design, exposure, or confounding
45 factors.
46 Specificity of association: As originally
47 intended, this refers to one cause
48 associated with one effect Current
49 understanding that many agents cause
50 multiple effects and many effects have
51 multiple causes make this a less
52 informative aspect of causation, unless
53 the effect is rare or unlikely to have
54 multiple causes.
55 Temporal relationship: A causal
56 interpretation requires that exposure
57 precede development of the effect
58 Biologic gradient (exposure-response
59 relationship): Exposure-response
60 relationships strongly suggest causation.
61 A monotonic increase is not the only
62 pattern consistent with causation. The
63 presence of an exposure-response
64 gradient also weighs against bias and
65 confounding as the source of an
66 association.
67 Biologic plausibility: An inference of
68 causation is strengthened by data
69 demonstrating plausible biologic
70 mechanisms, if available. Plausibility
71 may reflect subjective prior beliefs if
72 there is insufficient understanding of the
73 biologic process involved.
74 Coherence: An inference of causation is
75 strengthened by supportive results from
76 animal experiments, toxicokinetic
77 studies, and short-term tests. Coherence
78 may also be found in other lines of
79 evidence, such as changing disease
80 patterns in the population.
81 "Natural experiments": A change in
82 exposure that brings about a change in
83 disease frequency provides strong
84 evidence, as it tests the hypothesis of
85 causation. An example would be an
86 intervention to reduce exposure in the
87 workplace or environment that is
88 followed by a reduction of an adverse
89 effect.
90 Analogy: Information on structural
91 analogues or on chemicals that induce
92 similar mechanistic events can provide
93 insight into causation.
This document is a draft for review purposes only and does not constitute Agency policy.
xxi DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 These considerations are consistent with
2 guidelines for systematic reviews that
3 evaluate the quality and weight of evidence.
4 Confidence is increased if the magnitude of
5 effect is large, if there is evidence of an
6 exposure-response relationship, or if an
7 association was observed and the plausible
8 biases would tend to decrease the magnitude
9 of the reported effect. Confidence is
10 decreased for study limitations,
11 inconsistency of results, indirectness of
12 evidence, imprecision, or reporting bias
13 [Guyattetal.. 2008b: Guyattetal.. 2008a).
14 5.2. Evaluating evidence in humans
15 For each effect, the assessment evaluates
16 the evidence from the epidemiologic studies
17 as a whole. The objective is to determine
18 whether a credible association has been
19 observed and, if so, whether that association
20 is consistent with causation. In doing this,
21 the assessment explores alternative
22 explanations (such as chance, bias, and
23 confounding) and draws a conclusion about
24 whether these alternatives can satisfactorily
25 explain any observed association.
26 To make clear how much the
27 epidemiologic evidence contributes to the
28 overall weight of the evidence, the
29 assessment may select a standard descriptor
30 to characterize the epidemiologic evidence
31 of association between exposure to the agent
32 and occurrence of a health effect.
33 Sufficient epidemiologic evidence of an
34 association consistent with causation:
35 The evidence establishes a causal
36 association for which alternative
37 explanations such as chance, bias, and
38 confounding can be ruled out with
39 reasonable confidence.
40 Suggestive epidemiologic evidence of an
41 association consistent with causation:
42 The evidence suggests a causal
43 association but chance, bias, or
44 confounding cannot be ruled out as
45 explaining the association.
46 Inadequate epidemiologic evidence to
47 infer a causal association: The available
48 studies do not permit a conclusion
49 regarding the presence or absence of an
50 association.
51 Epidemiologic evidence consistent with no
52 causal association: Several adequate
53 studies covering the full range of human
54 exposures and considering susceptible
55 populations, and for which alternative
56 explanations such as bias and
57 confounding can be ruled out, are
58 mutually consistent in not finding an
59 association.
60 5.3. Evaluating evidence in animals
61 For each effect, the assessment evaluates
62 the evidence from the animal experiments as
63 a whole to determine the extent to which
64 they indicate a potential for effects in
65 humans. Consistent results across various
66 species and strains increase confidence that
67 similar results would occur in humans.
68 Several concepts discussed by Hill [1965]
69 are pertinent to the weight of experimental
70 results: consistency of response, dose-
71 response relationships, strength of response,
72 biologic plausibility, and coherence [U.S.
73 EPA. 2005a. §2.2.1.7: 1994. Appendix C]}.
74 In weighing evidence from multiple
75 experiments, U.S. EPA f2005a). S2.5
76 distinguishes:
77 Conflicting evidence (that is, mixed positive
78 and negative results in the same sex and
79 strain using a similar study protocol)
80 from
81 Differing results (that is, positive results
82 and negative results are in different
83 sexes or strains or use different study
84 protocols).
85 Negative or null results do not invalidate
86 positive results in a different experimental
87 system. The EPA regards all as valid
88 observations and looks to explain differing
89 results using mechanistic information (for
90 example, physiologic or metabolic
91 differences across test systems) or
92 methodological differences (for example,
93 relative sensitivity of the tests, differences in
This document is a draft for review purposes only and does not constitute Agency policy.
xxii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 dose levels, insufficient sample size, or
2 timing of dosing or data collection).
3 It is well established that there are
4 critical periods for some developmental and
5 reproductive effects [U.S. EPA. 2006b.
6 2005a. b, 1998. 1996. 1991cl Accordingly,
7 the assessment determines whether critical
8 periods have been adequately investigated.
9 Similarly, the assessment determines
10 whether the database is adequate to
11 evaluate other critical sites and effects.
12 In evaluating evidence of genetic
13 toxicity:
14 - Demonstration of gene mutations,
15 chromosome aberrations, or aneuploidy
16 in humans or experimental mammals
17 (in vivo) provides the strongest
18 evidence.
19 - This is followed by positive results in
20 lower organisms or in cultured cells
21 (in vitro) or for other genetic events.
22 - Negative results carry less weight, partly
23 because they cannot exclude the
24 possibility of effects in other tissues
25 flARC. 20061
26 For germ-cell mutagenicity, the EPA has
27 defined categories of evidence, ranging from
28 positive results of human germ-cell
29 mutagenicity to negative results for all
30 effects of concern (U.S. EPA. 1986a. §2.3).
31 5.4. Evaluating mechanistic data
32 Mechanistic data can be useful in
33 answering several questions.
34 - The biologic plausibility of a causal
35 interpretation of human studies.
36 - The generalizability of animal studies to
37 humans.
38 - The susceptibility of particular
39 populations or lifestages.
40 The focus of the analysis is to describe, if
41 possible, mechanistic pathways that lead to a
42 health effect These pathways encompass:
43 - Toxicokinetic processes of absorption,
44 distribution, metabolism, and
45 elimination that lead to the formation of
46 an active agent and its presence at the
47 site of initial biologic interaction.
48 - Toxicodynamic processes that lead to a
49 health effect at this or another site (also
50 known as a mode of action}.
51 For each effect, the assessment discusses
52 the available information on its modes of
53 action and associated key events (key events
54 being empirically observable, necessary
55 precursor steps or biologic markers of such
56 steps; mode of action being a series of key
57 events involving interaction with cells,
58 operational and anatomic changes, and
59 resulting in disease). Pertinent information
60 may also come from studies of metabolites
61 or of compounds that are structurally similar
62 or that act through similar mechanisms.
63 Information on mode of action is not
64 required for a conclusion that the agent is
65 causally related to an effect (U.S. EPA. 2005a.
66 §2.51
67 The assessment addresses several
68 questions about each hypothesized mode of
69 action fU.S. EPA. 2005a. §2.4.3.41
70 1) Is the hypothesized mode of action
71 sufficiently supported in test animals?
72 Strong support for a key event being
73 necessary to a mode of action can come
74 from experimental challenge to the
75 hypothesized mode of action, in which
76 studies that suppress a key event
77 observe suppression of the effect
78 Support for a mode of action is
79 meaningfully strengthened by consistent
80 results in different experimental models,
81 much more so than by replicate
82 experiments in the same model. The
83 assessment may consider various
84 aspects of causation in addressing this
85 question.
86 2) Is the hypothesized mode of action
87 relevant to humans? The assessment
88 reviews the key events to identify critical
89 similarities and differences between the
90 test animals and humans. Site
91 concordance is not assumed between
92 animals and humans, though it may hold
This document is a draft for review purposes only and does not constitute Agency policy.
xxiii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 for certain effects or modes of action.
2 Information suggesting quantitative
3 differences in doses where effects would
4 occur in animals or humans is
5 considered in the dose-response
6 analysis. Current levels of human
7 exposure are not used to rule out human
8 relevance, as IRIS assessments may be
9 used in evaluating new or unforeseen
10 circumstances that may entail higher
11 exposures.
12 3) Which populations or lifestages can
13 be particularly susceptible to the
14 hypothesized mode of action? The
15 assessment reviews the key events to
16 identify populations and lifestages that
17 might be susceptible to their occurrence.
18 Quantitative differences may result in
19 separate toxicity values for susceptible
20 populations or lifestages.
21 The assessment discusses the likelihood
22 that an agent operates through multiple
23 modes of action. An uneven level of support
24 for different modes of action can reflect
25 disproportionate resources spent
26 investigating them [U.S. EPA.
27 2005a. §2.4.3.3). It should be noted that in
28 clinical reviews, the credibility of a series of
29 studies is reduced if evidence is limited to
30 studies funded by one interested sector
31 [Guyattetal.. 2008a).
32 For cancer, the assessment evaluates
33 evidence of a mutagenic mode of action to
34 guide extrapolation to lower doses and
35 consideration of susceptible lifestages. Key
36 data include the ability of the agent or a
37 metabolite to react with or bind to DNA,
38 positive results in multiple test systems, or
39 similar properties and structure-activity
40 relationships to mutagenic carcinogens [U.S.
41 EPA. 2005a.§2.3.5).
42 5.5. Characterizing the overall weight
43 of the evidence
44 After evaluating the human, animal, and
45 mechanistic evidence pertinent to an effect,
46 the assessment answers the question: Does
47 the agent cause the adverse effect [NRG.
48 2009. 19831? In doing this, the assessment
49 develops a narrative that integrates the
50 evidence pertinent to causation. To provide
51 clarity and consistency, the narrative
52 includes a standard hazard descriptor. For
53 example, the following standard descriptors
54 combine epidemiologic, experimental, and
55 mechanistic evidence of carcinogenicity [U.S.
56 EPA. 2005a. §2.51
57 Carcinogenic to humans: There is
58 convincing epidemiologic evidence of a
59 causal association (that is, there is
60 reasonable confidence that the
61 association cannot be fully explained by
62 chance, bias, or confounding); or there is
63 strong human evidence of cancer or its
64 precursors, extensive animal evidence,
65 identification of key precursor events in
66 animals, and strong evidence that they
67 are anticipated to occur in humans.
68 Likely to be carcinogenic to humans: The
69 evidence demonstrates a potential
70 hazard to humans but does not meet the
71 criteria for carcinogenic. There may be a
72 plausible association in humans,
73 multiple positive results in animals, or a
74 combination of human, animal, or other
75 experimental evidence.
76 Suggestive evidence of carcinogenic
77 potential: The evidence raises concern
78 for effects in humans but is not sufficient
79 for a stronger conclusion. This
80 descriptor covers a range of evidence,
81 from a positive result in the only
82 available study to a single positive result
83 in an extensive database that includes
84 negative results in other species.
85 Inadequate information to assess
86 carcinogenic potential: No other
87 descriptors apply. Conflicting evidence
88 can be classified as inadequate
89 information if all positive results are
90 opposed by negative studies of equal
91 quality in the same sex and strain.
92 Differing results, however, can be
93 classified as suggestive evidence or as
94 likely to be carcinogenic.
This document is a draft for review purposes only and does not constitute Agency policy.
xxiv DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Wot likely to be carcinogenic to humans:
2 There is robust evidence for concluding
3 that there is no basis for concern. There
4 may be no effects in both sexes of at least
5 two appropriate animal species; positive
6 animal results and strong, consistent
7 evidence that each mode of action in
8 animals does not operate in humans; or
9 convincing evidence that effects are not
10 likely by a particular exposure route or
11 below a defined dose.
12 Multiple descriptors may be used if there
13 is evidence that carcinogenic effects differ by
14 dose range or exposure route [U.S. EPA.
15 2005a.32.5).
16 Another example of standard descriptors
17 comes from EPA's Integrated Science
18 Assessments, which evaluate causation for
19 the effects of the criteria pollutants in
20 ambient air [U.S. EPA. 2010b. 31.61.
21 Causal relationship: Sufficient evidence to
22 conclude that there is a causal
23 relationship. Observational studies
24 cannot be explained by plausible
25 alternatives, or they are supported by
26 other lines of evidence, for example,
27 animal studies or mechanistic
28 information.
29 Likely to be a causal relationship:
30 Sufficient evidence that a causal
31 relationship is likely, but important
32 uncertainties remain. For example,
33 observational studies show an
34 association but coexposures are difficult
35 to address or other lines of evidence are
36 limited or inconsistent; or multiple
37 animal studies from different
38 laboratories demonstrate effects and
39 there are limited or no human data.
40 Suggestive of a causal relationship: At
41 least one high-quality epidemiologic
42 study shows an association but other
43 studies are inconsistent.
44 Inadequate to infer a causal relationship:
45 The studies do not permit a conclusion
46 regarding the presence or absence of an
47 association.
48 Not likely to be a causal relationship:
49 Several adequate studies, covering the
50 full range of human exposure and
51 considering susceptible populations, are
52 mutually consistent in not showing an
53 effect at any level of exposure.
54 The EPA is investigating and may on a
55 trial basis use these or other standard
56 descriptors to characterize the overall
57 weight of the evidence for effects other than
58 cancer.
59 6. Selecting studies for derivation
60 of toxicity values
61
For each effect where there is credible
62 evidence of an association with the agent,
63 the assessment derives toxicity values if
64 there are suitable epidemiologic or
65 experimental data. The decision to derive
66 toxicity values may be linked to the hazard
67 descriptor.
68 Dose-response analysis requires
69 quantitative measures of dose and response.
70 Then, other factors being equal:
71 - Epidemiologic studies are preferred over
72 animal studies, if quantitative measures
73 of exposure are available and effects can
74 be attributed to the agent.
75 - Among experimental animal models,
76 those that respond most like humans are
77 preferred, if the comparability of
78 response can be determined.
79 - Studies by a route of human
80 environmental exposure are preferred,
81 although a validated toxicokinetic model
82 can be used to extrapolate across
83 exposure routes.
84 - Studies of longer exposure duration and
85 follow-up are preferred, to minimize
86 uncertainty about whether effects are
87 representative of lifetime exposure.
88 - Studies with multiple exposure levels are
89 preferred for their ability to provide
90 information about the shape of the
91 exposure-response curve.
This document is a draft for review purposes only and does not constitute Agency policy.
xxv DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 - Studies with adequate power to detect
2 effects at lower exposure levels are
3 preferred, to minimize the extent of
4 extrapolation to levels found in the
5 environment.
6 Studies with nonmonotonic exposure-
7 response relationships are not necessarily
8 excluded from the analysis. A diminished
9 effect at higher exposure levels may be
10 satisfactorily explained by factors such as
11 competing toxicity, saturation of absorption
12 or metabolism, exposure misclassification,
13 or selection bias.
14 If a large number of studies are suitable
15 for dose-response analysis, the assessment
16 considers the study characteristics in this
17 section to focus on the most informative
18 data. The assessment explains the reasons
19 for not analyzing other groups of studies. As
20 a check on the selection of studies for dose-
21 response analysis, the EPA asks peer
22 reviewers to identify studies that were not
23 adequately considered.
24 7. Deriving toxicity values
25 7.1. General framework for dose-
26 response analysis
27 The EPA uses a two-step approach that
28 distinguishes analysis of the observed dose-
29 response data from inferences about lower
30 doses fU.S. EPA. 2005a. 531
31 Within the observed range, the preferred
32 approach is to use modeling to incorporate a
33 wide range of data into the analysis. The
34 modeling yields a point of departure (an
35 exposure level near the lower end of the
36 observed range, without significant
37 extrapolation to lower doses; see Sections
38 7.2 and 7.3).
39 Extrapolation to lower doses considers
40 what is known about the modes of action for
41 each effect (see Sections 7.4 and 7.5). If
42 response estimates at lower doses are not
43 required, an alternative is to derive reference
44 values, which are calculated by applying
45 factors to the point of departure in order to
46 account for sources of uncertainty and
47 variability (see Section 7.6).
48 For a group of agents that induce an
49 effect through a common mode of action, the
50 dose-response analysis may derive a relative
51 potency factor for each agent. A full dose-
52 response analysis is conducted for one well-
53 studied index chemical in the group, then the
54 potencies of other members are expressed in
55 relative terms based on relative toxic effects,
56 relative absorption or metabolic rates,
57 quantitative structure-activity relationships,
58 or receptor binding characteristics (U.S. EPA.
59 2005a. 33.2.6: 2000b. 34.41.
60 Increasingly, EPA is basing toxicity
61 values on combined analyses of multiple
62 data sets or multiple responses. The EPA
63 also considers multiple dose-response
64 approaches if they can be supported by
65 robust data.
66 7.2. Modeling dose to sites of biologic
67 effects
68 The preferred approach for analysis of
69 dose is toxicokinetic modeling because of its
70 ability to incorporate a wide range of data.
71 The preferred dose metric would refer to the
72 active agent at the site of its biologic effect or
73 to a close, reliable surrogate measure. The
74 active agent may be the administered
75 chemical or a metabolite. Confidence in the
76 use of a toxicokinetic model depends on the
77 robustness of its validation process and on
78 the results of sensitivity analyses (U.S. EPA.
79 2006a: 2005a. §3.1: 1994. §4.3).
80 Because toxicokinetic modeling can
81 require many parameters and more data
82 than are typically available, the EPA has
83 developed standard approaches that can be
84 applied to typical data sets. These standard
85 approaches also facilitate comparison across
86 exposure patterns and species.
87 - Intermittent study exposures are
88 standardized to a daily average over the
89 duration of exposure. For chronic effects,
90 daily exposures are averaged over the
91 lifespan. Exposures during a critical
92 period, however, are not averaged over a
This document is a draft for review purposes only and does not constitute Agency policy.
xxvi DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 longer duration (U.S. EPA. 2005a. 53.1.1:
2 1991c.S3.21
3 - Doses are standardized to equivalent
4 human terms to facilitate comparison of
5 results from different species.
6 - Oral doses are scaled allometrically
7 using mg/kg3/4-d as the equivalent
8 dose metric across species.
9 Allometric scaling pertains to
10 equivalence across species, not
11 across lifestages, and is not used to
12 scale doses from adult humans or
13 mature animals to infants or children
14 fU.S. EPA. 2011: 2005a. 33.1.31.
15 - Inhalation exposures are scaled
16 using dosimetry models that apply
17 species-specific physiologic and
18 anatomic factors and consider
19 whether the effect occurs at the site
20 of first contact or after systemic
21 circulation [U.S. EPA. 2012a:
22 1994.331.
23 It can be informative to convert doses
24 across exposure routes. If this is done, the
25 assessment describes the underlying data,
26 algorithms, and assumptions [U.S. EPA.
27 2005a. 33.1.41.
28 In the absence of study-specific data on,
29 for example, intake rates or body weight, the
30 EPA has developed recommended values for
31 use in dose-response analysis [U.S. EPA,
32 19881
33 7.3. Modeling response in the range
34 of observation
35 Toxicodynamic ("biologically based")
36 modeling can incorporate data on biologic
37 processes leading to an effect. Such models
38 require sufficient data to ascertain a mode of
39 action and to quantitatively support model
40 parameters associated with its key events.
41 Because different models may provide
42 equivalent fits to the observed data but
43 diverge substantially at lower doses, critical
44 biologic parameters should be measured
45 from laboratory studies, not by model fitting.
46 Confidence in the use of a toxicodynamic
47 model depends on the robustness of its
48 validation process and on the results of
49 sensitivity analyses. Peer review of the
50 scientific basis and performance of a model
51 is essential [U.S. EPA. 2005a. 33.2.21.
52 Because toxicodynamic modeling can
53 require many parameters and more
54 knowledge and data than are typically
55 available, the EPA has developed a standard
56 set of empirical ("curve-fitting") models
57 (http://www.epa.gov/ncea/bmds/) that can
58 be applied to typical data sets, including
59 those that are nonlinear. The EPA has also
60 developed guidance on modeling dose-
61 response data, assessing model fit, selecting
62 suitable models, and reporting modeling
63 results (U.S. EPA. 2012c1. Additional
64 judgment or alternative analyses are used if
65 the procedure fails to yield reliable results,
66 for example, if the fit is poor, modeling may
67 be restricted to the lower doses, especially if
68 there is competing toxicity at higher doses
69 fU.S. EPA. 2005a. 33.2.31.
70 Modeling is used to derive a point of
71 departure fU.S. EPA. 2012c: 2005a. 33.2.41.
72 (See Section 7.6 for alternatives if a point of
73 departure cannot be derived by modeling.)
74 - If linear extrapolation is used, selection
75 of a response level corresponding to the
76 point of departure is not highly
77 influential, so standard values near the
78 low end of the observable range are
79 generally used (for example, 10% extra
80 risk for cancer bioassay data, 1% for
81 epidemiologic data, lower for rare
82 cancers).
83 - For nonlinear approaches, both
84 statistical and biologic considerations
85 are taken into account.
86
87
88
89
90
91
92
93
94
95
For dichotomous data, a response
level of 10% extra risk is generally
used for minimally adverse effects,
5% or lower for more severe effects.
For continuous data, a response level
is ideally based on an established
definition of biologic significance. In
the absence of such definition, one
control standard deviation from the
control mean is often used for
This document is a draft for review purposes only and does not constitute Agency policy.
xxvii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2
3
minimally adverse effects, one-half
standard deviation for more severe
effects.
4 The point of departure is the 95% lower
5 bound on the dose associated with the
6 selected response level.
7 7.4. Extrapolating to lower doses and
8 response levels
9 The purpose of extrapolating to lower
10 doses is to estimate responses at exposures
11 below the observed data. Low-dose
12 extrapolation, typically used for cancer data,
13 considers what is known about modes of
14 action [U.S. EPA. 2005a. §3.3.1 and §3.3.2).
15 1) If a biologically based model has been
16 developed and validated for the agent,
17 extrapolation may use the fitted model
18 below the observed range if significant
19 model uncertainty can be ruled out with
20 reasonable confidence.
21 2) Linear extrapolation is used if the dose-
22 response curve is expected to have a
23 linear component below the point of
24 departure. This includes:
25 - Agents or their metabolites that are
26 DNA-reactive and have direct
27 mutagenic activity.
28 - Agents or their metabolites for which
29 human exposures or body burdens
30 are near doses associated with key
31 events leading to an effect
32 Linear extrapolation is also used when
33 data are insufficient to establish mode of
34 action and when scientifically plausible.
35 The result of linear extrapolation is
36 described by an oral slope factor or an
37 inhalation unit risk, which is the slope of
38 the dose-response curve at lower doses
39 or concentrations, respectively.
40 3) Nonlinear models are used for
41 extrapolation if there are sufficient data
42 to ascertain the mode of action and to
43 conclude that it is not linear at lower
44 doses, and the agent does not
45 demonstrate mutagenic or other activity
46 consistent with linearity at lower doses.
47 Nonlinear approaches generally should
48 not be used in cases where mode of
49 action has not been ascertained. If
50 nonlinear extrapolation is appropriate
51 but no model is developed, an alternative
52 is to calculate reference values.
53 4) Both linear and nonlinear approaches
54 may be used if there are multiple modes
55 of action. For example, modeling to a low
56 response level can be useful for
57 estimating the response at doses where a
58 high-dose mode of action would be less
59 important
60 If linear extrapolation is used, the
61 assessment develops a candidate slope
62 factor or unit risk for each suitable data set.
63 These results are arrayed, using common
64 dose metrics, to show the distribution of
65 relative potency across various effects and
66 experimental systems. The assessment then
67 derives or selects an overall slope factor and
68 an overall unit risk for the agent, considering
69 the various dose-response analyses, the
70 study preferences discussed in Section 6,
71 and the possibility of basing a more robust
72 result on multiple data sets.
73 7.5. Considering susceptible
74 populations and lifestages
75 The assessment analyzes the available
76 information on populations and lifestages
77 that may be particularly susceptible to each
78 effect. A tiered approach is used [U.S. EPA.
79 2005a.53.5).
80 5) If an epidemiologic or experimental
81 study reports quantitative results for a
82 susceptible population or lifestage, these
83 data are analyzed to derive separate
84 toxicity values for susceptible
85 individuals.
86 6) If data on risk-related parameters allow
87 comparison of the general population
88 and susceptible individuals, these data
89 are used to adjust the general-population
90 toxicity values for application to
91 susceptible individuals.
This document is a draft for review purposes only and does not constitute Agency policy.
xxviii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 7) In the absence of chemical-specific data,
2 the EPA has developed age-dependent
3 adjustment factors for early-life exposure
4 to potential carcinogens that have a
5 mutagenic mode of action. There is
6 evidence of early-life susceptibility to
7 various carcinogenic agents, but most
8 epidemiologic studies and cancer
9 bioassays do not include early-life
10 exposure. To address the potential for
11 early-life susceptibility, the EPA
12 recommends [U.S. EPA. 2005b. §5):
13 - 10-fold adjustment for exposures
14 before age 2 years.
15 - 3-fold adjustment for exposures
16 between ages 2 and 16 years.
17 7.6. Reference values and uncertainty
18 factors
19 An oral reference dose or an inhalation
20 reference concentration is an estimate of an
21 exposure (including in susceptible
22 subgroups) that is likely to be without an
23 appreciable risk of adverse health effects
24 over a lifetime fU.S. EPA. 2002.34.21.
25 Reference values are typically calculated for
26 effects other than cancer and for suspected
27 carcinogens if a well characterized mode of
28 action indicates that a necessary key event
29 does not occur below a specific dose.
30 Reference values provide no information
31 about risks at higher exposure levels.
32 The assessment characterizes effects
33 that form the basis for reference values as
34 adverse, considered to be adverse, or a
35 precursor to an adverse effect. For
36 developmental toxicity, reproductive
37 toxicity, and neurotoxicity there is guidance
38 on adverse effects and their biologic markers
39 fU.S. EPA. 1998.1996.1991c1.
40 To account for uncertainty and
41 variability in the derivation of a lifetime
42 human exposure where adverse effects are
43 not anticipated to occur, reference values are
44 calculated by applying a series of uncertainty
45 factors to the point of departure. If a point of
46 departure cannot be derived by modeling, a
47 no-observed-adverse-effect level or a
48 lowest-observed-adverse-effect level is used
49 instead. The assessment discusses scientific
50 considerations involving several areas of
51 variability or uncertainty.
52 Human variation: The assessment accounts
53 for variation in susceptibility across the
54 human population and the possibility
55 that the available data may not be
56 representative of individuals who are
57 most susceptible to the effect. A factor of
58 10 is generally used to account for this
59 variation. This factor is reduced only if
60 the point of departure is derived or
61 adjusted specifically for susceptible
62 individuals (not for a general population
63 that includes both susceptible and non-
64 susceptible individuals) (U.S. EPA,
65 2002. §4.4.5: 1998. §4.2: 1996. §4:
66 1994. §4.3.9.1: 1991c. 33.41.
67 Animal-to-human extrapolation: If animal
68 results are used to make inferences
69 about humans, the assessment adjusts
70 for cross-species differences. These may
71 arise from differences in toxicokinetics
72 or toxicodynamics. Accordingly, if the
73 point of departure is standardized to
74 equivalent human terms or is based on
75 toxicokinetic or dosimetry modeling, a
76 factor of 101/2 (rounded to 3) is applied
77 to account for the remaining uncertainty
78 involving toxicokinetic and
79 toxicodynamic differences. If a
80 biologically based model adjusts fully for
81 toxicokinetic and toxicodynamic
82 differences across species, this factor is
83 not used. In most other cases, a factor of
84 10 is applied fU.S. EPA. 2011:
85 2002. §4.4.5: 1998. §4.2: 1996. §4:
86 1994. §4.3.9.1: 1991c. §3.4).
87 Adverse-effect level to no-observed-
88 adverse-effect level: If a point of
89 departure is based on a lowest-
90 observed-adverse-effect level, the
91 assessment must infer a dose where
92 such effects are not expected. This can be
93 a matter of great uncertainty, especially
94 if there is no evidence available at lower
95 doses. A factor of 10 is applied to
96 account for the uncertainty in making
This document is a draft for review purposes only and does not constitute Agency policy.
xxix DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 this inference. A factor other than 10
2 may be used, depending on the
3 magnitude and nature of the response
4 and the shape of the dose-response
5 curve fU.S. EPA. 2002. §4.4.5: 1998. §4.2:
6 1996. §4: 1994. §4.3.9.1: 1991c.§3.41
7 Subchronic-to-chronic exposure: If a point
8 of departure is based on subchronic
9 studies, the assessment considers
10 whether lifetime exposure could have
11 effects at lower levels of exposure. A
12 factor of 10 is applied to account for the
13 uncertainty in using subchronic studies
14 to make inferences about lifetime
15 exposure. This factor may also be
16 applied for developmental or
17 reproductive effects if exposure covered
18 less than the full critical period. A factor
19 other than 10 may be used, depending
20 on the duration of the studies and the
21 nature of the response (U.S. EPA. 2002.
22 §4.4.5: 1998. §4.2: 1994. §4.3.9.11
23 Incomplete database: If an incomplete
24 database raises concern that further
25 studies might identify a more sensitive
26 effect, organ system, or lifestage, the
27 assessment may apply a database
28 uncertainty factor [U.S. EPA.
29 2002. §4.4.5: 1998. §4.2: 1996. §4:
30 1994. §4.3.9.1: 1991c. §3.4). The size of
31 the factor depends on the nature of the
32 database deficiency. For example, the
33 EPA typically follows the suggestion that
34 a factor of 10 be applied if both a
35 prenatal toxicity study and a two-
36 generation reproduction study are
37 missing and a factor of 101/2 if either is
38 missing [U.S. EPA. 2002. §4.4.5).
39 In this way, the assessment derives
40 candidate values for each suitable data set
41 and effect that is credibly associated with the
42 agent These results are arrayed, using
43 common dose metrics, to show where effects
44 occur across a range of exposures [U.S. EPA.
45 1994. §4.3.9).
46 The assessment derives or selects an
47 organ- or system-specific reference value for
48 each organ or system affected by the agent.
49 The assessment explains the rationale for
50 each organ/system-specific reference value
51 (based on, for example, the highest quality
52 studies, the most sensitive outcome, or a
53 clustering of values). By providing these
54 organ/system-specific reference values, IRIS
55 assessments facilitate subsequent
56 cumulative risk assessments that consider
57 the combined effect of multiple agents acting
58 at a common site or through common
59 mechanisms [NRG. 2009).
60 The assessment then selects an overall
61 reference dose and an overall reference
62 concentration for the agent to represent
63 lifetime human exposure levels where
64 effects are not anticipated to occur. This is
65 generally the most sensitive organ/system-
66 specific reference value, though
67 consideration of study quality and
68 confidence in each value may lead to a
69 different selection.
70 7.7. Confidence and uncertainty in the
71 reference values
72 The assessment selects a standard
73 descriptor to characterize the level of
74 confidence in each reference value, based on
75 the likelihood that the value would change
76 with further testing. Confidence in reference
77 values is based on quality of the studies used
78 and completeness of the database, with more
79 weight given to the latter. The level of
80 confidence is increased for reference values
81 based on human data supported by animal
82 data [U.S. EPA. 1994. §4.3.9.2).
83 High confidence: The reference value is not
84 likely to change with further testing,
85 except for mechanistic studies that might
86 affect the interpretation of prior test
87 results.
88 Medium confidence: This is a matter of
89 judgment, between high and low
90 confidence.
91 Low confidence: The reference value is
92 especially vulnerable to change with
93 further testing.
This document is a draft for review purposes only and does not constitute Agency policy.
xxx DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 These criteria are consistent with
2 guidelines for systematic reviews that
3 evaluate the quality of evidence. These also
4 focus on whether further research would be
5 likely to change confidence in the estimate of
6 effect [Guyattetal.. 2008b).
7 All assessments discuss the significant
8 uncertainties encountered in the analysis.
9 The EPA provides guidance on
10 characterization of uncertainty [U.S. EPA.
11 2005a. §3.6]. For example, the discussion
12 distinguishes model uncertainty (lack of
13 knowledge about the most appropriate
14 experimental or analytic model) and
15 parameter uncertainty (lack of knowledge
16 about the parameters of a model).
17 Assessments also discuss human variation
18 (interpersonal differences in biologic
19 susceptibility or in exposures that modify
20 the effects of the agent).
21
22
References
23 CDC (Centers for Disease Control and
24 Prevention). (2004). The health
25 consequences of smoking: A report
26 of the Surgeon General. Washington,
27 DC: U.S. Department of Health and
28 Human Services.
29 http://www.surgeongeneral.gov/lib
30 rary/smokingconsequences/
31 Guyatt. GH: Oxman. AD: Kunz. R: Vist. GE:
32 Falck-Ytter. Y: Schiinemann. HI.
33 (2008a). GRADE: What is "quality of
34 evidence" and why is it important to
35 clinicians? [Review]. BMJ 336: 995-
36 998.
37 http://dx.doi.org/10.1136/bmi.3949
38 0.551019.BE
39 Guyatt. GH: Oxman. AD: Vist. GE: Kunz. R:
40 Falck-Ytter. Y: Alonso-Coello. P:
41 Schiinemann. HI. (2008b). GRADE:
42 An emerging consensus on rating
43 quality of evidence and strength of
44 recommendations. BMJ 336: 924-
45 926.
46 http://dx.doi.org/10.1136/bmi.3948
47 9.470347.AD
48 HEW (U.S. Department of Health, Education
49 and Welfare). (1964). Smoking and
50 health: Report of the advisory
51 committee to the surgeon general of
52 the public health service.
53 Washington, DC: U.S. Department of
54 Health, Education, and Welfare.
55 http://profiles.nlm.nih.gov/ps/retrie
56 ve/ResourceMetadata/NNBBMO
57 Hill. AB. (1965). The environment and
58 disease: Association or causation?
59 ProcRSocMed 58: 295-300.
60 IARC (International Agency for Research on
61 Cancer). (2006). Preamble to the
62 IARC monographs. Lyon, France.
63 http://monographs.iarc.fr/ENG/Pre
64 amble/
65 IOM (Institute of Medicine). (2008).
66 Improving the presumptive disability
67 decision-making process for
68 veterans. In JM Samet; CC Bodurow
69 (Eds.). Washington, DC: National
70 Academies Press.
71 http://www.nap.edu/openbook.php
72 ?record id=11908
73 NRC (National Research Council). (1983).
74 Risk assessment in the federal
75 government: Managing the process.
76 Washington, DC: National Academies
77 Press.
78 http://www.nap.edu/openbook.php
79 ?record id=366&page=Rl
80 NRC (National Research Council). (2009).
81 Science and decisions: Advancing
82 risk assessment. Washington, DC:
83 National Academies Press.
84 http://www.nap.edu/catalog/12209
85 .html
86 Rothman. KT: Greenland. S. (1998). Modern
87 epidemiology (2nd ed.). Philadelphia,
88 PA: Lippincott, Williams, & Wilkins.
89 U.S. EPA (U.S. Environmental Protection
90 Agency). (1986a). Guidelines for
91 mutagenicity risk assessment
92 (EPA/630/R-98/003). Washington,
93 DC: U.S. Environmental Protection
94 Agency, Risk Assessment Forum.
95 http://www.epa.gov/iris/backgrd.ht
96 ml
This document is a draft for review purposes only and does not constitute Agency policy.
xxxi DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
U.S. EPA (U.S. Environmental Protection
Agency). (1986b). Guidelines for the
health risk assessment of chemical
mixtures. (EPA/630/R-98/002).
Washington, DC: U.S. Environmental
Protection Agency, Risk Assessment
Forum.
http://cfpub.epa.gov/ncea/cfm/reco
rdisplay.cfm?deid=22567
U.S. EPA (U.S. Environmental Protection
U.S.
Agency). (1986c). Guidelines for the
health risk assessment of chemical
mixtures. Fed Reg 51: 34014-34025.
EPA (U.S. Environmental Protection
Agency). (1988). Recommendations
for and documentation of biological
values for use in risk assessment.
(EPA/600/6-87/008). Cincinnati,
OH: U.S. Environmental Protection
Agency, National Center for
Environmental Assessment
http://cfpub.epa.gov/ncea/cfm/reco
rdisplay.cfm?deid=34855
U.S. EPA (U.S. Environmental Protection
Agency). (199 Ic). Guidelines for
developmental toxicity risk
assessment. (EPA/600/FR-91/001).
Washington, DC: U.S. Environmental
Protection Agency, Risk Assessment
Forum.
http://www.epa.gov/raf/publication
s/guidelines-dev-toxicity-risk-
assessmenthtm
U.S. EPA (U.S. Environmental Protection
Agency). (1994). Methods for
derivation of inhalation reference
concentrations and application of
inhalation dosimetry. (EPA/600/8-
90/066F). Research Triangle Park,
NC: U.S. Environmental Protection
Agency, Environmental Criteria and
Assessment Office.
http://cfpub.epa.gov/ncea/cfm/reco
rdisplay.cfm?deid=71993
U.S. EPA (U.S. Environmental Protection
Agency). (1996). Guidelines for
reproductive toxicity risk
assessment. (EPA/630/R-96/009).
Washington, DC: U.S. Environmental
Protection Agency, Risk Assessment
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
U.S.
Forum.
http://www.epa.gov/raf/publication
s/pdfs/REPR051.PDF
EPA (U.S. Environmental Protection
Agency). (1998). Guidelines for ne
urotoxicity risk assessment
(EPA/630/R-95/001F). Washington,
DC: U.S. Environmental Protection
Agency, Risk Assessment Forum.
http://www.epa.gov/raf/publication
s/pdfs/NEUROTOX.PDF
U.S. EPA (U.S. Environmental Protection
Agency). (2000b). Supplementary
guidance for conducting health risk
assessment of chemical mixtures.
(EPA/630/R-00/002). Washington,
DC: U.S. Environmental Protection
Agency, Risk Assessment Forum.
http://cfpub.epa.gov/ncea/cfm/reco
rdisplay.cfm?deid=20533
U.S. EPA (U.S. Environmental Protection
U.S.
U.S.
Agency). (2002). A review of the
reference dose and reference
concentration processes.
(EPA/630/P-02/002F). Washington,
DC: U.S. Environmental Protection
Agency, Risk Assessment Forum.
http://cfpub.epa.gov/ncea/cfm/reco
rdisplay.cfm?deid=51717
EPA (U.S. Environmental Protection
Agency). (2005a). Guidelines for
carcinogen risk assessment
(EPA/630/P-03/001F). Washington,
DC: U.S. Environmental Protection
Agency, Risk Assessment Forum.
http: / /www. epa. gov/cancer guidelin
eg
EPA
(U.S. Environmental Protection
Agency). (2005b). Supplemental
guidance for assessing susceptibility
from early-life exposure to
carcinogens. In US Environmental
Protection Agency, Risk Assessment
Forum (pp. 1125-1133).
(EPA/630/R-03/003F). Washington,
DC: U.S. Environmental Protection
Agency, Risk Assessment Forum.
http://www.epa.gov/cancerguidelin
es/guidelines-carcinogen-
supplementhtm
This document is a draft for review purposes only and does not constitute Agency policy.
xxxii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
U.S. EPA (U.S. Environmental Protection
Agency). (2006a). Approaches for the
application of physiologically based
pharmacokinetic (PBPK) models and
supporting data in risk assessment
(Final Report) [EPA Report].
(EPA/600/R-05/043F). Washington,
DC: U.S. Environmental Protection
Agency, National Center for
Environmental assessment
http://cfpub.epa.gov/ncea/cfm/reco
rdisplay.cfm?deid=157668
U.S. EPA (U.S. Environmental Protection
U.S.
U.S.
U.S.
Agency). (2006b). A framework for
assessing health risk of
environmental exposures to
children. (EPA/600/R-05/093F).
Washington, DC: U.S. Environmental
Protection Agency, National Center
for Environmental Assessment
http://cfpub.epa.gov/ncea/cfm/reco
rdisplay.cfm?deid=158363
EPA (U.S. Environmental Protection
Agency). (2010b). Integrated science
assessment for carbon monoxide
[EPA Report]. (EPA/600/R-
09/019F). Research Triangle Park,
NC.
http://cfpub.epa.gov/ncea/cfm/reco
rdisplay.cfm?deid=218686
EPA (U.S. Environmental Protection
Agency). (2011). Recommended use
of body weight 3/4 as the default
method in derivation of the oral
reference dose.
(EPA/100/R11/0001). Washington,
DC: U.S. Environmental Protection
Agency, Risk Assessment Forum.
http://www.epa.gov/raf/publication
s/interspecies-extrapolation.htm
EPA (U.S. Environmental Protection
Agency). (2012a). Advances in
inhalation gas dosimetry for
derivation of a reference
concentration (RFC) and use in risk
assessment. (EPA/600/R-12/044).
Washington, DC.
http://cfpub.epa.gov/ncea/cfm/reco
rdisplay.cfm?deid=244650
50 U
51
52
53
54
55
56
57
58 U
59
60
61
62
63
S. EPA (U.S. Environmental Protection
S.
Agency). (2012b). Benchmark dose
technical guidance. (EPA/100/R-
12/001). Washington, DC: Risk
Assessment Forum.
http://www.epa.gov/raf/publication
s/pdfs/benchmark dose guidance.p
df
EPA (U.S. Environmental Protection
Agency). (2014). EPAs Integrated
Risk Information System:
Assessment development process
[EPA Report]. Washington, DC.
http://epa.gov/iris/process.htm
64
65 August 2013
This document is a draft for review purposes only and does not constitute Agency policy.
xxxiii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2 EXECUTIVE SUMMARY
3 Occurrence and Health Effects
4 Benzo[a]pyrene is a five-ring polycyclic aromatic hydrocarbon (PAH).
5 Benzo[a]pyrene (along with other PAHs) is released into the atmosphere as a
6 component of smoke from forest fires, industrial processes, vehicle exhaust,
7 cigarettes, and through the burning of fuel (such as wood, coal, and petroleum
8 products). Oral exposure to benzo[a]pyrene can occur by eating certain food
9 products, such as charred meats, where benzo[a]pyrene is formed during the
10 cooking process or by eating foods grown in areas contaminated with
11 benzo[a]pyrene (from the air and soil). Dermal exposure may occur from contact
12 with soils or materials that contain soot, tar, or crude petroleum products or by
13 using certain pharmaceutical products containing coal tars, such as those used to
14 treat the skin conditions, eczema and psoriasis. The magnitude of human exposure
15 to benzo[a]pyrene and other PAHs depends on factors such as lifestyle (e.g., diet,
16 tobacco smoking), occupation, and living conditions (e.g., urban versus rural setting,
17 domestic heating, and cooking methods).
18 Animal studies demonstrate that exposure to benzo[a]pyrene may be
19 associated with developmental, reproductive, and immunological effects. In
20 addition, epidemiology studies involving exposure to PAH mixtures have reported
21 associations between internal biomarkers of exposure to benzo[a]pyrene
22 (benzo[a]pyrene diol epoxide-DNA adducts) and adverse birth outcomes (including
23 reduced birth weight, postnatal body weight, and head circumference) and
24 decreased fertility.
25 Studies in multiple animal species demonstrate that benzo[a]pyrene is
26 carcinogenic at multiple tumor sites (alimentary tract, liver, kidney, respiratory
27 tract, pharynx, and skin) by all routes of exposure. In addition, there is strong
28 evidence of carcinogenicity in occupations involving exposure to PAH mixtures
29 containing benzo[a]pyrene, such as aluminum production, chimney sweeping, coal
30 gasification, coal-tar distillation, coke production, iron and steel founding, and
31 paving and roofing with coal tar pitch. An increasing number of occupational
32 studies demonstrate a positive exposure-response relationship with cumulative
33 benzo[a]pyrene exposure and lung cancer.
34
35 Effects Other Than Cancer Observed Following Oral Exposure
36 In animals, oral exposure to benzo[a]pyrene has been shown to result in developmental
37 toxicity, reproductive toxicity, and immunotoxicity. Developmental effects in rats and mice include
38 neurobehavioral changes and cardiovascular effects following gestational exposures. Reproductive
39 and immune effects include decreased sperm counts, ovary weight, and follicle numbers, and
40 decreased immunoglobulin and B-cell numbers and thymus weight following oral exposures in
41 adult animals. In humans, benzo[a]pyrene exposure occurs in conjunction with other PAHs and, as
This document is a draft for review purposes only and does not constitute Agency policy.
xxxiv DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 such, attributing the observed effects to benzo[a]pyrene is complicated. However, human studies
2 report associations between particular health endpoints and internal measures of exposure, such as
3 benzo[a]pyrene-deoxyribonucleic acid (DNA) adducts, or external measures of benzo[a]pyrene
4 exposure. Overall, the human studies report developmental and reproductive effects that are
5 generally analogous to those observed in animals, and provide qualitative, supportive evidence for
6 hazards associated with benzo[a]pyrene exposure.
7 Oral Reference Dose (RfD) for Effects Other Than Cancer
8 Organ- or system-specific RfDs were derived for hazards associated with benzo[a]pyrene
9 exposure where data were amenable (see Table ES-1). These organ- or system-specific reference
10 values may be useful for subsequent cumulative risk assessments that consider the combined effect
11 of multiple agents acting at a common site.
12 Developmental toxicity, represented by neurobehavioral changes following neonatal
13 exposure, was chosen as the basis for the proposed overall oral RfD as the available data indicate
14 that neurobehavioral changes represent the most sensitive hazard of benzo[a]pyrene exposure.
15 The neurodevelopmental study by Chen etal. [2012] was used to derive the RfD. The endpoint of
16 altered anxiety-like behavior, as measured in the elevated plus maze, was selected as the critical
17 effect due to the sensitivity of this endpoint and the observed dose-response relationship of effects
18 across dose groups. Benchmark dose (BMD) modeling was utilized to derive the BMDLiso of
19 0.09 mg/kg-day that was used as the point of departure (POD) for RfD derivation.
20 The proposed overall RfD was calculated by dividing the POD for altered anxiety-like
21 behavior as measured in the elevated plus maze by a composite uncertainty factor (UF) of 300 to
22 account for the extrapolation from animals to humans (10), for interindividual differences in
23 human susceptibility (10), and for deficiencies in the toxicity database (3).
24 Table ES-1. Organ/system-specific RfDs and proposed overall RfD for
25 benzo[a]pyrene
Effect
Developmental
Reproductive
Immunological
Proposed Overall RfD
Basis
Neurobehavioral changes
Gavage neurodevelopmental study in rats (postnatal days
[PNDs] 5-11)
Chen etal. (2012)
Decreased ovary weight
Gavage subchronic (60 d) reproductive toxicity study in rats
Xu etal. (2010)
Decreased thymus weight and serum IgM
Gavage subchronic (35 d) study in rats
De Jong etal. (1999)
Developmental toxicity
RfD
(mg/kg-d)
3 x ID'4
4 x 10"4
2 x 10"3
3 x ID'4
Confidence
Medium
Medium
Low
Medium
This document is a draft for review purposes only and does not constitute Agency policy.
xxxv DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Confidence in the Overall Oral RfD
2 The overall confidence in the RfD is medium. Confidence in the principal study [Chenetal..
3 2012] is medium-to-high. The design, conduct, and reporting of this neurodevelopmental study
4 was good and a wide variety of neurotoxicity endpoints were measured. Some informative
5 experimental details were, however, omitted including the sensitivity of some assays at the
6 indicated developmental ages and lack of reporting gender-specific data for all outcomes. Several
7 subchronic and developmental studies covering a wide variety of endpoints are also available;
8 however, the lack of a multigeneration toxicity study with exposure throughout development is not
9 available. Therefore, confidence in the database is medium.
10 Effects Other Than Cancer Observed Following Inhalation Exposure
11 In animals, inhalation exposure to benzo[a]pyrene has been shown to result in
12 developmental and reproductive toxicity. Studies in rats following inhalation exposure show
13 decreased fetal survival and brain effects in offspring, and decreased testes weight and sperm
14 counts in adult animals. Overall, the available human PAH mixtures studies report developmental
15 and reproductive effects that are generally analogous to those observed in animals, and provide
16 qualitative, supportive evidence for the hazards associated with benzo[a]pyrene exposure.
17 Inhalation Reference Concentration (RfC) for Effects Other Than Cancer
18 An attempt was made to derive organ- or system-specific RfCs for hazards associated with
19 benzo[a]pyrene exposure where data were amenable (see Table ES-2). These organ- or system-
20 specific reference values may be useful for subsequent cumulative risk assessments that consider
21 the combined effect of multiple agents acting at a common site.
22 Developmental toxicity, represented by decreased fetal survival, was chosen as the basis for
23 the proposed inhalation RfC as the available data indicate that developmental effects represent a
24 sensitive hazard of benzo[a]pyrene exposure. The developmental inhalation study in rats by
25 Archibong et al. [2002] and the observed decreased fetal survival following exposure to
26 benzo[a]pyrene on gestation days (CDs] 11-20 were used to derive the overall RfC. The lowest-
27 observed-adverse-effect level (LOAEL] of 25 |J.g/m3 based on decreased fetal survival was selected
28 as the POD. The LOAEL was adjusted to account for the discontinuous daily exposure to derive the
29 PODADj and the human equivalent concentration (HEC] was calculated from the PODADj by
30 multiplying by the regional deposited dose ratio (RDDRER] for extrarespiratory (i.e., systemic]
31 effects, as described in Methods/or Derivation of Inhalation Reference Concentrations and
32 Application of Inhalation Dosimetry (U.S. EPA. 1994]. These adjustments resulted in a PODnEc of
33 4.6 [ig/m3, which was used as the POD for RfC derivation.
34 The RfC was calculated by dividing the POD by a composite UF of 3,000 to account for
35 toxicodynamic differences between animals and humans (3], interindividual differences in human
36 susceptibility (10], LOAEL-to-no-observed-adverse-effect level (NOAEL] extrapolation (10], and
37 deficiencies in the toxicity database (10].
This document is a draft for review purposes only and does not constitute Agency policy.
xxxvi DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2
Table ES-2. Organ/system-specific RfCs and proposed overall RfC for
benzo[a]pyrene
Effect
Developmental
Reproductive
Proposed Overall
RfC
Basis
Decreased fetal survival
Developmental toxicity study in rats (GDs 11-20)
Archibongetal. (2002)
Reductions in testes weight and sperm parameters
Subchronic (60 d) reproductive toxicity study in rats
(Archibong et al. (2008); Ramesh et al. (2008))
Developmental toxicity
RfC(mg/m3)
2 x 10"6
Not
calculated3
2 x 10"6
Confidence
Low-medium
NA
Low-medium
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
aNot calculated due to UF >3,000.
Confidence in the Overall Inhalation RfC
The overall confidence in the RfC is low-to-medium. Confidence in the principal study
[Archibong et al., 2002] is medium. The conduct and reporting of this developmental inhalation
study were adequate; however, a NOAEL was not identified. Confidence in the database is low due
to the lack of a multigeneration toxicity study and the lack of information on varied toxicity
endpoints following subchronic and chronic inhalation exposure. However, confidence in the RfC is
bolstered by consistent systemic effects observed by the oral route (including reproductive and
developmental effects) and similar effects observed in human populations exposed to PAH
mixtures.
Evidence for Human Carcinogenicity
Under EPA's Guidelines for Carcinogen Risk Assessment [U.S. EPA. 2005a). benzo[a]pyrene is
"carcinogenic to humans" based on strong and consistent evidence in animals and humans. The
evidence includes an extensive number of studies demonstrating carcinogenicity in multiple animal
species exposed via all routes of administration and increased cancer risks, particularly in the lung
and skin, in humans exposed to different PAH mixtures containing benzo[a]pyrene. Mechanistic
studies provide strong supporting evidence that links the metabolism of benzo[a]pyrene to DNA-
reactive agents with key mutational events in genes that can lead to tumor development These
events include formation of specific DNA adducts and characteristic mutations in oncogenes and
tumor suppressor genes that have been observed in humans exposed to PAH mixtures. This
combination of human, animal, and mechanistic evidence provides the basis for characterizing
benzo[a]pyrene as "carcinogenic to humans."
26 Quantitative Estimate of Carcinogenic Risk From Oral Exposure
27 Lifetime oral exposure to benzo[a]pyrene has been associated with forestomach, liver, oral
28 cavity, jejunum or duodenum, and auditory canal tumors in male and female Wistar rats,
This document is a draft for review purposes only and does not constitute Agency policy.
xxxvii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 forestomach tumors in male and female Sprague-Dawley rats, and forestomach, esophagus, tongue,
2 and larynx tumors in female B6C3Fi mice (male mice were not tested). Less-than-lifetime oral
3 exposure to benzo[a]pyrene has also been associated with forestomach tumors in more than
4 10 additional bioassays with several strains of mice. The Kroese etal. [2001] and Beland and Gulp
5 [1998] studies were selected as the best available studies for dose-response analysis and
6 extrapolation to lifetime cancer risk following oral exposure to benzo[a]pyrene. These studies
7 included histological examinations for tumors in many different tissues, contained three exposure
8 levels and controls, contained adequate numbers of animals per dose group (~50/sex/group],
9 treated animals for up to 2 years, and included detailed reporting methods and results (including
10 individual animal data].
11 Time-weighted, average daily doses were converted to human equivalent doses (HEDs] on
12 the basis of (body weight]3/4 scaling [U.S. EPA. 1992]. EPA then used the multistage-Weibull model
13 for the derivation of the oral slope factor. This model was used because it incorporates the time at
14 which death-with-tumor occurred and can account for differences in mortality observed between
15 the exposure groups. Using linear extrapolation from the BMDLio, human equivalent oral slope
16 factors were derived for each gender/tumor site combination (slope factor = 0.1/BMDLi0] reported
17 by Kroese etal. [2001] and Beland and Gulp [1998]. The oral slope factor of 1 per mg/kg-day
18 based on the tumor response in the alimentary tract (forestomach, esophagus, tongue, and larynx]
19 of female B6C3Fi mice [Beland and Gulp. 1998] was selected as the factor with the highest value
20 (most sensitive] among a range of slope factors derived.
21 Quantitative Estimate of Carcinogenic Risk From Inhalation Exposure
22 Inhalation exposure to benzo[a]pyrene has been associated with squamous cell neoplasia in
23 the larynx, pharynx, trachea, nasal cavity, esophagus, and forestomach of male Syrian golden
24 hamsters exposed for up to 130 weeks to benzo[a]pyrene condensed onto NaCl particles [Thyssen
25 etal.. 1981]. Supportive evidence for the carcinogenicity of inhaled benzo[a]pyrene comes from
26 additional studies with hamsters exposed to benzo[a]pyrene via intratracheal instillation. The
27 Thyssen et al. [1981] bioassay represents the only study of lifetime exposure to inhaled
28 benzo[a]pyrene.
29 A time-to-tumor dose-response model was fit to the time-weighted average [TWA]
30 continuous exposure concentrations and the individual animal incidence data for the overall
31 incidence of tumors in the upper respiratory tract or pharynx. The inhalation unit risk of
32 6 x lO-4 per ng/m3 was calculated by linear extrapolation (slope factor = 0.1/BMCLio] from a
33 BMCLio of 0.16 mg/m3for the occurrence of upper respiratory and upper digestive tract tumors in
34 male hamsters chronically exposed by inhalation to benzo[a]pyrene [Thyssen et al.. 1981].
35 Quantitative Estimate of Carcinogenic Risk From Dermal Exposure
36 Skin cancer in humans has been documented to result from occupational exposure to
37 complex mixtures of PAHs including benzo[a]pyrene, such as coal tar, coal tar pitches, unrefined
This document is a draft for review purposes only and does not constitute Agency policy.
xxxviii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 mineral oils, shale oils, and soot In animal models, numerous dermal bioassays have demonstrated
2 an increased incidence of skin tumors with increasing dermal exposure of benzo[a]pyrene in all
3 species tested (mice, rabbits, rats, and guinea pigs), although most benzo[a]pyrene bioassays have
4 been conducted in mice. Due to the evidence supporting a hazard from exposure to benzo[a]pyrene
5 by the dermal route (see Section 1.1.5) and the availability of quantitative information, a cancer
6 slope factor for the dermal route was developed. The analysis in this assessment focuses on
7 lifetime carcinogenicity bioassays in several strains of mice demonstrating increasing incidence of
8 benign and malignant skin tumors following repeated dermal exposure to benzo[a]pyrene.
9 The National Institute for Occupational Safety and Health (NIOSH) study (Sivaketal.. 1997:
10 NIOSH, 1989) was selected as the best available study for dose-response analysis and extrapolation
11 to lifetime cancer risk following dermal exposure to benzo[a]pyrene. This study used three
12 exposure levels that highlighted the low-dose region and reported a number of attributes not
13 available for the older studies, including single housing of mice, blinded assessment of tumor status,
14 and time of appearance of tumors for each animal.
15 This mouse skin tumor incidence data was modeled using the multistage-Weibull model.
16 The resulting BMDLio was adjusted for interspecies differences by allometric scaling. The dermal
17 slope factor of 0.006 per ng/day was calculated by linear extrapolation (slope factor =
18 0.1/BMDLio-HEo) from the human equivalent POD for the occurrence of skin tumors in male mice
19 exposed dermally to benzo[a]pyrene for 104 weeks. As this slope factor has been developed for a
20 local effect, it is not intended to estimate systemic risk of cancer following dermal absorption of
21 benzo[a]pyrene into the systemic circulation.
22 Susceptible Populations and Lifestages
23 Benzo[a]pyrene has been determined to be carcinogenic by a mutagenic mode of action in
24 this assessment According to the Supplemental Guidance for Assessing Susceptibility from Early Life
25 Exposure to Carcinogens (U.S. EPA. 2005b). individuals exposed during early life to carcinogens with
26 a mutagenic mode of action are assumed to have an increased risk for cancer. The oral slope factor
27 of 1 per mg/kg-day, inhalation unit risk of 0.0006 per [ig/m3, and dermal slope factor of 0.006 per
28 Mg/day for benzo[a]pyrene, calculated from data applicable to adult exposures, do not reflect
29 presumed early life susceptibility to this chemical. Although some chemical-specific data exist for
30 benzo[a]pyrene that demonstrate increased early life susceptibility to cancer, these data were not
31 considered sufficient to develop separate risk estimates for childhood exposure. In the absence of
32 adequate chemical-specific data to evaluate differences in age-specific susceptibility, the
33 Supplemental Guidance (U.S. EPA. 2005b) recommends that age-dependent adjustment factors
34 (ADAFs) be applied in estimating cancer risk. The ADAFs are 10- and 3-fold adjustments that are
35 combined with age specific exposure estimates when estimating cancer risks from early life
36 (<16 years of age) exposures to benzo[a]pyrene.
37 Regarding effects other than cancer, there are epidemiological studies that report
38 associations between developmental effects (decreased postnatal growth, decreased head
This document is a draft for review purposes only and does not constitute Agency policy.
xxxix DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 circumference, and neurodevelopmental delays), reproductive effects and internal biomarkers of
2 exposure to benzo[a]pyrene. Studies in animals also indicate alterations in neurological
3 development and heightened susceptibility to reproductive effects following gestational or early
4 postnatal exposure to benzo[a]pyrene.
5 Key Issues Addressed in Assessment
6 The overall RfD and RfC were developed based on effects observed following exposure to
7 benzo[a]pyrene during a critical window of development The derivation of a general population
8 toxicity value based on exposure during development has implications regarding the evaluation of
9 populations exposed outside of the developmental period and the averaging of exposure to
10 durations outside of the critical window of susceptibility. Discussion of these considerations is
11 provided in Sections 2.1.5 and 2.2.5.
12 The dermal slope factor was developed based on data in animals. Because there is no
13 established methodology for extrapolating dermal toxicity from animals to humans, several
14 alternative approaches were evaluated (see Appendix D in Supplemental Information). Allometric
15 scaling using body weight to the % power was selected based on known species differences in
16 dermal metabolism and penetration of benzo[a]pyrene.
17
This document is a draft for review purposes only and does not constitute Agency policy.
xl DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
LITERATURE SEARCH STRATEGY| STUDY SELECTION
3 The literature search strategy used to identify primary, peer-reviewed literature pertaining
4 to benzo[a]pyrene was conducted using the databases listed in Table LS-1 (see Appendix C for the
5 complete list of keywords used). References from previous assessments by the U.S. Environmental
6 Protection Agency (EPA) and other national and international health organizations were also
7 examined. EPA conducted a comprehensive, systematic literature search for benzo[a]pyrene
8 through February, 2012. In addition, a search of the online database PubMed was conducted for the
9 timeframe January 2012 through August 2014, to ensure inclusion of critical studies published
10 since the initial literature search.
11
Table LS-1. Summary of the search strategy employed for benzo[a]pyrene
Database
Keywords
Pubmed
Toxcenter
Toxline
Chemical name (CASRN): benzo[a]pyrene (50-32-8)
Synonyms: benzo[d,e,f]chrysene, benzo[def]chrysene, 3,4-benzopyrene, 1,2-benzpyrene, 3,4-bp,
benz(a)pyrene, 3,4-benzpyren, 3,4-benzpyrene, 4,5-benzpyrene, 6,7-benzopyrene, benzopirene,
benzo(alpha)pyrene
Standard toxicology search keywords
Toxicity (including duration, effects to children and occupational exposure); development;
reproduction; teratogenicity; exposure routes; pharmacokinetics; toxicokinetics; metabolism; body
fluids; endocrinology; carcinogenicity; genotoxicity; antagonists; inhibitors
TSCATS
ChemID
Chemfinder
CCRIS
HSDB
GENETOX
RTECS
Searched by CASRNs and chemical names (including synonyms)
12
13
14
15
16
17
18
19
20
21
22
aPrimary and secondary keywords used for the Pubmed, Toxcenter, and Toxline databases can be found in the
Supplemental Information.
Figure LS-1 depicts the literature search, study selection strategy, and number of references
obtained at each stage of literature screening. Approximately 20,700 references were identified
with the initial keyword search. Based on a secondary keyword search followed by a preliminary
manual screen of titles or abstracts by a toxicologist, approximately 1,190 references were
identified that provided information potentially relevant to characterizing the health effects or
physical and chemical properties of benzo[a]pyrene. A more detailed manual review of titles,
abstracts, and/or papers was then conducted. Notable exclusions from the Toxicological Review
This document is a draft for review purposes only and does not constitute Agency policy.
xli DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 are large numbers of animal in vivo or in vitro studies designed to identify potential therapeutic
2 agents that would prevent the carcinogenicity or genotoxicity of benzo[a]pyrene and toxicity
3 studies of benzo[a]pyrene in nonmammalian species (e.g., aquatic species, plants).
4 For the updated literature search conducted for the timeframe January 2012 through
5 August 2014, the search terms included benzo(a)pyrene AND (rat OR mouse OR mice) and results
6 were screened manually by title, abstract, and/or full text using the exclusion criteria outlined in
7 Figure LS-1. Relevant studies that could potentially impact the hazard characterization and dose-
8 response assessment were identified and considered. No studies were identified that would impact
9 the assessment's major conclusions. Several pertinent studies, published since the last
10 comprehensive literature search, were identified and incorporated into the text where relevant
11
This document is a draft for review purposes only and does not constitute Agency policy.
xlii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
References identified based on initial keyword search (see Table LS-1): "21,000 references
References identified based on
secondary keyword search (see Table
LS-1): ~6,100 references
30 references submitted by American
Petroleum Institute
1
2
Secondary keyword searching (see Table
LS-1): ~14,600 references excluded
Manual screen of titles/abstracts:
~4,900 references excluded
• Not relevant to BaP toxicity in mammals
(e.g., toxicity in aquatic species, plants)
• Site-specific risk assessments
• Chemical analytical methods
• Cancer chemotherapy studies
Considered for inclusion in the Toxicological Review: - 1,000
references; references subsequently evaluated based on
Preamble Section 3
Manual screen of manuscripts excluded: ~ 600 references
• Not relevant to BaP toxicity in mammals
• Inadequate basis to infer exposure
• Inadequate reporting of study methods or results
• Animal toxicity studies with mixtures of chemicals
• Abstracts
• Duplicates
Approximately 700 references cited in the Draft Toxicological Review
• Developmental toxicity: 37 references
• Reproductive toxicity: 70 references
* Immunotoxicity: 58 references
• Other Toxicological Effects: 27 references
• Forestomach toxicity: 5 references
• Hematological toxicity: 3 references
• Liver toxicity: 3 references
• Kidney toxicity: 3 references
• Cardiovascular toxicity: 11 references
• Neurological toxicity: 12 references
• Carcinogenicity: 171 references
•Toxicokinetic: 115 references
•Genotoxicity: 196 references
Figure LS-1. Study selection strategy.
This document is a draft for review purposes only and does not constitute Agency policy,
xliii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Selection of studies for inclusion in the Toxicological Review was based on consideration of
2 the extent to which the study was informative and relevant to the assessment and general study
3 quality considerations. In general, the relevance of health effect studies was evaluated as outlined
4 in the Preamble and EPA guidance (/I Review of the Reference Dose and Reference Concentration
5 Processes [U.S. EPA. 2002] and Methods for Derivation of Inhalation Reference Concentrations and
6 Application of Inhaled Dosimetry [U.S. EPA. 1994]]. The reasons for excluding epidemiological and
7 animal studies from the references identified by the keyword search are provided in Figure LS-1.
8 The available studies examining the health effects of benzo[a]pyrene exposure in humans
9 are discussed and evaluated in the hazard identification sections of the assessment (Section 1], with
10 specific limitations of individual studies and of the collection of studies noted. The common major
11 limitation of the human epidemiological studies (with respect to identifying potential adverse
12 health outcomes specifically from benzo[a]pyrene] is that they all involve exposures to complex
13 mixtures containing other PAHs and other compounds. The evaluation of the epidemiological
14 literature focuses on studies in which possible associations between external measures of exposure
15 to benzo[a]pyrene or biomarkers of exposure to benzo[a]pyrene (e.g., benzo[a]pyrene-DNA
16 adducts or urinary biomarkers] and potential adverse health outcomes were evaluated. Pertinent
17 mechanistic studies in humans (e.g., identification of benzo[a]pyrene-DNA adducts and
18 characteristics of mutations in human tumors] were also considered in assessing the weight of
19 evidence for the carcinogenicity of benzo[a]pyrene.
20 The health effects literature for benzo[a]pyrene is extensive. All animal studies of
21 benzo[a]pyrene involving repeated oral, inhalation, or dermal exposure that were considered to be
22 of acceptable quality, whether yielding positive, negative, or null results, were considered in
23 assessing the evidence for health effects associated with chronic exposure to benzo[a]pyrene.
24 These studies were evaluated for aspects of design, conduct, or reporting that could affect the
25 interpretation of results and the overall contribution to the synthesis of evidence for determination
26 of hazard potential using the study quality considerations outlined in the Preamble. Discussion of
27 study strengths and limitations (that ultimately supported preferences for the studies and data
28 relied upon] were included in the text where relevant.
29 Animal toxicity studies involving short-term duration and other routes of exposure were
30 also evaluated to inform conclusions about health hazards, especially regarding mode of action.
31 The references considered and cited in this document, including bibliographic information and
32 abstracts, can be found on the Health and Environmental Research Online (HERO] website2
33 (http://hero.epa.gov/benzoapyrene].
2HERO (Health and Environmental Research On-line] is a database of scientific studies and other references
used to develop EPA's risk assessments aimed at understanding the health and environmental effects of
pollutants and chemicals. It is developed and managed in EPA's Office of Research and Development (ORD]
by the National Center for Environmental Assessment (NCEA]. The database includes more than
300,000 scientific articles from the peer-reviewed literature. New studies are added continuously to HERO.
This document is a draft for review purposes only and does not constitute Agency policy.
xliv DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2 1. HAZARD IDENTIFICATION
3 1.1. PRESENTATION AND SYNTHESIS OF EVIDENCE BY ORGAN/SYSTEM
4 NOTE: In the environment, benzo[a]pyrene occurs in conjunction with other structurally
5 related chemical compounds known as polycyclic aromatic hydrocarbons (PAHs).3 Accordingly,
6 there are few epidemiologic studies designed to solely investigate the effects of benzo[a]pyrene.
7 There are, however, many epidemiologic studies that have investigated the effects of exposure to
8 PAH mixtures. Benzo[a]pyrene is universally present in these mixtures and is routinely analyzed
9 and detected in environmental media contaminated with PAH mixtures, thus, it is often used an an
10 indicator chemical to measure exposure to PAH mixtures [Bostrom et al.. 2002].
11 1.1.1. Developmental Toxicity
12 Human and animal studies provide evidence for PAH- and benzo[a]pyrene-induced
13 developmental effects. Effects on fetal survival, postnatal growth, and development have been
14 demonstrated in human populations exposed to PAH mixtures during gestation. Animal studies
15 demonstrate various effects including changes in fetal survival, pup weight, blood pressure, fertility,
16 reproductive organ weight and histology, and neurological function in gestationally and/or early
17 postnatally treated animals.
18 Altered Birth Outcomes
19 Human and animal studies provide evidence that benzo[a]pyrene exposure may lead to
20 altered outcomes reflecting growth and development in utero or in early childhood. Two cohort
21 studies in pregnant women in China and the United States examined cord blood levels of
22 benzo[a]pyrene-7,8-diol-9,10 epoxide (BPDE)-deoxyribonucleic acid (DNA) adducts in relation to
23 measures of child growth following exposure to PAH mixtures [Tangetal.. 2006: Perera etal..
24 2005b: Perera etal., 2004] (Table 1-1). In the Chinese cohort, high benzo[a]pyrene-adduct levels
25 were associated with reduced weight at 18, 24, and 30 months of age, but not at birth [Tangetal.,
26 2006). In the U.S. cohort, an independent effect on birth weight was not observed with either
27 benzo[a]pyrene-adducts or environmental tobacco smoke (ETS) exposure; however, a doubling of
28 cord blood adducts in combination with ETS exposure in utero was seen, corresponding to an 8%
29 reduction in birth weight [Perera etal.. 2005b]. ETS, also called secondhand smoke, is the smoke
30 given off by a burning tobacco product and the smoke exhaled by a smoker that contains over
3PAHs are a large class of chemical compounds formed during the incomplete combustion of organic matter.
This document is a draft for review purposes only and does not constitute Agency policy.
1-1 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 7,000 chemicals including benzo[a]pyrene. No associations were seen with birth length (or height
2 at later ages) in either of these cohort studies.
3 A Chinese case-control study indicated that PAH exposure may be associated with increased
4 risk of fetal death [Wu etal.. 2010). A strong association was seen between maternal blood
5 benzo[a]pyrene-DNA adduct levels and risk of delayed miscarriage (fetal death before 14 weeks of
6 gestation), with a fourfold increased risk for levels above compared with below the median.
7 However, no significant difference in adduct levels was detected between fetal tissue from cases
8 compared to controls.
9 Decreased fetal survival has also been noted in gestationally treated animals at relatively
10 high doses by the oral and inhalation routes. An approximate 40% decrease in fetal survival was
11 noted in mouse dams treated by gavage on CDs 7-16 at doses of 160 mg/kg-day, but no decreases
12 were observed at 10 or 40 mg/kg-day (Mackenzie and Angevine. 1981). Several lower dose studies
13 of rats treated on CDs 14-17 with doses of up to 1.2 mg/kg-day benzo[a]pyrene did notobserve
14 any difference in fetal survival (Jules etal.. 2012: McCallister etal.. 2008: Brown et al.. 2007). By
15 the inhalation route, fetal survival was decreased by 19% following exposure to 25 ug/m3
16 benzo[a]pyrene on CDs 11-20 in F344 rats (Archibong et al., 2002). Another publication from the
17 same group of collaborators Wuetal. (2003a) graphically reported fetal survival as part of a study
18 analyzing metabolites of benzo[a]pyrene and activation of the aryl hydrocarbon receptor (AhR) and
19 cytochrome P450 (CYP450) 1A1 fWuetal.. 2003al This study did notreportthe number of dams
20 or litters, and no numerical data were reported. The study authors reported statistically significant
21 decreases in fetal survival at 75 and 100 ug/m3 benzo[a]pyrene on CDs 11-20 compared to the
22 25 ug/m3 group. An apparent decrease in fetal survival was also seen at 25 ug/m3, but it was
23 unclear whether or not this change was statistically significant compared to the vehicle control.
24 In animals (Table 1-2 and Figure 1-1), reduced body weight in offspring has also been noted
25 in some developmental studies. Decreases in body weight (up to 13%) were observed in mice
26 following prenatal gavage exposure (gestation days [CDs] 7-16), and as time from exposure
27 increased (postnatal days [PNDs] 20-42), the dose at which effects were observed decreased (from
28 40 to 10 mg/kg-day, respectively) (Mackenzie and Angevine. 1981). In addition, decreases in body
29 weight (approximately 10-15%) were observed in rats on PNDs 36 and 71 following gavage
30 exposure at only 2 mg/kg-day on PNDs 5-11 (Chen etal., 2012). At doses up to 1.2 mg/kg-day and
31 follow-up to PND 30, two developmental studies in rats did not observe decrements in pup body
32 weight following treatment from GD 14 to 17 flules etal.. 2012: McCallister etal.. 20081 Maternal
33 toxicity was not observed in mouse or rat dams exposed to up to 160 mg/kg-day benzo[a]pyrene
34 flules etal.. 2012: McCallister etal.. 2008: Brown etal.. 2007: Kristensen etal.. 1995: Mackenzie and
35 Angevine. 1981).
36 Fertility in Offspring
37 Several studies suggest that gestational exposure to maternal tobacco smoke decreases the
38 future fertility of female offspring fYe etal.. 2010: Tensen etal.. 1998: Weinberg etal.. 19891
This document is a draft for review purposes only and does not constitute Agency policy.
1-2 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 (Table 1-1). In animal models, marked effects on the development of male and female reproductive
2 organs and the fertility of animals exposed gestationally has also been demonstrated [Kristensen et
3 al.. 1995: Mackenzie and Angevine. 1981] (Table 1-2 and Figure 1-1). In two studies examining
4 reproductive effects in mice, decreased fertility and fecundity in Fl animals was observed following
5 exposure to doses >10 mg/kg-day during gestation (Kristensen etal.. 1995: Mackenzie and
6 Angevine. 1981). When Fl females were mated with untreated males, a dose-related decrease in
7 fertility of >30% was observed, in addition to a 20% decrease in litter size starting at the lowest
8 dose tested of 10 mg/kg-day (Mackenzie and Angevine. 1981). A dose-related decrease in fertility
9 was also observed in male mice treated gestationally with benzo[a]pyrene. At the lowest dose
10 tested (10 mg/kg-day), a 35% decrease in fertility was observed when gestationally exposed
11 animals were mated with untreated females (Mackenzie and Angevine. 1981). Similar effects on
12 fertility were observed in another developmental study in mice (Kristensen et al.. 1995). Fl
13 females (bred continuously for 6 months) in this study had 63% fewer litters, and litters were 30%
14 smaller as compared to control animals. The fertility of male offspring was not assessed in this
15 study.
16 Reproductive Organ Effects in Offspring
17 The above-mentioned studies also demonstrated dose-related effects on male and female
18 reproductive organs in animals exposed gestationally to benzo[a]pyrene (Table 1-2 and Figure 1-
19 1). Testicular weight was decreased and atrophic seminiferous tubules and vacuolization were
20 increased at >10 mg/kg-day in male mice exposed to benzo[a]pyrene gestationally from GD 7 to 16;
21 severe atrophic seminiferous tubules were observed at 40 mg/kg-day (Mackenzie and Angevine.
22 1981). Testicular weight was also statistically significantly decreased in Sprague-Dawley rats
23 treated on PNDs 1-7 at doses >10 mg/kg-day, when examined at PND 8 (Liang etal.. 2012).
24 In female mice treated with doses >10 mg/kg-day during gestation, ovarian effects were
25 observed including decreases in ovary weight, numbers of follicles, and corpora lutea (Kristensen et
26 al.. 1995: Mackenzie and Angevine. 1981). Specifically, ovary weight in Fl offspring was reduced
27 31% following exposure to 10 mg/kg-day benzo[a]pyrene (Kristensen et al.. 1995). while in
28 another gestational study at the same dose level, ovaries were so drastically reduced in size (or
29 absent) that they were not weighed (Mackenzie and Angevine, 1981). Hypoplastic ovaries with few
30 or no follicles and corpora lutea (numerical data not reported), and ovaries with few or no small,
31 medium, or large follicles and corpora lutea (numerical data not reported) have also been observed
32 in mouse offspring exposed gestationally to benzo[a]pyrene (Kristensen et al.. 1995: Mackenzie and
33 Angevine. 1981).
34 Cardiovascular Effects in Offspring
35 Increased systolic and diastolic blood pressure was observed in adult animals following
36 gestational treatment with benzo[a]pyrene (Tules etal., 2012] (Table 1-2 and Figure 1-1).
37 Approximate elevations in systolic and diastolic blood pressure of 20-30 and 50-80% were noted
This document is a draft for review purposes only and does not constitute Agency policy.
1-3 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 in the 0.6 and 1.2 mg/kg-day dose groups, respectively. Heart rate was decreased at 0.6 mg/kg-day,
2 but was increased at 1.2 mg/kg-day.
3 Immune Effects in Offspring
4 Several injection studies in laboratory animals suggest that immune effects may occur
5 following gestational or early postnatal exposure to benzo[a]pyrene. These studies are discussed in
6 Section 1.1.3.
7
8
Table 1-1. Evidence pertaining to developmental effects of benzo[a]pyrene in
humans
Study design and reference
Results
Tangetal. (2006) (Tongliang, China)
Birth cohort
150 nonsmoking women who delivered
babies between March 2002 and June 2002
Exposure: Mean hours per day exposed to
ETS 0.42 (SD 1.19); lived within 2.5 km of
power plant that operated from December
2001 to May 2002; benzo[a]pyrene-DNA
adducts from maternal and cord blood
samples; cord blood mean 0.33 (SD 0.14)
(median 0.36) adducts/10 8 nucleotides;
maternal blood mean 0.29 (SD 0.13)
adducts/10~8 nucleotides
Relation between cord blood benzo[a]pyrene-DNA adducts and log-
transformed weight and height
Weight
Beta (p-value)
-0.007 (0.73)
Birth
18 mo
24 mo
30 mo
-0.048 (0.03)
-0.041 (0.027)
-0.040 (0.049)
Length (height)
Beta (p-value)
-0.001 (0.89)
-0.005 (0.48)
-0.007 (0.28)
-0.006 (0.44)
Adjusted for ETS, sex of child, maternal height, maternal weight, and
gestational age (for measures at birth)
(Perera et al. (2005b); Perera et al. (2004))
(New York, United States)
Birth cohort
265 pregnant African-American and
Dominican nonsmoking women who
delivered babies between April 1998 and
October 2002 (214 and 208 for weight and
length analysis, respectively);
approximately 40% with a smoker in the
home
Exposure: Benzo[a]pyrene-DNA adducts in
cord blood samples; mean 0.22 (SD 0.14)
adducts/10~8 nucleotides; median of
detectable values 0.36 adducts/
10 8 nucleotides
Relation between cord blood benzo[a]pyrene-DNA adducts and log-
transformed weight and length
Interaction
term
Benzo[a]-
pyrene-DNA
adducts
Weight
Beta (p-value)
-0.088 (0.05)
-0.020 (0.49)
Length
Beta (p-value)
-0.014 (0.39)
-0.005 (0.64)
ETS in home -0.003 (0.90) -0.007 (0.32)
Adjusted for ethnicity, sex of newborns, maternal body mass index,
dietary PAHs, and gestational age
This document is a draft for review purposes only and does not constitute Agency policy,
1-4 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Study design and reference
Results
Wu et al. (2010) (Tianjin, China)
Case control study: 81 cases (96%
participation rate)—fetal death confirmed
by ultrasound before 14 wks of gestation;
81 controls (91% participation rate)—
elective abortions; matched by age,
gestational age, and gravidity; excluded
smokers and occupational PAH exposure
Exposure: Benzo[a]pyrene in aborted tissue
and maternal blood samples (51 cases and
controls; 2 of 4 hospitals)
Benzo[a]pyrene adduct levels (/Iff nucleotides), mean (± SD)
Cases Controls (p-value)
Maternal blood 6.0 (±4.7) 2.7 (± 2.2) (<0.001)
Aborted tissue 4.8 (± 6.0) 6.0 (± 7.4) (0.29)
Low correlation between blood and tissue levels (r = -0.02 in cases,
r = -0.21 in controls)
Association between benzo[a]pyrene adducts and miscarriage3
OR (95% Cl)
Per unit increase in adducts 1.37 (1.12,1.67)
Dichotomized at median 4.56 (1.46,14.3)
Conditional logistic regression, adjusted for maternal education,
household income, and gestational age; age also considered as
potential confounder
1
2
3
4
Cl = confidence interval; OR = odds ratio; SD = standard deviation.
Table 1-2. Evidence pertaining to developmental effects of benzo[a]pyrene in
animals
Study design and reference
Results
Birth outcomes and postnatal growth
Mackenzie and Angevine (1981)
CD-I mice, 30 or 60 FO females/
dose
0,10, 40, or 160 mg/kg-d by
gavage
CDs 7-16
4, number of FO females with viable litters: 46/60, 21/30,44/60, and 13/30*
4, Fl body weight at PND 20
% change from control: 0, 4, -7*, and -13*
^ Fl body weight at PND 42
% change from control: 0, -6*, -6*, and -10*
(no difference in pup weight at PND 4)
Kristensenetal. (1995)
NMRI mice, 9 FO females/dose
0 or 10 mg/kg-d by gavage
CDs 7-16
Exposed FO females showed no gross signs of toxicity and no effects on
fertility (data not reported)
Jules etal. (2012)
Long-Evans rats, 6-17 FO
females/dose
0, 0.15, 0.3, 0.6, or 1.2 mg/kg-d by
gavage
CDs 14-17
No overt signs of toxicity in dams or offspring, differences in pup body
weight, or number of pups/litter
This document is a draft for review purposes only and does not constitute Agency policy,
1-5 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Study design and reference
Results
McCallisteretal. (2008)
Long-Evans Hooded rats,
5-6/group
0 or 0.3 mg/kg-d by gavage
CDs 14-17
No difference in number of pups/litter
No overt maternal or pup toxicity
No difference in liverbody weight
Increased brain:body weight ratio at PNDs 15 and 30 (data not shown)
Brown et al. (2007)
Long-Evans Hooded rats, 6/group
0, 0.025, or 0.15 mg/kg-d by
gavage
CDs 14-17
No difference in number of pups/litter or overt maternal or pup toxicity
Chen etal. (2012)
Sprague-Dawley rats, 20 pups
(10 male and 10 female)/group
0, 0.02, 0.2, or 2 mg/kg-d by
gavage
PNDs 5-11
Statistically significant decrease in pup body weight (approximate 10-15%
decrease) at 2 mg/kg-d measured on PNDs 36 and 71 (no significant
alteration of pup weight during treatment period)
No differences among treatment groups in developmental milestones:
incisor eruption, eye opening, development of fur, testis decent, or vaginal
opening
Archibong et al. (2002)
F344 rats, 10 females/group
0, 25, 75, or 100 u.g/m3 nose-only
inhalation for 4 hrs/d
CDs 11-20
•i, fetal survival ([pups/litter]/[implantation sites/litter] x 100)
% fetal survival: 97, 78*, 38*, and 34*%
Reproductive effects in offspring
Mackenzie and Angevine (1981)
CD-I mice, 30 or 60 FO females/
dose
0,10, 40, or 160 mg/kg-d by
gavage
CDs 7-16
^ number of Fl females with viable litters: 35/35, 23/35*, 0/55*, and 0/20*
4/F1 female fertility index (females pregnant/females exposed to males x
100): 100, 66*, 0*, and 0*
4, Fl male fertility index (females pregnant/females exposed to males x
100): 80, 52*, 5*, and 0*
•^ F2 litter size from Fl dams (20%) at 10 mg/kg-d (no litters were produced
at high doses)
4, size or absence of Fl ovaries (weights not collected)
hypoplastic ovaries with few or no follicles and corpora lutea (numerical data
not reported)
•^ testicular weight in Fl offspring
% change from control: 0, -42, -82, and ND (statistical significance not
reported)
/T" atrophic seminiferous tubules and vacuolization at >10 mg/kg-d; severe
atrophic seminiferous tubules at 40 mg/kg-d (numerical data not reported)
This document is a draft for review purposes only and does not constitute Agency policy,
1-6 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Study design and reference
Results
Kristensenetal. (1995)
NMRI mice, 9 FO females/dose
0 or 10 mg/kg-d by gavage
CDs 7-16
4, number of F2 litters (-63%)
4, F2 litter size (-30%)
4' ovary weight (-31%) in Fl females
Few or no small, medium, or large follicles and corpora lutea
Cardiovascular effects in offspring
Jules etal. (2012)
Long-Evans rats, 6-17 FO
females/dose
0, 0.15, 0.3, 0.6, or 1.2 mg/kg-d by
gavage
CDs 14-17
/T" systolic blood pressure (measured at PND 53)
15%* increase at 0.6 mg/kg-d
52%* increase at 1.2 mg/kg-d
(other dose groups not reported)
/T" diastolic blood pressure (measured at PND 53)
33%* increase at 0.6 mg/kg-d
83% increase at 1.2 mg/kg-d
(other dose groups not reported)
Altered heart rate
10%* increase at 0.6 mg/kg-d
8%* decrease at 1.2 mg/kg-d
1
2
3
*Statistically significantly different from the control (p< 0.05).
a% change from control calculated as: (treated value - control value)/control value x 100.
This document is a draft for review purposes only and does not constitute Agency policy,
1-7 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
1000 i
100 =
T 10,
ra :
E
0 0.1 !
n m
• LOAEL ANOAEL • Doses >LOAEL c
Doses «
£
v
i
4- number
viable litters
^ ox ^
ro rn o o r^
ci 2 « Q ° „ " -0
rH t-l*jLJrsl^f*l ~
njOgZ .^" g
CJ ^r^^.li-2g^uira
= 0 Q •^c^^-^o^S
>0^rarTfOmLHSlN'-,,
MJliiJ3i3?l
m
c
1
5s
I P
o r-
l£3 V
^H C
r^ -
g
0 ^
0
x|/ ovarian size
-H LO
co cn
2 2 o
O — p
.E c * c.
? •- o ^
0) C ^H
&0 C C i
4
; mice
ID
rH
IS
Q
o
/
esticular
weight
and t
atrophy
Reproductive Effects
in Offspring
m
m
SSQ
o 2 z
(N Q_
"^ ^ O
0) rH OJ
* Q ^
^ ij n
Cardio-
vascular
Effects in
Offspring
2 Figure 1-1. Exposure-response array for developmental effects following oral
3 exposure to benzo[a]pyrene.
4 Neurodevelopmental Effects
5 There is evidence in humans and animals that benzo[a]pyrene induces developmental
6 neurotoxicity. In addition to the persistent reductions in cognitive ability observed in epidemiology
7 studies of prenatal PAH exposure, the two epidemiology studies that examined benzo[a]pyrene-
8 specific measures observed effects on neurodevelopment and behavior in young children. Altered
9 learning and memory, motor activity, anxiety-like behavior, and electrophysiological changes have
10 also been observed in animals following oral and inhalation exposure to benzo[a]pyrene.
11 The mammalian brain undergoes periods of rapid brain growth, particularly during the last
12 3 months of pregnancy in humans, which has been compared to the first 1-2 weeks of life in the rat
13 and mouse neonate [Dobbing and Sands, 1979,1973]. This period is characterized by axonal and
14 dendritic outgrowth and the establishment of mature neuronal connections. Also during this
15 critical period, animals acquire many new motor and sensory abilities [Kolb and Whishaw. 1989).
16 There is a growing literature of animal studies that shows subtle changes in motor and cognitive
This document is a draft for review purposes only and does not constitute Agency policy.
1-8 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 function following acute or repeated perinatal or lactational exposure to benzo[a]pyrene [Bouayed
2 etal.. 2009a: McCallister etal.. 2008: Wormley etal.. 20041 These effects are described below.
3 Cognitive function
4 Head circumference at birth is associated with measures of intelligence in children, even
5 among term infants [Broekman et al.. 2009: Gale etal.. 2006). The two birth cohort studies that
6 examined maternal or cord blood levels of benzo[a]pyrene-DNA adducts in relation to head
7 circumference provide some evidence of an association, most strongly within the context of an
8 interaction with ETS [Tang etal.. 2006: Pereraetal.. 2005b: Perera etal.. 2004] (Table 1-3). The
9 cohort in Tongliang, China also examined intelligence quotient scores at age 5 years [Pereraetal.,
10 2012a). An interaction with ETS was seen in this analysis, with larger decrements seen on the full
11 scale and verbal scales with increased benzo[a]pyrene-DNAadduct levels in the presence of
12 prenatal exposure to ETS compared to the effects seen in the absence of prenatal exposure.
13 Animal studies have also provided evidence of altered learning and memory behaviors
14 following lactational or direct postnatal exposure to benzo[a]pyrene [Chen etal.. 2012: Bouayed et
15 al., 2009a] (Table 1-4). In mice, spatial working memory was measured using the Y-maze
16 spontaneous alternation test [Bouayed etal., 2009a). This test records alternations between arm
17 entries in a Y-shaped maze as a measure of memory, as rodents typically prefer to investigate a new
18 arm of the maze. To a lesser extent, this test can also reflect changes in sensory processing, novelty
19 preference, and anxiety-related responses in rodents. An improvement in working memory was
20 evident in mice, as exhibited by significant increases in spontaneous alternations in the Y-maze test
21 in mice on PND 40 following lactational exposure to 2 mg/kg-day benzo[a]pyrene (but not
22 20 mg/kg-day) from PND 0 to 14 [Bouayed et al., 2009a). The total number of arm entries in the
23 Y-maze was unaffected by lactational exposure, suggesting that changes in motor function were not
24 driving this response. Similarly, CPRlox/loxtransgenic mice (a mouse model with low brain
25 expression of CYP450 reductase) exhibit deficiencies in novel object recognition tests following in
26 utero exposure to benzo[a]pyrene during late gestation [Li etal.. 2012: Shengetal.. 2010). This
27 finding suggests impairment in short-term memory, although these tests also reflect locomotor
28 exploratory behavior, response to novelty, and attention. In rats, spatial learning and memory was
29 measured using the Morris water maze, which measures the ability of a rat to navigate to a target
30 platform using external spatial cues. Increased escape latency (time to find the hidden platform), as
31 well as decreased time in the target quadrant and decreased number of platform crossings during a
32 probe trial with the platform removed were observed in PND 39-40 rats following postnatal
33 exposure to 2 mg/kg-day benzo[a]pyrene [Chen etal.. 2012). These effects were more pronounced
34 in animals tested at PNDs 74-75, with effects observable at >0.2 mg/kg-day. No difference in swim
35 speed was observed during the probe trial tests (swim speed did not appear to be analyzed during
36 the hidden platform trials) between treatment groups, suggesting that the observed changes are
37 not attributable to general motor impairment These observations may indicate primary effects of
38 benzo[a]pyrene on learning and/or memory processes; however, the presented data were
This document is a draft for review purposes only and does not constitute Agency policy.
1-9 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 insufficient to attribute these findings to learning and memory processes alone. Specifically, visual
2 examinations of the improvements in escape latency (slopes) over the four learning trial days were
3 not noticeably affected by treatment dose, suggesting that all groups learned at a similar rate. As
4 four trials/day were averaged for each animal at each trial day, it is unclear whether the dose-
5 related increases in escape latency already observable at trial day 1 reflect effects on learning
6 across those first four trials or other effects (e.g., altered anxiety or vision responses). As it is not
7 clear that the groups learned to a comparable extent in hidden platform tests, the results of the
8 probe trial cannot be conclusively attributed to memory retention; however, the decreased
9 performance of the benzo[a]pyrene-exposed rats in the Morris water maze still represents a
10 persistent neurobehavioral change.
11 Neuromuscular function, coordination, and sensorimotor development
12 Motor behavior and coordination, assessed by locomotion, reaching, balance,
13 comprehension, drawing, and hand control was one of the specific domains assessed in the Chinese
14 birth cohort evaluated by Tangetal. (2008). In children aged 2 years, decreased scores were seen
15 in relation to increasing benzo[a]pyrene-DNA adducts measured in cord blood, with a Beta per unit
16 increase in adducts of -16 (p = 0.04), and an approximate twofold increased risk of development
17 delay per unit increase in adducts (Table 1-3).
18 In laboratory animals (Table 1-4 and Figure 1-2), impaired performance in neuromuscular
19 and sensorimotor tests have been consistently observed in mice lactationally exposed to >2 mg/kg-
20 day benzo[a]pyrene from PND 0 to 14 (Bouayed et al.. 2009a) and in rat pups postnatally exposed
21 to >0.02 mg/kg-day benzo[a]pyrene from PND 5 to 11 (Chenetal.. 2012). In the righting reflex test,
22 significant increases in righting time were observed in PNDs 3-5 mice and in PNDs 12-16 rats.
23 These decrements did not show a monotonic dose response. In another test of sensorimotor
24 function and coordination, dose-dependent increases in latency in the negative geotaxis test were
25 observed in PND 5-9 mice and in PND 12-14 rats. The forelimb grip strength test of
26 neuromuscular strength was also evaluated in both mice and rats, but alterations were only
27 observed in mice. In mice, a dose-dependent increase in duration of forelimb grip was observed on
28 PNDs 9 and 11 during lactational exposure to benzo[a]pyrene. The Water Escape Pole Climbing
29 test was also used to evaluate neuromuscular function and coordination in mice (Bouayed etal.,
30 2009a). No effect on climbing time was observed, suggesting no change in muscle strength.
31 However, increased latency in pole grasping and pole escape in PND 20 male pups was observed,
32 highlighting potential decrements in visuomotor integration and/or coordination, although anxiety
33 or fear-related responses cannot be ruled out Treatment-dependent increases in pup body weight
34 around the testing period complicate the interpretation of these results.
35 Chen etal. (2012) observed statistically significant delays on the order of ~0.2-0.3 seconds
36 in the surface righting test and ~3-4 seconds in the negative geotaxis test. Differences due to
37 exposure were notable between PND 12 and 16, with equal performance in righting observed by
38 PND 18 and in geotaxis by PND 16. The authors found no effect on gender, therefore the data for
This document is a draft for review purposes only and does not constitute Agency policy.
1-10 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 male and female rats was pooled for these measures. However, it should be noted that differences
2 in the maturation of these developmental landmarks following challenge have been shown to exist
3 between males and females. Negative geotaxis and surface righting are discrete endpoints
4 routinely used as part of a neurobehavioral test battery to assess acquisition of behavioral reflexes.
5 Chenetal. [2012] used the surface righting and negative geotaxis tests as quantitative measures of
6 sensorimotor function at PND 12 and beyond. Typically in these tests, animals are observed on
7 consecutive days (e.g., PNDs 3-12) and time to acquisition of these phenotypes is measured. Chen
8 etal. [2012] did not measure performance over consecutive days of development or measure
9 baseline acquisition of these behaviors. Although these tests as conducted by Chen etal. [2012]
10 cannot discern a developmental delay, the data support a transient impairment of sensorimotor
11 function in animals that have already developed this reflex (e.g., able to orient 180 degrees and able
12 to right within 2 seconds]. Thus, these data indicate that benzo [ajpyrene may affect sensorimotor
13 function in developing animals.
14 Anxiety and activity
15 Anxiety/depression and attention/hyperactivity symptoms in children ages 6-7 years were
16 examined via questionnaire in relation to prenatal air monitoring of benzo[a]pyrene and other
17 PAHs, and in relation to benzo[a]pyrene-specific DNA adducts measured at birth in a follow-up of a
18 birth cohort study conducted in New York City [Perera et al.. 2012b]. PAH exposure levels (based
19 on personal air monitoring, n = 253] and benzo[ajpyrene-specific DNA adducts measured in cord
20 blood samples (n = 138] were both positively associated with symptoms of anxiety/depression and
21 attention problems (see Table 1-3]. Given the limited sample size, however, the cord blood results
22 are based on relatively sparse data (<5 in the borderline or clinical range in the low exposure
23 referent group]. Associations with maternal blood adducts were similar to or slightly smaller than
24 those seen with cord blood adducts. Exposure was treated as a dichotomy (i.e., for adducts,
25 detectable compared with non-detectable levels] in these analyses.
26 Decreased anxiety-like behavior was reported in both rats and mice weeks to months
27 following postnatal oral exposure to benzo[a]pyrene (Chenetal.. 2012: Bouayed et al.. 2009a]
28 (Table 1-4]. Anxiety-like behaviors were measured in both species using an elevated plus maze,
29 where an increase in the time spent in the closed arms of the maze is considered evidence of
30 anxious behavior. In mice, significant increases in the percent open arm entries and percent time
31 spent in open arms of the maze, as well as significantly decreased entries into closed arms of the
32 maze (in the 2 mg/kg-day group], were observed on PND 32 following lactational exposure to
33 >2 mg/kg-day benzo[a]pyrene (Bouayed et al.. 2009a]. To rule out potential differences in total
34 activity or general motivation and exploration, the authors expressed the open arm data as
35 percentages, and they also demonstrated that there were no exposure-related effects on the total
36 number of arm entries. The mice also exhibited decreased latency of the first entry into an open
37 arm following lactational exposure to 20 mg/kg-day benzo[a]pyrene. Similar results were reported
38 for rats, with decreased anxiety-like behavior following oral benzo[a]pyrene exposure from PND 5
This document is a draft for review purposes only and does not constitute Agency policy.
1-11 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 to 11, although sex-specific differences were observed [Chenetal., 2012]. In females, postnatal
2 exposure to >0.2 mg/kg-day benzo[a]pyrene was associated with a significant increase in the
3 number of open arm entries and a significant decrease in the number of closed arm entries on
4 PND 70. Significantly increased time in open arms of the maze was reported in PND 70 female rats
5 following postnatal exposure to >0.02 mg/kg-day. Male rats also showed decreased anxiety-like
6 behavior on PND 70, although the doses needed to detect these responses were higher than females
7 (i.e., increases at >2 mg/kg-day for open arm entries and >0.2 mg/kg-day for time spent in open
8 arms). A significant decrease in latency to enter an open arm of the maze was observed in both
9 male and female rat pups exposed to 2 mg/kg-day benzo[a]pyrene. Similar to the observations in
10 mice, exposure did not appear to have an effect on total activity or general motivation of the rats, as
11 total arm entries were unchanged by treatment
12 Increased spontaneous locomotor activity in the open field on PNDs 34 and 69 has been
13 reported in rats postnatally exposed to 2 and >0.2 mg/kg-day, respectively [Chenetal.. 2012). but
14 not in mice exposed lactationally to doses up to 20 mg/kg-day and tested on PND 15 [Bouayed et
15 al.. 2009a). Interestingly, no differences in the open field test were observed in rats that were
16 postnatally exposed and tested on PNDs 18 and 20, suggesting either that longer latencies between
17 exposure and testing may be required, or that these developmental effects may only manifest in
18 more mature rats [Chenetal.. 2012). An apparent increased sensitivity of older animals was also
19 present in the elevated plus maze and Morris water maze tests performed by Chenetal. [2012].
20 Elevated activity in an open field is attributable primarily to either increased motor activity or
21 decreased anxiety-like behavior. However, the relative contributions of these two components
22 could not be separated in either of these studies, as the authors did not evaluate activity in central
23 versus peripheral regions of the field (i.e., anxious rodents will spend less time in the center of the
24 field).
25 Electrophysiological changes
26 Electrophysiological effects of gestational exposure to benzo[a]pyrene have been examined
27 in two studies (by the same research group) through implanted electrodes in the rat cortex and
28 hippocampus (Table 1-4). Maternal inhalation exposure to 0.1 mg/m3 resulted in reduced long-
29 term potentiation in the dentate gyrus of male offspring between PND 60 and 70 (Wormley etal.,
30 2004): however, significant fetal toxicity at this exposure level complicates interpretation of these
31 results. Oral exposure of dams to 0.3 mg/kg-day for 4 days during late gestation resulted in
32 decreased evoked neuronal activity in male offspring following mechanical whisker stimulation
33 between PND 90 and 120 (McCallister etal.. 2008). Specifically, the authors noted reduced spike
34 numbers in both short and long latency responses following whisker stimulation. These effects
35 were observed several months post-exposure, suggesting that gestational benzo[a]pyrene exposure
36 may have long-lasting functional effects on neuronal activity elicited by sensory stimuli.
This document is a draft for review purposes only and does not constitute Agency policy.
1-12 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2
Table 1-3. Evidence pertaining to the neurodevelopmental effects of
benzo[a]pyrene from PAH mixtures
Reference and study design
Results
(Tang et al. (2008): Tang et al. (2006))
(Tongliang, China)
Birth cohort
150 nonsmoking women, delivered
March 2002-June 2002; lived within
2.5 km of power plant that operated from
December 2001 to May 2002
Outcomes: Head circumference at birth;
Gesell Developmental Schedule,
administered by physicians at 2 yrs of age
(four domains: motor, adaptive, language,
and social); standardized mean
score = 100 ± SD 15 (score
<85 = developmental delay)
Exposure: Mean hrs/d exposed to ETS
0.42 (SD 1.19); lived within 2.5 km of
power plant that operated from
December 2001 to May 2002;
benzo[a]pyrene-DNA adducts from
maternal and cord blood samples; cord
blood mean 0.33 (SD 0.14) (median 0.36)
adducts/10~8 nucleotides; maternal blood
mean 0.29 (SD 0.13)
adducts/10~8 nucleotides
Relation between cord blood benzo[a]pyrene-DNA adducts and log-
transformed head circumference
Birth
18 mo
24 mo
30 mo
Beta (p-value)
-0.011 (0.057)
-0.012 (0.085)
-0.006 (0.19)
-0.005 (0.31)
High versus low, dichotomized at median, adjusted for ETS, sex of
child, maternal height, maternal weight, Cesarean section delivery,
maternal head circumference, and gestational age (for measures at
birth)
(Tang et al. (2008); Tang etal. (2006)) (see
above for population and exposure
details)
n = 110 for Developmental Quotient
analysis; no differences between the
110 participants in this analysis and the
nonparticipants with respect to maternal
age, gestational age, birth weight, birth
length, or birth head circumference;
higher maternal education (direction not
reported, p = 0.056)
Outcomes: Gesell Developmental
Schedule, administered by physicians at
2 yrs of age (four domains: motor,
adaptive, language, and social);
standardized mean score = 100 ± SD 15
(score <85 = developmental delay)
Association between benzo[a]pyrene adducts and development
Beta (95% Cl)a
OR (95% Cl)
Motor
Adaptive
Language
Social
Average
-16.0 (-31.3, -0.72)*
-15.5 (-35.6, 4.61)
-16.6 (-33.7, 0.46)
-9.29 (-25.3, 6.70)
-14.6 (-28.8, -0.37)*
1.91(1.22,2.97)*
1.16 (0.76,1.76)
1.31(0.84,2.05)
1.52 (0.93, 2.50)
1.67 (0.93, 3.00)
aLinear regression of change in Developmental Quotient per unit
increase in benzo[a]pyrene adducts
bLogistic regression of risk of developmental delay (defined as
normalized score <85) per 1 unit (0.1 adducts/10~8 nucleotides)
increase in adducts
Both analyses adjusted for sex, gestational age, maternal education,
ETS, and cord lead levels
This document is a draft for review purposes only and does not constitute Agency policy,
1-13 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Reference and study design
Results
Perera et al. (2012a); (Tang et al. (2008);
Tang et al. (2006)) (see above for
population and exposure details)
132 (83%) followed through age 5; 100 of
these had complete data for analysis; no
differences between the 100 participants
in this analysis and the nonparticipants
with respect to adduct levels, ETS
exposure, IQ measures, maternal age,
gestational age, or infant gender; higher
maternal education (60 and 35% with
> high school, respectively, in participants
and nonparticipants, p < 0.05)
Outcomes: Wechsler Preschool and
Primary Intelligence Quotient scale
(Shanghai version)
ETS measure not correlated with benzo[a]pyrene adduct measures
(i.e., absolute value of Spearman r < 0.10)
Relation between cord blood benzo[a]pyrene-DNA adducts, ETS
exposure, and IQ measures
Beta (95% Cl)
Main effect
Full scale -2.42 (-7.96, 3.13)
Verbal -1.79 (-7.61,4.03)
Performance -2.57 (-8.92, 3.79)
Beta per 1 unit increase in log-transformed cord adducts, adjusted for
ETS exposure, gestational age, maternal education, cord lead, maternal
age, and gender
With ETS interaction term
-10.10 (-18.90, -1.29)
-10.35 (-19.61, -1.10)
-7.78 (-18.03,2.48)
(Perera et al. (2012b); Perera et al.
(2005b); Perera et al. (2004)) (United
States, New York)
Birth cohort
265 pregnant women: African-American
and Dominican nonsmoking women who
delivered babies between April 1998 and
October 2002 (253 and 207 for behavior
and head circumference analysis,
respectively); approximately 40% with a
smoker in the home
Outcomes: Head circumference at birth
Exposure: Benzo[a]pyrene-DNA adducts
from maternal and cord blood samples;
mean 0.22 (SD 0.14)
adducts/10 8 nucleotides; median of
detectable values
0.36 adducts/10"8 nucleotides
Relation between cord blood benzo[a]pyrene-DNA adducts,
environmental tobacco smoke exposure (ETS), and log-transformed
head circumference
Interaction term
benzo[a]pyrene-DNA adducts
ETS in home
Beta (p-value)
-0.032 (0.01)
-0.007 (0.39)
-0.005 (0.43)
High versus low, dichotomized at 0.36 adducts/10 nucleotides,
adjusted for ethnicity, sex of newborns, maternal body mass index,
dietary PAHs, and gestational age
This document is a draft for review purposes only and does not constitute Agency policy,
1-14 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Reference and study design
Results
Perera et al. (2012b)
n = 215 with outcome data and no missing
covariate data); no differences between
the participants in this analysis and the
nonparticipants with respect to adduct
levels, ETS exposure, maternal age,
gestational age, and socioeconomic
variables; participants more likely to be
female and African-American
Outcomes: Child Behavior Checklist
(118 items), completed by mothers for
children ages 6-7 yrs. Two domains:
anxious/depression, attention problems
(normalized T-score <65 = borderline or
clinical syndrome); also used for scales of
anxiety problems and attention deficit
hyperactivity problems based on DSM
classification
Logistic regression of risk of borderline or clinical status in relation PAH
levels and to detectable levels of benzo[a]pyrene adducts
Cord blood
Anxious/depressed
Attention problems
Anxiety (DSM)
Attention deficit -
hyperactivity (DSM)
Prevalence
6.3 %
6.7%
9.5%
7.9%
PAH
OR (95% Cl)
8.9(1.7,46.5)
3.8(1.1,12.7)
4.6 (1.5, 14.3)
2.3 (0.79, 6.7)
OR (95% Cl)
2.6(0.69,9.4)
4.1 (0.99, 16.6)
2.5(0.84,7.7)
2.6 (0.68, 10.3)
Exposure dichotomized for PAH as above and below median
(2.273 ng/m3) for parent population and for cord blood benzo[a]pyrene
adducts as detectable (n = 56 cord blood samples) versus non-
detectable (n = 92); adjusted for sex, gestational age, maternal
education, maternal IQ, prenatal ETS, ethnicity, age, heating season,
prenatal demoralization, and HOME inventory
1
2
3
4
5
6
7
*Statistically significantly different from the control (p< 0.05).
DSM = Diagnostic and Statistical Manual of Mental Disorders; HOME = Home Observation for Measurement of the
Environment; IQ= intelligence quotient.
Table 1-4. Evidence pertaining to the neurodevelopmental effects of
benzo[a]pyrene in animals
Reference and study design
Results3
Cognitive fun ction
Chen etal. (2012)
Sprague-Dawley rats, 20 pups
(10 male and 10 female)/group
0, 0.02, 0.2, or 2 mg/kg-d by gavage
PNDsS-11
Hidden Platform test in Morris water maze:
Adolescent test period (PNDs 36-39): significant increase in escape
latency at 2 mg/kg-d only
Adult test period (PNDs 71-74): significant increase in escape
latency at >0.2 mg/kg-d
Increases in latency were already ~30% greater than controls at
2 mg/kg-d on the first trial day (i.e., on PND 36 or 71)
All experimental groups exhibited similar improvements in escape
latency, as slopes were visually equivalent across the 4 trial days
Probe test in the Morris water maze (d 5):
Time spent in the target quadrant:
PND 40: significant decrease at 2 mg/kg-d only
PND 75: significant decrease at >0.2 mg/kg-d
Number of platform crossings:
PND 40: significant decrease at 2 mg/kg-d only
PND 75: significant decrease at >0.2 mg/kg-d (in females) and
This document is a draft for review purposes only and does not constitute Agency policy,
1-15 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Reference and study design
Results3
2 mg/kg-d (in males)
Bouayed et al. (2009a)
Female Swiss albino mice, 5/group
0, 2, or 20 mg/kg-d maternal
gavage
PNDs 0-14 (lactational exposure)
Significant increase in the percent of spontaneous alternations in the Y-
maze alternation test at 2 mg/kg-d but not at 20 mg/kg-d
No effect on the total number of arm entries in the Y-maze alternation test
Neuromuscularfunction, coordination, and sensohmotor development
Chen etal. (2012)
Sprague-Dawley rats, 20 pups
(10 male and 10 female)/group
0, 0.02, 0.2, or 2 mg/kg-d by gavage
PNDs 5-11
Latency in the surface righting reflex test
PND 12: significant increase at 0.2 mg/kg-d only
PND 14: significant increase at 0.02 and 2 mg/kg-d only
PND 16: significant difference at 2 mg/kg-d only
PND 18: no significant difference
Latency in the negative geotaxis test
PND 12: significant increase at all doses
PND 14: significant increase at 2 mg/kg-d only
PNDs 16 and 18: no significant difference
No effect on duration of forelimb grip in forelimb grip strength test
No effect on the latency to retract from the edge in cliff aversion test
Note: Males and females were pooled for all analyses
Bouayed et al. (2009a)
Female Swiss albino mice, 5/group
0, 2, or 20 mg/kg-d by maternal
gavage
PNDs 0-14 (lactational exposure)
Significant increase in righting time in the surface righting reflex test at both
doses on PNDs 3 and 5 (but not PNDs 7 and 9)
Significant increase in latency in the negative geotaxis time for 20 mg/kg-d
dose group at PNDs 5, 7, and 9 (no significant difference at PND 11)
Significant increase in duration of forelimb grip in forelimb grip strength
test at both dose groups on PND 9 (statistically significant at PND 11 only at
high dose)
Significant increase in pole grasping latency in male pups in the water
escape pole climbing test at 20 mg/kg-d
No effect on climbing time in the water escape pole climbing test
Significant increase in pole escape latency in the water escape pole climbing
test in male rats at 20 mg/kg-d
Anxiety and/or motor activity
Chen etal. (2012)
Sprague-Dawley rats, 20 pups
(10 male and 10 female)/group
0, 0.02, 0.2, or 2 mg/kg-d by gavage
PNDs 5-11
Elevated plus maze:
Significant increase in the number of entries into open arms at PND
70 at >0.2 mg/kg-d (in females) and 2 mg/kg-d (in males) (no
difference at PND 35)
Significant decrease in the number of entries into closed arms at
PND 70 at >0.2 mg/kg-d (in females) and 2 mg/kg-d (in males) (no
This document is a draft for review purposes only and does not constitute Agency policy,
1-16 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Reference and study design
Results3
difference at PND 35)
Significant increase in the time spent in open arms at PND 35 at
2 mg/kg-d in females and at PND 70 at doses >0.02 mg/kg-d in
females and >0.2 mg/kg-d in males
Significant decrease in latency time to first enter an open arm on
PND 70 at >0.2 mg/kg-d (no difference at PND 35)
No effect on the total number of arm entries between treatment
groups
(calculated by EPA from graphically reported open and closed arm
entries)
Open field test:
Significant increase in the number of squares: PND 34, significant
increase at 2 mg/kg-d; PND 69, significant increase at >0.02 mg/kg-
d (no difference at PNDs 18 and 20)
Significant increase in rearing activity at 0.2 mg/kg-d on PND 69 (no
difference at PNDs 18, 20, and 34)
Bouayed et al. (2009a)
Female Swiss albino mice, 5/group
0, 2, or 20 mg/kg-d by maternal
gavage
PNDs 0-14 (lactational exposure)
Elevated plus maze:
Significantly increased time in open arms at >2 mg/kg-d
Significantly increased percentage of entries into open arms at
>2 mg/kg-d
Significantly decreased entries into closed at 2 mg/kg-d, but not at
20 mg/kg-d
Significantly decreased latency time to enter an open arm at
20 mg/kg-d
No effect on the total number of arm entries
No significant effect of gender on performance was detected, so
males and females were pooled for analyses
Open field test:
No significant change in activity on PND 15, but data not provided
Electrophysiological changes
McCallister et al. (2008)
Long-Evans Hooded rats, 5-6/group
0 or 0.3 mg/kg-d by gavage
CDs 14-17
Statistically significant decreases in stimulus-evoked cortical neuronal
activity on PNDs 90-120
Reduction in the number of spikes in both the short and long latency
periods on PNDs 90-120 (numerical data not presented)
This document is a draft for review purposes only and does not constitute Agency policy,
1-17 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Reference and study design
Results3
Wormley et al. (2004)
F344 rats, 10 females/group
0 or 100 u.g/m3 by nose-only
inhalation for 4 hrs/d
CDs 11-21
Electrophysiological changes in the hippocampus:
Consistently lower long term potentiation following gestational
exposure (statistical analysis not reported)
% change relative to control: -26%
Note: significant fetal toxicity observed (99 versus 34% birth index)
1
2
a% change from control calculated as: (treated value - control value)/control value x 100.
100
10 ,
o
Q
0.1 -.
0.01
ILOAEL ANOAEL • Doses > LOAEL O Doses < NOAEL
• A A
y. Q
I?
j~- LTi "C3
O ,/! CJ
= 0 !=
•5 "S
Q '=
c Q '-
5-|
p
5 Q '5
3? 3
O r-
CD =
elevated plus
maze test
4
5
6
7
8
9
10
11
12
13
Figure 1-2. Exposure-response array for neurodevelopmental effects
following oral exposure.
Mode of Action Analysis—Developmental Toxicity and Neurodevelopmental Toxicity
Data regarding the potential mode of action for the various manifestations of developmental
toxicity associated with benzo[a]pyrene exposure are limited, and the mode of action for
developmental toxicity is not known. General hypothesized modes of action for the various
observed developmental effects include, but are not limited to, altered cell signaling, cytotoxicity,
and oxidative stress.
It is plausible that developmental effects of benzo[a]pyrene may be mediated by altered cell
signaling through the AhR. Benzo[a]pyrene is a ligand for the AhR, and activation of this receptor
This document is a draft for review purposes only and does not constitute Agency policy,
1-18 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 regulates downstream gene expression including the induction of GYP enzymes important in the
2 conversion of benzo[a]pyrene into reactive metabolites. Studies in AhR knock-out mice indicate
3 that AhR signaling during embryogenesis is essential for normal liver, kidney, vascular,
4 hematopoietic, and immune development [Schmidt etal.. 1996: Fernandez-Salguero etal.. 1995).
5 In experiments in AhR responsive and less-responsive mice, the mice with the less-responsive AhR
6 were protected from renal injury as adults following gavage treatment with 0.1 or 0.5 mg/kg-day
7 benzo[a]pyrene from GD 10 to 13. Renal injury was indicated by an increase in urinary albumin
8 and a decrease in glomerular number [Nanez etal.. 2011].
9 Low birth weight has been associated with prenatal exposure to PAHs in human
10 populations [Perera et al., 2005b]. Several epidemiology studies have revealed an inverse
11 association between low birth weight and increased blood pressure, hypertension, and measures of
12 decreased renal function as adults [Zandi-Nejadetal.. 2006). It has been hypothesized that this
13 may be attributable to a congenital nephron deficit associated with intrauterine growth restriction
14 [Zandi-Nejad et al.. 2006].
15 No clear mode(s) of action for the observed neurodevelopmental changes following
16 benzo[a]pyrene exposure have been demonstrated. General hypothesized mechanisms with
17 limited support are related to altered central nervous system neurotransmission. These
18 mechanisms involve altered neurotransmitter gene expression, neurotransmitter levels, and
19 neurotransmitter receptor signaling in regions associated with spatial learning, anxiety, and
20 aggression, such as the hippocampus, striatum, amygdala, andhypothalamus [Li etal.. 2012: Oiu et
21 al.. 2011: Tang etal.. 2011: Xia etal.. 2011: Bouayed etal.. 2009a: Grova etal.. 2008: Brown etal..
22 2007: Grova etal.. 2007: Stephanou etal.. 1998).
23 Mechanistic studies in rodents exposed as adults, which exhibit some of the same
24 behavioral changes as animals exposed during development, may also inform potential mode(s) of
25 action for the observed neurodevelopmental changes. Specifically regarding potential changes in
26 spatial learning and memory processes, multiple studies in developing [Li etal.. 2012: McCallister
27 etal.. 2008: Wormley etal.. 2004] and adult [Maciel etal.. 2014: Oiu etal.. 2013: Tang etal.. 2011:
28 Grova etal.. 2008: Grova etal.. 2007] rodents suggest that changes in N-methyl-D-aspartate
29 (NMDA] receptor signaling seen with benzo[a]pyrene exposure (e.g., changes in expression
30 patterns of NR2A and NR2B subunits] may be responsible for apparent effects on learning and
31 memory.
32 In relation to potential changes in anxiety-like behaviors (and also relevant to effects on
33 learning and memory processes], many commonly used anti-anxiety medications work by
34 increasing brain serotonin levels (e.g., selective serotonin reuptake inhibitors], increasing brain
35 dopamine levels (e.g., dopamine reuptake inhibitors], or by targeting gamma-aminobutyric acid
36 (GABA] receptors (e.g., benzodiazepines]. Although GABAA receptor messenger ribonucleic acid
37 (mRNA] in whole-brain homogenates was unchanged following lactational benzo[a]pyrene
38 exposure, exposure at >2 mg/kg-day from PND 1 to 14 caused dose-dependent decreases in
This document is a draft for review purposes only and does not constitute Agency policy.
1-19 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 serotonin receptor (5HTiA) expression [Bouayed et al., 2009a: Stephanou etal., 1998]. Additional
2 support for identifying changes in monoamine neurotransmitter signaling (serotonin and dopamine
3 signaling, in particular) as a potential mechanism(s) in the altered anxiety-like behaviors observed
4 following benzo[a]pyrene exposure is provided in multiple studies of rodents exposed as adults
5 f Bouayed etal.. 2012: Oiu etal.. 2011: Xia etal.. 2011: Stephanou etal.. 1998: Tayasekaraetal..
6 1992] and in a single study of blood neurotransmitter levels in occupationally-exposed men [Niu et
7 al.. 2010]. Overall, these data suggest possible effects of benzo[a]pyrene exposure on NMDA
8 receptor expression and regulation of monoamine neurotransmitters including serotonin and
9 dopamine, but these findings require additional studies to clarify and extend understanding of
10 these events.
11 Summary of Developmental Effects
12 Developmental effects following in utero exposure to PAH mixtures or benzo[a]pyrene
13 alone have been reported in humans and in animal models. In human populations, decreased head
14 circumference, decreased birth weight, and decreased postnatal weight have been reported.
15 Analogous effects in laboratory animals, including decreased pup weight and decreased fetal
16 survival, have been noted following gestational or early postnatal exposure to benzo[a]pyrene by
17 the oral or inhalation route [Chen etal.. 2012: Archibongetal.. 2002: Mackenzie and Angevine.
18 1981]. Reproductive function is also altered in mice treated gestationally with benzo[a]pyrene
19 [Kristensen etal.. 1995: Mackenzie and Angevine. 1981]. These effects include impaired
20 reproductive performance in Fl offspring (male and female] and alterations of the weight and
21 histology of reproductive organs (ovaries and testes].
22 The available human and animal data also support the conclusion thatbenzo[a]pyrene is a
23 developmental neurotoxicant. Human studies of environmental PAH exposure in two cohorts have
24 observed neurotoxic effects, including suggestions of reduced head circumference (Tang etal..
25 2006: Perera etal.. 2005b: Perera etal.. 20041. impaired cognitive ability fPerera etal.. 2009: Tang
26 etal.. 2008]. impaired neuromuscular function (Tang etal.. 2008]. and increased attention
27 problems and anxious/depressed behavior following prenatal exposure (Perera etal.. 2012b].
28 These effects were seen in birth cohort studies in different populations (New York City and China],
29 in studies using specific benzo[a]pyrene measures (i.e., adduct levels measured in cord blood
30 samples] (Perera et al.. 2012b: Tang etal.. 2008: Tang etal.. 2006: Perera et al.. 2005b: Perera etal..
31 2004]. The available evidence from mice and rats also demonstrates significant and persistent
32 developmental impairments following exposure to benzo[a]pyrene. Impaired learning and
33 memory behaviors, decreased anxiety-like behaviors, and impaired neuromuscular function were
34 consistently observed in multiple neurobehavioral tests in two separate species at comparable oral
35 doses, and in the absence of maternal or neonatal toxicity (Chen etal.. 2012: Bouayed et al.. 2009a].
36 A decrease in anxiety, indicative of a change in nervous system function, can impair an
37 organism's ability to react to a potentially harmful situation. This decreased ability of an organism
38 to adapt to the environment is considered to be an adverse effect according to EPA's Neurotoxicity
This document is a draft for review purposes only and does not constitute Agency policy.
1-20 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 guidelines [U.S. EPA, 1998]. EPA's Developmental Toxicity guidelines state that an alteration in a
2 functional outcome following developmental exposure indicates the potential for altered
3 development in humans, although the types of developmental effects in animals will not necessarily
4 be the same as those produced in humans [U.S. EPA. 1991c].
5 Studies in humans also suggest that behavior is adversely affected by benzo[a]pyrene
6 exposure. In studies where symptoms were self-reported, exposure to benzo[a]pyrene was
7 positively associated with increased symptoms of anxiety/depression and/or attention problems in
8 prenatally-exposed children [Pereraetal.. 2012b]. In addition, one study of an occupationally-
9 exposed cohort of coke oven workers also reported increased symptoms of anxiety/depression
10 and/or attention problems [Qiu etal., 2013]: however, these results were inconsistent with a
11 second cohort of occupationally exposed male workers [Niu etal.. 2010). While these results could
12 suggest that benzo[a]pyrene exposure may adversely affect anxiety-related processes in a different
13 manner across species (i.e., increasing anxiety in humans, but decreasing anxiety in rodents),
14 additional studies in humans are necessary to draw such a conclusion. Importantly, however, it is
15 recognized that animals and humans may not necessarily experience the same effects or functional
16 changes, and that limitations in the information available across species can sometimes prevent
17 endpoint-specific comparisons [Francis etal., 1990].
18 In conclusion, EPA identified developmental toxicity and developmental neurotoxicity as
19 human hazards of benzo[a]pyrene exposure.
20 Susceptible Populations and Lifestages
21 Childhood susceptibility to benzo[a]pyrene toxicity is indicated by epidemiological studies
22 reporting associations between adverse birth outcomes and developmental effects and internal
23 biomarkers of exposure to benzo[a]pyrene, presumably via exposure to complex PAH mixtures
24 fPereraetal..2012b: Pereraetal.. 2009: Tang etal.. 2008: Tang etal.. 2006: Pereraetal.. 2005b:
25 Pereraetal.. 2005a: Pereraetal.. 2004). The occurrence of benzo[a]pyrene-specific DNAadducts in
26 maternal and umbilical cord blood in conjunction with exposure to ETS was associated with
27 reduced birth weight and head circumference in offspring of pregnant women living in New York
28 City [Pereraetal.. 2005b]. In other studies, elevated levels of BPDE-DNA adducts in umbilical cord
29 blood were associated with: [1] reduced birth weights or reduced head circumference [Perera et
30 al., 2005a: Perera etal., 2004]: and (2] decreased body weight at 18, 24, and 30 months [Tang etal.,
31 2008: Tang etal.. 20061
32 Studies in humans and experimental animals indicate that exposure to PAHs in general, and
33 benzo[a]pyrene in particular, may impact neurological development. Observational studies in
34 humans have suggested associations between gestational exposure to PAHs and later measures of
35 neurodevelopment [Pereraetal.. 2009: Tang etal.. 2008]. In the Perera et al. [2009] study, the
36 exposure measures are based on a composite of eight PAHs measured in air. In Tang etal., [2008],
37 increased levels of benzo[a]pyrene-DNA adducts in cord blood were associated with decreased
38 developmental quotients in offspring [Tang etal.. 2008].
This document is a draft for review purposes only and does not constitute Agency policy.
1-21 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Evidence in animals of the effects of benzo[a]pyrene on neurological development includes:
2 (1] decrements in reflex-related behaviors associated with neuromuscular coordination and
3 sensorimotor function [Chenetal.. 2012: Bouayedetal.. 2009a]: (2) disrupted learning and/or
4 short-term memory processes [Chenetal.. 2012: Lietal.. 2012: Shengetal.. 2010: Bouayedetal..
5 2009a); and (3) decreased anxiety-related responses [Chenetal.. 2012: Bouayedetal.. 2009a].
6 Mechanistic studies also support findings of benzo[a]pyrene induced alterations in
7 electrophysiological response to stimulation of the dentate gyrus of the hippocampus [Wormley et
8 al.. 2004: Wu etal.. 2003a] and decreased evoked response in the field cortex [McCallister etal..
9 20081.
10 1.1.2. Reproductive Toxicity
11 Human and animal studies provide evidence for benzo[a]pyrene-induced male and female
12 reproductive toxicity. Effects on sperm quality and male fertility have been demonstrated in human
13 populations highly exposed to PAH mixtures [Spares and Melo. 2008: Hsu etal.. 2006]. The use of
14 internal biomarkers of exposure in humans (e.g., BPDE-DNA adducts) support associations between
15 benzo[a]pyrene exposure and these effects. In females, numerous epidemiological studies indicate
16 that cigarette smoking reduces fertility; however, few studies have specifically examined levels of
17 benzo[a]pyrene exposure and female reproductive outcomes. Animal studies demonstrate
18 decrements in sperm quality, changes in testicular histology, and hormone alterations following
19 benzo[a]pyrene exposure in adult male animals, and decreased fertility and ovotoxic effects in adult
20 females following exposure to benzo[a]pyrene.
21 Male Reproductive Effects
22 Fertility
23 Effects on male fertility have been demonstrated in populations exposed to mixtures of
24 PAHs. Spermatozoa from smokers have reduced fertilizing capacity, and embryos display lower
25 implantation rates [Spares and Melo. 2008). Occupational PAH exposure has been associated with
26 higher levels of PAH-DNA adducts in sperm and male infertility [Gaspari et al.. 2003]. In addition,
27 men with higher urinary levels of PAH metabolites have been shown to be more likely to be infertile
28 [Xiaetal., 2009]. Studies were not identified that directly examined the reproductive capacity of
29 adult animals following benzo[a]pyrene exposure. However, a dose-related decrease in fertility
30 was observed in male mice treated in utero with benzo[a]pyrene, as discussed in Section 1.1.1.
31 Sperm parameters
32 Effects on semen quality have been demonstrated in populations exposed to mixtures of
33 PAHs including coke oven workers and smokers [Spares and Melo, 2008: Hsu etal., 2006]. Coke
34 oven workers had higher frequency of oligospermia (19 versus 0% in controls] and twice the
35 number of morphologically abnormal sperm (Hsu etal.. 2006]. Elevated levels of BPDE-DNA
This document is a draft for review purposes only and does not constitute Agency policy.
1-22 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 adducts have been measured in the sperm of populations exposed to PAHs occupationally [Gaspari
2 etal.. 20031 and through cigarette smoke fPhillips. 2002: Zenzes etal.. 19991 A higher
3 concentration of BPDE-DNA adducts was observed in sperm not selected for intrauterine
4 insemination or in vitro fertilization based on motility and morphology in patients of fertility clinics
5 [Perrinetal.. 201 Ib: Perrinetal.. 2011a). An association between benzo[a]pyrene exposure levels
6 and increased sperm DNA fragmentation using the sperm chromatin structure assay was observed
7 by Rubes etal. [2010]. However, it is currently unclear whether the sperm chromatin structure
8 assay, which measures sperm fragmentation following denaturation, is predictive of fertility
9 [Sakkas and Alvarez. 2010: ASRM. 20081.
10 In several studies in rats and mice, a decrease in sperm count, motility, and production and
11 an increase in morphologically abnormal sperm have been reported (Table 1-5 and Figure 1-3).
12 Alterations in these sperm parameters have been observed in different strains of rats and mice and
13 across different study designs and routes of exposure.
14 Decreases in epididymal sperm counts (25-50% compared to controls) have been reported
15 in Sprague-Dawley rats and C57BL6 mice treated with 1-5 mg/kg-day benzo[a]pyrene by oral
16 exposure for 42 or 90 days (Chen etal., 2011: Mohamed etal., 2010). Another subchronic study
17 noted a 44% decrease in epididymal sperm concentration at doses >50 mg/kg-day in Hsd:ICR
18 (GDI) mice (Tengetal.. 2013). Additionally, a 15% decrease in epididymal sperm count was
19 observed at a much lower dose in Sprague-Dawley rats exposed to benzo[a]pyrene for 90 days
20 (Chung etal.. 2011). However, confidence in this study is limited because the authors dosed the
21 animals with 0.001, 0.01, and 0.1 mg/kg-day benzo[a]pyrene, but only reported on sperm
22 parameters at the mid-dose, and no other available studies demonstrated findings in the range of
23 the mid- and high-dose. In rats, an oral short-term study and a subchronic inhalation study lend
24 support for the endpoint of decreased sperm count (Arafaetal., 2009: Archibong et al., 2008:
25 Ramesh et al.. 2008). Significantly decreased sperm count and daily sperm production (20-40%
26 decrease from control in each parameter) were observed in rats following 10 days of gavage
27 exposure to 50 mg/kg-day (Arafaetal.. 2009) and following gavage dosing with 10 mg/kg-day on
28 PNDs 1-7 (Liang etal.. 2012). In addition, a 69% decrease from controls in sperm count was
29 observed in rats following inhalation exposure to 75 |ig/m3 benzo[a]pyrene for 60 days (Archibong
30 etal.. 2008: Ramesh etal.. 20081.
31 Both oral and inhalation exposure of rodents to benzo[a]pyrene have been shown to lead to
32 decreased epididymal sperm motility and altered morphology. Decreased motility of 20-30%
33 compared to controls was observed in Hsd:ICR (GDI) mice (>100 mg/kg-day), C57BL6 mice (>1
34 mg/kg-day), and Sprague-Dawley rats (0.01 mg/kg-day) following subchronic oral exposure (Teng
35 etal.. 2013: Chung etal.. 2011: Mohamed etal.. 2010). The effective doses spanned several orders
36 of magnitude; however, as noted above, reporting is limited in the study that observed effects at
37 0.01 mg/kg-day benzo[a]pyrene (Chung etal., 2011). A short-term oral study in rats also reported
38 a significantly decreased number of motile sperm (~40% decrease) following 10 days of gavage
This document is a draft for review purposes only and does not constitute Agency policy.
1-23 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 exposure to 50 mg/kg-day benzo[a]pyrene [Arafaetal., 2009]. In addition, decreased sperm
2 motility was observed following inhalation exposure to 75 [ig/m3 benzo[a]pyrene in rats for 60
3 days [Archibongetal.. 2008: Rameshetal.. 2008] and to >75 [ig/m3 for 10 days [Inyangetal..
4 2003]. Abnormal sperm morphology was observed in Sprague-Dawley rats treated with
5 5 mg/kg-day benzo[a]pyrene by gavage for 84 days [Chenetal.. 2011] and in rats exposed to
6 75 [J.g/m3 benzo[a]pyrene by inhalation for 60 days [Archibongetal.. 2008: Ramesh et al.. 2008].
7 Testicular changes
8 Several studies have demonstrated dose-related effects on male reproductive organs in
9 adult animals exposed subchronically to benzo[a]pyrene (Table 1-5 and Figure 1-3]. Decreases in
10 testicular weight of approximately 35% have been observed in a 60-day gavage study in Hsd:ICR
11 (GDI] mice at 100 mg/kg-day [Tengetal.. 2013]. in a 10-day gavage study in adult Swiss albino rats
12 at 50 mg/kg-day [Arafaetal.. 2009]. and following subchronic inhalational exposure of adult F344
13 rats to 75 [ig/m3 [Archibong et al.. 2008: Ramesh et al.. 2008]. No effects on testes weight were
14 observed in Wistar rats exposed for 35 days to gavage doses up to 50 mg/kg-day [Kroese etal..
15 2001], F344 rats exposed for 90 days to dietary doses up to 100 mg/kg-day [Knuckles etal., 2001],
16 or Sprague-Dawley rats exposed for 90 days to gavage doses up to 0.1 mg/kg-day [Chung etal.,
17 2011]. Strain differences may have contributed to differences in response; however, F344 rats
18 exposed to benzo[a]pyrene via inhalation showed effects on testicular weight [Archibongetal..
19 2008: Ramesh et al.. 2008]. In addition, decreased testicular weight has also been observed in
20 offspring following in utero and early postnatal exposure to benzo[a]pyrene as discussed in Section
21 1.1.1.
22 Histological changes in the testis have often been reported to accompany decreases in
23 testicular weight. Apoptosis, as evident by increases in terminal deoxynucleotidyl transferase dUTP
24 nick end labeling [TUNEL] positive germ cells and increases in caspase-3 staining, was evident in
25 seminiferous tubules of Sprague-Dawley rats following 90 days of exposure to >0.001 and
26 0.01 mg/kg-day, respectively, benzo[a]pyrene by gavage [Chung etal.. 2011]. However, the study
27 authors did not observe testicular atrophy or azospermia in any dose group. Seminiferous tubules
28 were reported to look qualitatively similar between controls and animals exposed to
29 benzo[a]pyrene by inhalation doses of 75 [ig/m3 for 60 days [Archibongetal., 2008: Ramesh etal.,
30 2008]. However, when histologically examined, statistically significantly reduced tubular lumen
31 size and length were observed in treated animals. Seminiferous tubule diameters also appeared to
32 be reduced in exposed animals, although this difference did not reach statistical significance
33 [Archibong et al.. 2008: Ramesh etal.. 2008]. In addition, histological changes in the seminiferous
34 tubules have also been observed in offspring following in utero exposure to benzo[a]pyrene as
35 discussed in Section 1.1.1.
This document is a draft for review purposes only and does not constitute Agency policy.
1-24 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Epididymal changes
2 In addition to testicular effects, histological effects in the epididymis have been observed
3 following 90-day gavage exposure to benzo[a]pyrene [Chung etal.. 2011] (Table 1-5 and Figure 1-
4 3). Specifically, statistically significant decreased epididymal tubule diameter (for caput and cauda)
5 was observed at doses >0.001 mg/kg-day. At the highest dose tested (0.1 mg/kg-day), diameters
6 were reduced approximately 25%. A 60 day gavage study in Hsd:ICR(CDl) mice observed a 27%
7 decrease in cauda epididymis weight at 100 mg/kg-day (Jengetal.. 2013]: however, no change in
8 epididymis weight was observed following an 84-day treatment in Sprague-Dawley rats of 5
9 mg/kg-day benzo[a]pyrene (Chenetal., 2011].
10 Hormone changes
11 Several animal models have reported decreases in testosterone following both oral and
12 inhalation exposure to benzo[a]pyrene (Table 1-5 and Figure 1-3]. In male Sprague-Dawley rats,
13 decreases in testosterone have been observed following 90-day oral exposures (Chung etal.. 2011:
14 Zheng etal., 2010]. Statistically significant decreases of 15% in intratesticular testosterone were
15 observed at 5 mg/kg-day in one study (Zheng etal., 2010], while a second study in the same strain
16 of rats reported statistically significant decreases of approximately 40% in intratesticular
17 testosterone and 70% in serum testosterone at 0.1 mg/kg-day (Chung etal.. 2011]. In addition,
18 Sprague-Dawley rats treated with 10 mg/kg-day by gavage on PNDs 1-7 exhibited statistically
19 significantly decreased serum testosterone (>40%] when examined at PND 8 and PND 35 (Liang et
20 al.. 2012]. Statistically significant decreases in intratesticular testosterone (80%] and serum
21 testosterone (60%] were also observed following inhalation exposure to 75 |ig/m3 benzo[a]pyrene
22 in F344 rats for 60 days (Archibongetal., 2008: Ramesh etal., 2008]. Statistically significant
23 increases in serum luteinizing hormone (LH] have also been observed in Sprague-Dawley rats
24 following gavage exposure to benzo[a]pyrene at doses of >0.01 mg/kg-day (Chung etal.. 2011] and
25 in F344 rats following inhalation exposure to 75 |ig/m3 benzo[a]pyrene for 60 days (Archibong et
26 al.. 2008: Ramesh etal.. 2008].
27 Table 1-5. Evidence pertaining to the male reproductive toxicity of
28 benzo[a]pyrene in adult animals
Reference and study design
Results
Sperm quality
Mohamed et al. (2010)
C57BL/6 mice, 10 males/dose (treated
before mating with unexposed females)
0,1, or 10 mg/kg-d by gavage (FO males
only)
42 d
•^ epididymal sperm count in FO mice
Approximate % change from control (data reported graphically)
0, -50*, and -70*
4, epididymal sperm motility in FO mice
Approximate % change from control (data reported graphically):
0, -20*, and -50*
This document is a draft for review purposes only and does not constitute Agency policy,
1-25 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Reference and study design
Results
4/ epididymal sperm count in untreated Fl and F2 generations (data
reported graphically)
No effects were observed in the F3 generation
Chen etal. (2011)
Sprague-Dawley rats, 10 males/dose
0 or 5 mg/kg-d by gavage
84 d
epididymal sperm count (% change from control)
0 and-29*
% abnormal epididymal sperm
5 and 8*
Chung etal. (2011)
Sprague-Dawley rats, 20-25 males/dose
0, 0.001, 0.01, or 0.1 mg/kg-d by gavage
90 d
4/ epididymal sperm motility (% change relative to control; reported
only for 0.01 mg/kg-d)
0 and -30*
No statistically significant decrease in epididymal sperm count
(Archibong et al. (2008); Ramesh et al.
(2008))
F344 rats, 10 males/group
0 or 75 u.g/m3, 4 hrs/d by inhalation
60 d
•^ epididymal sperm motility (% change from control)
0 and-73*
4, epididymal sperm count (% change from control)
0 and-69*
/T" % abnormal epididymal sperm
33 and 87*
4, spermatids/g testis (approximate % change from control;
numerical data not reported)
0 and -45*
Testicular changes (weight, histology)
Mohamed et al. (2010)
C57BL/6 mice, 10 males/dose (treated
before mating with unexposed females)
0,1, or 10 mg/kg-d by gavage (FO males
only)
42 d
4/ seminiferous tubules with elongated spermatids (approximate %
change from control; numerical data not reported)
0, -20*, and -35*
No statistically significant change in area of seminiferous epithelium
of testis (approximate % change from control; numerical data not
reported)
0, 5, and 20
Chung etal. (2011)
Sprague-Dawley rats, 20-25 males/dose
0, 0.001, 0.01, or 0.1 mg/kg-d by gavage
90 d
/T" number of apoptoticgerm cells per tubule (TUNELor caspase 3
positive)
No change in testis weight or histology
Chen etal. (2011)
Sprague-Dawley rats, 10/dose
0 or 5 mg/kg-d by gavage
84 d
1" testicular lesions characterized as irregular arrangement of germ
cells and absence of spermatocytes (numerical data not reported)
No change in testis weight
(Archibong et al. (2008); Ramesh et al.
(2008))
F344 rats, 10 adult males/group
decreased testis weight (% change from control)
0 and 34*
This document is a draft for review purposes only and does not constitute Agency policy,
1-26 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Reference and study design
0 or 75 u.g/m3, 4 hrs/d by inhalation
60 d
Kroese et al. (2001)
Wistar rats, 10/sex/dose
0, 1.5, 5, 15, or 50 mg/kg-d by gavage,
5d/wk
35 d
Knuckles etal. (2001)
F344 rats, 20/sex/dose
0, 5, 50, or 100 mg/kg-d in diet
90 d
Epididymal changes (weight, histology)
Chung etal. (2011)
Sprague-Dawley rats, 20-25 males/dose
0, 0.001, 0.01, or 0.1 mg/kg-d by gavage
90 d
Chen etal. (2011)
Sprague-Dawley rats, 10/dose
0 or 5 mg/kg-d by gavage
84 d
Hormone changes
Chung etal. (2011)
Sprague-Dawley rats, 20-25 males/dose
0, 0.001, 0.01, 0.1 mg/kg-d by gavage
90 d
Zheng etal. (2010)
Sprague-Dawley rats, 8 males/dose
0, 1, or 5 mg/kg-d by gavage
90 d
(Archibong et al. (2008); Ramesh et al.
(2008))
F344 rats, 10 adult males/group
0 or 75 u.g/m3, 4 hrs/d by inhalation
Results
•i, size of seminiferous tubule lumens and reduced tubular length
No change in % of tubules with elongated spermatids
No change in testis weight
No change in testis weight
•i, diameter of caput epididymal tubule (n = 5; numerical data not
reported)
•^ diameter of cauda epididymal tubule (n = 5; numerical data not
reported)
No change in epididymis weight
•i, Intratesticular testosterone (approximate % change from control;
data reported graphically)
0, -12, -25, and -40*
•^ Serum testosterone (approximate % change from control;
numerical data not reported)
0, 0, -35, and -70*
/T" serum LH (approximate % change from control; numerical data not
reported)
0,33, 67*, and 87*
•i, Human chorionic gonadotropin (hCG) or dibutyl cyclic adenosine
monophosphate (dbcAMP)-stimulated testosterone production in
Leydig cells
•^ Intratesticular testosterone (approximate % change from control;
numerical data not reported)
0, -15, and -15*
•i, intratesticular testosterone (approximate % change from control;
numerical data not reported)
0 and -80*
This document is a draft for review purposes only and does not constitute Agency policy,
1-27 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2
3
4
5
6
Reference and study design Results
60 d 4, serum testosterone (approximate % change from control)
0 and -60*
1" serum LH (approximate % change from control)
0 and 50*
*Statistically significantly different from the control (p< 0.05).
a% change from control calculated as: (treated value - control value)/control value x 100.
100 :
1 10 =
CUD
-Si
E ! '-
o
0
0.1 =
0.01 =
n nm
A NUALL
o Doses < NOAEL
=•=
B B JL
-1- =t±=
T T T
^ i= ^ _i i_
O __i ^ ^H Wl . t/l ^H yi ^H yi O ^
o S g2 32 g2 32 32 32
r-j^ r\l'* rvj . fc rsj ' " r\j.* fN-* rN-*
_ d >• -.>- >- ^>- -.>- ^>*
: "T3 ^~O "~O :~O ^~O :~C
<"Q ^^ ~3^ 3 fU ^ ~7^ 3 fO ^ <^ ^ fO 3
If || || g| || || ||
~ -JS i^i^ _c o 1*1 ^ _co _co ^=0
i^V "^ 00 LJCTi WDO O O LJCTi MC^i
{9
sU epididymal -J, epididymal %b epididymal Irregular ^/ TT (serum ^ LH \|/ TT
sperm count sperm count sperm motility germ cell and (intratesticular)
and motility INnatanormal organization intratesticular}
sperm
Sperm Quality Testicular Hormone Changes
Effects
Figure 1-3. Exposure-response array for male reproductive effects following
oral exposure in adult animals.
This document is a draft for review purposes only and does not constitute Agency policy,
1-28 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Mode-of-action analysis—male reproductive effects
2 Exposure to benzo[a]pyrene in laboratory animals induces male reproductive effects
3 including decreased levels of testosterone and increased levels of LH, decreased sperm quality, and
4 histological changes in the testis. Decrements in sperm quality and decreased fertility have also
5 been demonstrated in populations highly exposed to PAH mixtures [Spares and Melo. 2008: Hsu et
6 al.. 20061
7 Numerous studies have indicated that benzo[a]pyrene reduces testosterone levels following
8 oral or inhalation exposure [Chung etal., 2011: Zheng etal., 2010: Archibongetal., 2008: Ramesh et
9 al., 2008]. It is plausible that the effects on sperm quality and histological changes of the
10 reproductive organs are secondary to an insufficiency of testosterone [Inyangetal.. 2003). One
11 study has hypothesized that benzo[a]pyrene perturbs the production of testosterone by Leydig
12 cells [Chung etal.. 2011). This study found a statistically significant reduction in testicular
13 testosterone in rats treated with 0.1 mg/kg-day benzo[a]pyrene for 90 days and found that
14 testosterone production in isolated Leydig cells was also inhibited approximately 50%, even in
15 cultures stimulated with human chorionic gonadotropin and dibutryl cyclic adenosine
16 monophosphate.
17 Leydig cell function is thought to be regulated by testicular macrophages [Hales. 2002).
18 When testicular macrophages are activated and produce inflammatory mediators, Leydig cell
19 testosterone production is inhibited [Hales. 2002). Zheng etal. [2010] treated rats with 5 mg/kg-
20 day benzo[a]pyrene for 90 days and reported a statistically significant increase in ED-1 type
21 testicular macrophages and a statistically significant decrease in intratesticular testosterone.
22 Arafaetal. [2009] reported that male reproductive effects observed following
23 benzo[a]pyrene exposure could be ameliorated by antioxidant pre-treatment. This study reported
24 decreased sperm count, motility, and production, in addition to decreased testis weight following a
25 10 day oral administration in rats of 50 mg/kg-day benzo[a]pyrene. Pretreatment with the citris
26 flavonoid hesperidin protected rats from all of these effects except the decrease in sperm motility.
27 A study in tobacco smokers suggests that direct DNA damage from the reactive metabolite
28 BPDE may decrease sperm motility [Perrinetal.. 2011a]. In this study, motile sperm were
29 separated from non-motile sperm using a "swim-up" self-migration technique. The investigators
30 found that the motile sperm selected by this method had significantly fewer BPDE-adducts than
31 non-selected sperm.
32 Other hypothesized modes of action of the observed male reproductive effects include
33 benzo[a]pyrene-mediated DNA damage to male germ cells leading to genotoxicity, cytotoxicity, and
34 apoptosis [Chungetal.. 2011: Perrinetal.. 2011b: Perrinetal.. 201 la: Olsenetal.. 2010: Revel et
35 al.. 2001]. compromised function of Sertoli cells [Raychoudhury and Kubinski. 2003]. and
36 decreased embryo viability post-fertilization associated with sperm DNA damage [Borinietal.,
37 2006: Seli etal.. 20041
This document is a draft for review purposes only and does not constitute Agency policy.
1-29 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Female Reproductive Effects
2 Fertility
3 In women, exposure to cigarette smoke has been shown to affect fertility, including effects
4 related to pregnancy, ovulatory disorders, and spontaneous abortion (as reviewed in Waylen etal..
5 2009: Cooper and Moley. 2008: Spares and Melo. 2008). In addition, several studies suggest that in
6 utero exposure to maternal tobacco smoke also decreases the future fertility of female offspring [Ye
7 etal.. 2010: Jensen etal.. 1998: Weinbergetal.. 1989]. Benzo[a]pyrene levels in follicular fluid and
8 benzo[a]pyrene-DNA adducts in granulosa-lutein cells and oocytes and in human cervical cells have
9 been associated with smoking status and with amount smoked [Neal etal.. 2008: Mancinietal..
10 1999: Melikian etal.. 1999: Zenzes etal.. 1998: Shamsuddin and Can. 19881
11 Few epidemiological studies have examined the specific influence of components of PAH
12 mixtures on fertility or other reproductive outcomes; EPA identified only two studies with specific
13 data on benzo[a]pyrene (Table 1-6). One of these studies addressed the probability of conception
14 among women undergoing in vitro fertilization [Neal etal., 2008]. Follicular fluid benzo[a]pyrene
15 levels were significantly higher among the women who did not conceive compared with women
16 who did get pregnant. No association was seen between conception and serum levels of
17 benzo[a]pyrene. The other study examined risk of delayed miscarriage (fetal death before
18 14 weeks of gestation), using a case-control design with controls selected from women undergoing
19 elective abortion (Wu etal.. 2010). A strong association was seen between maternal blood
20 benzo[a]pyrene-DNA adduct levels and risk of miscarriage, with a fourfold increased risk for levels
21 above compared with below the median. Benzo[a]pyrene-DNA adduct levels were similar in the
22 aborted tissue of cases compared with controls.
23 Experimental studies in mice also provide evidence thatbenzo[a]pyrene exposure affects
24 fertility (Table 1-7 and Figure 1-4). Decreased fertility and fecundity (decreased number of FO
25 females producing viable litters at parturition) was statistically significantly reduced by about 35%
26 in adult females exposed to 160 mg/kg-day of benzo[a]pyrene (Mackenzie and Angevine. 1981). In
27 another study, FO females showed no signs of general toxicity or effects on fertility following gavage
28 exposure to 10 mg/kg-day on CDs 7-16 (Kristensen et al., 1995). Decrements infertility were
29 more striking in the offspring from these studies, as described in Section 1.1.1.
30 Ovarian effects
31 Human epidemiological studies that directly relate ovotoxicity and benzo[a]pyrene
32 exposure are not available; however, smoking, especially during the time of the peri-menopausal
33 transition, has been shown to accelerate ovarian senescence (Midgette and Baron. 1990).
34 Benzo[a]pyrene-induced ovarian toxicity has been demonstrated in animal studies. In adult female
35 rats treated by gavage, statistically significant, dose-related decreases in ovary weight have been
36 observed in female rats treated for 60 days at doses >5 mg/kg (2.5 mg/kg-day adjusted) (Xu etal..
37 2010). At 10 mg/kg in adult rats (5 mg/kg-day adjusted), ovary weight was decreased 15% (Xu et
This document is a draft for review purposes only and does not constitute Agency policy.
1-30 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 al., 2010]. Changes in ovary weight were not observed in two subchronic studies in rats.
2 Specifically, no changes in ovary weight were seen in Wistar rats exposed for 35 days to gavage
3 doses up to 50 mg/kg-day [Kroese etal.. 2001] or in F344 rats exposed for 90 days to dietary doses
4 up to 100 mg/kg-day (Knuckles etal.. 2001).
5 In adult female rats treated by gavage, dose-related decreases in the number of primordial
6 follicles have been observed in female rats treated for 60 days at doses >2.5 mg/kg-day, with a
7 statistically significant decrease of approximately 2 0% at the high dose (Xu etal.. 2010] (Table 1-7
8 and Figure 1-4). No notable differences in other follicle populations and corpora lutea were
9 observed. However, in utero studies exposing dams to the same doses produced offspring with few
10 or no follicles or corpora lutea (Kristensenetal., 1995: Mackenzie and Angevine, 1981]. Additional
11 support for the alteration of female reproductive endpoints comes from intraperitoneal (i.p.)
12 experiments in animals and in vitro experiments. Several studies have observed ovarian effects
13 (decreased numbers of ovarian follicles and corpora lutea, absence of folliculogenesis, oocyte
14 degeneration, and decreased fertility) in rats and mice exposed via i.p. injection (Borman etal..
15 2000: Miller etal.. 1992: Swartz and Mattison. 1985: Mattison etal.. 1980). Further evidence is
16 available from in vitro studies showing inhibition of antral follicle development and survival, as
17 well as decreased production of estradiol, in mouse ovarian follicles cultured with benzo[a]pyrene
18 for 13 days (Sadeu and Foster. 2011). Likewise, follicle stimulating hormone (FSH)-stimulated
19 growth of cultured rat ovarian follicles was inhibited by exposure to benzo[a]pyrene (Neal etal..
20 20071
21 Hormone levels
22 Alteration of hormone levels has been observed in female rats following oral or inhalation
23 exposure to benzo[a]pyrene (Table 1-7 and Figure 1-4). Inhalation exposure to
24 benzo[a]pyrene:carbon black particles during gestation resulted in decreases in plasma
25 progesterone, estradiol, and prolactin in pregnant rats (Archibongetal.. 2002). In addition,
26 statistically significant, dose-related decreases in estradiol along with altered estrus cyclicity was
27 observed in female rats treated for 60 days at doses >2.5 mg/kg-day by gavage (Xu etal.. 2010).
28 Mechanistic experiments have also noted decreased estradiol output in murine ovarian follicles
29 cultured with benzo[a]pyrene in vitro for 13 days, but did not find any decrease in progesterone
30 (Sadeu and Foster. 2011).
31 Cervical effects
32 One subchronic animal study is available that investigated effects in the cervix following
33 oral exposure to benzo[a]pyrene (Table 1-7 and Figure 1-4). Statistically-significant dose-related
34 increases in the incidence of cervical inflammatory cells were observed in mice exposed twice a
35 week for 98 days to benzo[a]pyrene via gavage at doses >2.5 mg/kg (Gao etal., 2011). Cervical
36 effects of increasing severity, including epithelial hyperplasia, atypical hyperplasia, apoptosis, and
37 necrosis, were observed at higher doses. There are no data on cervical effects in other species or in
This document is a draft for review purposes only and does not constitute Agency policy.
1-31 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 other mouse strains. Gao etal. [2011] considered the hyperplasia responses to be preneoplastic
2 lesions. Cervical neoplasia was not reported in the available chronic bioassays, but this tissue was
3 not subjected to histopathology examination in either bioassay [Kroese etal.. 2001: Beland and
4 Gulp. 1998). Thus, the relationship of the cervical lesions to potential development of neoplasia is
5 uncertain.
6
7
Table 1-6. Evidence pertaining to the female reproductive effects of
benzo[a]pyrene in humans
Reference and study design
Results
Probability of conception
Neal etal. (2008)
36 women undergoing in vitro fertilization
(19 smokers, 7 passive smokers, and
10 nonsmokers)
Exposure: benzo[a]pyrene in serum and follicular
fluid
Benzo[a]pyrene levels (ng/mL)
Follicular fluid
Serum
Conceived
0.1
0.01
Did not
conceive
1.7
0.05
p-value
<0.001
Not reported
Fetal death
Wu et al. (2010) (Tianjin, China)
Case control study: 81 cases (96% participation
rate)—fetal death confirmed by ultrasound before
14 wks gestation; 81 controls (91% participation
rate)—elective abortions; matched by age,
gestational age, and gravidity; excluded smokers
and occupational PAH exposure
Exposure: benzo[a]pyrene in aborted tissue and
maternal blood samples (51 cases and controls,
2 of 4 hospitals)
Benzo[a]pyrene adduct levels (/10s nucleotides), mean
(±SD)
Cases Controls p-value
Maternal blood 6.0 (±4.7) 2.7 (± 2.2) <0.001
Aborted tissue 4.8 (± 6.0) 6.0 (± 7.4) 0.29
Low correlation between blood and tissue levels (r = -0.02 in
cases, r = -0.21 in controls)
Association between benzo[a]pyrene adducts and
miscarriage3
OR 95% Cl
Per unit increase in adducts 1.37 1.12,1.67
Dichotomized at median 4.56 1.46,14.3
Conditional logistic regression, adjusted for maternal
education, household income, and gestational age; age also
considered as potential confounder
This document is a draft for review purposes only and does not constitute Agency policy,
1-32 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2
Table 1-7. Evidence pertaining to the female reproductive effects of
benzo[a]pyrene in adult animals
Reference and study design
Results3
Fertility
Mackenzie and Angevine (1981)
CD-I mice, 30 or 60 FO females/dose
0,10, 40, or 160 mg/kg-d by gavage
CDs 7-16
number of FO females with viable litters
46/60, 21/30, 44/60, and 13/30*
Kristensenetal. (1995)
NMRI mice, 9 females/dose
0 or 10 mg/kg-d by gavage
CDs 7-16
No changes in fertility of FO females
Ovarian effects (weight, histology, follicle numbers)
Xuetal. (2010)
Sprague-Dawley rats, 6 females/
dose
0, 5 or 10 mg/kg by gavage every
other day (2.5 and 5 mg/kg-d,
adjusted)
60 d
•^ ovary weight (% change from control)
0, -11*, and -15*
•^ number of primordial follicles (20%* decrease at high dose)
1" increased apoptosis of ovarian granulosa cells (approximate %
apoptosis)
2, 24*, and 14*
Knuckles etal. (2001)
F344 rats, 20/sex/dose
0, 5, 50, or 100 mg/kg-d in diet
90 d
No changes in ovary weight
Kroese et al. (2001)
Wistar rats, 10/sex/dose
0,1.5, 5,15, or 50 mg/kg-d by gavage
5d/wk
35 d
No changes in ovary weight
Hormone levels
Xuetal. (2010)
Sprague-Dawley rats, 6 females/
dose
0, 5, or 10 mg/kg by gavage every
other day (2.5 and 5 mg/kg-d,
adjusted)
60 d
•^ serum estradiol (approximate % change from control)
0, -16, and -25*
Altered estrous cyclicity
Archibong et al. (2002)
F344 rats, 10 females/group
0, 25, 75, or 100 u.g/m3 by inhalation
4 hrs/d
GDs 11-20 (serum hormones tested
at GD 15 and 17 in 0, 25, and
75 u.g/m3 dose groups)
4> FO estradiol, approximately 50% decrease at 75 |Jg/m3 at GD 17
•^ FO prolactin, approximately 70% decrease at 75 u.g/m3at GD 17
/T" FO plasma progesterone approximately 17% decrease at 75 u.g/m3 at
GD17
This document is a draft for review purposes only and does not constitute Agency policy,
1-33 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Reference and study design
Results3
Cervical effects
Gaoetal. (2011)
ICR mice, 26 females/dose
0, 2.5, 5, or 10 mg/kg by gavage
2d/wk
98 d
-t cervical epithelial hyperplasia: 0/26, 4/26, 6/25*, and 7/24*
T* cervical atypical hyperplasia: 0/26, 0/26, 2/25, and 4/24*
T* inflammatory cells in cervical epithelium: 3/26, 10/26, 12/25*, and
18/24*
1
2
3
*Statistically significantly different from the control (p< 0.05).
a% change from control calculated as: (treated value - control value)/control value x 100.
1000 T 1
**" 1 nn
CUD
(U
o
Q
10 -
i
3
L
<
1
/
_
A
L
\
I
3 A •
0
m .
• <
c
^H u^
00 Cl
O CTi „
^ 8
.E 01 ra .-
S 'E ° S
QD C C i
e .- oj r-
1? P
QJ O -i'
c
o
•^3
sl/FO female fertility
Fertility Effects
A '
j
V
• LOAEL
ANOAEL
• Doses > LOAEL
O Doses < NOAEL
^ A
o ^^ ^
)
^ »H (/I T— 1
0 CJ g - 0 «
,— | - O j_ O ^
o >; ^ ^ '- ; :
:• ^ "ro 3 "ra TD
® -g 2^ ^ ^ S"
X o O LTI t_) i
LO >; ro =! O
— c tJi
4^ ovarian weight
Ovarian Effects
O ra
T— 1
^ "S
*^ >-
o ^
x 6
4- ovarian
follicles
• •
T f
A M
$ m
A
o 2
^ "S
O ^3
x 6
4- serum
Estradiol
Hormone
Changes
o ^_
CJ >.
(5 ob
T cervical
epithelial
hyperplasia
Cervical
Effects
5
6
Figure 1-4. Exposure-response array for female reproductive effects following
oral exposure in adult animals.
This document is a draft for review purposes only and does not constitute Agency policy,
1-34 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Mode-of-action analysis—female reproductive effects
2 Although the mechanisms underlying female reproductive effects following benzo[a]pyrene
3 exposure are not fully established, associations with stimulation of apoptosis, impairment of
4 steroidogenesis, and cytotoxicity have been made. Ovarian lesions in benzo[a]pyrene-exposed rats
5 have been associated with increased apoptosis in ovarian granulosa cells and alteration in
6 hormone-mediated regulation of folliculogenesis [Xu etal.. 2010). and results from in vitro
7 experiments provide support for an association between benzo[a]pyrene exposure and impaired
8 folliculogenesis, steroidogenesis, and oocyte maturation [Sadeu and Foster, 2011: Nealetal., 2007].
9 A growing body of research suggests that benzo[a]pyrene triggers the induction of apoptosis in
10 oocytes through AhR-driven expression of pro-apoptotic genes, including Bax [Kee etal.. 2010: Neal
11 etal.. 2010: Pru etal.. 2009: Matikainen etal.. 2002: Matikainen etal.. 2001: Robles etal.. 20001
12 Other proposed mechanisms include the impairment of folliculogenesis from reactive metabolites
13 [Takizawa et al.. 1984: Mattison and Thorgeirsson. 1979.1977] or by a decreased sensitivity to
14 FSH-stimulated follicle growth [Neal etal.. 2007]. Based on findings that an ERa antagonist
15 counteracted effects of subcutaneously administered benzo[a]pyrene on uterine weight (decreased
16 in neonatal rats and increased in immature rats], interactions with ERa have been proposed,
17 possibly via occupation of ERa binding sites or via AhR-ER-crosstalk [Kummer et al.. 2008: Kummer
18 etal.. 2007]. However, several in vitro studies have demonstrated low affinity binding of
19 benzo[a]pyrene to the estrogen receptor and alteration of estrogen-dependent gene expression
20 [Liu etal.. 2006: van Lipzigetal.. 2005: Vondraceketal.. 2002: Fertucketal.. 2001: Charles etal..
21 2000]. so the role of the ER in benzo [a]pyrene-induced reproductive toxicity is unclear.
22 Summary of Reproductive Effects
23 Male reproductive effects
24 Exposure to benzo [a]pyrene in laboratory animals induces male reproductive effects
25 including decreased levels of testosterone and increased levels of LH, decreased sperm count and
26 motility, histological changes in the testis, and decreased reproductive success. These findings in
27 animals are supported by decrements in sperm quality and decreased fertility in human
28 populations exposed to PAH mixtures [Spares and Melo, 2008: Hsu etal., 2006]. In laboratory
29 animals, male reproductive toxicity has been observed after oral and inhalation exposure to rats or
30 mice. Effects seen after oral exposures include impaired fertility, effects on sperm parameters,
31 decreased reproductive organ weight, testicular lesions, and hormone alterations [Chen etal.. 2011:
32 Chung etal.. 2011: Mohamedetal.. 2010: Zheng etal.. 2010: Mackenzie and Angevine. 1981]. In
33 addition to oral exposure, male reproductive effects of benzo [a] pyrene have also been observed
34 following inhalation exposure in rats [Archibongetal., 2008: Rameshetal., 2008: Inyangetal.,
35 2003]. The male reproductive effects associated with benzo[a]pyrene exposure are considered to
36 be biologically plausible and adverse.
This document is a draft for review purposes only and does not constitute Agency policy.
1-35 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 In conclusion, EPA identified male reproductive system effects as a human hazard of
2 benzo[a]pyrene exposure.
3 Female reproductive effects
4 A large body of mechanistic data, both in vivo and in vitro, suggests that benzo[a]pyrene
5 impacts fertility through the disruption of folliculogenensis. This finding is supported, albeit
6 indirectly, by observations of premature ovarian senescence in women exposed to cigarette smoke
7 [Midgette and Baron. 1990]. Evidence for female reproductive toxicity of benzo[a]pyrene comes
8 from studies of human populations exposed to PAH mixtures as well as laboratory animal and in
9 vitro studies. In addition, two human studies observed associations specifically between
10 benzo[a]pyrene measures and two fertility-related endpoints: decreased ability to conceive [Neal et
11 al.. 2008: Neal etal.. 2007] and increased risk of early fetal death (i.e., before 14 weeks of gestation]
12 [Wuetal.. 2010]. Studies in multiple strains of rats and mice indicate fertility-related effects
13 including decreases in ovarian follicle populations and decreased fecundity. Decreased serum
14 estradiol has also been noted in two different strains of rats exposed by oral or inhalation exposure.
15 The reproductive effects associated with benzo[a]pyrene exposure are biologically supported and
16 relevant to humans.
17 In conclusion, EPA identified female reproductive effects as a human hazard of
18 benzo[a]pyrene exposure.
19 Susceptible Populations and Lifestages
20 Epidemiological studies indicate that exposure to complex mixtures of PAHs, such as
21 through cigarette smoke, is associated with measures of decreased fertility in humans [Neal etal.,
22 2008: El-Nemr etal., 1998] and that prenatal exposure to cigarette smoking is associated with
23 reduced fertility of women later in life [Weinberg et al.. 1989]. A case-control study in a Chinese
24 population has also indicated that women with elevated levels of benzo[a]pyrene-DNA adducts in
25 maternal blood were 4 times more likely to have experienced a miscarriage [Wu etal.. 2010].
26 Inhalation exposure of pregnant female rats to benzo[a]pyrene:carbon black aerosols on
27 CDs 11-20 caused decreased fetal survival and number of pups per litter associated with decreased
28 levels of plasma progesterone, estradiol, and prolactin [Archibongetal., 2002]. Decreased numbers
29 of live pups were also seen in pregnant mice following i.p. exposure to benzo[a]pyrene [Mattison et
30 al.. 1980]. These results indicate that benzo[a]pyrene exposure can decrease the ability of females
31 to maintain pregnancy.
32 Oral multigenerational studies of benzo[a]pyrene exposure in mice demonstrated effects on
33 fertility and the development of reproductive organs (decreased ovary and testes weight] in both
34 male and female offspring of pregnant mice exposed to 10-160 mg/kg-day on CDs 7-16
35 (Kristensen etal., 1995: Mackenzie and Angevine, 1981].
36 Reductions in female fertility associated with decreased ovary weight and follicle number
37 following gestational exposure (as discussed in Section 1.1.1] are supported by observations of:
This document is a draft for review purposes only and does not constitute Agency policy.
1-36 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 (1) destruction of primordial follicles [Borman et al., 2000: Mattisonetal., 1980] and decreased
2 corpora lutea [Miller etal.. 1992: Swartz and Mattison. 1985] in adult female mice following i.p.
3 exposure; (2] decreased ovary weight in adult female rats following oral exposure [Xu etal.. 2010]:
4 and (3] stimulation of oocyte apoptosis [Matikainenetal.. 2002: Matikainen etal.. 2001] or by a
5 decreased sensitivity to FSH-stimulated follicle growth [Neal etal.. 2007].
6 Reductions in male fertility associated with decreased testes weight following gestational
7 exposure (as discussed in Section 1.1.1] are supported by observations of: (1] decreased sperm
8 count, altered serum testosterone levels, testicular lesions, and/or increased numbers of apoptotic
9 germ cells in adult rats following repeated oral exposure to benzo[a]pyrene [Chung etal., 2011:
10 Chen etal., 2010: Zheng etal., 2010: Arafa etal., 2009]: (2] decreased epididymal sperm counts in
11 adult FO and Fl generations of male mice following 6 weeks of oral exposure of the FO animals to
12 benzo[a]pyrene [Mohamed etal.. 2010]: and (3] decreased testis weight, decreased testicular or
13 plasma testosterone levels, and/or decreased sperm production, motility, and density in adult male
14 rats following repeated inhalation exposure to aerosols of benzo[a]pyrene:carbon black [Archibong
15 etal.. 2008: Ramesh etal.. 2008: Inyangetal.. 2003].
16 1.1.3. Immunotoxicity
17 Human studies evaluating immune effects following exposure to benzo[a]pyrene alone are
18 not available for any route of exposure. However, a limited number of occupational human studies,
19 particularly in coke oven workers [Zhang etal.. 2012: Wu etal.. 2003b: Winker et al.. 1997:
20 Szczekliketal.. 1994]. show effects on immune parameters associated with exposure to PAH
21 mixtures. These studies are of limited utility because effects associated specifically with
22 benzo[a]pyrene cannot be distinguished from other constituents of the PAH mixture. Subchronic
23 and short-term animal studies have reported immunotoxic effects of benzo[a]pyrene by multiple
24 routes of exposure (Table 1-8 and Figure 1-5]. Effects include changes in thymus weight and
25 histology, decreased B cell percentages and other alterations in the spleen, and immune
26 suppression. Data obtained from subchronic oral gavage studies are supported by short-term, i.p.,
27 intratracheal, and subcutaneous (s.c.] studies. Additionally, there is evidence in animals for effects
28 of benzo[a]pyrene on the developing immune system. No studies were located that examined
29 immune system endpoints following inhalation exposure of animals to benzo[a]pyrene.
30 Thymus Effects
31 Decreased thymus weights (up to 62% compared to controls] were observed in male and
32 female Wistar rats exposed by gavage to 10-90 mg/kg-day benzo[a]pyrene for 35 or 90 days
33 (Kroese etal.. 2001: De Jong etal.. 1999]. This effect may be due to thymic atrophy. The incidence
34 of slight thymic atrophy was increased in males (6/10] and females (3/10] at a dose of
35 30 mg/kg-day in a 90-day study, although there was no evidence of atrophy at any lower dose
36 (Kroese et al., 2001]. Additionally, at the highest dose tested (90 mg/kg-day] in one of the 35-day
37 studies, the relative cortex surface area of the thymus and thymic medullar weight were
This document is a draft for review purposes only and does not constitute Agency policy.
1-37 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 significantly reduced [De Jongetal.. 1999]. Other histopathological changes in the thymus
2 (increased incidence of brown pigmentation of red pulp; hemosiderin) were observed in Wistar
3 rats of both sexes at 5 0 mg/kg-day in a 3 5-day study; however, this tissue was not examined in
4 intermediate-dose groups [Kroese etal.. 2001). Consistent with the effects observed in these
5 studies, decreased thymus weights and reduced thymic cellularity were observed in i.p. injection
6 studies that exposed mice to doses ranging from 50 to 150 mg/kg in utero [Holladay and Smith.
7 1995.1994: Urso and Johnson. 1988).
8 Spleen Effects
9 Reduced splenic cellularity, indicated by decreased relative and absolute number of B cells
10 in the spleen (decreased up to 41 and 61% compared to controls, respectively) and decreased
11 absolute number of splenic cells (31% decrease atthe highest dose), was observed in a subchronic
12 study in male Wistar rats administered 3-90 mg/kg-day benzo[a]pyrene by gavage for 35 days (De
13 Jongetal.. 1999). While the effect on the relative number of B cells was dose-related, the lower
14 doses did not affect the number of B cells or the absolute splenic cell number. The reduced splenic
15 cell count at the highest dose was attributed by the study authors to the decreased B cells, and
16 suggests a possible selective toxicity of benzo[a]pyrene to B cell precursors in the bone marrow.
17 The spleen effects observed in De long et al. (1999) are supported by observations of reduced
18 spleen cellularity and decreased spleen weights following i.p. injection or in utero benzo[a]pyrene
19 exposure to doses ranging from 50 to 150 mg/kg (Holladay and Smith. 1995: Urso etal.. 1988).
20 In addition to physical effects on the spleen, several studies have demonstrated functional
21 suppression of the spleen following benzo[a]pyrene exposure. Dose-related decreases in sheep red
22 blood cell (SRBC) specific serum IgM levels after SRBC challenge were reported in rats (10 or
23 40 mg/kg-day) and mice (5, 20,or 40 mg/kg-day) following s.c. injection of benzo[a]pyrene for
24 14 days (Temple et al.. 1993). Similarly, reduced spleen cell responses, including decreased
25 numbers of plaque forming cells and reduced splenic phagocytosis to SRBC and lipopolysaccharide
26 challenge, were observed in B6C3Fi mice exposed to doses >40 mg/kg-day benzo[a]pyrene by i.p.
27 or s.c. injection for 4-14 days (Lyte and Bick. 1985: Dean etal.. 1983: Munson and White. 1983] or
28 by intratracheal instillation for 7 days (Schnizlein et al.. 1987).
29 Immunoglobulin Alterations
30 Alterations in immunoglobulin levels have been associated with exposure to PAH mixtures
31 in a limited number of human studies. Some occupational studies have reported evidence of
32 immunosuppression following PAH exposure. For example, reductions in serum IgM and/or IgA
33 titers were reported in coke oven workers (Wuetal.. 2003b: Szczeklik et al.. 1994). Conversely,
34 immunostimulation of immunoglobulin levels has also been observed in humans, specifically
35 elevated IgG (Karakaya et al., 1999] and elevated IgE (Wu etal., 2003b] following occupational PAH
36 exposure.
This document is a draft for review purposes only and does not constitute Agency policy.
1-38 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Decreases in serum IgM (13-33% compared to controls) and IgA levels (22-61% compared
2 to controls) were observed in male Wistar rats exposed to 3-90 mg/kg-day benzo[a]pyrene by
3 gavage for 35 days (De long etal.. 1999): however, these reductions were not dose-dependent.
4 Similarly, reductions in IgA (9-38% compared to controls) were also observed in male and female
5 B6C3Fi mice exposed to doses of 5-40 mg/kgbenzo[a]pyreneby s.c. injection for 14 days (Munson
6 and White. 1983). Reductions in serum IgG levels of 18-24%, although not statistically significant,
7 were observed in female B6C3Fi mice exposed to doses >50 mg/kg benzo[a]pyrene by i.p. injection
8 for 14 days (Dean etal.. 1983).
9 Immune Suppression and Sensitization
10 Some occupational studies of coke oven emissions have reported evidence of
11 immunosuppression following PAH exposure. Reduced mitogenic responses in T cells (Winker et
12 al.. 1997) and reduced T-lymphocyte proliferative responses (Karakaya et al.. 2004) have been
13 observed following occupational exposure to PAH. Increased levels of apoptosis were observed in
14 the peripheral blood mono nuclear cells (a population of lymphocytes and monocytes) of
15 occupationally exposed coke oven workers, which is a response that may contribute to
16 immunodeficiency in this population (Zhang etal., 2012). However, a limitation of this study is that
17 it does not attribute the proportion of apoptotic activity to a specific class of cells and does not
18 include assessment of other potential markers of immunotoxicity in peripheral blood.
19 Results of functional immune assays in laboratory animals following short-term i.p. and s.c.
20 exposures add to the evidence for benzo[a]pyrene immunotoxicity. Resistance to Streptococcus
21 pneumonia or Herpes simplex type 2 was dose dependently reduced in B6C3Fi mice following s.c.
22 injection of >5 mg/kg-day benzo[a]pyrene for 14 days (Munson etal., 1985). Reduced cell
23 proliferation, IFN-y release, and IL-4 release were observed in male and female C56BL/6 mice
24 following short-term exposure to a gavage dose of 13 mg/kg benzo[a]pyrene as measured in a
25 modified local lymph node assay (van den Berg et al.. 2005). A statistically significant decrease in
26 natural killer cell activity was observed in male Wistar rats (Effector:Target cell ratio was
27 40.9 ± 28.4% that of controls) exposed to 90 mg/kg-day by gavage for 35 days (De Jong etal..
28 1999): however, splenic natural killer cell activity was not affected in B6C3Fi mice after s.c.
29 injection of 40mg/kg-day benzo[a]pyrene for 14 days (Munson etal., 1985). The magnitude of the
30 dose and duration of the exposure may account for the discrepancy between these two studies.
31 Single i.p. injections of 50 mg/kg benzo[a]pyrene decreased pro- and/or pre-B-lymphocytes and
32 neutrophils in the bone marrow of C57BL/6J mice without affecting the numbers of immature and
33 mature B-lymphocytes or GR-1+ myeloid cells (Galvan et al.. 2006).
34 In contrast to studies that have shown immunosuppression, benzo[a]pyrene may also
35 induce sensitization responses. Epicutaneous abdominal application of 100 [ig benzo[a]pyrene to
36 C3H/HeN mice, followed by ear challenge with 20 |ig benzo[a]pyrene 5 days later, produced a
37 contact hypersensitivity (a significant ear swelling) response (Klemme etal.. 1987).
This document is a draft for review purposes only and does not constitute Agency policy.
1-39 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Developmental Immunotoxicity
2 As noted above, several i.p. injection studies suggest that cell-mediated and humoral
3 immunity may be altered by exposure to high doses of benzo[a]pyrene during gestation.
4 Suppression of the mixed lymphocyte response, the graft-versus-host response, and suppression of
5 the plaque-forming cell response to SRBCs was observed in mice exposed in utero to 150 mg/kg
6 during mid (CDs 11-13), late (CDs 16-18), or both (CDs 11-17) stages of gestation; these effects
7 persisted until 18 months of age [Urso and Gengozian. 1984.1982.1980). Fetal thymic atrophy, as
8 assessed by reductions in cellularity (74-95%, compared to controls), was observed in mice
9 exposed to 50-150 mg/kg benzo[a]pyrene from GD 13 to 17, when examined on GD 18 [Holladay
10 and Smith. 1994). Analysis of cell surface markers (e.g., CD4, CDS) from the same study indicate
11 that benzo [a]pyrene may inhibit and/or delay thymocyte maturation, possibly contributing to the
12 observed thymic atrophy (Holladay and Smith. 1994). Consistent with these findings, several other
13 studies have noted decreased thymocyte numbers and disrupted T cell maturation after in utero
14 exposure to benzo[a]pyrene (Rodriguez etal.. 1999: Holladay and Smith. 1995: Lummus and
15 Henningsen. 1995: Urso etal.. 1992: Urso and Tohnson. 1987).
16 The fetal liver is the primary hematopoietic organ during gestation and a major source of
17 thymocyte precursors beginning around GD 10 or 11 in mice (Landreth and Dodson. 2005: Penit
18 andVasseur. 1989). Statistically significant reductions in total cellularity in the fetal liver of 54 and
19 67% were reported in offspring after i.p. exposures of 50 or 100 mg/kg benzo[a]pyrene,
20 respectively, to the dams on CDs 13-17 (Holladay and Smith. 1994). The decreased fetal liver
21 cellularity was accompanied by decreased expression of terminal deoxynucleotidyl transferase and
22 CD45R cellular markers, which are known to be present in cortical thymocyte progenitors in the
23 fetal liver (Holladay and Smith. 1994: Fine et al.. 1990: Silverstone et al.. 1976). These data also
24 suggest that benzo[a]pyrene disrupts liver hematopoiesis during gestation and may interfere with
25 prolymphoid seeding of the thymus, possibly contributing to thymic atrophy and cell-mediated
26 immunosuppression. Decreased numbers of CD4+ T-cells have been reported in the spleen of
27 1-week-old mice following in utero benzo[a]pyrene exposure by i.p. injection to the dams,
28 demonstrating the potential for downstream effects on T-cell development (Rodriguez etal.. 1999).
29 The decreased numbers of CD4+ T-cells correspond with observations of decreased proliferation in
30 the presence of Concanavalin A and a weak response compared to controls in an allogeneic mixed
31 lymphocyte reaction assay (Urso and Kramer. 2008).
32 Postnatal exposure to benzo[a]pyrene has also been suggested to cause immune effects.
33 Dose-dependent decreases in erythrocytes (attributed to reduced bone marrow erythropoiesis), as
34 well as reduced expression of IL-4 and IFN-y were observed in the pups of Wistar rats exposed to
35 0.1-10 mg/kg-day benzo[ajpyrene by subcutaneous injection for 14 days (Matiasovic et al.. 2008).
36 This finding suggests that benzo[a]pyrene may alter the immune response to infection or
37 vaccination in developing animals.
This document is a draft for review purposes only and does not constitute Agency policy.
1-40 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2
Table 1-8. Evidence pertaining to the immune effects of benzo[a]pyrene in
animals
Reference and study design
Results3
Thy m us effects
Kroese et al. (2001)
Wistar rats, 10/sex/dose
0, 3, 10, or 30 mg/kg-d by gavage 5 d/wk
90 d
De Jong etal. (1999)
Wistar rats, 8 males/dose
0, 3, 10, 30, or 90 mg/kg-d by gavage 5 d/wk
35 d
Kroese et al. (2001)
Wistar rats, 10/sex/dose
0, 1.5, 5, 15, or 50 mg/kg-d by gavage 5 d/wk
35 d
4, thymus weight
Females (% change from control): 0, -3, -6, and -28*
Males (% change from control): 0, 0, -13, and -29*
/T" slight thymic atrophy
Females (incidence): 0/10, 0/10, 0/10, and 3/10
Males (incidence): 0/10, 2/10, 1/10, and 6/10*
4> thymus weight
% change from control: 0, -9, -15*, -25*, and -62*
4> thymus weight
Females (% change from control): 0, 13, 8, -3, and -17*
Males (% change from control): 0, -8, -11, -27*, and -33*
Spleen effects
De Jong etal. (1999)
Wistar rats, 8 males/dose
0, 3, 10, 30, or 90 mg/kg-d by gavage 5 d/wk
35 d
•^ relative number (%) of B cells in spleen
% change from control: 0, -8, -13*, -18*, and -41*
4, total number of B cells in spleen
% change from control: 0, 13, -13, -13, and -61*
Change in total cell number in the spleen
% change from control: 0, 20, 0, +7, and -31*
Immunoglobulin alterations
De Jong etal. (1999)
Wistar rats, 8 males/dose
0, 3, 10, 30, or 90 mg/kg-d by gavage 5 d/wk
35 d
4, serum IgM
% change from control: 0, -13, -14, -33*, and -19
4, serum IgA
% change from control: 0, -27, -22, -28, and -61*
3
4
5
*Statistically significantly different from the control (p< 0.05).
a% change from control calculated as: (treated value - control value)/control value x 100.
This document is a draft for review purposes only and does not constitute Agency policy,
1-41 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 n
"j;
ro
DO
OJ3
0
Q
•1
• LOAEL
ANOAEL
• Doses > LOAEL
0 Doses < NOAEL
1
z
.
Kroese etal., 2001
rm
k ,
A L
5 ,
90-day study; females
Krocscctdl., 2001
A
4
•
> <
Tr
L
m
!•
T
M
,
v
i_l
c
A •
Z
) <
^
A
'
> L
r.
v .
* • • «
T T
1 L
\ ji
T
k
A L
V .
f
T
.
3 z
\ <
, ,
1
>
>
t/i */> t/i
CD — i-H — Q —
CD '~ O ^3 O ^
£ ^-* E o E (~g E
"? -^ ~? g ra" 0 ~?
N^- Thymus Weight
Thymus Effects
>/> ^>
S| ° 1
^ >£ f >:
|| ||
^Slight Thymic
Atrophy
cr> ^
en -^
o 3
«, -
° -3
tu LO
Q m
•^ n
relative
number
{%) of B
cells in
spleen
s
cr, ^ cr, ^
CD ^ en -^
CD '"3 CD -~
CJ ^ O zi
on ^ QO ^
1 -S l-o
(LJ LTl CU |_^)
O m O ro
vL-in total Change in
number of total cell
B cells in number in
spleen the spleen
oleen Effects
Cl •"
3 E
c >-
^ ip
Q m
vUSerum
IfiM
ImmunG
Alter,
U1
0 -±
CJ ^
r&
-=., T3
Q ro
IRA
globulin
tions
2 Figure 1-5. Exposure-response array for immune effects following oral
3 exposure.
4 Mode-of-Action Analysis—Immune Effects
5 Exposure to benzo[a]pyrene induces immunosuppressive effects such as decreased
6 numbers of B cells in the spleen and decreased thymus weight and cellularity following oral, i.p.,
7 s.c., or intratracheal exposure in experimental animals. However, the key events underlying
8 benzo[a]pyrene immunotoxicity have not been identified.
9 Benzo[a]pyrene is a well-known ligand for the AhR [Okey etal.. 1994: Nebertetal.. 1993:
10 Postlind et al.. 1993). Ligands of the AhR have been shown to have a role in regulating
11 hematopoietic stem cells in the bone marrow, a major site of B-cell proliferation and antibody
12 production [Esser. 2009). Benzo[a]pyrene reduced B-cell lymphopoiesis at concentrations as low
13 as 10~8M [Hardin et al.. 1992]. Furthermore, Ah-responsive (C57BL/6) mice showed greater dose-
14 dependent reductions in B-cell lymphopoiesis than those observed in Ah-nonresponsive (DBA/2)
This document is a draft for review purposes only and does not constitute Agency policy.
1-42 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 mice [Hardinetal., 1992]. Addition of the AhR antagonist and CYP450 inhibitor, a-naphthaflavone,
2 inhibited the benzo[a]pyrene-induced suppression of B-cell lymphopoiesis in a concentration-
3 dependent fashion. Similarly, the CYP1A1 inhibitor, l-(l-propynyl) pyrene, blocked
4 benzo[a]pyrene-induced B-cell growth inhibition but not growth inhibition caused by the
5 benzo[a]pyrene metabolite, BPDE; these data suggest that a GYP 1A1-dependent metabolite of
6 benzo[a]pyrene is responsible for the B-cell growth suppressive effects observed after
7 benzo[a]pyrene exposure [Allan etal.. 2006]. Altogether, these data suggest that benzo[a]pyrene
8 may regulate B-cell proliferation and antibody production in the bone marrow via the AhR.
9 Summ ary oflmm un e Effects
10 Evidence for immunotoxic effects of benzo[a]pyrene exposure comes from animal studies
11 that vary in route and duration of exposure. There are no human epidemiological studies that
12 provide specific support for benzo [a]pyrene immunotoxicity; however, immunosuppression has
13 been observed in studies following occupational exposure to PAH mixtures. However, these
14 findings are limited by co-exposures to other constituents of PAH mixtures.
15 Effects such as altered thymus weight and histology, spleen effects, and altered
16 immunoglobulin levels observed by the oral route reported in animal bioassays provide some
17 evidence of immunotoxicity following benzo[a]pyrene exposure; however, in vivo functional assays
18 provide stronger support for immunotoxicity [WHO. 2012]. The immunological changes observed
19 in the available subchronic gavage studies are supported by a larger database of in vivo studies of
20 benzo[a]pyrene (by parenteral exposure] indicating functional immunosuppression such as
21 decreased proliferative responses to antigens and decreased resistance to pathogens or tumor cells
22 [Kong etal.. 1994: Blanton etal.. 1986: Munson etal.. 1985: White etal.. 1985: Dean etal.. 1983:
23 Munson and White, 1983]. Although the key events underlying the mode of action of
24 benzo[a]pyrene immunotoxicity are not firmly established, there is evidence of physical alterations
25 to tissues/organs of the immune system, as well as decreases in immune function. Evidence of
26 benzo[a]pyrene-associated immunotoxicity is supported by consistent thymic effects observed in
27 two oral studies, as well as splenic effects, and varying immunosuppressive responses observed in
28 short-term or in vitro tests.
29 EPA concluded there was suggestive evidence that immunotoxicity is a potential human
30 hazard of benzo [a]pyrene exposure.
31 Suscep tible Pop illations an d Lifestages
32 The severity and persistence of immune effects observed during in utero studies suggests
33 that immunotoxicity may be greater during gestation than adulthood [Dietertand Piepenbrink.
34 2006: Holladay and Smialowicz. 2000: Urso and Gengozian. 1982]. Urso and Gengozian [1982]
35 provide experimental support demonstrating that immunosuppression from benzo [a]pyrene
36 exposure during gestation was greater than for mice exposed after birth to a 25-fold higher dose.
37 There is also substantial literature indicating that disruption of the immune system during certain
This document is a draft for review purposes only and does not constitute Agency policy.
1-43 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 critical periods of development (e.g., initiation of hematopoiesis, migration of stem cells, expansion
2 of progenitor cells) may have significant and lasting impacts on lifetime immune function (e.g.,
3 Burns-NaasetaL 2008: Dietert. 2008: Landreth. 2002: DietertetaL 20001 In addition, chemical-
4 specific studies show increased dose sensitivity and disease persistence from developmental versus
5 adult chemical exposure (reviewed in Luebke et al.. 2006).
6 Thymus toxicity is a sensitive and specific effect of benzo[a]pyrene and has been observed
7 in both prenatal and adult exposure studies. The thymus serves as a major site of thymocyte
8 proliferation and selection for maturation, and impairment can lead to cell-mediated immune
9 suppression (Kuper etal., 2002: De Waal et al., 1997: Kuper etal., 1992]. The thymus is believed to
10 be critical for T lymphocyte production during early life and not in adulthood (Hakim etal., 2005:
11 Schonland etal.. 2003: Petrie. 2002: Mackall etal.. 1995). Therefore, the decreases in thymus
12 weight observed in studies of adult animals exposed to benzo[a]pyrene suggest that
13 immunosuppression may be a heightened concern for individuals developmentally exposed to
14 benzo[a]pyrene.
15 1.1.4. Other Toxicity
16 There is some evidence thatbenzo[a]pyrene can produce effects in the forestomach, liver,
17 kidney, and cardiovascular system, as well as alter hematological parameters. However, there is
18 less evidence for these effects compared to organ systems described earlier in Sections 1.1.1-1.1.3.
19 Overall, EPA concluded that the available evidence does not support these noncancer effects as
20 potential human hazards.
21 Forestomach Toxicity
22 Lesions have been observed in the forestomach following subchronic and chronic oral
23 exposure to benzo[a]pyrene (Table 1-9). Increases in the incidence of forestomach hyperplasia
24 have been observed in Wistar rats following shorter-term, subchronic, and chronic gavage exposure
25 (Kroese etal.. 2001: De long et al.. 1999] and in B6C3Fi mice following chronic dietary exposure
26 (Beland and Gulp. 1998: Gulp etal.. 1998).
27 Following chronic gavage exposure, increased incidences of forestomach hyperplasia were
28 observed in male and female rats at 3 and 10 mg/kg-day; at the highest dose, a lower incidence of
29 hyperplasia was reported (Kroese etal., 2001). However, only the highest-level lesion (hyperplasia,
30 papilloma, or carcinoma) observed in each organ was scored, such that hyperplasia observed in the
31 forestomach, in which tumors were also observed, was not scored. The majority of animals in the
32 high-dose group exhibited forestomach tumors; therefore, the hyperplasia was not scored and the
33 incidence of forestomach hyperplasia in the study is more uncertain at the highest dose. Shorter-
34 term studies (Kroese etal.. 2001: De Jong etal.. 1999] showed dose-related increases in
35 forestomach hyperplasia at doses >10 mg/kg-day in Wistar rats. In addition, following chronic
36 dietary exposure, a dose-dependent increase in the incidence of forestomach hyperplasia and
37 hyperkeratosis was observed in female mice at >0.7 mg/kg-day (Beland and Gulp. 1998: Gulp etal..
This document is a draft for review purposes only and does not constitute Agency policy.
1-44 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 1998]. Forestomach tumors were also observed at >0.7 mg/kg-day by Beland and Gulp [1998] and
2 Gulp etal.f 19981.
3 Although humans do not have a forestomach, forestomach effects observed in rodents are
4 believed to be supportive of a human hazard, as humans have similar squamous epithelial tissue in
5 their oral cavity [IARC. 2003: Wester and Kroes. 1988). Mechanistic investigations suggest that
6 bioactivation of benzo[a]pyrene leads to reactive intermediates that can lead to mutagenic events,
7 as well as to cytotoxic and apoptotic events. The available human, animal, and in vitro evidence
8 best supports a mutagenic mode of action as the primary mode by which benzo[a]pyrene induces
9 carcinogenesis. Available data indicate that forestomach hyperplasia may be a histological
10 precursor to neoplasia observed at this site after chronic exposure to benzo[a]pyrene [Kroese etal.,
11 2001: De long et al.. 1999). Dose-response data show that forestomach hyperplasia occurs at
12 shorter durations and at lower doses than tumors in rats and mice exposed to benzo[a]pyrene for
13 up to 2 years [Kroese etal.. 2001: Beland and Gulp. 1998). Kroese etal. [2001] reported that the
14 forestomach lesions demonstrated a progression over the course of intercurrent sacrifices; the
15 authors described early lesions as focal or confluent basal hyperplasia, followed by more advanced
16 hyperplasia with squamous cell papilloma, culminating in squamous cell carcinoma. The
17 description of the progression of forestomach lesions provided by Kroese etal. [2001], coupled
18 with the observation that hyperplasia occurs before tumors and at lower doses than tumors,
19 suggests that forestomach hyperplasia induced by benzo[a]pyrene is likely a preneoplastic lesion.
20 Hematological Toxicity
21 Altered hematological parameters, including decreases in red blood cell [RBC] count,
22 hemoglobin, and hematocrit have been observed in laboratory animals following benzo[a]pyrene
23 exposure (Table 1-9]. Statistically significant decreases in RBC count, hemoglobin, and hematocrit
24 were observed in male Wistar rats at doses >10 mg/kg-day for 35 days [De long etal.. 1999]. A
25 minimal, but statistically significant increase in mean cell volume and a decrease in mean cell
26 hemoglobin were observed at the highest dose (90 mg/kg-day], which may indicate dose-related
27 toxicity for the RBCs and/or RBC precursors in the bone marrow [De Jong etal.. 1999]. Similarly,
28 male and female F344 rats also showed maximal decreases in RBC counts, hematocrit, and
29 hemoglobin levels between 10 and 12% in a 90-day dietary study [Knuckles etal., 2001]. Findings
30 were significant for RBC counts and hematocrit in males at >50 mg/kg-day, while decreased RBC
31 counts and hematocrit in females and hemoglobin levels in both sexes were only significant in the
32 100 mg/kg-day group [Knuckles et al.. 2001]. Small, but not statistically significant, decreases in
33 RBC counts and hemoglobin were observed in both 35- and 90-day studies in Wistar rats [Kroese et
34 al.. 2001]. It should be noted that when observed, the magnitudes of the decreases in RBCs,
35 hemoglobin, and hematocrit were generally small; about 18% at 90 mg/kg-day and <10% at lower
36 doses [De long etal.. 1999] and about 10% in F344 rats [Knuckles etal.. 2001]. A decrease in white
37 blood cells [WBCs], attributed to reduced numbers of lymphocytes and eosinophils, was also
38 observed at 90 mg/kg-day following gavage exposure for 35 days [De long etal.. 1999]. The mode
This document is a draft for review purposes only and does not constitute Agency policy.
1-45 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 of action by which benzo[a]pyrene exposure may lead to altered hematological parameters is
2 undetermined.
3 Liver Toxicity
4 Liver effects other than cancer associated with benzo[a]pyrene exposure primarily include
5 changes in liver weight and abnormal histopathology (Table 1-9). Increased liver weight was
6 reported in a 90-day study in both male and female Wistar rats given benzo[a]pyrene by gavage
7 [Kroese et al.. 2001]. Both females (17% increase) and males (29% increase) demonstrated
8 statistically significant increased liver weights at the highest dose tested (30 mg/kg-day); a
9 statistically significant increase (15%) was also reported in males at 10 mg/kg-day. Similar to the
10 findings in the 90-day study by Kroese etal. (2001). increased liver:body weight ratios were
11 observed at the highest dose in a 90-day dietary study in male F344 rats, although there was no
12 change observed in female liver weights (Knuckles etal.. 2001). Increased liver:body weight ratios
13 were also observed in both sexes at high doses (600 and 1,000 mg/kg) in an accompanying acute
14 study (Knuckles etal.. 2001). A statistically significant increase in liver weight was also observed in
15 male Wistar rats given 90 mg/kg-day benzo[a]pyrene by gavage for 35 days (De long et al., 1999).
16 Consistent with the findings by De long et al. (1999), a statistically significant increased liver weight
17 (about 18%) was also observed in both male and female Wistar rats at the highest dose
18 (50 mg/kg-day) given by gavage in a 3 5-day study (Kroese etal.. 2001).
19 Limited exposure-related differences in clinical chemistry parameters associated with liver
20 toxicity were observed; no differences in alanine aminotransferase or serum aspartate
21 transaminase levels were observed, and a small dose-related decrease in y-glutamyl transferase
22 was observed in males only exposed to benzo[a]pyrene for 90 days (Kroese etal., 2001).
23 Treatment-related lesions in the liver (oval cell hyperplasia) were identified as statistically
24 significantly increased following exposure to 90 mg/kg-day benzo[a]pyrene for 35 days; however,
25 incidence data were not reported (De long etal.. 1999). A 2-year carcinogenicity study (Kroese et
26 al.. 2001) observed some histopathological changes in the liver; however, organs with tumors were
27 not evaluated. Since many of the animals in the highest two doses developed liver tumors, the dose
28 responsiveness of the histological changes is unclear.
29 A dose-dependent increase in liver microsomal ethoxyresorufin-o-deethylase (EROD)
30 activity, indicative of CYP1A1 induction, was observed in both sexes at doses >1.5 mg/kg-day in a
31 35-day study (Kroese et al.. 2001). However, at the highest dose tested, with the greatest fold
32 induction in EROD activity, there was no evidence of associated adverse histopathologic findings.
33 The finding of increased liver weight across multiple studies of varying exposure durations, as well
34 as histopathological changes in the liver provide evidence of the liver as a target of benzo[a]pyrene-
35 induced toxicity. The mode of action by which benzo[a]pyrene induces these effects is unknown.
This document is a draft for review purposes only and does not constitute Agency policy.
1-46 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Kidney Toxicity
2 There is minimal evidence of kidney toxicity following exposure to benzo[a]pyrene
3 (Table 1-9). Statistically significant decreases in kidney weight were observed at doses of 3, 30, and
4 90 mg/kg-day, but not at 10 mg/kg-day, in a 35-day gavage study in male Wistar rats [De long etal..
5 1999). In a 35-day gavage study with a similar dose range in male and female Wistar rats, no
6 statistically significant changes in kidney weights were observed at any dose [Kroese et al.. 2001).
7 Histopathological analysis of kidney lesions revealed an apparent dose-responsive increase in the
8 incidence of abnormal tubular casts in the kidney in male F344 rats exposed by diet for 90 days
9 [Knuckles etal., 2001]. The casts were described as molds of distal nephrons lumen and were
10 considered by the study authors to be indicative of renal dysfunction. However, the statistical
11 significance of the kidney lesions is unclear. Several gaps and inconsistencies in the reporting make
12 interpretation of the kidney effects difficult, including: (1] no reporting of numerical data; (2) no
13 indication of statistical significance in the accompanying figure for kidney lesions; (3) discrepancies
14 between the apparent incidences and sample sizes per dose group; and (4) uncertainty in how
15 statistical analysis of histopathological data was applied. As such, the significance of the abnormal
16 tubular casts is unclear.
17 Cardiovascular Toxicity
18 Atherosclerotic vascular disease and increased risk of cardiovascular mortality have been
19 associated with cigarette smoking [Ramos and Moorthy. 2005: Miller and Ramos. 2001: Thirman et
20 al.. 1994] and, to a more limited degree, occupational exposure to PAH mixtures [Friesenetal..
21 2010: Friesen et al.. 2009: Burstyn et al.. 2005: Chau et al.. 1993]. Elevated mortality due to
22 cardiovascular disease was observed in a PAH-exposed occupational population (coke oven plant
23 workers], but elevated cardiovascular mortality was also observed in the non-exposed or slightly
24 exposed populations [Chau etal.. 1993]. Elevated risks of ischemic heart disease (IHD] were
25 associated with past cumulative benzo[a]pyrene exposure among aluminum smelter workers (with
26 a 5-year lag], although the trend was not statistically significant; there was no observed association
27 with more recent benzo[a]pyrene exposure (Friesen et al.. 2010]. Elevated risk of mortality from
28 IHD was also associated with cumulative benzo[a]pyrene exposure in a cohort of male asphalt
29 workers (although not statistically significant]; the trend in average benzo[a]pyrene exposure and
30 association with IHD was statistically significant, with an approximately 60% increase in risk
31 between the lowest and highest exposure groups (Burstyn et al.. 2005]. The two studies that
32 associate benzo[a]pyrene exposure with cardiovascular effects (Friesenetal.. 2010: Burstyn etal..
33 2005] rely on statistical models to create exposure groups rather than direct measurement of the
34 cohort under examination. Additionally, while these studies used benzo[a]pyrene exposure
35 groupings for analysis, they cannot address co-exposures that may have occurred in the
36 occupational setting (asphalt or aluminum smelters] or exposures that occurred outside the
37 workplace.
This document is a draft for review purposes only and does not constitute Agency policy.
1-47 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Increased systolic and diastolic blood pressure has been observed in the offspring of dams
2 exposed to increasing concentrations ofbenzo[a]pyrene (Tules etal.. 2012] (Table 1-1). At the
3 highest dose tested (1.2 mg/kg body weight by gavage to the dams), systolic pressures were
4 elevated approximately 50% and diastolic pressures were elevated approximately 80% above
5 controls. An intranasal exposure of 0.01 mg/kg-day benzo[a]pyrene in adult male rats also
6 produced an increase in blood pressure following a 7-day exposure (Centner and Weber. 2011).
7 Reduced endothelial integrity and increased smooth muscle cell mass, both related to
8 atherosclerosis, have been observed in Sprague-Dawley rats exposed to 10 mg/kg benzo[a]pyrene
9 by i.p. injection (once/week for 8 weeks) (Zhang and Ramos, 1997). The molecular mechanisms
10 underlying PAH-induced vascular injury and the development of atherosclerosis are not well
11 established, but current hypotheses include cell proliferative responses to injury of endothelial cells
12 from reactive metabolites (including reactive oxygen species [ROS]) and genomic alterations in
13 smooth muscle cells from reactive metabolites leading to transformed vasculature cells and
14 eventual plaque formation (Ramos and Moorthy. 2005). However, while the link between PAHs
15 and atherosclerotic disease has been studied, experiments specifically looking at the relationship
16 between levels of exposure to benzo[a]pyrene (via environmentally relevant routes) and the
17 development of aortic wall lesions related to atherosclerosis have not generally been performed.
18 One exception to this observation comes from a series of experiments on Apolipoprotein E
19 knock-out (ApoE-/-) mice exposed orally to benzo[a]pyrene. ApoE-/- mice develop spontaneous
20 atherosclerosis, which is thought to be due to enhanced oxidative stress from the lack of ApoE
21 (Godschalketal.. 2003). Overall, these studies suggest that benzo[a]pyrene exposure in ApoE-/-
22 mice enhances the progression of atherosclerosis through a general local inflammatory process.
23 Nervous System Effects
24 Neurobehavioral function and mood state were evaluated in two studies of men
25 occupationally exposed to PAH mixtures (Oiuetal.. 2013: Niu etal.. 2010). Alterations in
26 neurobehavioral function was evaluated in coke oven workers using the Neurobehavioral Core Test
27 Battery (self-reported symptoms by questionnaire). These studies also measured urinary levels of
28 the PAH metabolite, 1-hydroxypyrene, as markers of PAH exposure. In both studies, exposure was
29 associated with decrements in short-term memory and/or attention in digit span tests. In addition,
30 Qiu etal. (2013) reported an association between benzo[a]pyrene exposure and decrements in
31 tests related to sensorimotor coordination (i.e., reaction time, digit symbol, and pursuit aiming
32 tests), as well as lower health ratings for the tension-anxiety mood category; Niu etal. (2010)
33 performed the same test battery but did not detect these associations.
34 Alterations in neuromuscular, autonomic, sensorimotor, and electrophysiological endpoints
35 have been reported in rats and mice following acute or short-term exposure to benzo[a]pyrene
36 (Bouayedetal..2009b: Grova etal.. 2008: Grova etal.. 2007: Saunders etal.. 2006: Liu etal.. 2002:
37 Saunders etal.. 2002: Saunders etal.. 2001). Impaired learning and memory (as measured by
38 Morris water maze performance or novel object recognition) was observed following subchronic
This document is a draft for review purposes only and does not constitute Agency policy.
1-48 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 gavage in adult rats [Maciel etal., 2014: Chenetal., 2011: Chengzhietal., 2011] and following
2 subchronic or short-term i.p. exposure in adult mice (Oiuetal.. 2011: Xiaetal.. 2011: Grovaetal..
3 2007). Decreased anxiety-like behavior in hole board and elevated plus maze tests has been
4 observed following short-term i.p. exposure [Grovaetal.. 2008). while decreased depressive-like
5 activity was observed in the tail suspension test (but not the forced swim test) following short-term
6 oral exposure [Bouayedetal.. 2012). In addition, a 28-day gavage study in male mice observed an
7 increase in aggressive behavior (as measured by the resident intruder test) and an increase in
8 consummatory sexual behavior in mice treated with 0.02 mg/kg-day (Bouayed et al.. 2009b).
9 These data are consistent with the neurobehavioral effects observed following developmental
10 exposure, and they suggest that benzo[a]pyrene exposure could be neurotoxic in adults; however,
11 only limited data are available to inform the neurotoxic potential of repeated subchronic or chronic
12 exposure to benzo[a]pyrene via the oral route (Table 1-9).
13 Table 1-9. Evidence pertaining to other toxicities of benzo[a]pyrene in
14 animals
Reference and study design
Results3
Forestomach toxicity
Kroese et al. (2001)
Wistar (Riv:TOX) rats: male and female (52/sex/dose
group)
0, 3,10, or 30 mg/kg-d by gavage 5 d/wk
104 wks (chronic)
Wistar (Riv:TOX) rats: male and female
(10/sex/dose group)
0, 3,10, or 30 mg/kg-d by gavage 5 d/wk
90 d (subchronic)
Wistar (specific pathogen-free Riv:TOX) rats
(10/sex/dose group)
0,1.5, 5,15, or 50 mg/kg body weight by gavage
5 d/wk
5 wks (shorter-term)
Forestomach hyperplasia (basal cell hyperplasia)
incidences'5:
M: 2/50; 8/52; 8/52; and 0/52
F: 1/52; 8/51; 13/51; and 2/52
Forestomach hyperplasia (slight basal cell hyperplasia)
incidences:
M: 2/10; 0/10; 6/10; and 7/10
F: 0/10; 2/10; 3/10; and 7/10
Forestomach hyperplasia (basal cell hyperplasia)
incidences:
M: 1/10; 1/10; 4/10; 3/10; and 7/10
F: 0/10; 1/10; 1/10; 3/10; and 7/10*
(Beland and Gulp (1998); Gulp et al. (1998))
BSCSFi mice: female (48/dose group)
0, 5, 25, or 100 ppm in the diet (average daily
doses": 0, 0.7, 3.3, and 16.5 mg/kg-d)
2 years
Forestomach hyperplasia
Incidences: 13/48; 23/47; 33/46*; and 38/47*
Forestomach hyperkeratosis
Incidences: 13/48, 22/47, 33/46*, 38/47*
De Jong etal. (1999)
Wistar rats: male (8/dose group)
0, 3,10, 30, or 90 mg/kg-d by gavage 5 d/wk
5 wks
Forestomach hyperplasia (basal cell hyperplasia)
statistically significantly increased incidences at 30 and
90 mg/kg-d were reported, but incidence data were not
provided
This document is a draft for review purposes only and does not constitute Agency policy,
1-49 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Reference and study design
Results3
Hematological toxicity
Kroese et al. (2001)
Wistar rats, 10/sex/dose
0, 3,10, or 30 mg/kg-d by gavage 5 d/wk
90 d
Wistar rats, 10/sex/dose
0,1.5, 5,15, or 50 mg/kg-d by gavage 5 d/wk for
35 d
RBC count and hemoglobin changes not statistically
significant in males or females at any dose (numerical data
not reported)
RBC count: changes not statistically significant (numerical
data not reported)
Hemoglobin: changes not statistically significant
(numerical data not reported)
Knuckles etal. (2001)
F344 rats, 20/sex/dose
0, 5, 50, or 100 mg/kg-d by diet
90 d
4, RBC count
Females (% change from control): statistically significant at
100 mg/kg-d (numerical data not reported)
Males (% change from control): statistically significant at
50 and 100 mg/kg-d (numerical data not reported)
•i, hematocrit
Females (% change from control): statistically significant at
100 mg/kg-d (numerical data not reported)
Males (% change from control): statistically significant at
50 and 100 mg/kg-d (numerical data not reported)
•i, hemoglobin
Females: statistically significant at 100 mg/kg-d (numerical
data not reported)
Males: statistically significant at 100 mg/kg-d (numerical
data not reported)
De Jong etal. (1999)
Wistar rats, 8 males/dose
0, 3,10, 30, or 90 mg/kg-d by gavage 5 d/wk
35 d
4, RBC count
% change from control: 0, -1, -5*, -10*, and -18*
•^ hemoglobin
% change from control: 0, -1, -7*, -10*, and -18*
•i, hematocrit
% change from control: 0,0, -6*, -8*, and -14*
4, WBC count
% change from control: 0, -8, -9, -9, and -43*
1" mean cell volume
% change from control: 0,0, -3, 0, and 3*
•i, mean corpuscular hemoglobin concentration
% change from control: 0, -1, -1, -1, and -3*
This document is a draft for review purposes only and does not constitute Agency policy,
1-50 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Reference and study design
Results3
Liver toxicity
Kroese et al. (2001)
Wistar rats, 10/sex/dose
0, 3, 10, or 30 mg/kg-d by gavage 5 d/wk
90 d
Wistar rats, 10/sex/dose
0, 1.5, 5, 15, or 50 mg/kg-d by gavage 5 d/wk for
35 d
Knuckles etal. (2001)
F344 rats, 20/sex/dose
0, 5, 50, or 100 mg/kg-d by diet
90 d
De Jong etal. (1999)
Wistar rats, 8 males/dose
0, 3, 10, 30, or 90 mg/kg-d by gavage 5 d/wk
35 d
1" liver weight
Females (% change from control): 0, -2, 4, and 17*
Males (% change from control): 0, 7, 15*, and 29*
Liver histopathology: no effects reported
/T" liver weight
Females (% change from control): 0, 3, 2, 9, and 18*
Males (% change from control): 0, 2, 1, 3, and 18*
Liver histopathology: no effects reported
1" liverbody weight ratio
Females: no change (numerical data not reported)
Males (% change from control): 23% change reported at
100 mg/kg-d (numerical data not reported)
1" liver weight
% change from control: 0, -9, 7, 5, and 15*
/T" liver oval cell hyperplasia (numerical data not reported)
reported as significant at 90 mg/kg-d;
Kidney effects
Knuckles etal. (2001)
F344 rats, 20/sex/dose
0, 5, 50, or 100 mg/kg-d by diet
90 d
De Jong etal. (1999)
Wistar rats, 8 males/dose
0, 3, 10, 30, or 90 mg/kg-d by gavage 5 d/wk
35 d
Kroese et al. (2001)
Wistar rats, 10/sex/dose
0, 1.5, 5, 15, or 50 mg/kg-d by gavage 5 d/wk
35 d
1" abnormal tubular casts
Females: not statistically significant (numerical data not
reported)
Males: apparent dose-dependent increase (numerical data
not reported)
•i, kidney weight
% change from control: 0, -11*, -4, -10*, and -18*
Kidney weight: no change (data not reported)
Nervous system effects
Chengzhietal. (2011)
Sprague-Dawley rats, male, 32/dose
0 or 2 mg/kg-d by gavage
90 d
/T" time required for treated rats to locate platform in
water maze (data reported graphically)
This document is a draft for review purposes only and does not constitute Agency policy,
1-51 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Reference and study design
Results3
Bouayed et al. (2009b)
Swiss albino mice, male, 9/group
0, 0.02, or 0.2 mg/kg-d by gavage
28 d
Significant decrease in latency to attack and increase in the
number of attacks in the resident-intruder test at
0.02 mg/kg-d (but not at high dose)
Significant increase in mount number in the copulatory
behavior test at 0.02 and 0.2 mg/kg-d
1
2 *Statistically significantly different from the control (p < 0.05).
3 a% change from control calculated as: (treated value - control value)/control value x 100.
4 bReported incidences may not fully account for the occurrence of hyperplasias due to the scoring of only the
5 highest-level lesion in an individual animal (e.g., animals with forestomach tumors that also showed hyperplasia
6 would not have the observation of hyperplasia recorded).
7 °Based on the assumption that daily benzo[a]pyrene intake at 5 ppm was one-fifth of the 25-ppm intake (about
8 21 u.g/d) and using TWA body weights of 0.032 kg for the control, 5- and 25-ppm groups and 0.026 kg for the
9 100-ppm group.
10 1.1.5. Carcinogenicity
11 Eviden ce in Hum ans
12 Numerous epidemiologic studies indicate an association between PAH related occupations
13 and lung, bladder, and skin cancer (Table 1-10). This discussion primarily focuses on epidemiologic
14 studies that included a direct measure of benzo[a]pyrene exposure. All identified studies have co-
15 exposures to other PAHs. The identified studies were separated into tiers according to the extent
16 and quality of the exposure analysis and other study design features:
17 Tier 1: Detailed exposure assessment conducted (using a benzo(a)pyrene metric), large
18 sample size (>~50 exposed cases), and adequate follow-up period to account for expected
19 latency (e.g., >20 years for lung cancer).
20 Tier 2: Exposure assessment, sample size, or follow-up period did not meet the criteria for
21 Tier 1, or only a single-estimate exposure analysis was conducted.
22 For lung cancer, each of the Tier 1 studies observed increasing risks of lung cancer with
23 increasing cumulative exposure to benzo[a]pyrene (measured in |ig/m3-years), and each of these
24 studies addressed in the analysis the potential for confounding by smoking (Armstrong and Gibbs.
25 2009: Spinelli etal.. 2006: Xu etal.. 19961 (Table 1-11). These three studies represent different
26 geographic locations and two different industries. The pattern of results in the Tier 2 studies was
27 mixed, as would be expected for studies with less precise exposure assessments or smaller sample
28 sizes: one of the standardized mortality ratio (SMR) estimates was <1.0, with the other eight
29 estimates ranging from 1.2 to 2.9 (Table 1-12). In considering all of the available studies,
30 particularly those with the strongest methodology, there is considerable support for an association
31 between benzo[a]pyrene exposure and lung cancer, although the relative contributions of
32 benzo[a]pyrene and of other PAHs cannot be established.
This document is a draft for review purposes only and does not constitute Agency policy,
1-52 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 For bladder cancer, the cohort and nested case-control studies observed a much smaller
2 number of cases compared with lung cancer; this limits their ability to examine exposure-response
3 relationships. Three cohort studies with detailed exposure data, however, identified 48-90 cases
4 fBurstynetal..20Q7: Gibbs and Sevigny. 2007a: GibbsetaL 2007: Gibbs and Sevigny. 2007bl
5 [Spinellietal.. 2006] (Tier 1 studies, Table 1-13). Although cumulative exposure (up to
6 approximately 2 ug/m3-years] was not related to increasing risk in the study of asphalt workers by
7 Burstyn et al. (2007). an exposure-response was seen with the wider exposure range (i.e.,
8 >80 ug/m3-years] examined in two studies of aluminum smelter workers by (Gibbs and Sevigny
9 (2007aj: Gibbs etal. (2007]: Gibbs and Sevigny (2007b]]: and (Spinellietal.. 2006]. This difference
10 in response is not surprising, given that the highest exposure group in the asphalt worker studies
11 corresponded to the exposures seen in the lowest exposure categories in the studies of aluminum
12 smelter workers. The five studies with more limited exposure information or analyses each
13 included between 2 and 16 bladder cancer cases, with relative risk (RR] estimates ranging from
14 0.6 to 2.9. None of these individual effect estimates was statistically significant (Tier 2 studies,
15 Table 1-13].
16 Two of the identified occupational studies contained information on risk of mortality from
17 melanoma. Neither of these studies observed increased risks of this type of cancer, with an SMR of
18 0.91 (95% confidence interval [CI] 0.26, 2.48] (22 cases] in Spinelli et al. f 20061 and 0.58 (95% CI
19 0.12,1.7] in Gibbs etal. (2007](3 cases]. These studies did not include information on non-
20 melanoma skin cancers.
21 Non-melanoma skin cancer, specifically squamous cell carcinoma, is of particular interest
22 with respect to dermal PAH exposures. The literature pertaining to this kind of cancer and PAH
23 exposure goes back to the 18th century work of Sir Percivall Pott describing scrotal cancer, a
24 squamous cell skin cancer, in English chimney sweeps (Brown and Thornton, 1957]. Recent studies
25 of chimney sweeps in several Nordic countries have not found increases in non-melanoma skin
26 cancer incidence (Hogstedtetal.. 2013: Pukkalaetal.. 2009: Evanoff etal.. 1993]. likely due to
27 greatly reduced exposure associated with better occupational hygiene (IARC. 2012]. A study
28 among asphalt workers (roofers] reported an increased risk of mortality from non-melanoma skin
29 cancer among asphalt workers (roofers], with an SMR of 4.0 (95% CI: 1.0,10.9] among workers
30 employed >20 years (Hammond etal., 1976]. In addition to this study, two studies in Scandinavian
31 countries examined non-melanoma skin cancer risk in relation to occupations with likely dermal
32 exposure to creosote (i.e., timber workers and brick makers] using incidence data from population
33 registries (Karlehagenetal.. 1992: Tornqvistetal.. 1986]. The standardized incidence ratio (SIR]
34 estimates were 1.5 (95% CI: 0.7, 2.6] based on five exposed cases and 2.37 (95% CI: 1.08, 4.50]
35 based on nine cases in Tornqvistetal. (1986] and Karlehagen et al. (1992]. Because non-melanoma
36 skin cancers are rarely fatal if caught early, and the preventative excision of precancerous lesions is
37 common, the available occupational studies and cancer registries likely underestimate the risk of
This document is a draft for review purposes only and does not constitute Agency policy.
1-53 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 squamous cell carcinoma (Car0e etal.. 2013: Voelter-Mahlknechtetal.. 2007: ONS. 2003: Letzel and
2 Drexler. 19981
3 In addition to cohorts of workers occupationally exposed to PAH mixtures, populations
4 exposed to benzo[a]pyrene through topical coal tar formulations for the treatment of psoriasis,
5 eczema, and dermatitis have also been studied [Roelofzenetal.. 2010: Mitropoulos and Norman.
6 2005: Stern et al.. 1998: Stern and Laird. 1994: Lindelof and Sigurgeirsson. 1993: Torinuki and
7 Tagami. 1988: Pittelkowetal.. 1981: MaughanetaL 1980: Stern etal.. 1980). Epidemiological
8 studies examining skin cancer risk in relation to various types of topical tar exposure are
9 summarized in the Supplemental Information, Table D-6. Case reports, reviews, and studies that
10 did not include a measure of coal tar use are not included. The available studies examining
11 therapeutic topical coal tar use and risk of skin cancer were limited by low quality exposure data
12 with high potential of exposure misclassification (e.g., Roelofzenetal.. 2010: Mitropoulos and
13 Norman. 2005: Lindelof and Sigurgeirsson. 1993). small size and short duration of follow-up up
14 [e.g.. Torinuki and Tagami. 1988]. and choice of referent rates and differences in disease
15 ascertainment between cases and the reference population (e.g., Pittelkowetal.. 1981: Maughan et
16 al., 1980]. In addition, clinic-based studies focused on the regimen of coal tar in conjunction with
17 ultraviolet-B (UVB] therapy, although some appear to indicate increased risk with coal tar
18 exposure, cannot distinguish effects of coal tar from the effects of UVB (e.g., Stern etal.. 1998: Stern
19 and Laird. 1994: Lindelof and Sigurgeirsson. 1993: Stern etal.. 1980]. Therefore, because of the
20 limitations with respect to study design and analysis, EPA did not consider these studies further in
21 the evaluation of the risk of skin cancer from exposure to benzo[a]pyrene. Although EPA does not
22 consider the available studies sufficient to evaluate the risk of skin cancer, acute studies of coal tar
23 treated patients provide in vivo evidence of benzo[a]pyrene-specific genotoxicity (increased BPDE-
24 DNA adducts] in human skin (Godschalk etal.. 2001: Rojas etal.. 2001: Zhang etal.. 1990], an early
25 key eventin the carcinogenic mode of action of benzo[a]pyrene (see Figure 1-6 of Section 1.1.5).
26 Lung, bladder, and skin cancers are the cancers that have been observed in occupational
27 studies of PAH mixtures (Benbrahim-Tallaa etal.. 2012: Baan etal.. 2009: Secretan etal.. 2009).
28 The reproducibility of lung, bladder, and skin cancers in different populations and exposure
29 settings after occupational exposure to PAH mixtures (see Table 1-10) adds plausibility to the
30 hypothesis that common etiologic factors may be operating. The potential role that benzo[a]pyrene
31 may play as a causal agent is further supported by the observation that these same sites are also
32 increased in the studies that included a direct measure of benzo[a]pyrene.
This document is a draft for review purposes only and does not constitute Agency policy.
1-54 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Table 1-10. Cancer sites for PAH-related agents reviewed by IARC
PAH-related mixture or
occupation
Aluminum production
Carbon electrode manufacture
Coal gasification
Coal tar distillation
Coal tar pitch (paving and roofing)
Coke production
Creosotes
Diesel exhaust
Indoor emissions from household
combustion of biomassfuel (primarily
wood)
Indoor emissions from household
combustion of coal
Mineral oils, untreated or mildly
treated
Shale oils
Soot (chimney sweeping)
Sites with sufficient
evidence in humans
Lung, urinary bladder
Lung
Skin
Lung
Lung
Lung
Lung
Skin
Skin
Lung, skin
Sites with limited
evidence in humans
Lung
Urinary bladder
Skin
Urinary bladder
Lung
Urinary bladder
Reference
Baanetal. (2009)
IARC (2010)
Baanetal. (2009)
Baanetal. (2009)
Baanetal. (2009)
Baanetal. (2009)
IARC (2010)
Benbrahim-Tallaa et al.
(2012)
Secretan et al. (2009)
Secretan et al. (2009)
Baanetal. (2009)
Baanetal. (2009)
Baanetal. (2009)
2
3
4
5
Source: Adapted from IARC (2010).
Table 1-11. Summary of epidemiologic studies of benzo[a]pyrene (direct
measures) in relation to lung cancer risk: Tier 1 studies
Reference and study design
Results
Armstrong and Gibbs (2009) (Quebec, Canada)
Cohort, aluminum smelter workers, seven plants
16,431 (15,703 men; 728 women); duration
minimum 1 yr, began work 1966-1990; follow-up
through 1999 (mean ~30 yrs); smoking information
collected from medical records
Exposure: Job exposure matrix ~5,000 personal
benzo[a]pyrene measures from the 1970s to 1999
Related references: Lavoue et al. (2007) (exposure
data); (Gibbs and Sevigny (2007a); Gibbs etal.
(2007); Gibbs and Sevigny (2007b); Armstrong et al.
SMR 1.32 (1.22, 1.42) [677 cases]
Lung cancer risk by cumulative benzo[a]pyrene exposure
Median
benzo[a]-
pyrene n
u.g/m3-yrs cases SMR(95%CI) RR (95% Cl)
0 35 0.62(0.44,0.87) 1.0 (referent)
10 266 1.09 (0.96, 1.23) 1.75 (1.23,2.48)
30 70 1.88(1.47,2.38) 3.02(2.01,4.52)
60 53 1.21 (0.91, 1.59) 1.94 (1.27,2.97)
120 114 1.93(1.59,2.32) 3.09(2.12,4.51)
This document is a draft for review purposes only and does not constitute Agency policy,
1-55 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
(1994))
Spinelli et al. (2006) (British Columbia, Canada)
Cohort, aluminum smelter workers; 6,423 (all men);
duration minimum >3 yrs; began work 1954-1997;
follow-up through 1999 (14% loss to follow-up;
mean ~24 yrs); smoking information from self-
administered questionnaire
Exposure: Job exposure matrix using 1,275 personal
benzo[a]pyrene measures from 1977 to 2000 (69%
for compliance monitoring)
Related references: Friesen et al. (2006) (exposure
data); Spinelli etal. (1991)
Xuetal. (1996) (China)
Nested case-control in iron-steel worker cohort
610 incident cases (96% participation); 959 controls
(94% participation) (all men); duration data not
reported; smoking information collected from
interviews; next-of-kin interviews with 30% of lung
cancer cases and 5% of controls
Exposure: Job exposure matrix 82,867 historical
monitoring records, 1956-1992
240 116 1.79(1.48,2.15) 2.86(1.96,4.18)
480 23 2.36(1.49,3.54) 3.77(2.23,6.38)
No evidence of confounding by smoking
Additional modeling as continuous variable: RR 1.35 (95%
Cl 1.22, 1.51) at 100 u.g/m3-yrs (0.0035 per u.g/m3-yrs
increase); other shapes of exposure-response curve
examined.
SMR: 1.07 (0.89, 1.28) [120 cases]
SIR: 1.10 (0.93, 1.30) [147 cases]
Lung cancer risk by cumulative benzo[a]pyrene exposure
Benzo[a]pyrene
u.g/m3-yrs n cases RR (95% Cl)a
0-0.5 25 1.0 (referent)
0.5-20 42 1.23 (0.74,2.03)
20-40 23 1.35 (0.76,2.40)
40-80 25 1.36 (0.78,2.39)
>80 32 1.79 (1.04,3.01)
Adjusting for smoking category; trend p < 0.001.
Lung cancer risk by cumulative benzo[a]pyrene exposure
Benzo[a]-pyrene
(u.g/m3-yrs) n cases RR (95% Cl)a
<0.84 72 1.1(0.8,1.7)
0.85-1.96 117 1.6(1.2,2.3)
1.97-3.2 96 1.6(1.1,2.3)
>3.2b 105 1.8(1.2,2.5)
Adjusting for birth year and smoking category; trend
p < 0.004. Referent group is "nonexposed" (employed in
administrative or low-exposure occupations).
bStudy table IV unclear; could be >3.0 for this category.
1
2
Table 1-12. Summary of epidemiologic studies of benzo[a]pyrene (direct
measures) in relation to lung cancer risk: Tier 2 studies
Reference and study design
Results
Limited follow-up period (<20 yrs)
Friesen et al. (2009) (Australia)
Cohort, aluminum smelter workers; 4,316 (all men);
duration minimum 90 d; began work after 1962;
follow-up through 2002, mean 16 yrs (maximum
20 yrs); Smoking information from company records
RR 1.2 (0.7, 2.3) [19 cases in exposed; 20 in unexposed]
Lung cancer risk by cumulative benzo[a]pyrene exposure
Benzo[a]-pyrene
u.g/m3-yrs n cases RR (95% Cl)a
0 20 1.0 (referent)
This document is a draft for review purposes only and does not constitute Agency policy,
1-56 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Reference and study design
if employed before 1995 and study interviews if
employed after 1994
Exposure: Job/task exposure matrix using TWA
benzo[a]pyrene measures (n = 655), 1977-2004 (79%
from 1990 to 2004)
Results
>0-0.41 6
0.41-10.9 6
>10.9 7
aPoisson regression, adjusting for smoking;
0.7 (0.3, 1.8)
1.4(0.6,3.5)
1.7(0.7,4.2)
trend p = 0.22.
Proxy measure
Olsson et al. (2010) (Denmark, Norway, Finland,
Israel)
Nested case-control, asphalt workers; 433 lung
cancer cases (65% participation); 1,253 controls (58%
participation), matched by year of birth, country (all
men); duration: minimum >2 seasons, median
8 seasons; began work 1913-1999; follow-up: from
1980 to 2002-2005 (varied by country); smoking
information from interviews
Exposure: Compilation of coal tar exposure
measures, production characteristics, and repeat
measures in asphalt industry in each country used to
develop exposure matrix
Related references: (Boffetta et al. (2003); Burstyn
etal. (2000))
Costantino et al. (1995) (United States, Pennsylvania)
Cohort, coke oven workers; 5,321 and 10,497
unexposed controls (non-oven steel workers;
matched by age, race, date of first employment) (all
men); duration data not reported; worked in 1953;
follow-up through 1982 (length data not reported)
Exposure: Average daily exposure coal tar pitch
volatiles: 3.15 mg/m3 top-side full-time jobs,
0.88 mg/m3 side jobs; used to calculate weighted
cumulative exposure index
Related reference: Dong et al. (1988) (exposure data)
Lung cancer risk by cumulative coal tar exposure3
Coal tar n
unit-yrsa cases RR
0.39-4.29 43 1.31
4.30-9.42 32 0.98
9.43-16.88 30 0.97
16.89-196.48 54 1.60
(trend p-value)
(95% Cl)
(0.87, 2.0)
(0.62, 1.6)
(0.61, 1.6)
(1.09, 2.4)
(0.07)
Adjusting for set, age, country, tobacco pack-years.
SMR 1.95 (1.59, 2.33) [255 cases]
Lung cancer risk by cumulative exposure
Coal tar pitch
volatiles n
(mg/m3-mo) cases
0 203
1-49 34
50-199 43
200-349 59
350-499 39
500-649 27
>650 56
Adjusting for age, race, coke plant, period
trend p< 0.001.
RR (95% Cl)a
1.0 (referent)
1.2 (0.85, 1.8)
1.6(1.1,2.3)
2.0(1.5,2.8)
2.0(1.6,3.2)
2.7(2.0,4.6)
3.1(2.4,4.6)
of follow-up;
This document is a draft for review purposes only and does not constitute Agency policy,
1-57 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Reference and study design
Results
Limited exposure information
Liuetal. (1997)(China)
Cohort, various carbon plants and aluminum smelter
workers; 6,635 (all men); duration minimum 15 yrs;
began work before 1971; follow-up: through 1985
(mean ~14 yrs); smoking information from
questionnaire
Exposure: Area samples from one carbon plant,
1986-1987
SMR 2.2 (1.1, 2.8) [50 cases]
Lung cancer risk by exposure category
Exposure
category
None
Low
Moderate
High
Mean
benzo[a]-
pyrene
u.g/m3
0.30
1.19
n cases
13
6
5
26
SMR(95%CI)a
1.49 (0.83, 2.5)
1.19(0.48,2.5)
1.52 (0.55, 3.4)
4.30(2.9,6.2)
'Calculated by EPA from data in paper.
Bergerand Manz (1992) (Germany)
SMR 2.88 (2.28, 3.59) [78 cases]
Cohort, coke oven workers; 789 (all men); duration
minimum 10 yrs (mean 27 yrs); began work
1900-1989; follow-up through 1989 (length data not
reported); smoking information from plant records
and interviews
Exposure: Mean benzo[a]pyrene: 28 u.g/m3 (range
0.9-89 u.g/m3)
(Hansen (1991); Hansen, 1989) (Denmark)
Cohort, asphalt workers; 679 workers (applicators)
(all men); duration data not reported; employed
1959 to 1980; follow-up to 1986 (mean ~11 yrs);
smoking information from 1982 surveys of industry
and general population
Exposure: Asphalt fume condensate, 35 personal
samples during flooring: median 19.7 mg/m3 (range
0.5-260 mg/m3)
SMR 2.90 (1.88, 4.3) [25 cases] (ages 40-89)
SMR 2.46 (1.59, 3.6) [25 cases] (with smoking adjustment)
This document is a draft for review purposes only and does not constitute Agency policy,
1-58 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Reference and study design
Results
Gustavsson etal. (1990) (Sweden)
Cohort, gas production (coke oven) workers; 295 (all
men); duration minimum 1 yr, median 15 yrs;
employed 1965-1972; follow-up: 1966-1986
(mortality); 1966-1983 (incidence; mean ~15yrs);
smoking information from interviews with older
workers
Exposure: Area sampling - top of ovens;
benzo[a]pyrene, 1,964 mean 4.3 u.g/m3 (range
0.007-33 u.g/m3); 1,965 mean 0.52 u.g/m3
(0.021-1.29 u.g/m3)
SMR 0.82 (0.22, 2.1) [4 cases] (referent group = employed
men)
SIR 1.35 (0.36, 3.5) [4 cases]
Moulin etal. (1989) (France)
Cohort and nested case-control, two carbon
electrode plants; 1,302 in Plant A (all men),
employed in 1975; follow-up 1975-1985 (incidence);
smoking information from plant records; 1,115 in
Plant B (all men); employed in 1957; follow-up
1957-1984 (mortality); duration of employment and
follow-up data not reported
Plant A: SMR 0.79 (0.32,1.6) [7 cases]
Plant B: SMR 1.18 (0.63, 2.0) [13 cases]
Internal comparison (case-control), >1 yr duration:
Plant A: OR 3.42 (0.35, 33.7) [7 cases, 21 controls]
Plant B: OR 0.49 (0.12, 2.0) [13 cases, 33 controls]
Exposure: Benzo[a]pyrene, 19 area samples and
16 personal samples in Plant A (personal sample
mean 2.7 u.g/m3; range 0.59-6.2 u.g/m3); 10 area
samples and 7 personal samples in Plant B; personal
sample mean 0.17 u.g/m3, range 0.02-0.57 u.g/m3
Hammond etal. (1976) (United States)
SMR 1.6 (1.3,1.9) [99 cases] (>20 yrs since joining union)
(CIs calculated by EPA from data in paper)
Cohort, asphalt roofers; 5,939 (all men); duration
minimum 9 yrs, began before 1960; follow-up
through 1971
Exposure: 52 personal samples (masks with filters)
during specific jobs and tasks; mean benzo[a]pyrene
16.7 u.g per 7-hr d
This document is a draft for review purposes only and does not constitute Agency policy,
1-59 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2
Table 1-13. Summary of epidemiologic studies of benzo[a]pyrene (direct
measures) in relation to bladder cancer risk
Reference and study design
Results
Tier 1 studies
Burstyn et al. (2007) (Denmark, Norway, Finland,
Israel)
Cohort, asphalt workers; 7,298 (all men); duration
minimum >2 seasons, median 8 seasons; began
work 1913-1999; follow-up began around 1960,
ended around 2000 (years varied by country);
median 21 yrs; smoking information not collected
Exposure: Compilation of benzo[a]pyrene
measures, production characteristics, and repeat
measures in asphalt industry in each country used
to develop exposure matrix
Related references: (Boffetta etal., 2003; Burstyn
etal. (2000))
48 incident bladder cancer cases (39 cases in analyses with
15-yr lag)
Bladder cancer risk by cumulative benzo[a]pyrene exposure3
Benzo[a]-
pyrene
u.g/m3-yrsa
0-0.253
0.253-0.895
0.895-1.665
>1.665
RR (95% Cl)
n cases
12
12
12
12
RR (95% Cl)
(15-yr lag)0
(no lag)0
1.0 (referent) 1.0 (referent)
0.69 (0.29,1.6) 1.1 (0.44, 2.9)
1.21(0.45,3.3) 1.7(0.62,4.5)
0.84(0.24,2.9) 1.1(0.30,4.0)
Adjusting for age, calendar period, total duration of
employment, country.
bTrend p = 0.9.
'Trend p = 0.63.
Stronger pattern seen with average exposure in 15-yr lag
(RR 1.5, 2.7,1.9 in second through fourth quartile; trend
p = 0.15)
(Gibbs and Sevigny (2007a); Gibbs et al. (2007);
Gibbs and Sevigny (2007b)) (Quebec, Canada)
Hired before 1950: SMR 2.24 (1.77, 2.79) [78 cases]
Bladder cancer risk by cumulative benzo[a]pyrene exposure
Cohort, aluminum smelter workers, seven plants
16,431 (15,703 men; 728 women); duration
minimum 1 yr, began work 1966-1990; follow-up:
through 1999 (mean ~30 yrs); smoking information
collected from medical records
Exposure: Job exposure matrix using
~5,000 personal benzo[a]pyrene measures from the
1970s to 1999
Related references: Lavoue et al. (2007) (exposure
data); (Armstrong etal. (1994); Gibbs (1985); Gibbs
and Horowitz (1979))
Benzo[a]-
pyrene
u.g/m3-yrsa
0
10
30
60
120
240
480
n cases
3
14
3
1
15
30
12
Smoking-
adjusted
SMR(95%CI) RRb
0.73(0.15,2.1) 1.0 (referent)
0.93 (0.45, 1.4)
1.37 (0.28, 4.0)
0.35 (0.9,1.9)
4.2 (2.4, 6.9)
6.4 (4.3, 9.2)
23.9 (12.2,
41.7)
1.11
1.97
0.49
8.49
aCategory midpoint.
bCls not reported; highest category is >80 |jg/m3-yrs
(n observed = 57).
Mortality risk reduced in cohort hired in 1950-1959,
SMR =1.23.
Similar patterns seen in analysis of bladder cancer
incidence.
This document is a draft for review purposes only and does not constitute Agency policy,
1-60 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Reference and study design
Spinelli et al. (2006) (British Columbia, Canada)
See Table 1-11 for study details; this study is
considered a 'Tier 2") study for bladder cancer
because of the smaller number of bladder cancer
cases (n = 12) compared with lung cancer cases
(n = 120)
Tier 2 studies
Friesen et al. (2009) (Australia)
See Table 1-12 for study details
Costantino et al. (1995) (United States,
Pennsylvania)
See Table 1-12 for study details
Hammond et al. (1976) (United States)
See Table 1-12 for study details
Moulin etal. (1989) (France)
See Table 1-12 for study details
Gustavsson et al. (1990) (Sweden)
See Table 1-12 for study details
Results
SMR 1.39 (0.72, 2.43) [12 cases]
SIR 1.80; Cl 1.45-2.21 [90 cases, including in situ]
Bladder cancer risk by cumulative benzo[a]pyrene exposure
Benzo[a]-
pyrene
u.g/m3-years n cases RR (95% Cl)a
0-0.5 17 1.0 (referent)
0.5-20 20 0.83 (0.43, 1.59)
20-40 13 1.16 (0.56,2.39)
40-80 18 1.50 (0.77,2.94)
>80 22 1.92 (1.02,3.65)
Adjusting for smoking category; trend p < 0.001.
RR 0.6 (0.2, 2.0) [five cases in exposed; eight in unexposed]
Bladder cancer risk by cumulative benzo[a]pyrene exposure
Benzo[a]-
pyrene
u.g/m3-yrs n cases RR (95% Cl)a
0 8 1.0 (referent)
>0-0.41 1 0.2 (0.03,1.9)
0.41-10.9 2 0.7 (0.2,3.7)
>10.9 2 1.2 (0.2,5.6)
aPoisson regression, adjusting for smoking category; trend
p = 0.22.
SMR 1.14 (0.61, 2.12) (16 cases)
SMR 1.7 (0.94, 2.8) (13 cases) (>20 yrs since joining union)
(CIs calculated by EPA from data in paper)
Plant A: 0 observed cases; expected <1.0
Plant B: SMR 1.94 (0.40, 5.0) (3 cases)
SMR 2.85 (0.30, 10.3) (2 cases) (referent group - employed
men)
This document is a draft for review purposes only and does not constitute Agency policy,
1-61 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Evidence in Animals
2 Oral exposure
3 Evidence of tumorigenicity following oral exposure to benzo[a]pyrene has been
4 demonstrated in rats and mice. As summarized in Table 1-14, oral exposure to benzo[a]pyrene has
5 resulted in an increased incidence of tumors in the alimentary tract in male and female rats [Kroese
6 etal..2Q01: Brune etal.. 1981] and female mice [Beland and Gulp. 1998: Gulp etal.. 1998]. liver
7 carcinomas in male and female rats, kidney adenomas in male rats [Kroese etal.. 2001]. and
8 auditory canal tumors in both sexes [Kroese etal., 2001].
9 Forestomach tumors have been observed in several lifetime cancer bioassays in rats and
10 mice following both gavage and dietary exposure to benzo[a]pyrene at doses ranging from
11 0.016 mg/kg-day in Sprague-Dawley rats to 3.3 and 10 mg/kg-day in B6C3Fi mice and Wistar rats,
12 respectively fKroese etal.. 2001: Beland and Gulp. 1998: Gulp etal.. 1998: Brune etal.. 19811 In
13 addition, multiple less-than-lifetime oral exposure cancer bioassays in mice provide supporting
14 evidence that oral exposure to benzo[a]pyrene is associated with an increased incidence of
15 forestomach tumors [Weyandetal., 1995: Benjamin etal., 1988: Robinson etal., 1987: El-Bayoumy,
16 1985: Triolo etal.. 1977: Wattenberg. 1974: Roe etal.. 1970: Biancifiori etal.. 1967: Chouroulinkov
17 etal.. 1967: Fedorenko and Yansheva. 1967: Neal and Rigdon. 1967: Berenblum andHaran. 1955].
18 Although humans do not have a forestomach, similar squamous epithelial tissue is present in the
19 oral cavity [IARC. 2003: Wester and Kroes. 1988]: therefore, EPA concluded that forestomach
20 tumors observed in rodents following benzo[a]pyrene exposure are relevant in the assessment of
21 carcinogenicity [Beland and Gulp, 1998]. For further discussion, see Sections 1.2 and 2.3.4.
22 Elsewhere in the alimentary tract, dose-related increases of benign and malignant tumors
23 were observed. In rats, oral cavity tumors were induced in both sexes and adenocarcinomas of the
24 jejunum were induced in males [Kroese etal.. 2001]. In mice, tumors were induced in the tongue,
25 esophagus, and larynx of females (males were not tested] [Beland and Gulp. 1998: Gulp etal.. 1998].
26 Chronic oral exposure to benzo[a]pyrene resulted in a dose-dependent increased incidence
27 of liver carcinomas in both sexes of Wistar rats, with the first liver tumors detected in week 35 in
28 high-dose male rats; liver tumors were described as complex, with a considerable proportion
29 (59/150 tumors] metastasizingto the lungs (Kroese etal., 2001]. Treatment-related hepatocellular
30 tumors were not observed in mice (Beland and Gulp. 1998: Gulp etal.. 1998].
31 A statistically significantly increased incidence of kidney tumors (cortical adenomas] was
32 observed in male Wistar rats following chronic gavage exposure (Kroese etal.. 2001] (Table 1-14].
33 The kidney tumors were observed at the mid- and high-dose groups. Treatment-related kidney
34 tumors were not observed in two other chronic studies (Beland and Gulp. 1998: Brune etal.. 1981].
35 Lung tumors were also observed following almost nine months of dietary exposure to
36 approximately 10 mg/kg-day in female AJ mice (Weyand et al., 1995]. Other lifetime exposure
37 studies did not report treatment-related increases in lung tumors (Kroese etal.. 2001: Beland and
38 Gulp. 1998: Gulp etal.. 19981
This document is a draft for review purposes only and does not constitute Agency policy.
1-62 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Table 1-14. Tumors observed in chronic oral animal bioassays
Study design and reference
Results
Kroese et al. (2001)
Wistar (Riv:TOX) rats (52/sex/dose
group)
0, 3,10, or 30 mg/kg-d by gavage 5 d/wk
2 yrs
Forestomach
incidences:
M: 0/52; 7/52*; 18/52*; and 17/52* (papilloma)
M: 0/52; 1/52; 25/52*; and 35/52* (squamous cell carcinoma)
F: 1/52; 3/51; 20/51*; and 25/52* (papilloma)
F: 0/52; 3/51; 10/51*; and 25/52* (squamous cell carcinoma)
Oral cavity
incidences:
M: 0/24; 0/24; 2/37; and 10/38* (papilloma)
M: 1/24; 0/24; 5/37; and 11/38* (squamous cell carcinoma)
F: 0/19; 0/21; 0/9; and 9/31*(papilloma)
F: 1/19; 0/21; 0/9; and 9/31* (squamous cell carcinoma)
Jejunum (adenocarcinomas)
incidences:
M: 0/51; 0/50; 1/51; and 8/49*
F: 0/50; 0/48; 0/50; and 2/51
Duodenum (adenocarcinomas)
incidences:
M: 0/51; 0/50; 0/51; and 1/49
F: 0/49; 0/48; 0/50; and 2/51
Liver (adenomas and carcinomas)
incidences:
M: 0/52; 3/52; 15/52*; and 4/52 (adenoma)
M: 0/52; 1/52; 23/52*; and 45/52* (carcinoma)
F: 0/52; 2/52; 7/52*; and 1/52 (adenoma)
F: 0/52; 0/52; 32/52*; and 50/52* (carcinoma)
Kidney (cortical adenoma)
incidences:
M: 0/52; 0/52; 7/52*; and 8/52*
F: increase not observed
Auditory canalb (Zymbal gland) (carcinomas)
incidences:
M: 0/1; 0/0; 2/7; and 19/33*
F: 0/0; 0/1; 0/0; and 13/20*
This document is a draft for review purposes only and does not constitute Agency policy,
1-63 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
(Beland and Gulp (1998); Gulp et al.
(1998))
BSCSFi mice: female (48/dose group)
0, 5, 25, or 100 ppm (average daily
doses3: 0, 0.7, 3.3, and 16.5 mg/kg-d) in
the diet
2 yrs
Forestomach (papillomas and squamous cell carcinomas)
incidences: 1/48; 3/47; 36/46*; and 46/47*
Esophagus (papillomas and carcinomas)
incidences: 0/48; 0/48; 2/45; and 27/46*
Tongue (papillomas and carcinomas)
incidences: 0/49; 0/48; 2/46; and 23/48*
Larynx (papillomas and carcinomas)
incidences: 0/35; 0/35; 3/34; and 5/38
Bruneetal. (1981)
Sprague-Dawley rats: male and female
(32/sex/dose)
Gavage: 0, 6,18, 39 mg/kg-yr (0, 0.016,
0.049, 0.107 mg/kg-d)
Diet: 0, 6, 39 mg/kg-yr (0, 0.016,
0.107 mg/kg-d)
Treated until moribund or dead
2 yrs
Forestomach (papillomas and carcinomas0); gavage
incidences: 3/64; 12/64*; 26/64*; and 14/64*
Forestomach (papillomas); diet
incidences: 2/64; 1/64; and 9/64*
Larynx and esophagus (papillomas); gavage
incidences: 3/64; 1/64; 0/64; and 0/64
Larynx and esophagus (papillomas); diet
incidences: 1/64; 2/64; and 1/64
1
2 * Indicates statistical significance as identified in study.
3 aBased on the assumption that daily benzo[a]pyrene intake at 5 ppm was one-fifth of the 25-ppm intake (about
4 21 u.g/day) and using TWA body weights of 0.032 kg for the control, 5- and 25-ppm groups and 0.026 kg for the
5 100-ppm group.
6 blncidencesarefor number of rats with tumors compared with number of tissues examined histologically.
7 Auditory canal tissue was examined histologically when abnormalities were observed on macroscopic
8 examination.
9 °Two malignant forestomach tumors were observed (one each in the mid- and high-dose groups).
10 Inhalation exposure
11 The inhalation database of benzo[a]pyrene carcinogenicity studies consists of one lifetime
12 inhalation bioassay in male hamsters [Thyssenetal.. 1981]. Intratracheal instillation studies in
13 hamsters are also available [Feron and Kruysse. 1978: Ketkar etal.. 1978: Feronetal.. 1973: Henry
14 etal.. 1973: Saffiottietal.. 19721
15 Several long term intratracheal installation studies in hamsters evaluated the
16 carcinogenicity of benzo[a]pyrene [Feron and Kruysse. 1978: Feronetal.. 1973: Henry etal.. 1973:
17 Saffiotti etal.. 1972). These studies treated animals with benzo[a]pyrene once a week in a saline
18 solution (0.5-0.9%) for >8 months and observed animals for 1-2 years following cessation of
19 exposure. Tumors in the larynx, trachea, bronchi, bronchioles, and alveoli were observed.
20 Individual studies also reported tumors in the nasal cavity and forestomach. These intratracheal
21 instillation studies support the carcinogenicity of benzo[a]pyrene in the respiratory tract; however,
22 direct extrapolation from a dose delivered by intratracheal instillation to an inhalation
23 concentration expected to result in similar responses is not recommended [Driscoll etal.. 2000).
This document is a draft for review purposes only and does not constitute Agency policy,
1-64 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Lifetime inhalation exposure to benzo[a]pyrene resulted in the development of tumors in
2 the respiratory tract and pharynx in Syrian golden hamsters [Thyssen et al.. 1981). The authors
3 stated that the rates of tumors of other organs generally corresponded to the rates in controls. U.S.
4 EPA [1990] obtained individual animal data (including individual animal pathology reports for the
5 respiratory and upper digestive tracts, time-to-death data, and exposure chamber monitoring data)
6 from the study authors [Clement Associates. 1990): this information is summarized in Table 1-15.
7 Concentration-dependent increased incidences of tumors in the upper respiratory tract, including
8 the larynx and trachea, were seen at measured exposure concentrations of >9.5 mg/m3. In addition,
9 a decrease in mean tumor latency was observed in the larynx and trachea. Nasal cavity tumors
10 were observed at the mid- and high-concentration, but the incidences were not dose-dependent. A
11 concentration-related increase in tumors in the upper digestive tract (pharynx and esophagus) was
12 also reported. In addition, a single forestomach tumor was observed in each of the mid- and high-
13 concentration groups. Also, in animals with a tumor in either the larynx or pharynx, forestomach
14 tumors were not observed in control animals. The study authors suggested that the upper digestive
15 tract tumors were a consequence of mucociliary particle clearance. All nasal, forestomach,
16 esophageal, and tracheal tumors occurred in hamsters that also had tumors in the larynx or
17 pharynx, except in the mid-concentration group, where two animals with nasal tumors had no
18 tumors in the pharynx or larynx.
19 A re-analysis of the individual animal pathology reports and the exposure chamber
20 monitoring data provided by the study authors yielded estimates of average continuous lifetime
21 exposures for each individual hamster. Group averages of individual average continuous lifetime
22 exposure concentrations were 0, 0.25,1.01, and 4.29 mg/m3 for the control through high-exposure
23 groups (U.S. EPA. 1990).
This document is a draft for review purposes only and does not constitute Agency policy.
1-65 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Table 1-15. Tumors observed in chronic inhalation animal bioassays
Reference and study design
Results'1
Thyssenetal. (1981)
Syrian golden hamsters: male
(26-34 animals/group placed on study)
0, 2.2, 9.5, or 46.5 mg/m3 on NaCI particles by nose only
inhalation for 3-4.5 hrs, 5-7 d/wk
(TWA exposure concentrations3: 0, 0.25,1.01, and
4.29 mg/m3)
Treated until moribund or dead (up to 130 wks)
MMAD: not reported
Larynx
incidences: 0/26; 0/21; 11/26; and 11/25
mean tumor latency0: 107 and 68 wks
Pharynx
incidences: 0/23; 0/19; 9/22; and 18/23
mean tumor latency: 97 and 68 wks
Trachea
incidences: 0/27; 0/21; 2/26; and 3/25
mean tumor latency: 115 and 63 wks
Nasal cavity
incidences: 0/26; 0/22; 4/26; and 1/34
mean tumor latency: 116 and 79 wks
Esophagus
incidences: 0/27; 0/22; 0/26; and 2/34
mean tumor latency: 71 wks
Forestomach
incidences: 0/27; 0/22; 1/26; and 2/34
mean tumor latency: 119 and 72 wks
2
3 aDuration-adjusted inhalation concentrations calculated from exposure chamber monitoring data and exposure
4 treatment times. Daily exposure times: 4.5 hours/day, 5 days/week on weeks 1-12; 3 hours/day, 5 days/week on
5 weeks 13-29; 3.7 hours/day, 5 days/week on week 30; 3 hours/day, 5 days/week on weeks 31-41; and
6 3 hours/day, 7 days/week for reminder of the experiment.
7 bThyssen et al. (1981) reported only the incidences of malignant tumors, confirmed by comparison with the
8 original study pathology data (Clement Associates, 1990). The incidences summarized here include relevant
9 benign tumors (papillomas, polyps, and papillary polyps). The malignant tumors were squamous cell carcinomas,
10 with the exception of one in situ carcinoma of the larynx and one adenocarcinoma of the nasal cavity, both in the
11 9.5 mg/m3 group. Denominators reflect the number of animals examined for histopathology for each tissue. See
12 Section D.4.2 and Table E-17 in the Supplemental Material for study details and a complete listing of individual
13 data, respectively.
14 cMean time of observation of tumor, 9.5 and 46.5 mg/m3 concentration groups.
15 dThyssen et al. (1981) did not report statistical significance testing. See Section D.4.2.
16 Dermal exposure
17 Repeated application of benzo[a]pyrene to skin (in the absence of exogenous promoters)
18 has been demonstrated to induce skin tumors in mice, rats, rabbits, and guinea pigs. These studies
19 have been reviewed by multiple national and international health agencies [IARC. 2010: IPCS. 1998:
20 ATSDR. 1995: IARC. 1983.1973]. Mice have been the most extensively studied species in dermal
21 carcinogenesis studies of benzo[a]pyrene because of evidence thatthey may be more sensitive than
22 other animal species; however, comprehensive comparisons of species differences in sensitivity to
This document is a draft for review purposes only and does not constitute Agency policy,
1-66 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 lifetime dermal exposure are not available. Systemic tumors in benzo[a]pyrene-treated mice were
2 not increased compared to controls in lifetime dermal bioassays in which macroscopic examination
3 of internal organs was included [Higginbotham etal.. 1993: Habsetal.. 1980: Schmahl etal.. 1977:
4 Schmidt etal.. 1973: Roe etal.. 1970: Poel. 19591
5 The analysis in this document focuses on lifetime carcinogenicity bioassays in several
6 strains of mice following repeated dermal exposure to benzo[a]pyrene (Table 1-16). These studies
7 involved 2- or 3-times/week exposure protocols, at least two exposure levels plus controls, and
8 histopathological examinations of the skin and other tissues [Sivaketal.. 1997: Grimmer etal..
9 1984: Habsetal.. 1984: Grimmer etal.. 1983: Habsetal.. 1980: Schmahl etal.. 1977: Schmidt etal..
10 1973: Roe etal.. 1970: Poel. 1963. 19591.
11 Numerous studies in mice observed skin tumors following benzo[a]pyrene exposure, but
12 were not considered further in this assessment because of the availability of the lifetime studies
13 identified above. These studies included several "skin painting" studies in mouse skin that did not
14 report the doses applied (e.g., Wynder and Hoffmann. 1959: Wynderetal.. 1957]: several shorter-
15 term studies (Albertetal.. 1991: Nesnowetal.. 1983: Emmettetal.. 1981: Levin etal.. 1977]:
16 initiation-promotion studies utilizing acute dosing of benzo[a]pyrene followed by repeated
17 exposure to a potent tumor promoter; and studies involving vehicles expected to interact with or
18 enhance benzo[a]pyrene carcinogenicity (e.g.. Bingham and Falk. 1969].
19 One study applied benzo[a]pyrene (topically once a week for 6 months] to immuno-
20 compromised mice with human skin grafts (n = 10] and did not observe tumors, whereas all three
21 control mice (mice with no skin grafts] developed skin tumors (Urano etal.. 1995]. The authors
22 concluded this result indicates that human skin is much less susceptible to benzo[a]pyrene than
23 mouse skin. Though some studies indicate that the skin grafts maintain some metabolic function
24 (Das etal., 1986], it is unclear whether the human skin grafts maintain the same viability,
25 vascularization, and full metabolic capacity as human skin in vivo (Kappes etal.. 2004]. Another
26 concern is the short amount of time allowed for tumor development All of the mice with human
27 skin grafts treated with benzo [ajpyrene died within 6 months of the start of treatment (Urano etal..
28 1995]. While 6 months is generally sufficient for the development of tumors in mouse skin, human
29 latency for squamous cell carcinoma in PAH-exposed occupational cohorts is thought to be
30 >20 years (Youngetal.. 2012: Voelter-MahlknechtetaL 2007: Everall and Dowd. 1978]. Potent
31 mutagenic carcinogens such as 7,12-dimethylbenz[a]anthracene, methylcholanthrene, and
32 methylnitronitrosoguanidine also fail to produce skin tumors in this model system (Soballe etal..
33 1996: Urano etal.. 1995: Graem. 1986]. Therefore, the ability of this model system to predict
34 hazard for human skin cancer risk (particularly from metabolically active carcinogens] is unclear.
This document is a draft for review purposes only and does not constitute Agency policy.
1-67 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Table 1-16. Tumors observed in chronic dermal animal bioassays
Reference and study design
Results3
Poel (1959)
C57Lmice: male (13-56/dose)
0, 0.15, 0.38, 0.75, 3.8, 19, 94, 188, 376, or 752 u.g
Dermal; 3 times/wk for up to 103 wks or until the
appearance of a tumor by gross examination
Skin tumors (gross skin tumors and epidermoid carcinoma);
dose-dependent decreased time of tumor appearance
incidences:
Gross skin tumors: 0/33; 5/55; 11/55; 7/56; 41/49;
38/38; 35/35; 12/14; 14/14; and 13/13
Epidermoid carcinoma: 0/33; 0/55; 2/55; 4/56;
32/49; 37/38; 35/35; 10/14; 12/14; and 13/13
Cytotoxicity: information not provided
Poel (1963)
SWR, CSHeB, or A/He mice: male (14-25/dose)
0, 0.15, 0.38, 0.75, 3.8, 19.0, 94.0, or 470 u.g
Dermal; 3 times/wk until mice died or a skin tumor
was observed
Skin tumors and dose-dependent decreased time of first
tumor appearance
incidences:
SWR: 0/20; 0/25; 2/22; 15/18; 12/17; 16/16; 16/17;
and 14/14
CSHeB: 0/17; 0/19; 3/17; 4/17; 11/18; 17/17; 18/18;
and 17/17
A/He mice: 0/17; 0/18; 0/19; 0/17; 0/17; 21/23;
11/16; and 17/17
Cytotoxicity: information not provided
Roe etal. (1970)
Swiss mice: female (50/dose)
0, vehicle, 0.1,0.3,1, 3, or 9 |jg
Dermal; 3 times/wk for up to 93 wks
Skin tumors; malignant skin tumors were observed in 4/41 and
31/40 mice in the two high-dose groups, respectively
incidences:
0/43; 0/47; 1/42; 0/42; 1/43; 8/41; and 34/46
Cytotoxicity: information not provided
Schmidt etal. (1973)
NMRI mice: female (100/group)
Swiss mice: female (100/group)
0, 0.05, 0.2, 0.8, or 2 u.g
Dermal; 2 times/wk until spontaneous death
occurred or until an advanced carcinoma was
observed
Skin tumors (carcinomas)
incidences:
NMRI:
2/100 at 2 |jg (papillomas);
2/100 at 0.8 u.g and 30/100 at 2 u.g (carcinomas)
Swiss:
3/80 at 2 |jg (papillomas);
5/80 at 0.8 u.g and 45/80 at 2 u.g (carcinomas)
Cytotoxicity: information not provided
Schmahl etal. (1977)
NMRI mice: female (100/group)
0,1,1.7, or 3 u.g
Dermal; 2 times/wk until natural death or until they
developed a carcinoma at the site of application
Skin tumors (papillomas and carcinomas)
incidences:
0/81; 1/77; 0/88; and 2/81 (papillomas)
0/81; 10/77; 25/88; and 43/81 (carcinomas)
Cytotoxicity: information not provided
Habs etal. (1980)
NMRI mice: female (40/group)
0,1.7, 2.8, or 4.6 |jg
Dermal; 2 times/wk until natural death or gross
observation of infiltrative tumor growth
Skin tumors and dose-dependent increase in age-standardized
tumor incidence
incidences:
0/35; 8/34; 24/35; and 22/36
age-standardized tumor incidence:
0, 24.8, 89.3, and 91.7%
Cytotoxicity: information not provided
This document is a draft for review purposes only and does not constitute Agency policy,
1-68 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Reference and study design
Results3
(Grimmer et al. (1984); Grimmer et al. (1983))
CFLP mice: female (65-80/group)
0, 3.9, 7.7, or 15.4 u.g (1983 study)
0, 3.4, 6.7, or 13.5 u.g (1984 study)
Dermal; 2 times/wk for 104 wks
Skin tumors (papillomas and carcinomas) with a decrease in
tumor latency
incidences:
1983: 0/80; 7/65; 5/64; and 2/64 (papillomas)
0/80; 15/65; 34/64; and 54/64 (carcinomas)
1984: 0/65; 6/64; 8/65; and 4/65 (papillomas)
0/65; 37/64; 45/65; and 53/65 (carcinomas)
Cytotoxicity: information not provided
Habsetal. (1984)
NMRI mice: female (20/group)
0, 2, or 4 u.g
Dermal; 2 times/wk for life
Skin tumors (papillomas and carcinomas) with a decrease in
mean survival time
incidences:
0/20; 2/20; and 0/20 (papillomas)
0/20; 7/20; and 17/20 (carcinomas)
Cytotoxicity: information not provided
(Sivak et al. (1997); Arthur D Little, 1989; NIOSH
(1989))
C3H/HeJ mice: male (30/group)
0, 0.05, 0.5, or 5 u.g
Dermal; 2 times/wk for up to 104 wks
Skin tumors (papillomas and carcinomas)
incidences:
0/30; 0/30; 5/30 (1 papilloma, 1 keratoacanthoma,
3 carcinomas); and 27/30 (1 papilloma,
28 carcinomas)
Cytotoxicity: 80% incidence of scabs and sores in highest dose
group; no Cytotoxicity noted at lower doses
1
2 Statistical significance not reported by study authors.
3 Mode-of-Action Analysis—Carcinogenicity
4 The carcinogenicity of benzo[a]pyrene, the most studied PAH, is well documented [IARC.
5 2010: Xuetal.. 2009: TiangetaL 2007: TiangetaL 2005: Xue and Warshawsky. 2005: RameshetaL
6 2004: Bostr6metal..20Q2: Penningetal.. 1999: IPCS. 1998: Harvey. 1996: ATSDR. 1995: Cavalieri
7 andRogan. 1995: U.S. EPA. 1991b). The primary mode of action by which benzo[a]pyrene induces
8 carcinogenicity is via a mutagenic mode of action. This mode of action is presumed to apply to all
9 tumor types and is relevant for all routes of exposure. The general sequence of key events
10 associated with a mutagenic mode of action for benzo[a]pyrene is: (1) bioactivation of
11 benzo[a]pyrene to DNA-reactive metabolites via three possible metabolic activation pathways: a
12 diol epoxide pathway, a radical cation pathway, and an o-quinone and ROS pathway; (2) direct DNA
13 damage by reactive metabolites, including the formation of DNA adducts and ROS-mediated
14 damage; (3) formation and fixation of DNA mutations, particularly in tumor suppressor genes or
15 oncogenes associated with tumor initiation; and (4) clonal expansion of mutated cells during the
16 promotion and progression phases of cancer development. These events are depicted as stages of
17 benzo[a]pyrene-induced carcinogenesis in Figure 1-6.
18 Benzo[a]pyrene is a complete carcinogen, in that it can act as both an initiator and a
19 promoter of carcinogenesis. Initiation via direct DNA damage (key event 2) can occur via all three
20 metabolites of benzo[a]pyrene. DNA damage that is not adequately repaired leads to mutation (key
This document is a draft for review purposes only and does not constitute Agency policy,
1-69 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
event 3), and these mutations can undergo clonal expansion (key event 4) enabled by multiple
mechanisms also induced by benzo[a]pyrene, including AhR binding leading to an upregulation of
genes related to biotransformation, growth, and differentiation, and regenerative cell proliferation
resulting from cytotoxicity and a sustained inflammatory response. However, there is not sufficient
evidence that these mechanisms, which contribute to the promotion and progression phases of
cancer development, act independently of DNA damage and mutation to produce benzo[a]pyrene-
induced tumors (please see Other possible modes of action, below). The available human, animal,
and in vitro evidence supports a mutagenic mode of action as the primary mode by which
benzo[a]pyrene induces carcinogenesis.
Key events in the mode of action for benzo[a]pyrene carcinogenicity
Mutation
(transversion
K-ras,H-ras,
andpSJ targets
Upregulation of genes
related to
biotransformation,
growth, and
differentiation
Mutation
(depurination
H-rastarget
ci-qumone
andROS
Mutation
(depurination
oxidative
damage and
strand scission
DNAadducts
and oxidative
base damage
Figure 1-6. Proposed metabolic activation pathways and key events in the
carcinogenic mode of action for benzo[a]pyrene.
Data in Support of the Mode of Action
Summary of metabolic activation pathways
Dial epoxide pathway. Benzo[a]pyrene diol epoxide metabolites, believed to be the most
potent DNA-binding metabolites of benzo[a]pyrene, are formed through a series of Phase I
metabolic reactions (see Appendix D of the Supplemental Information). The initial metabolism is
carried out primarily by the inducible activities of GYP enzymes including CYP1A1, CYP1B1, and
CYP1A2. Further metabolism by epoxide hydrolase and the mixed function oxidase system yields
This document is a draft for review purposes only and does not constitute Agency policy,
1-70 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 (+)-anti-BPDE, one of the mostpotentDNA-binding metabolites of benzo[a]pyrene.
2 Benzo[a]pyrene diol epoxide metabolites interact preferentially with the exocyclic amino groups of
3 deoxyguanine and deoxyadenine [Geacintovetal.. 1997: TerinaetaL 1991). Adducts may give rise
4 to mutations unless these adducts are removed by DNA repair processes prior to replication. The
5 stereochemical nature of the diol epoxide metabolite (i.e., anti- versus syn-diol epoxides) affects the
6 number and type of adducts and mutation that occurs [Geacintovetal.. 1997}. Transversion
7 mutations (e.g., GC^TA or AT^TA) are the most common type of mutation found in mammalian
8 cells following diol epoxide exposure (Bostrometal.. 2002].
9 Radical cation pathway. Radical cation formation involves a one-electron oxidation by GYP
10 or peroxidase enzymes (i.e., horseradish peroxidase, prostaglandin H synthetase) that produce
11 electrophilic radical cation intermediates (Cavalieri and Rogan. 1995.1992}. Radical cations can be
12 further metabolized to phenols and quinones (Cavalieri et al.. 1988e: Cavalieri etal.. 1988d). or they
13 can form unstable adducts with DNA that ultimately result in depurination. The predominant
14 depurinating adducts occur at the N-3 and N-7 positions of adenine and the C-8 and N-7 positions of
15 guanine (Cavalieri and Rogan. 1995].
16 o-Quinone/ROS pathway. The o-quinone metabolites of PAHs are formed by enzymatic
17 dehydroge nation of dihydrodiols (Bolton etal.. 2000: Penning etal.. 1999: Harvey. 1996: ATSDR.
18 1995] (see Appendix D of the Supplemental Information]. Dihydrodiol dehydrogenase enzymes are
19 members of the a-keto reductase gene superfamily. o-Quinone metabolites are potent cytotoxins,
20 are weakly mutagenic, and are capable of producing a broad spectrum of DNA damage. These
21 metabolites can interact directly with DNA as well as result in the production of ROS (i.e., hydroxyl
22 and superoxide radicals] that may produce further cytotoxicity and DNA damage. The
23 o-quinone/ROS pathway also can produce depurinated DNA adducts from benzo[a]pyrene
24 metabolites. In this pathway, and in the presence of NAD(P]+, aldo-keto reductase oxidizes
25 benzo[a]pyrene-7,8-diol to a ketol, which subsequently forms benzo[a]pyrene-7,8-dione. This and
26 other PAH o-quinones react with DNA to form unstable, depurinating DNA adducts. In the presence
27 of cellular reducing equivalents, o-quinones can also activate redox cycles, which produce ROS
28 (Penning etal.. 1996]. DNA damage in in vitro systems following exposure to benzo[a]pyrene-
29 7,8-dione or other o-quinone PAH derivatives occurs through the aldo-keto reductase (AKR]
30 pathway and can involve the formation of stable DNA adducts (Baluetal., 2004], N-7 depurinated
31 DNA adducts (Mccoull etal.. 1999]. DNA damage from ROS (8-oxo-7,8-dihydro-2'-deoxyguanosine
32 adducts] (Park etal.. 2006). and strand scission (Flowers etal.. 1997: Flowers etal.. 19961
33 Summary of genotoxicity and mutagenicity
34 The ability of metabolites of benzo[a]pyrene to cause mutations and other forms of DNA
35 damage in both in vivo and in vitro studies is well documented (see genotoxicity tables in
36 Appendix D in Supplemental Information). With metabolic activation (e.g., the inclusion of S9),
37 benzo[a]pyrene is consistently mutagenic in the prokaryotic Salmonella/Ames and Escherichia coli
38 assays. In mammalian in vitro studies, benzo[a]pyrene is consistently mutagenic and clastogenic,
This document is a draft for review purposes only and does not constitute Agency policy.
1-71 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 and induces cell transformation both with and without metabolic activation. Cytogenetic damage in
2 the form of chromosomal aberrations (CAs), micronuclei (MN), sister chromatid exchanges (SCEs),
3 and aneuploidy are commonplace following benzo[a]pyrene exposure as are DNA adduct
4 formation, single-strand breaks (SSB), and induction of DNA repair and unscheduled DNA synthesis
5 (UDS). In vitro mammalian cell assays have been conducted in various test systems, including
6 human cell lines.
7 In the majority of in vivo studies, benzo[a]pyrene has tested positive in multiple species and
8 strains and under various test conditions for cell transformation, CAs, DNA adducts, DNA strand
9 breaks, MN formation, germline mutations, somatic mutations (H-ras, K-ras, p53, lacZ, hprt), and
10 SCEs. Human studies are available following exposures to PAH mixtures through cigarette smoke
11 or occupational exposure in which benzo[a]pyrene-specific DNA adducts have been detected, and it
12 has been demonstrated qualitatively that benzo[a]pyrene metabolites damage DNA in exposed
13 humans.
14 Experimental support for the hypothesized mode of action
15 EPA's Guidelines for Carcinogen Risk Assessment [Section 2.4; [2005a]] describe a procedure
16 for evaluating mode-of-action data for cancer. A framework for analysis of mode of action
17 information is provided and followed below.
18 Strength, consistency, and specificity of association. Strong evidence links the
19 benzo[a]pyrene diol epoxide metabolic activation pathway with key mutational events in genes that
20 are associated with tumor initiation (mutations in the p53 tumor suppressor gene and H-ras or
21 K-ras oncogenes) (Table 1-17). Results in support of a mutagenic mode of action via
22 benzo[a]pyrene diol epoxide include observations of frequent G^T transversion mutations in p53
23 and ras genes in lung tumors of human cancer patients exposed to coal smoke (Keohavongetal.,
24 2003: DeMarini et al.. 2001). These results are consistent with evidence thatbenzo[a]pyrene diol
25 epoxide is reactive with guanine bases in DNA; that G^T transversions, displaying strand bias, are
26 the predominant type of mutations caused by benzo[a]pyrene in several biological systems (Liu et
27 al.. 2005: Hainaut and Pfeifer. 2001: Marshall etal.. 1984): and that sites of DNA adduction at
28 guanine positions in cultured human HeLa or bronchial epithelial cells exposed to benzo[a]pyrene
29 diol epoxide correspond to p53 mutational hotspots observed in human lung cancers (Denissenko
30 etal., 1996: Puisieuxetal., 1991). In addition, mice exposed to benzo[a]pyrene in the diet (Culp et
31 al.. 2000) or by i.p. injection (Nesnowetal.. 1998a: Nesnowetal.. 1998b: Nesnowetal.. 1996.1995:
32 Mass etal.. 1993) had forestomach or lung tumors, respectively, showing frequent G—>T or C
33 transversions in the K-ras gene. Supporting evidence includes observations that benzo[a]pyrene
34 diol epoxide (specifically (+)-anti-BPDE) is more potent than benzo[a]pyrene itself, benzo[a]pyrene
35 phenols, or benzo[a]pyrene diols in mutagenicity assays in bacterial and in vitro mammalian
36 systems (Malaveille etal., 1977: Newbold and Brookes, 1976) and in producing lung tumors in
37 newborn mice following i.p. administration. Other supporting evidence includes observations of
38 elevated BPDE-DNA adduct levels in WBCs of groups of coke oven workers and chimney sweeps,
This document is a draft for review purposes only and does not constitute Agency policy.
1-72 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 occupations with known elevated risks of cancer [Vineisetal., 2007: Pavanello etal., 1999], and in
2 lung tissue from tobacco smokers with lung cancer [Rojas etal.. 2004: Godschalketal.. 2002: Rojas
3 etal.. 1998: Andreassen etal.. 1996: Alexandrovetal.. 1992}. Several epidemiological studies have
4 indicated that PAH-exposed individuals who are homozygous for a CYP1A1 polymorphism, which
5 increases the inducibility of this enzyme (thus increasing the capacity to produce benzo[a]pyrene
6 diol epoxide), have increased levels of PAH or BPDE-DNA adducts [Aklillu etal.. 2005: Alexandrov
7 etal.. 2002: Bartsch etal.. 2000: Perera and Weinstein. 2000).
8 Additional supporting evidence of a mutagenic mode of action for benzo[a]pyrene
9 carcinogenicity is the extensive database of in vitro and in vivo studies demonstrating the
10 genotoxicity and mutagenicity of benzo[a]pyrene following metabolic activation (Table 1-17). In
11 vitro studies overwhelmingly support the formation of DNA adducts, mutagenesis in bacteria, yeast,
12 and mammalian cells, several measures of cytogenetic damage (CA, SCE, MN), and DNA damage. In
13 vivo systems in animal models are predominantly positive for somatic mutations following
14 benzo[a]pyrene exposure.
15 Support for the radical cation activation pathway contributing to tumor initiation through
16 mutagenic events includes observations that depurinated DNA adducts (expected products from
17 reactions of benzo[a]pyrene radical cations with DNA) accounted for 74% of identified DNA
18 adducts in mouse skin exposed to benzo[a]pyrene (Roganetal.. 1993] and that 9 of 13 tumors
19 examined from mice exposed to dermal applications of benzo[a]pyrene had H-ras oncogene
20 mutations attributed to depurinated DNA adducts from benzo[a]pyrene radical cations
21 (Chakravarti et al.. 1995).
22 Support for the o-quinone/ROS pathway contributing to tumor initiation via mutagenic
23 events includes in vitro demonstration that several types of DNA damage can occur from
24 o-quinones and ROS (Parketal.. 2006: Balu etal.. 2004: Mccoull etal.. 1999: Flowers etal.. 1997:
25 Flowers etal.. 1996). In addition, benzo[a]pyrene-7,8-dione can induce mutations in the p53 tumor
26 suppressor gene using an in vitro yeast reporter gene assay (Park etal.. 2008: Shenetal.. 2006: Yu
27 etal.. 2002). and dominant p53 mutations induced by benzo[a]pyrene-7,8-dione in this system
28 corresponded to p53 mutational hotspots observed in human lung cancer tissue (Park etal.. 2008).
29 Dose-response concordance and temporal relationship. Studies in humans demonstrating
30 that benzo [a]pyrene-induced mutational events in p53 or ras oncogenes precede tumor formation
31 are not available, but there is evidence linking benzo [a] pyrene exposure to signature mutational
32 events in humans. In vitro exposure of human p53 knock-in murine fibroblasts to 1 |iM
33 benzo[a]pyrene for 4-6 days induced p53 mutations with similar features to those identified in p53
34 mutations in human lung cancer; i.e., predominance of G^T transversions with strand bias and
35 mutational hotspots atcodons 157-158 (Liu etal.. 2005).
36 Bennett etal. (1999) demonstrated a dose-response relationship between smoking
37 history/intensity and the types of p53 mutations associated with benzo[a]pyrene (G—>T
38 transversions) in human lung cancer patients (Table 1-17). In lung tumors of nonsmokers, 10% of
This document is a draft for review purposes only and does not constitute Agency policy.
1-73 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 p53 mutations were G^T transversions, versus 40% in lung tumors from smokers with >60 pack-
2 years of exposure.
3 In mice, dose-response and temporal relationships have been described between the
4 formation of BPDE-DNA adducts and skin and forestomach tumors (Table 1-17). In a study using
5 mice treated dermally with benzo[a]pyrene once or twice per week for up to 15 weeks (10, 25, or
6 50 nmol benzo[a]pyrene per application), levels of benzo[a]pyrene-DNA adducts in the skin, lung,
7 and liver increased with increasing time of exposure and increasing dose levels (Talaskaetal..
8 2006). Levels at the end of the exposure period were highest in the skin; levels in the lung and liver
9 at the same time were 10- and 20-fold lower, respectively. Levels of benzo[a]pyrene-DNA adducts
10 in skin and lung increased in an apparent biphasic manner showing a lower linear slope between
11 the two lowest dose levels, compared with the slope from the middle to the highest dose.
12 Another study examined the dose-response relationship and the time course of
13 benzo[a]pyrene-induced skin damage (Table 1-17), DNA adduct formation, and tumor formation in
14 female mice. Mice were treated dermally with 0,16, 32, or 64 |ig of benzo[a]pyrene once per week
15 for 29 weeks (Albert etal.. 1991). Indices of skin damage and levels of BPDE-DNA adducts in skin
16 reached plateau levels in exposed groups by 2 -4 weeks of exposure. With increasing dose level,
17 levels of BPDE-DNA adducts (fmol/ng DNA) initially increased in a linear manner and began to
18 plateau at doses >32 [ig/week. Tumors began appearing after 12-14 weeks of exposure for the
19 mid- and high-dose groups and at 18 weeks for the low-dose group. At study termination
20 (3 5 weeks after start of exposure), the mean number of tumors per mouse was approximately one
21 per mouse in the low- and mid-dose groups and eight per mouse in the high-dose group. The time-
22 course data indicate thatbenzo[a]pyrene-induced increases in BPDE-DNA adducts preceded the
23 appearance of skin tumors, consistent with the formation of DNA adducts as a precursor event in
24 benzo [a] pyrene-induced skin tumors. A follow-up to this study by the same authors (Albert etal.,
25 1996) measured DNA adducts, necrosis, and inflammation (marked by an increase in leukocytes) in
26 the skin of treated mice after 5 weeks of dermal exposure. In the 64 [ig/week dose group,
27 statistically elevated levels of DNA adducts, inflammation, and necrosis were reported; however, in
28 the lower dose group (16 [ig/week), DNA adducts were statistically significantly elevated without
29 increases in inflammation and necrosis.
30 Gulp etal. (1996) compared dose-response relationships for BPDE-DNA adducts and
31 tumors in female B6C3Fi mice exposed to benzo [a] pyrene in the diet at 0,18.5, 90, or 350 [J.g/day
32 for 28 days (to examine adducts) or 2 years (to examine tumors) (Table 1-17). The benzo[a]pyrene
33 dose-tumor response data showed a sharp increase in forestomach tumor incidence between the
34 18.5 ^g/day group (6% incidence) and the 90 ^g/day group (78% incidence). The BPDE-DNA
35 adduct levels in forestomach showed a relatively linear dose-response throughout the
36 benzo[a]pyrene dose range tested. The appearance of increased levels of BPDE-DNA adducts in the
37 target tissue at 2 8 days is temporally consistent with the contribution of these adducts to the
38 initiation of forestomach tumors. Furthermore, about 60% of the examined tumors had mutations
This document is a draft for review purposes only and does not constitute Agency policy.
1-74 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 in the K-ras oncogene at codons 12 and 13, which were G^T or G^C transversions indicative of
2 BPDE reactions with DNA (Gulp etal.. 1996).
3 Biological plausibility and coherence. The evidence for a mutagenic mode of action for
4 benzo[a]pyrene is consistent with the current understanding that mutations in p53 and ras
5 oncogenes are associated with increased risk of tumor initiation (Table 1-17). The benzo[a]pyrene
6 database is internally consistent in providing evidence for BPDE-induced mutations associated with
7 tumor initiation in cancer tissue from humans exposed to complex mixtures containing
8 benzo[a]pyrene [Keohavongetal.. 2003: Pfeifer andHainaut. 2003: Pfeifer etal.. 2002: DeMarini et
9 al., 2001: Hainaut and Pfeifer, 2001: Bennett etal., 1999), in animals exposed to benzo[a]pyrene
10 [Gulp etal.. 2000: Nesnowetal.. 1998a: Nesnowetal.. 1998b: Nesnowetal.. 1996.1995: Mass etal..
11 1993). and in in vitro systems [Denissenko etal.. 1996: Puisieux et al.. 1991). Consistent
12 supporting evidence includes: (1) elevated BPDE-DNA adduct levels in tobacco smokers with lung
13 cancer [Rojas etal.. 2004: Godschalk etal.. 2002: Rojasetal.. 1998: Andreassen et al.. 1996:
14 Alexandrovetal.. 1992): (2) demonstration of dose-response relationships between G—>T
15 transversions in p53 mutations in lung tumors and smoking intensity [Bennett etal.. 1999): (3) the
16 extensive database of in vitro and in vivo studies demonstrating the genotoxicity and mutagenicity
17 of benzo[a]pyrene following metabolic activation; and (4) general concordance between temporal
18 and dose-response relationships for BPDE-DNA adduct levels and tumor incidence in studies of
19 animals exposed to benzo[a]pyrene [Gulp etal.. 1996: Albert etal.. 1991). There is also supporting
20 evidence that contributions to tumor initiation through mutagenic events can be made by the
21 radical cation [Chakravartietal.. 1995: Roganetal.. 1993) and o-quinone/ROS metabolic activation
22 pathways [Park etal.. 2008: Park etal.. 2006: Shen etal.. 2006: Balu etal.. 2004: Yu etal.. 2002:
23 Mccoulletal.. 1999: Flowers etal.. 1997: Flowers etal.. 1996).
24 Table 1-17. Experimental support for the postulated key events for mutagenic
25 mode of action
1. Bioactivation of benzo[a]pyrene to DNA-reactive metabolites via three possible metabolic activation
pathways: a diol epoxide pathway, a radical cation pathway, and an o-quinone and ROS pathway
Evidence that benzo[a]pyrene metabolites induce key events:
• Metabolism of benzo[a]pyrene via all three pathways has been demonstrated in multiple in vitro
studies, and the diol epoxide and radical cation metabolic activation pathways have been demonstrated
in in vivo studies in humans and animals (see Metabolic Activation Pathways section)
• Multiple in vivo studies in humans and animals have demonstrated distribution of reactive metabolites
to target tissues
Human evidence that key events are necessary for carcinogenesis:
• Humans with CYP polymorphisms or lacking a functional GSTM1 gene form higher levels of benzo[a]pyrene
diol epoxides, leading to increased BPDE-DNA adduct formation and increased risk of cancer (Vineis et al.,
2007; Pavanello et al., 2005; Pavanello et al., 2004; Alexandrovetal., 2002; Perera and Weinstein, 2000)
This document is a draft for review purposes only and does not constitute Agency policy.
1-75 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
2. Direct DNA damage by the reactive metabolites, including the formation of DNA adducts and ROS-mediated
damage
Evidence that benzo[a]pyrene metabolites induce key events:
• Reactive benzo[a]pyrene metabolites have demonstrated genotoxicity in most in vivo and in vitro
systems in which they have been tested, including the bacterial mutation assay, transgenic mouse assay,
dominant lethal mutations in mice, BPDE-DNA adduct detection in humans and animals, and DNA
damage, CAs, MN formation, and SCE in animals (Appendix D in Supplemental Information)
• Multiple in vivo benzo[a]pyrene animal exposure studies have demonstrated DNA adduct formation in
target tissues that precede tumor formation and increase in frequency with dose (Gulp et al., 1996;
Talaska et al., 1996; Albert et al., 1991)
• Benzo[a]pyrene diol epoxide metabolites interact preferentially with the exocyclic amino groups of
deoxyguanine and deoxyadenine in DNA (Geacintov et al., 1997; Jerina et al., 1991; Koreeda et al., 1978;
Jeffrey et al., 1976)
• Benzo[a]pyrene o-quinone metabolites are capable of activating redox cycles and producing ROS that
cause oxidative base damage (Park et al., 2006; Balu etal., 2004; Mccoull etal., 1999; Flowers et al.,
1997; Flowers et al., 1996)
Human evidence that key events are necessary for carcinogenesis:
• Detection of benzo[a]pyrene diol epoxide-specific DNA adducts is associated with increased cancer risk
in humans that are occupationally exposed (see Evidence in Humans section)
• These benzo[a]pyrene diol epoxides formed BPDE-DNA adducts preferentially at guanine residues that
have been detected in tissues of humans with cancer who were exposed to PAHs (Vineis and Perera,
2007; Rojas et al., 2004; Godschalk et al., 2002; Li et al., 2001; Pavanello etal., 1999; Rojaset al., 1998;
Andreassen et al., 1996; Alexandrov et al., 1992)
3. Formation and fixation of DNA mutations, particularly in tumor suppressor genes or oncogenes associated
with tumor initiation
Evidence that benzo[a]pyrene metabolites induce key events:
• Several in vivo exposure studies have observed benzo[a]pyrene diol epoxide-specific mutational spectra
(e.g., G->T transversion mutations) in K-ras, H-ras, and p53 in forestomach or lung tumors (Gulp et al.,
2000; Nesnowetal., 1998a; Nesnowetal., 1998b; Nesnowetal., 1996,1995; Mass etal., 1993)
• Multiple animal exposure studies have identified benzo[a]pyrene-specific mutations in H-ras, K-ras, and
p53 in target tissues preceding tumor formation (Liu etal., 2005; Wei et al., 1999; Gulp etal., 1996)
(Chakravarti et al., 1995; Ruggeri et al., 1993)
Human evidence that key events are necessary for carcinogenesis:
• DNA adducts formed by the benzo[a]pyrene diol epoxide reacting with guanine bases lead
predominantly to G->T transversion mutations; these specific mutational spectra have been identified in
PAH-associated tumors in humans at mutational hotspots, including oncogenes (K-ras) and tumor
suppressor genes (p53) (Liu et al., 2005; Keohavong et al., 2003; Pfeifer and Hainaut, 2003; Pfeifer et al.,
2002; DeMarini etal., 2001; Hainaut and Pfeifer, 2001; Bennett et al., 1999; Denissenko et al., 1996;
Puisieuxetal., 1991; Marshall etal., 1984; Koreeda et al., 1978; Jeffrey et al., 1976)
This document is a draft for review purposes only and does not constitute Agency policy.
1-76 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
4. Clonal expansion of mutated cells during the promotion and progression phases of cancer development
Evidence that benzo[a]pyrene metabolites induce key events:
• Benzo[a]pyrene has been shown to be a complete carcinogen, in that skin tumors in mice, rats, rabbits,
and guinea pigs have been associated with repeated application of benzo[a]pyrene to skin in the
absence of exogenous promoters (IPCS, 1998; Sivak et al., 1997; ATSDR, 1995; Grimmer et al., 1984;
Habs et al., 1984; Grimmer et al., 1983; IARC, 1983, 1973; Habs et al., 1980; Schmahl et al., 1977;
Schmidt etal., 1973; Roe et al., 1970; Poel, 1960, 1959(IPCS, 1998; ATSDR, 1995; IARC, 1983, 1973)
• Mice exposed dermally to benzo[a]pyrene for 26 weeks were found to have increased frequencies of H-
ras mutations in exposure-induced hyperplastic lesions that were further increased in tumors (Wei etal.,
1999)
• AhR activation by PAHs (including benzo[a]pyrene) upregulates genes responsible for tumor promotion
and increases tumor incidence in mice (Ma and Lu, 2007; Talaska et al., 2006; Shimizu et al., 2000)
1 Other possible modes of action
2 The carcinogenic process for benzo[a]pyrene is likely to be related to some combination of
3 molecular events resulting from the formation of several reactive metabolites that interact with
4 DNA to form adducts and produce DNA damage resulting in mutations in cancer-related genes, such
5 as tumor suppressor genes or oncogenes. These events may reflect the initiation potency of
6 benzo[a]pyrene. However, benzo[a]pyrene possesses promotional capabilities that may be related
7 to AhR affinity, immune suppression, cytotoxicity and inflammation (including the formation of
8 ROS), as well as the inhibition of gap junctional intercellular communication (GJIC).
9 The ability of certain PAHs to act as initiators and promoters may increase their
10 carcinogenic potency. The promotional effects of PAHs appear to be related to AhR affinity and the
11 upregulation of genes related to growth and differentiation (Bostrom etal.. 2002]. The genes
12 regulated by this receptor belong to two major functional groups (i.e., induction of metabolism or
13 regulation cell differentiation and proliferation). PAHs bind to the cytosolic AhR in complex with
14 heat shock protein 90. The ligand-bound receptor is then transported to the nucleus in complex
15 with the AhR nuclear translocator protein. The AhR complex interacts with AhR elements of DNA
16 to increase the transcription of proteins associated with induction of metabolism and regulation of
17 cell differentiation and proliferation. Following benzo[a]pyrene exposure, disparities have been
18 observed in the tumor pattern and toxicity of Ah-responsive and Ah-nonresponsive mice, as
19 Ah-responsive mice were more susceptible to tumorigenicity in target tissues such as liver, lung,
20 and skin (Ma and Lu. 2007: Talaska etal.. 2006: Shimizu etal.. 20001.
21 Benzo[a]pyrene has both inflammatory and immunosuppressive effects that may function
22 to promote tumorigenesis. Inflammatory responses to cytotoxicity may contribute to the tumor
23 promotion process; for example, benzo[a]pyrene quinones (1,6-, 3,6-, and 6,12-benzo[a]pyrene-
24 quinone) generated ROS and increased cell proliferation by enhancing the epidermal growth factor
25 receptor pathway in cultured breast epithelial cells (Burdicketal.. 2003]. In addition, several
26 studies have demonstrated that exposure to benzo[a]pyrene increases the production of
This document is a draft for review purposes only and does not constitute Agency policy.
1-77 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 inflammatory cytokines, which may contribute to cancer progression [N'Diaye etal., 2006: Tamaki
2 et al.. 2004: Garcon et al.. 2001b: Garcon et al.. 2001a: Albert et al.. 1996: Albert et al.. 19911 One of
3 these studies, Albert etal. [1996]. measured DNA adducts, necrosis, and inflammation (marked by
4 an increase in leukocytes) in the skin of benzo[a]pyrene-treated mice after 5 weeks of dermal
5 exposure. In the highest dose group, statistically elevated levels of DNA adducts, inflammation, and
6 necrosis were reported; however, in the lower dose group, DNA adducts were statistically
7 significantly elevated without increases in inflammation and necrosis. It is likely that at high doses
8 of benzo[a]pyrene, inflammation promotes the formation of tumors.
9 In addition to inflammation, immunosuppressive effects of benzo[a]pyrene have been noted
10 [as reviewed in, Zaccaria and McClure, 2013]. Immune effects of benzo[a]pyrene exposure (see
11 Section 1.1.3] may provide an environment where tumor cells can evade detection by immune
12 surveillance mechanisms normally responsible for recognizing and eliminating nascent cancer cells
13 (Hanahan and Weinberg. 2011]. In addition, the developing fetus may be even more sensitive to
14 these effects; Urso and Gengozian (1980] found that mice exposed to benzo[a]pyrene in utero not
15 only had a significantly increased tumor incidence as adults but also a persistently suppressed
16 immune system.
17 Gap junctions are channels between cells that are crucial for differentiation, proliferation,
18 apoptosis, and cell death. Interruption of GJIC is associated with a loss of cellular control of growth
19 and differentiation, and consequently with the two epigenetic steps of tumor formation, promotion
20 and progression. Thus, the inhibition of gap junctional intercellular communication by
21 benzo[a]pyrene, observed in vitro (Sharovskaya et al.. 2006: Blahaetal.. 2002]. provides another
22 mechanism of tumor promotion.
23 In summary, there are tumor-promoting effects of PAH exposures that are not mutagenic.
24 Although these effects are observed following benzo[a]pyrene-specific exposures, the occurrence of
25 BPDE-DNA adducts and associated mutations that precede both cytotoxicity and tumor formation
26 and increase with dose provides evidence that mutagenicity is the primary event that initiates
27 tumorigenesis following benzo[a]pyrene exposures. A biologically plausible mode of action may
28 involve a combination of effects induced by benzo[a]pyrene, with mutagenicity as the initiating
29 tumorigenic event Subsequent AhR activation and cytotoxicity could then lead to increased ROS
30 formation, regenerative cell proliferation, and inflammatory responses, which, along with evasion
31 of immune surveillance and GJIC, would provide an environment where the selection for mutated
32 cells increases the rate of mutation, allowing clonal expansion and progression of these tumor cells
33 to occur. However, it was determined that, in comparison to the large database on the mutagenicity
34 of benzo[a]pyrene, there were insufficient data to develop a separate mode of action analysis for
35 these promotional effects.
36 Conclusions about the hypothesized mode of action
37 There is sufficient evidence to conclude that the major mode of action for benzo[a]pyrene
38 carcinogenicity involves mutagenicity mediated by DNA reactive metabolites. The evidence for a
This document is a draft for review purposes only and does not constitute Agency policy.
1-78 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 mutagenic mode of action for benzo[a]pyrene is consistent with the current understanding that
2 mutations in p53 and ras oncogenes are associated with increased risk of tumor initiation. The
3 benzo[a]pyrene database provides strong and consistent evidence for BPDE-induced mutations
4 associated with tumor initiation in cancer tissue from humans exposed to complex mixtures
5 containing benzo[a]pyrene, in animals exposed to benzo[a]pyrene, and in in vitro systems.
6 Supporting evidence suggests that contributions to tumor initiation through potential mutagenic
7 events can be made by the radical cation and o-quinone/ROS metabolic activation pathways. Other
8 processes may contribute to the carcinogenicity of benzo[a]pyrene via the promotion and
9 progression phases of cancer development (e.g., inflammation, cytotoxicity, sustained regenerative
10 cell proliferation).
11 Support for the Hypothesized Mode of Action in Test Animals
12 Benzo[a]pyrene induces gene mutations in a variety of in vivo and in vitro systems and
13 produces tumors in all animal species tested and by all routes of exposure (see Appendix D in
14 Supplemental Information). Strong, consistent evidence in animal models supports the postulated
15 key events: the metabolism of benzo[a]pyrene to DNA-reactive intermediates, the formation of
16 DNA adducts, the subsequent occurrence of mutations in oncogenes and tumor suppressor genes,
17 and the clonal expansion of mutated cells.
18 Relevance of the Hypothesized Mode of Action to Humans
19 A substantial database indicates that the postulated key events for a mutagenic mode of
20 action all occur in human tissues. Evidence is available from studies of humans exposed to PAH
21 mixtures (including coal smoke and tobacco smoke) indicating a contributing role for
22 benzo[a]pyrene diol epoxide in inducing key mutational events in genes that are associated with
23 tumor initiation (mutations in the p53 tumor suppressor gene and H-ras or K-ras oncogenes). The
24 evidence includes observations of a spectrum of mutations in ras oncogenes and the p53 gene in
25 lung tumors of human patients exposed to coal smoke or tobacco smoke) that are similar to the
26 spectrum of mutations caused by benzo[a]pyrene diol epoxide in several biological systems,
27 including tumors from mice exposed to benzo[a]pyrene. Additional supporting evidence includes
28 correspondence between hotspots of p53 mutations in human lung cancers and sites of DNA
29 adduction by benzo[a]pyrene diol epoxide in experimental systems, and elevated BPDE-DNA
30 adduct levels in respiratory tissue of lung cancer patients or tobacco smokers with lung cancer.
31 Populations or Lifestages Particularly Susceptible to the Hypothesized Mode of Action
32 A mutagenic mode of action for benzo[a]pyrene-induced carcinogenicity is considered
33 relevant to all populations and lifestages. The current understanding of biology of cancer indicates
34 that mutagenic chemicals, such as benzo[a]pyrene, are expected to exhibit a greater effect in early
35 life exposure versus later life exposure (U.S. EPA. 2005b: Vesselinovitchetal.. 1979). The EPA's
36 Supplemental Guidance for Assessing Susceptibility from Early-Life Exposure to Carcinogens (U.S. EPA.
This document is a draft for review purposes only and does not constitute Agency policy.
1-79 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 2005b] recommends the application of age-dependent adjustment factors (ADAFs) for carcinogens
2 that act through a mutagenic mode of action. Given that a determination benzo [a]pyrene acts
3 through a mutagenic mode of carcinogenic action has been made, ADAFs should be applied along
4 with exposure information to estimate cancer risks for early-life exposure.
5 Toxicokinetic information suggest early lifestages may have lower levels of some GYP
6 enzymes than adults [Ginsberg etal.. 2004: Cresteil. 1998). suggesting that lower levels of
7 mutagenic metabolites may be formed in early lifestages. Though expression of bioactivating
8 enzymes is believed to be lower in the developing fetus and children, metabolism of
9 benzo[a]pyrene still occurs, as indicated by the detection of benzo[a]pyrene-DNA or protein
10 adducts or urinary metabolites [Naufaletal., 2010: Ruchirawatetal., 2010: Suter etal., 2010:
11 Mielzynskaetal.. 2006: Pereraetal.. 2005a: Tang etal.. 1999: Whyattetal.. 1998). While
12 expression of GYP enzymes is lower in fetuses and infants, the greater liver to body mass ratio and
13 increased blood flow to liver in fetuses and infants may compensate for the decreased expression of
14 GYP enzymes [Ginsberg et al.. 2004]. Activity of Phase II detoxifying enzymes in neonates and
15 children is adequate for sulfation but decreased for glucuronidation and glutathione conjugation
16 [Ginsberg etal., 2004]. The conjugation of benzo[a]pyrene-4,5-oxide with glutathione was
17 approximately one-third less in human fetal than adult liver cytosol [Pacifici etal.,1988].
18 In addition, newborn or infant mice develop liver and lung tumors more readily than young
19 adult mice following acute i.p. exposures to benzo[a]pyrene [Vesselinovitch et al.. 1975]. These
20 results indicate that exposure to benzo[a]pyrene during early life stages presents additional risk for
21 cancer, compared with exposure during adulthood, despite lower metabolic activity in early
22 lifestages. Population variability in metabolism and detoxification of benzo [a]pyrene, in addition to
23 DNA repair capability, may affect cancer risk. Polymorphic variations in the human population in
24 CYP1A1, CYP1B1, and other GYP enzymes have been implicated as determinants of increased
25 individual cancer risk in some studies [Ickstadtetal.. 2008: Aklillu etal.. 2005: Alexandrov et al..
26 2002: Perera and Weinstein. 2000]. Some evidence suggests that humans lacking a functional
27 GSTM1 gene have higher BPDE-DNA adduct levels and are thus at greater risk for cancer [Binkova
28 etal.. 2007: Vineis etal.. 2007: Pavanello etal.. 2005: Pavanello etal.. 2004: Alexandrov et al.. 2 0 0 2:
29 Perera and Weinstein. 2000]. In addition, acquired deficiencies or inherited gene polymorphisms
30 that affect the efficiency or fidelity of DNA repair may also influence individual susceptibility to
31 cancer from environmental mutagens [Wangetal.. 2010: Ickstadtetal.. 2008: Binkova etal.. 2007:
32 Matulloetal..20Q3: Shen etal.. 2003: Cheng etal.. 2000: Perera and Weinstein. 2000: Wei etal..
33 2000: Amos etal.. 1999]. In general, however, available support for the role of single
34 polymorphisms in significantly modulating human PAH cancer risk from benzo[a]pyrene or other
35 PAHs is relatively weak or inconsistent Combinations of polymorphisms, on the other hand, may
36 be critical determinants of a cumulative DNA-damaging dose, and thus indicate greater
37 susceptibility to cancer from benzo [ajpyrene exposure [Vineis etal., 2007].
This document is a draft for review purposes only and does not constitute Agency policy.
1-80 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 An alysis of Toxicogen omicsData
2 An analysis of pathway-based transcriptomic data was conducted to help inform the cancer
3 mode of action for benzo[a]pyrene (see the Supplemental Information for details of this analysis).
4 These data support a mutagenic and cellular proliferation mode of action that follows three
5 candidate pathways: aryl hydrocarbon signaling; DNA damage regulation of the Gl/S phase
6 transition; and/or Nrf2 regulation of oxidative stress. Specifically, the analysis showed that
7 benzo[a]pyrene may activate the AhR, leading to the formation of oxidative metabolites and
8 radicals which may lead to oxidative damage and DNA damage. Subsequently, DNA damage can
9 occur and activate p53 andp53 target genes, including p21 and MDM2. In addition, the data
10 indicate that p53 signaling may be decreased under these conditions, as ubiquitin and MDM2 are
11 both upregulated, and work together to degrade p53. Furthermore, the transcriptional
12 upregulation of cyclin D may result in enough cyclin D protein to overcome the p21 inhibitory
13 competition for CDK4, allowing for Gl/S phase transition to occur. The data also support the
14 hypothesis that an upregulation of proliferating cell nuclear antigen (PCNA) in combination with
15 the upregulation of ubiquitin indicates that cells are moving towards the Gl/S phase transition.
16 Although the alterations to the Nrf2 pathway suggest cells are preparing for a pro-apoptotic
17 environment, there is no transcriptional evidence that the apoptotic pathways are being activated.
18 There are uncertainties associated with the available transcriptomics data. For instance,
19 the available studies only evaluate gene expression following benzo[a]pyrene exposure and do not
20 monitor changes in protein or metabolite expression, which would be more indicative of an actual
21 cellular state change. Further research is required at the molecular level to demonstrate that the
22 cellular signaling events being inferred from such data are actually operative and result in
23 phenotypic changes. In addition, this analysis relied upon two short term studies that evaluated
24 mRNA expression levels in a single tissue (liver) and species (mouse) and were conducted at
25 relatively high doses.
26 1.2. SUMMARY AND EVALUATION
27 1.2.1. Weight of Evidence for Effects Other than Cancer
28 The weight of the evidence from human and animal studies indicates that the strongest
29 evidence for human hazards following benzo[a]pyrene exposure is for developmental toxicity,
30 reproductive toxicity, and immunotoxicity. Most of the available human data on benzo[a]pyrene
31 report associations between particular health endpoints and concentrations of benzo[a]pyrene-
32 DNA adducts, with fewer noncancer studies correlating health effects with external measures of
33 exposure. In general, the available human studies report effects that are analogous to the effects
34 observed in animal toxicological studies, and provide qualitative, supportive evidence for the effect-
35 specific hazards identified in Sections 1.1.1-1.1.4.
36 In animals, evidence of developmental toxicity, reproductive toxicity, and immunotoxicity
37 has been observed across species and dosing regimens. The available evidence from mice and rats
This document is a draft for review purposes only and does not constitute Agency policy.
1-81 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 treated by gavage during gestation or in the early postnatal period demonstrate developmental
2 effects including decreased body weight, decreased fetal survival, decreased fertility, atrophy of
3 reproductive organs, and altered neurobehavioral outcomes [Chenetal.. 2012: Tules etal.. 2012:
4 Bouayed et al.. 2009a: Kristensenetal.. 1995: Mackenzie and Angevine. 1981). Several studies in
5 animals have indicated that exposure to benzo[a]pyrene in early life may result in altered
6 neurobehavioral outcomes and sensorimotor development [Chenetal.. 2012: Bouayed etal..
7 2009a]. Male and female reproductive toxicity, as evidenced by effects on sperm parameters,
8 decreased reproductive organ weights, histological changes, and hormone alterations, have been
9 observed after oral exposure in rats and mice [Chenetal., 2011: Chung etal., 2011: Mohamed etal.,
10 2010: Zheng etal., 2010: Mackenzie and Angevine, 1981]. Benzo[a]pyrene exposure has also been
11 shown to lead to altered immune cell populations and histopathological changes in immune system
12 organs [Kroese etal.. 2001: De long et al.. 1999). as well as thymic and splenic effects following
13 subchronic oral exposure. Varying immunosuppressive responses are also observed in short-term
14 oral and injection studies. Overall, the oral data support the conclusion that developmental toxicity
15 and reproductive toxicity are human hazards following exposure to benzo[a]pyrene and that
16 immunotoxicity is a potential human hazard of benzo[a]pyrene exposure.
17 Following inhalation exposure to benzo[a]pyrene in animals, evidence of developmental
18 and reproductive toxicity has been observed. Decreased fetal survival has been observed in rats
19 exposed to benzo[a]pyrene via inhalation during gestation [Wormley etal.. 2004: Archibongetal..
20 2002). Male reproductive toxicity, as evidenced by effects on sperm parameters, decreased testes
21 weight, and hormone alterations, has also been observed in rats following subchronic inhalation
22 exposure to benzo[a]pyrene [Archibongetal.. 2008: Ramesh etal.. 2008]. Female reproductive
23 toxicity, as evidenced by modified hormone levels in dams, has been observed following inhalation
24 exposure to benzo[a]pyrene during gestation [Archibongetal., 2002]. The inhalation data support
25 the conclusion that developmental toxicity and reproductive toxicity are human hazards following
26 exposure to benzo[a]pyrene.
27 Other types of effects were observed following benzo[a]pyrene exposure including
28 forestomach hyperplasia, hematological, hepatic, renal, cardiovascular, and adult neurological
29 toxicity (see Section 1.1.4]. Forestomach hyperplasia was observed following oral and inhalation
30 exposure; however, this endpoint most likely reflects early events in the neoplastic progression of
31 forestomach tumors following benzo[a]pyrene exposure (see Section 1.1.4], and was not
32 considered further for dose-response analysis and the derivation of reference values. For the
33 remaining effects, EPA concluded that the available evidence does not support these noncancer
34 effects as potential human hazards.
35 1.2.2. Weight of Evidence for Carcinogenicity
36 Under EPA's Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005b], benzo[a]pyrene is
37 "carcinogenic to humans." This guidance emphasizes the importance of weighing all of the evidence
38 in reaching conclusions about human carcinogenic potential. The descriptor of "carcinogenic to
This document is a draft for review purposes only and does not constitute Agency policy.
1-82 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 humans" can be used when the following conditions are met: (a) there is strong evidence of an
2 association between human exposure and either cancer or the key precursor events of the agent's
3 mode of action but not enough for a causal association; (b) there is extensive evidence of
4 carcinogenicity in animals; (c) the mode or modes of carcinogenic action and associated key
5 precursor events have been identified in animals; and (d) there is strong evidence that the key
6 precursor events that precede the cancer response in animals are anticipated to occur in humans
7 and progress to tumors, based on available biological information. The data supporting these four
8 conditions for benzo[a]pyrene are presented below and in Table 1-18.
9 a) Strong Human Evidence of Cancer or its Precursors
10 There is a large body of evidence for human carcinogenicity for complex PAH mixtures
11 containing benzo[a]pyrene, including soot, coal tars, coal-tar pitch, mineral oils, shale oils, and
12 smoke from domestic coal burning [IARC. 2010: Baanetal.. 2009). There is also evidence of
13 carcinogenicity, primarily of the lung and skin, in occupations involving exposure to PAH mixtures
14 containing benzo[a]pyrene, such as chimney sweeping, coal gasification, coal-tar distillation, coke
15 production, iron and steel founding, aluminum production, and paving and roofing with coal tar
16 pitch [IARC, 2010: Baan etal., 2009: Straif etal., 2005]. Increased cancer risks have been reported
17 among other occupations involving exposure to PAH mixtures such as carbon black and diesel
18 exhaust [Benbrahim-Tallaa et al.. 2012: Bosetti et al.. 2007). There is extensive evidence of the
19 carcinogenicity of tobacco smoke, of which benzo[a]pyrene is a notable constituent. The
20 methodologically strongest epidemiology studies (in terms of exposure assessment, sample size,
21 and follow-up period) provide consistent evidence of a strong association between benzo[a]pyrene
22 exposure and lung cancer. Three large epidemiology studies in different geographic areas,
23 representing two different industries, observed increasing risks of lung cancer with increasing
24 cumulative exposure to benzo[a]pyrene (measured in |ig/m3-years), with approximately a twofold
25 increased risk at the higher exposures; each of these studies addressed potential confounding by
26 smoking f Armstrong and Gibbs. 2009: Spinelli etal.. 2006: Xu etal.. 19961 (Table 1-11). Although
27 the relative contributions of benzo[a]pyrene and of other PAHs cannot be established, the
28 exposure-response patterns seen with the benzo[a]pyrene measures make it unlikely that these
29 results represent confounding by other exposures. Similarly, for bladder cancer, two of the three
30 cohort studies with detailed exposure data observed an increasing risk with exposures
31 >80 [ig/m3-years(Gibbs and Sevigny. 2007a: Gibbs etal.. 2007: Gibbs and Sevigny. 2007b: Spinelli et
32 al.. 2006] (Table 1-13). The exposure range was much lower in the third study (Burstyn etal..
33 2007: Gibbs and Sevigny. 2007a: Gibbs etal.. 2007: Gibbs and Sevigny. 2007b). such that the highest
34 exposure group only reached the level of exposure seen in the lowest exposure categories in the
35 other studies. Data pertaining to non-melanoma skin cancer is limited to studies with more indirect
36 exposure measures, e.g., based on occupations with likely dermal exposure to creosote (i.e., timber
37 workers, brick makers, and power linesmen); the RR estimates seen in the four available studies
38 that provide risk estimates for this type of cancer ranged from 1.5 to 4.6, with three of these four
This document is a draft for review purposes only and does not constitute Agency policy.
1-83 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 estimates >2.5 and statistically significant [Pukkala, 1995: Karlehagenetal., 1992: Tornqvistetal.,
2 1986: Hammond etal.. 1976). These four studies provide support for the association between
3 dermal PAH exposure, including benzo[a]pyrene exposure, and skin cancer. Although it is likely
4 that multiple carcinogens present in PAH mixtures contribute to the carcinogenic responses, strong
5 evidence is available from several studies of humans exposed to PAH mixtures supporting a
6 contributing role for benzo[a]pyrene diol epoxide in inducing key mutagenic precursor cancer
7 events in target tissues. Elevated BPDE-DNA adducts have been reported in smokers compared to
8 nonsmokers, and the increased adduct levels in smokers are typically increased twofold compared
9 with nonsmokers [Phillips, 2002]. Elevated BPDE-DNA adduct levels have been observed in WBCs
10 of groups of coke oven workers and chimney sweeps, occupations with known elevated risks of
11 cancer [Rojas etal.. 2000: Bartsch etal.. 1999: Pavanello etal.. 1999: Bartschetal.. 1998: Rojas et
12 al.. 1998). and in lung tissue from tobacco smokers with lung cancer [Rojas etal.. 2004: Godschalk
13 etal.. 2002: Bartschetal.. 1999: Rojas etal.. 1998: Andreassen et al.. 1996: Alexandrov etal.. 1992).
14 Mutation spectra distinctive to diol epoxides have been observed in the tumor suppressor
15 gene p53 and the K-ras oncogene in tumor tissues taken from lung cancer patients who were
16 chronically exposed to two significant sources of PAH mixtures: coal smoke and tobacco smoke.
17 Hackman et al. [2000] reported an increase of GC^TA transversions and a decrease of GC^AT
18 transitions at the hprt locus in T-lymphocytes of humans with lung cancer who were smokers
19 compared to non-smokers. Lung tumors from cancer patients exposed to emissions from burning
20 smoky coal showed mutations in p53 and K-ras that were primarily G^T transversions (76 and
21 86%, respectively) [DeMarini etal.. 2001). Keohavong etal. [2003] investigated the K-ras
22 mutational spectra from nonsmoking women and smoking men chronically exposed to emissions
23 from burning smoky coal, and smoking men who resided in homes using natural gas; among those
24 with K-ras mutations, 67, 86, and 67%, respectively, were G—>T transversions. Lung tumors from
25 tobacco smokers showed a higher frequency of p53 mutations that were G—>T transversions
26 compared with lung tumors in non-smokers [Pfeifer and Hainaut. 2003: Pfeifer etal.. 2002: Hainaut
27 and Pfeifer. 2001]. and the frequency of these types of p53 mutations in lung tumors from smokers
28 increased with increasing smoking intensity [Bennett etal.. 1999].
29 Similarly, investigations of mutagenesis following specific exposures to benzo[a]pyrene (as
30 opposed to PAH mixtures] have consistently observed that the benzo[a]pyrene diol epoxide is very
31 reactive with guanine bases in DNA, and that G^T transversions are the predominant type of
32 mutations caused by benzo[a]pyrene diol epoxide in several biological test [Pfeifer and Hainaut.
33 2003: Hainaut and Pfeifer. 2001]. Following treatment of human HeLa cells with benzo [a] pyrene
34 diol epoxide, Denissenko etal. [1996] reported that the distribution of BPDE-DNA adducts within
35 p53 corresponded to mutational hotspots observed in p53 in human lung cancers. Benzo[a]pyrene
36 exposure induced mutations in embryonic fibroblasts from human p53 "knock-in" mice that were
37 similar to those found in smoking related human cancers, with a predominance of G—>T
38 transversions that displayed strand bias and were also located in the same mutational hotspots
This document is a draft for review purposes only and does not constitute Agency policy.
1-84 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 found in p53 in human lung tumors [Liuetal., 2005]. These results, combined with a mechanistic
2 understanding that mutations in p53 (which encodes a key transcription factor in DNA repair and
3 regulation of cell cycle and apoptosis) may be involved in the initiation phase of many types of
4 cancer, are consistent with a common mechanism for mutagenesis following exposures to PAH
5 mixtures and provide evidence of a contributing role of benzo[a]pyrene diol epoxide in the
6 carcinogenic response of humans to coal smoke and tobacco smoke.
7 Therefore, while the epidemiological evidence alone does not establish a causal association
8 between human exposure and cancer, there is strong evidence that the key precursor events of
9 benzo[a]pyrene's mode of action are likely to be associated with tumor formation in humans.
10 b) Extensive Animal Evidence
11 In laboratory animals (rats, mice, and hamsters), exposures to benzo[a]pyrene via the oral,
12 inhalation, and dermal routes have been associated with carcinogenic responses both systemically
13 and at the site of administration. Three 2-year oral bioassays are available that associate lifetime
14 benzo[a]pyrene exposure with carcinogenicity at multiple sites. These bioassays observed
15 forestomach, liver, oral cavity, jejunum, kidney, auditory canal (Zymbal gland), and skin or
16 mammary gland tumors in male and female Wistar rats (Kroese et al., 2001): forestomach tumors
17 in male and female Sprague-Dawley rats (Brune etal.. 1981): and forestomach, esophagus, tongue,
18 and larynx tumors in female B6C3Fi mice (Beland and Gulp. 1998: Gulp et al.. 1998). Repeated or
19 short-term oral exposure to benzo[a]pyrene was associated with forestomach tumors in additional
20 bioassays with several strains of mice (Weyandetal.. 1995: Benjamin etal.. 1988: Robinson etal..
21 1987: El-Bayoumy. 1985: Triolo etal.. 1977: Wattenberg. 1974: Roe etal.. 1970: Biancifiori etal..
22 1967: Chouroulinkovetal.. 1967: Fedorenko and Yansheva. 1967: Neal andRigdon. 1967:
23 Berenblum and Haran, 1955). EPA has considered the uncertainty associated with the relevance of
24 forestomach tumors for estimating human risk from benzo[a]pyrene exposure. While humans do
25 not have a forestomach, squamous epithelial tissue similar to that seen in the rodent forestomach
26 exists in the oral cavity and upper two-thirds of the esophagus in humans (IARC. 2003). Human
27 studies, specifically associating exposure to benzo[a]pyrene with the alimentary tract tumors are
28 not currently available. However, benzo[a]pyrene-DNA adducts have been detected in oral and
29 esophageal tissue obtained from smokers (reviewed by Phillips, 2002] and several epidemiological
30 studies have identified increased exposure to PAHs as an independent risk factor for esophageal
31 cancer (Abedi-Ardekani et al.. 2010: Szymanska et al.. 2 010: Gustavsson etal.. 1998: Liu etal..
32 1997). Thus, EPA concluded that forestomach tumors in rodents are relevant for assessing the
33 carcinogenic risk to humans.
34 Lifetime inhalation exposure to benzo[a]pyrene was associated primarily with tumors in
35 the larynx and pharynx of male Syrian golden hamsters exposed to benzo[a]pyrene:NaCl aerosols
36 (Thyssenetal., 1981). Additionally, less-than-lifetime oral exposure cancer bioassays in mice
37 provide supporting evidence that exposure to benzo[a]pyrene is associated with an increased
38 incidence of lung tumors in mice (Weyandetal.. 1995: Robinson etal.. 1987: Wattenberg. 1974). In
This document is a draft for review purposes only and does not constitute Agency policy.
1-85 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 additional studies with hamsters, intratracheal instillation of benzo[a]pyrene was associated with
2 upper and lower respiratory tract tumors [Feron and Kruysse. 1978: Ketkar etal.. 1978: Feron et
3 al.. 1973: Henry etal.. 1973: Saffiotti etal.. 19721 Dermal application of benzo[a]pyrene
4 (2-3 times/week) has been associated with mouse skin tumors in numerous lifetime bioassays
5 fSivak etal.. 1997: Grimmer etal.. 1984: Habs etal.. 1984: Grimmer etal.. 1983: Habs etal.. 1980:
6 Schmahl etal.. 1977: Schmidt etal.. 1973: Roe etal.. 1970: Poel. 1963.19591 Skin tumors in rats,
7 rabbits, and guinea pigs have also been associated with repeated application of benzo[a]pyrene to
8 skin in the absence of exogenous promoters [IPCS. 1998: ATSDR. 1995: IARC. 1983.1973). When
9 followed by repeated exposure to a potent tumor promoter, acute dermal exposure to
10 benzo[a]pyrene induced skin tumors in numerous studies of mice, indicating that benzo[a]pyrene is
11 a strong tumor-initiating agent in the mouse skin model [Weyandetal.. 1992: Cavalierietal.. 1991:
12 Rice etal.. 1985: El-Bayoumy etal.. 1982: Lavoie etal.. 1982: Ravehetal.. 1982: Cavalierietal..
13 1981: Slaga etal.. 1980: Wood etal.. 1980: Slaga etal.. 1978: Hoffmann etal.. 19721
14 Carcinogenic responses in animals exposed to benzo [a] pyrene by other routes of
15 administration include: (1) liver or lung tumors in newborn mice given acute postnatal i.p.
16 injections (Lavoie etal.. 1994: Busby etal.. 1989: Weyand and Lavoie. 1988: Lavoie etal.. 1987:
17 Wislocki etal.. 1986: Busby etal.. 1984: Buening etal.. 1978: Kapitulnik etal.. 19781: (2) increased
18 lung tumor multiplicity in A/J adult mice given single i.p. injections [Mass etal.. 1993): (3) injection
19 site tumors in mice following s.c. injection [Nikonova. 1977: Pfeiffer. 1977: Homburger etal.. 1972:
20 Roe and Waters. 1967: Grant and Roe. 1963: Steiner. 1955: Rask-Nielsen. 1950: Pfeiffer and Allen.
21 1948: Bryan and Shimkin. 1943: Barry etal.. 1935]: (4) injection site sarcomas in mice following
22 intramuscular injection[Sugiyama. 1973]: (5) mammary tumors in rats with intramammilary
23 administration [Cavalierietal., 1991: Cavalierietal., 1988c: Cavalieri et al., 1988b: Cavalierietal.,
24 1988a]: (6) cervical tumors in mice with intravaginal application (Naslundetal., 1987]: and
25 (7) tracheal tumors in rats with intratracheal implantation [Topping etal.. 1981: Nettesheim etal..
26 19771
27 Therefore, the animal database provides extensive evidence of carcinogenicity in animals.
28 c) Key Precursor Events have been Identified in Animals
29 There is sufficient evidence to conclude that benzo[a]pyrene carcinogenicity involves a
30 mutagenic mode of action mediated by DNA-reactive metabolites. The benzo[a]pyrene database
31 provides strong and consistent evidence for BPDE-induced mutations associated with tumor
32 initiation in cancer tissue from humans exposed to complex mixtures containing benzo [a] pyrene, in
33 animals exposed to benzo[a]pyrene, and in in vitro systems. Other processes may contribute to the
34 carcinogenicity of benzo [ajpyrene via the promotion and progression phases of cancer
35 development (e.g., inflammation, cytotoxicity, sustained regenerative cell proliferation, anti-
36 apoptotic signaling), but the available evidence best supports a mutagenic mode of action as the
37 primary mode by which benzo[a]pyrene acts.
This document is a draft for review purposes only and does not constitute Agency policy.
1-86 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 d) Strong Evidence that the Key Precursor Events are Anticipated to Occur in Humans
2 Mutations in p53 and ras oncogenes have been observed in tumors from mice exposed to
3 benzo[a]pyrene in the diet [Gulp etal.. 2000] or by i.p. injection [Nesnowetal.. 1998a: Nesnow et
4 al.. 1998b: Nesnowetal.. 1996.1995: Mass etal.. 1993). Mutations in these same genes have also
5 been reported in lung tumors of human cancer patients, bearing distinctive mutation spectra (G—>T
6 transversions) that correlate with exposures to coal smoke [Keohavongetal.. 2003: DeMarini etal..
7 2001] or tobacco smoke [Pfeifer and Hainaut. 2003: Pfeifer etal.. 2002: Hainaut and Pfeifer. 2001:
8 Bennett etal.. 19991.
9 Table 1-18. Supporting evidence for the carcinogenic to humans cancer
10 descriptor for benzo[a]pyrene
Evidence
Reference
a) Strong human evidence of cancer or its precursors
• Increased risk of lung, bladder, and skin cancer
in humans exposed to complex PAH mixtures
containing benzo[a]pyrene
• Benzo[a]pyrene-specific biomarkers detected in
humans exposed to PAH mixtures associated
with increased risk of cancer
BPDE-DNA adducts in WBCs of coke oven
workers and chimney sweeps
BPDE-DNA adducts in smokers
• Benzo[a]pyrene-specific DNA adducts have
been detected in target tissues in humans
exposed to PAH mixtures
BPDE-DNA adducts in non-tumor lung
tissues of cigarette smokers with lung
cancer and in skin of eczema patients
treated with coal tar
BPDE-DNA adduct formation in p53 in
human cells in vitro corresponds to
mutational hotspots at guanine residues in
human lung tumors
• Benzo[a]pyrene-specific mutational spectra
identified in PAH-associated tumors in humans
(IARC (2010), 2004)); Secretan et al. (2009);Baan et al.
(2009); Benbrahim-Tallaa etal. (2012)
(Rojasetal. (2000); Bartsch et al. (1999); Pavanello et al.
(1999); Bartsch et al. (1998); Rojas et al. (1998))
Phillips (2002)
Rojasetal. (2004); (Godschalk et al. (2002); Bartsch et al.
(1999); Godschalk et al. (1998b); Rojas et al. (1998);
Andreassen et al. (1996); Alexandrov et al. (1992))
(Denissenko et al. (1996); Puisieux et al. (1991))
This document is a draft for review purposes only and does not constitute Agency policy,
1-87 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Evidence
- GC->TA transversions and GC->AT
transitions at hprt locus in T-lymphocytes
of humans with lung cancer
— G-^T transversions in exposed human-p53
knock-in mouse fibroblasts at the same
mutational hotspot in p53 from smoking-
related lung tumors in humans
- G->T transversions at the same mutational
hotspot in p53 and K-ras in human lung
tumors associated with smoky coal
exposures
- Increased percentage of G->T
transversions in p53 in smokers versus
nonsmokers
b) Extensive animal evidence
Oral exposures
• Forestomach tumors in male and female rats
and in female mice following lifetime exposure
• Forestomach tumors in mice following less-
than-lifetime exposures
• Alimentary tract and liver tumors in male and
female rats following lifetime exposure
• Kidney tumors in male rats following lifetime
exposure
• Auditory canal tumors in male and female rats
following lifetime exposure
• Esophageal, tongue, and laryngeal tumors in
female mice following lifetime exposure
• Lung tumors in mice following less-than-
lifetime exposure
Reference
Hackman et al. (2000)
Liu etal. (2005)
(Keohavong et al. (2003); DeMarini et al. (2001))
(Pfeifer and Hainaut (2003); Pfeifer et al. (2002); Hainaut
and Pfeifer (2001); Bennett etal. (1999))
(Kroese et al. (2001); Beland and Gulp (1998); Gulp et al.
(1998); Brune etal. (1981))
(Weyand et al. (1995); Benjamin et al. (1988); Robinson et
al. (1987); EI-Bayoumy (1985); Triolo et al. (1977);
Wattenberg (1974); Roe et al. (1970); Biancifiori et al.
(1967); Chouroulinkov et al. (1967); Fedorenko and
Yansheva (1967); Neal and Rigdon (1967); Berenblum and
Haran (1955))
Kroese et al. (2001)
Kroese et al. (2001)
Kroese et al. (2001)
(Beland and Gulp (1998); Gulp et al. (1998))
(Weyand et al. (1995); Robinson et al. (1987); Wattenberg
(1974))
This document is a draft for review purposes only and does not constitute Agency policy,
1-88 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Evidence
Reference
Inhalation exposures
• Upper respiratory tract tumors in male
hamsters following chronic exposure
Thyssenetal. (1981)
Dermal exposures
• Skin tumors in mice following lifetime
exposures without a promoter
• Skin tumors in rats, rabbits, and guinea pigs
following subchronic exposures
(Sivak et al. (1997); Grimmer et al. (1984); Habs et al.
(1984); Grimmer et al. (1983); Habs et al. (1980); Schmahl
et al. (1977); Schmidt et al. (1973); Roe et al. (1970); Poel
(1963), 1959))
(IPCS, 1998; ATSDR, 1995; IARC, 1983, 1973)
Other routes of exposure
• Respiratory tract tumors in hamsters following
intratracheal instillation
• Liver or lung tumors in newborn mice given
acute postnatal i.p. injections
• Lung tumor multiplicity in A/J adult mice given
single i.p. injections
(Feron and Kruysse, 1978; Ketkar et al., 1978; Feron et al.,
1973; Henry et al., 1973; Saffiotti et al., 1972)
(Lavoie et al., 1994; Busby et al., 1989; Weyand and
Lavoie, 1988; Lavoie et al., 1987; Wislocki et al., 1986;
Busby et al., 1984; Buening et al., 1978; Kapitulnik et al..
1978)
Mass et al. (1993)
c) Identification of key precursor events have been identified in animals
• Bioactivation of benzo[a]pyrene to DNA-
reactive metabolites has been shown to occur
in multiple species and tissues by all routes of
exposure
• Direct DNA damage by the reactive
metabolites, including the formation of DNA
adducts and ROS-mediated damage
• Formation and fixation of DNA mutations,
particularly in tumor suppressor genes or
oncogenes associated with tumor initiation
See 'Experimental Support for Hypothesized Mode of
Action' section
d) Strong evidence that the key precursor events are anticipated to occur in humans
• Mutations in p53 or ras oncogenes have been
observed in forestomach or lung tumors from
mice exposed to benzo[a]pyrene
(Gulp et al. (2000); Nesnow et al. (1998a); Nesnow et al.
(1998b); Nesnow et al. (1996), 1995); Mass et al. (1993))
This document is a draft for review purposes only and does not constitute Agency policy,
1-89 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Evidence
- G->T transversions in ras oncogenes or the
p53 gene have been observed in lung
tumors of human cancer patients exposed
to coal smoke
- Higher frequency of G->T transversions in
lung tumors from smokers versus
nonsmokers
Reference
(Keohavong et al. (2003); DeMarini et al. (2001))
(Pfeifer and Hainaut (2003); Pfeifer et al. (2002); Hainaut
and Pfeifer (2001); Bennett et al. (1999))
This document is a draft for review purposes only and does not constitute Agency policy,
1-90 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2 2. DOSE-RESPONSE ANALYSIS
3 2.1. ORAL REFERENCE DOSE FOR EFFECTS OTHER THAN CANCER
4 The oral reference dose (RfD) (expressed in units of mg/kg-day) is defined as an estimate
5 (with uncertainty spanning perhaps an order of magnitude) of a daily exposure to the human
6 population (including sensitive subgroups) that is likely to be without an appreciable risk of
7 deleterious effects during a lifetime. It can be derived from a no-observed-adverse-effect level
8 (NOAEL), lowest-observed-adverse-effect level (LOAEL), or the 95% lower bound on the
9 benchmark dose (BMDL), with uncertainty factors (UFs) generally applied to reflect limitations of
10 the data used.
11 2.1.1. Identification of Studies and Effects for Dose-Response Analysis
12 In Section 1.2.1, developmental, reproductive, and immunological toxicities were
13 highlighted as human hazards or potential human hazards of benzo[a]pyrene exposure by the oral
14 route. Studies within each effect category were evaluated using general study quality
15 characteristics (as discussed in Section 6 of the Preamble) to help inform the selection of studies
16 from which to derive toxicity values. Rationales for selecting the studies and effects to represent
17 each of these hazards are summarized below.
18 Human studies are preferred over animal studies when quantitative measures of exposure
19 are reported and the reported effects are determined to be associated with exposure. For
20 benzo[a]pyrene, human studies of environmental polycyclic aromatic hydrocarbon (PAH) mixtures
21 across multiple cohorts have observed effects following exposure to complex mixtures of PAHs.
22 The available data suggest that benzo[a]pyrene exposure may pose health hazards other than
23 cancer including reproductive and developmental effects such as infertility, miscarriage, and
24 reduced birth weight (Wuetal.. 2010: Neal etal.. 2008: Tang etal.. 2008: PereraetaL 2005b:
25 PereraetaL, 2005a), effects on the developing nervous system (PereraetaL, 2012a: PereraetaL,
26 2009), and cardiovascular effects (Friesenetal., 2010: Burstyn etal., 2005). However, the available
27 human studies that utilized benzo[a]pyrene-deoxyribonucleic acid (DNA) adducts as the exposure
28 metric do not provide external exposure levels of benzo[a]pyrene from which to derive a value, and
29 exposure is likely to have occurred by multiple routes. In addition, uncertainty exists due to
30 concurrent exposure to other PAHs and other components of the mixture (such as metals).
31 Animal studies were evaluated to determine which provided the most relevant routes and
32 durations of exposure; multiple exposure levels to provide information about the shape of the dose-
33 response curve; and power to detect effects at low exposure levels (U.S. EPA. 2002). The oral
34 database for benzo[a]pyrene includes a variety of studies and datasets that are suitable for use in
This document is a draft for review purposes only and does not constitute Agency policy.
2-1 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 deriving reference values. Specifically, chronic effects associated with benzo[a]pyrene exposure in
2 animals include observations of organ weight and histological changes and hematological
3 parameters observed in several oral cancer bioassays [Kroese etal.. 2001: Beland and Gulp. 1998).
4 Multiple subchronic studies are available that characterize a variety of effects other than cancer. In
5 addition, several developmental studies are available that help inform hazards of exposure during
6 sensitive developmental windows.
7 Developmental Toxicity
8 Numerous animal studies observed endpoints of developmental toxicity following oral
9 exposure during gestational or early postnatal development [Chen etal., 2012: Tules etal., 2012:
10 Shengetal.. 2010: BouayedetaL 2009a: Kristensen etal.. 1995: Mackenzie and Angevine. 19811
11 and were considered for dose-response analysis based on the above criteria. Kristensen et al.
12 [1995]. with only one dose group, was not considered further given its concordance with
13 Mackenzie and Angevine [1981]. which had multiple groups. From the remaining studies
14 demonstrating developmental toxicity, the studies conducted by Chen etal. [2012] and Jules et al.
15 [2012] were identified as the most informative studies for dose-response analysis. The
16 neurodevelopmental study by Chen etal. [2012] was a well-designed and well-conducted study
17 that evaluated multiple developmental endpoints and measures of neurotoxicity in neonatal,
18 adolescent, and adult rats after early postnatal exposure. The study randomly assigned a total of
19 10 male and 10 female pups per treatment group, with no more than one male and one female from
20 each litter for behavioral testing. In addition, the pups were cross-fostered with dams being rotated
21 among litters every 2-3 days to distribute any maternal caretaking differences randomly across
22 litters and treatment groups. Importantly, all tests were conducted by investigators blinded to
23 treatment, and test order was randomized each day.
24 In the neurobehavioral tests, Chen etal. [2012] observed increased locomotion in the open
25 field test, increased latency in negative geotaxis and surface righting tests, decreased anxiety-like
26 behaviors in the elevated plus maze test, and impaired performance in the Morris water maze test
27 at various time points following neonatal benzo[a]pyrene treatment Altered behaviors and
28 locomotion in open field tests could be attributed to anxiety responses due to open spaces and
29 bright light, as well as changes to motor system function. Chen etal. [2012] reported increased
30 quadrants crossed, which could indicate either increased motor activity or decreased anxiety (less
31 fear of the open spaces/bright lights]; thus, this response does not reflect a clear effect of exposure
32 on a discrete neurological function. In addition, the biological significance of the impairments
33 observed in the negative geotaxis and surface righting tests were considered somewhat uncertain
34 as these effects were not persistent as compared to the effects noted in the elevated plus maze and
35 Morris water maze (which persisted into adulthood]. As a result, EPA considered the elevated plus
36 maze and Morris water maze tests to be the most informative measures of neurobehavioral
37 function performed by Chen etal. [2012].
This document is a draft for review purposes only and does not constitute Agency policy.
2-2 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Significant, dose-related effects were reported in an established test of spatial learning and
2 memory (Morris water maze). Specifically, increased escape latency in hidden platform trials and
3 decreased time spent in the target quadrant during a probe trial were observed following
4 benzo[a]pyrene exposure in rats tested as adolescents or as adults. Due to the altered baseline
5 performance of treated animals on day 1 of the hidden platform trials these findings cannot be
6 specifically attributed to impaired learning. In fact, the slopes of the lines across trial days are
7 nearly identical for the treatment groups, suggesting the lack of a robust effect on learning. The
8 impaired Morris water maze performance of treated animals could be due to effects on several
9 other components of neurological function besides learning, including anxiety, vision, and
10 locomotion. Similarly, as escape latencies were not comparable across groups after learning
11 acquisition (i.e., the end of the hidden platform trials), differences in probe trial performance are
12 difficult to attribute to impaired memory retention alone. As a result of this lack of specificity,
13 although they identify significant effects of benzo[a]pyrene exposure, the Morris water maze data
14 were considered less informative than the results from the elevated plus maze test Chen et al.
15 [2012) reported an increase in the number of open arm entries in the elevated plus maze test, an
16 indicator of decreased anxiety-like behavior. These results indicate effects on a single, discrete
17 neurological function that are unlikely to be complicated by changes in other processes such as
18 motor activity (total activity, calculated by summing open and closed arm entries was unchanged
19 with treatment). This neurobehavioral endpoint is supported by similar observations in developing
20 [Bouayed et al.. 2009a) and adult [Grovaetal.. 2008) mice, and may be indirectly related to
21 observations of increased aggression in mice [Bouayed etal.. 2009b) and is considered adverse (see
22 discussion in Section 1.1.1).
23 Tules etal. (2012) was also identified for dose-response analysis. This study was of
24 sufficient duration, utilized multiple doses, did not observe maternal toxicity, and evaluated
25 multiple cardiovascular endpoints. The study authors reported increases in both systolic
26 (approximately 20-50%) and diastolic (approximately 33-83%) pressure and heart rate in adult
27 rats that were exposed gestationally to benzo[a]pyrene. A limitation of this study is that the
28 authors only reported effects at the two highest doses. However, given the magnitude of the
29 response and the appearance of these effects in adulthood following gestational exposure, these
30 endpoints were selected for dose-response analysis because of their sensitivity and biological
31 plausibility.
32 Bouayed et al. (2009a) and Mackenzie and Angevine (1981) were not selected for dose-
33 response analysis. Bouayed etal. (2009a) used the same tests as Chen etal. (2012). but the doses
34 evaluated were higher (2 and 20 mg/kg-day compared to 0.02, 0.2, and 2 mg/kg-day, respectively).
35 Similarly, Mackenzie and Angevine (1981) demonstrated developmental effects in a multi-dose
36 study with relevant routes and durations of exposure; however, the doses studied
37 (10-160 mg/kg-day) were much higher than those evaluated in other developmental toxicity
38 studies (Chenetal.. 2012: Tules etal.. 2012).
This document is a draft for review purposes only and does not constitute Agency policy.
2-3 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Reproductive Toxicity
2 Male reproductive toxicity was demonstrated in numerous subchronic studies [Chenetal..
3 2011: Chung etal.. 2011: Mohamedetal.. 2010: Zheng etal.. 2010). Chung etal. [2011] was not
4 included in the dose-response analysis because numerical data were not reported or were only
5 reported for the mid-dose of three doses. Chenetal. [2011] is a subchronic study that applied only
6 a single dose level. This study corroborated other available multi-dose studies and is considered
7 supportive, but was not considered for dose-response analysis due to the limited reporting of
8 numerical data. The studies conducted by Mohamedetal. [2010] and Zheng etal. [2010] were
9 identified as the most informative male reproductive toxicity studies for dose-response analysis.
10 Decreased sperm count and motility observed by Mohamedetal. [2010] and decreased
11 intratesticular testosterone levels observed by Zheng etal. [2010] were selected for dose-response
12 analysis as both represent sensitive endpoints of male reproductive toxicity and are indicators of
13 potentially decreased fertility. These effects are also consistent with human studies in PAH
14 exposed populations, as effects on male fertility and semen quality have been demonstrated in
15 epidemiological studies of smokers [reviewed by Spares and Melo, 2008].
16 Female reproductive toxicity was demonstrated in two subchronic studies [Gao etal., 2011:
17 Xu etal.. 2010]. Specifically, Xu etal.. 2010 demonstrated altered ovary weights and follicle
18 numbers, and Gao etal. [2011] demonstrated cervical epithelial cell hyperplasia following oral
19 exposure to benzo[a]pyrene. These studies were identified as the most informative studies on
20 female reproductive toxicity for dose-response analysis. Gao etal. [2011] identified statistically-
21 significant, dose-related increases in the incidence of cervical inflammatory cells in mice exposed to
22 low doses of benzo[a]pyrene for 98 days [Gao etal., 2011]. Cervical effects of increasing severity
23 (including epithelial hyperplasia, atypical hyperplasia, apoptosis, and necrosis] were also observed
24 at higher doses [Gao etal.. 2011]. There are no data on cervical effects in other species or in other
25 mouse strains. However, Gao etal. [2011] also evaluated cervical effects in separate groups of mice
26 exposed via intraperitoneal [i.p.] injection, and observed similar responses in these groups of mice,
27 providing support for the association between effects in this target organ and benzo[a]pyrene
28 exposure. Epidemiological studies have demonstrated an association between cigarette smoking
29 and increased risk of cervical cancer [Pate Capps etal., 2009]. In addition, benzo[a]pyrene
30 metabolites and benzo[a]pyrene-DNA adducts have been detected in human cervical mucus and
31 cervical tissues obtained from smokers [Phillips. 2002: Melikian et al.. 1999]. However, data to
32 support that cervical hyperplasia following oral benzo[a]pyrene exposure progresses to cervical
33 tumors were not available (no cervical tumors were noted in the two available lifetime oral cancer
34 bioassays]. Thus, in the absence of these data, cervical hyperplasia is presented as a noncancer
35 effect.
36 Xu etal. [2010] identified biologically and statistically significant decreases in ovary weight,
37 estrogen, and primordial follicles, and altered estrus cycling in treated animals. These reductions in
38 female reproductive parameters are supported by a large database of animal studies indicating that
This document is a draft for review purposes only and does not constitute Agency policy.
2-4 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 benzo[a]pyrene is ovotoxic with effects including decreased ovary weight, decreased primordial
2 follicles, and reduced fertility fBorman etal.. 2000: KristensenetaL 1995: Miller etal.. 1992:
3 Swartz and Mattison. 1985: Mackenzie and Angevine. 1981: Mattison et al.. 1980). Additionally,
4 epidemiology studies indicate that exposure to complex mixtures of PAHs, such as through cigarette
5 smoke, is associated with measures of decreased fertility in humans [Neal etal.. 2008: El-Nemr et
6 al.. 1998). Specific associations have also been made between infertility and increased levels of
7 benzo[a]pyrene in follicular fluid in women undergoing in vitro fertilization [Neal etal.. 2008].
8 Immunotoxicity
9 As described in Section 1.1.3, the immune system was identified as a potential human
10 hazard of benzo[a]pyrene exposure based on findings of organ weight and immunoglobulin
11 alterations, as well as effects on cellularity and functional changes in the immune system in animals.
12 The only available studies to support development of an RfD were conducted by Kroese etal.
13 [2001] and De Jong etal. [1999]. These are subchronic studies with multiple exposure levels and
14 adequate power to detect effects. In comparing these studies, the Kroese etal. [2001] study is
15 preferred for dose-response analysis due to its longer duration (90 days].
16 Decreasedthymus weight, observed in Kroese etal. [2001], decreased IgM and IgAlevels,
17 and decreased relative numbers of B-cells, observed in De long et al. [1999]. were selected for dose-
18 response analysis. It is recognized that thymus weight changes on their own have been noted to be
19 less reliable indicators of immunotoxicity [Luster etal.. 1992]. However, there are converging lines
20 of evidence that support the derivation of an organ/system-specific RfD for benzo[a]pyrene
21 immunotoxicity. Alterations in immunoglobulin levels have been noted in humans after exposure
22 to PAHs, as well as in animal studies after exposure to benzo[a]pyrene. Changes in B cell
23 populations in the spleen provide additional evidence of immunotoxicity. Finally, functional effects
24 on the immune system, including dose-related decreases in SRBC-specific IgM levels and dose-
25 dependent decreases in resistance to pneumonia or Herpes simplex type 2 following short-term
26 subcutaneous [s.c.] injection have been reported [Temple etal.. 1993: Munsonetal.. 1985]. The
27 observed decreases in thymus weight, IgM and IgA levels, and number of B cells associated with
28 exposure to benzo[a]pyrene were concluded to be representative of immunotoxicity following
29 benzo[a]pyrene exposure and were selected for dose-response analysis.
30 2.1.2. Methods of Analysis
31 No biologically based dose-response models are available for benzo[a]pyrene. In this
32 situation, EPA evaluates a range of dose-response models thought to be consistent with underlying
33 biological processes to determine how best to empirically model the dose-response relationship in
34 the range of the observed data. Consistent with this approach, all models available in EPA's
35 Benchmark Dose Software [BMDS] were evaluated. Consistent with EPA's Benchmark Dose
36 Technical Guidance Document [U.S. EPA, 2012c], the benchmark dose [BMD] and the 95% lower
37 confidence limit on the BMD [BMDL] were estimated using a benchmark response [BMR] of
This document is a draft for review purposes only and does not constitute Agency policy.
2-5 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 1 standard deviation (SD) from the control mean for continuous data or a BMR of 10% extra risk for
2 dichotomous data in the absence of information regarding what level of change is considered
3 biologically significant, and also to facilitate a consistent basis of comparison across endpoints,
4 studies, and assessments. The estimated BMDLs were used as points of departure (PODs). Further
5 details including the modeling output and graphical results for the best fit model for each endpoint
6 can be found in Appendix E of the Supplemental Information.
7 Among the endpoints identified as representative of the hazards of benzo[a]pyrene
8 exposure, the data for neurobehavioral changes in the elevated plus maze and Morris water maze
9 tests [Chenetal., 2012], decreased ovary weight [Xu etal., 2010], increased cervical hyperplasia
10 [Gao etal. [2011], and decreasedthymus weight [Kroese etal., 2001] were amenable to dose-
11 response modeling. Although the data for Morris water maze performance [Chenetal.. 2012] was
12 ultimately considered to be less informative than the elevated plus maze data (see Section 2.1.1],
13 EPA performed dose-response modeling on this endpoint to ensure that the elevated plus maze
14 data were an accurate representation of other sensitive, behavioral changes in this study. See
15 Appendix E of the Supplemental Information for details of statistical analyses.
16 The data for the remaining endpoints identified in Section 2.1.1 were not modeled.
17 Specifically, the data for cardiovascular effects observed in Tules etal. [2012] were limited due to
18 the reporting of results at only the two highest dose groups. The data for epididymal sperm counts
19 presented in the Mohamedetal. [2010] study were reported graphically only and requests for the
20 raw data were unsuccessful. The observed decrease in IgM and IgA [De long etal.. 1999] was
21 inconsistent and not amenable to dose-response modeling. NOAELs or LOAELs were used as the
22 POD for these endpoints.
23 Human equivalent doses [HEDs] for oral exposures were derived from the PODs estimated
24 from the laboratory animal data as described in EPA's Recommended Use of Body Weight3/4 as the
25 Default Method in Derivation of the Oral Reference Dose [U.S. EPA. 2011]. In this guidance, EPA
26 advocates a hierarchy of approaches for deriving HEDs from data in laboratory animals, with the
27 preferred approach being physiologically-based toxicokinetic modeling. Other approaches can
28 include using chemical-specific information in the absence of a complete physiologically-based
29 toxicokinetic model. As discussed in Appendix D of the Supplemental Information, several animal
30 physiologically based pharmacokinetic [PBPK] models for benzo[a]pyrene have been developed
31 and published, but a validated human PBPK model for benzo[a]pyrene for extrapolating doses from
32 animals to humans is not available. In lieu of either chemical-specific models or data to inform the
33 derivation of human equivalent oral exposures, a body weight scaling to the % power (i.e., B W3/4]
34 approach is applied to extrapolate toxicologically equivalent doses of orally administered agents
35 from adult laboratory animals to adult humans for the purpose of deriving an oral RfD. BW3/4
36 scaling was not employed for deriving HEDs from studies in which doses were administered
37 directly to early postnatal animals because of the absence of information on whether allometric
38 (i.e., body weight] scaling holds when extrapolating doses from neonatal animals to adult humans
This document is a draft for review purposes only and does not constitute Agency policy.
2-6 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
due to presumed toxicokinetic and/or toxicodynamic differences between lifestages [U.S. EPA,
2011: HattisetaL 20041
Consistent with EPA guidance fU.S. EPA. 20111. the points of departure (PODs) estimated
based on effects in adult animals are converted to HEDs employing a standard dosimetric
adjustment factor (DAF) derived as follows:
DAF =
where
BWa = animal body weight; and
BWh = human body weight.
Using BWa of 0.25 kg for rats and 0.035 kg for mice and BWh of 70 kg for humans [U.S. EPA.
1988). the resulting DAFs for rats and mice are 0.24 and 0.15, respectively. Applying this DAF to
the POD identified for effects in adult rats or mice yields a PODnED as follows (see Table 2-1):
PODHED = Laboratory animal dose (mg/kg-day) x DAF.
Table 2-1 summarizes the sequence of calculations leading to the derivation of a human-
equivalent POD for each data set discussed above.
Table 2-1. Summary of derivation of PODs
Endpoint and
reference
Species/
sex
Model3
BMR
BMD
(mg/kg-d)
BMDL
(mg/kg-d)
PODADJb
(mg/kg-d)
PODHEDC
(mg/kg-d)
Developmental
Neurobehavioral
changes
Chen etal. (2012)
Cardiovascular
effects
Jules etal. (2012)
Female
Sprague-
Dawley rats
Long-Evans
rats
Exponential
(M4)a
ISO
0.18
0.09
LOAEL(0.6mg/kg-d)
(15% T* in systolic blood pressure; 33% T* in
diastolic blood pressure)
0.09
0.6
0.09
0.15
Reproductive
Decreased ovary
weight
Xu etal. (2010)
Decreased
intratesticular
testosterone
Zheng etal. (2010)
Female
Sprague-
Dawley rats
Male
Sprague-
Dawley rats
Linear3
ISO
2.3
1.5
NOAEL (1 mg/kg-d)
(15% -^ in testosterone)
1.5
1
0.37
0.24
This document is a draft for review purposes only and does not constitute Agency policy,
2-7 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Endpointand
reference
Decreased sperm
count and motility
Mohamed et al.
(2010)
Cervical epithelial
hyperplasia
Gaoetal. (2011)
Species/
sex
Male C57BL/6
mice
Female ICR
mice
Model3
BMR
BMD
(mg/kg-d)
BMDL
(mg/kg-d)
LOAEL(lmg/kg-d)
(50% -^ in sperm count; 20% 4, in sperm motility)
Log-
logistic3
10%
0.58
0.37
PODADJb
(mg/kg-d)
1
0.37
PODHEDC
(mg/kg-d)
0.15
0.06
Immunological
Decreased thymus
weight
Kroese et al. (2001)
Decreased IgM levels
De Jong etal. (1999)
Decreased IgA levels
De Jong etal. (1999)
Decreased number
of B cells
De Jong etal. (1999)
Female
Wistar rats
Male
Wistar rats
Male
Wistar rats
Male
Wistar rats
Linear3
ISO
10.5
7.6
NOAEL(10mg/kg-d)
(14% 4, in IgM)
NOAEL(30mg/kg-d)
(28% 4, in IgA)
NOAEL(30mg/kg-d)
(7% -tin B cells at NOAEL; 31% ^ at LOAEL)
7.6
7.1
21
21
1.9
1.7
5.2
5.2
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
3For modeling details, see Appendix E in Supplemental Information.
bFor studies in which animals were not dosed daily, administered doses were adjusted to calculate the TWA daily
doses prior to BMD modeling.
CHED PODs were calculated using BW3/4scaling (U.S. EPA, 2011) for effects from dosing studies in adult animals
(i.e., Gao etal., 2011; Mohamed etal., 2010; Xu et al., 2010; De Jong etal., 1999) or for developmental effects
resulting from in utero exposures. BW3/4 scaling was not employed for deriving HEDs from studies in which doses
were administered directly to early postnatal animals (i.e., Chen et al., 2012) because of the absence of
information on whether allometric (i.e., body weight) scaling holds when extrapolating doses from neonatal
animals to adult humans due to presumed toxicokinetic and/or toxicodynamic differences between lifestages
(U.S. EPA, 2011; Hattis et al., 2004).
2.1.3. Derivation of Candidate Values
Under EPA's A Review of the Reference Dose and Reference Concentration Processes [U.S. EPA.
2002: Section 4.4.5). also described in the Preamble, five possible areas of uncertainty and
variability were considered. An explanation of the five possible areas of uncertainty and variability
follows:
An intraspecies uncertainty factor, UFn, of 10 was applied to account for variability and
uncertainty in toxicokinetic and toxicodynamic susceptibility within the subgroup of the human
population most sensitive to the health hazards of benzo[a]pyrene [U.S. EPA. 2002). In the case of
benzo[a]pyrene, the PODs were derived from studies in inbred animal strains and are not
considered sufficiently representative of the exposure and dose-response of the most susceptible
human subpopulations (in this case, the developing fetus). In certain cases, the toxicokinetic
component of this factor may be replaced when a PBPK model is available that incorporates the
This document is a draft for review purposes only and does not constitute Agency policy.
2-8 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 best available information on variability in toxicokinetic disposition in the human population
2 (including sensitive subgroups). In the case of benzo[a]pyrene, insufficient information is available
3 to quantitatively estimate variability in human susceptibility; therefore, the full value for the
4 intraspecies UF was retained.
5 An interspecies uncertainty factor, UFA, of 3 (101/2 = 3.16, rounded to 3) was applied, to all
6 PODs in Table 2-2 except Chen et al. [2012]. because BW3/4 scaling is being used to extrapolate oral
7 doses from laboratory animals to humans. Although BW3/4 scaling addresses some aspects of cross-
8 species extrapolation of toxicokinetic and toxicodynamic processes, some residual uncertainty
9 remains. In the absence of chemical-specific data to quantify this uncertainty, EPA's BW3/4 guidance
10 [U.S. EPA, 2011] recommends use of a UF of 3. BW3/4 scaling was not employed for deriving HEDs
11 from studies in which doses were administered directly to early postnatal animals [i.e. Chenetal..
12 2012] because of the absence of information on whether allometric (i.e., body weight] scaling holds
13 when extrapolating doses from neonatal animals to adult humans due to presumed toxicokinetic
14 and/or toxicodynamic differences between lifestages (U.S. EPA. 2011: Hattis etal.. 2004]. In this
15 case, a value of 10 was applied because of the absence of quantitative information to characterize
16 either the toxicokinetic or toxicodynamic differences between animals and humans at this lifestage.
17 A subchronic to chronic uncertainty factor, UFS, of 1 was applied when dosing occurred during
18 gestation (Tules etal.. 2012] or the early postnatal period (Chenetal.. 2012] that is relevant to
19 developmental effects. The developmental period is recognized as a susceptible lifestage and
20 repeated exposure is not necessary for the manifestation of developmental toxicity (U.S. EPA.
21 1991c]. A value of 10 was applied when the POD was based on a subchronic study (studies in
22 Table 2-2, other than the two developmental toxicity studies, were 42-90 days in duration] to
23 account for the possibility that longer exposure may induce effects at a lower dose.
24 A UF for extrapolation from a LOAEL to NOAEL, UFL, of 1 was applied when the POD was
25 based on a NOAEL (Zheng etal.. 2010: De long etal.. 1999]. A value of 1 was applied for LOAEL-to-
26 NOAEL extrapolation when a BMR of a 1 SD (Chenetal.. 2012: Kroese etal.. 2001] or 10% change
27 (Gao etal.. 2011] from the control was selected under an assumption that it represents a minimal
28 biologically significant response level. A NOAEL was not determined for the most sensitive effects
29 observed in Jules et al. (2012] and Mohamed etal. (2010]. At the LOAEL. Jules etal. (2012]
30 observed statistically significant increases in systolic (15%] and diastolic (33%] blood pressure
31 when measured in adulthood following gestational exposure. Regarding the study by Mohamed et
32 al. (2010]. the authors observed a statistically significant decrease sperm count (50%] and motility
33 (20%] in treated FO males at the LOAEL and the observed decrements in sperm count persisted in
34 untreated Fl male offspring. The data reported in these studies were not amenable to dose-
35 response modeling, which would have allowed for extrapolation to a minimally biologically
36 significant response level. Therefore, a full UF of 10 was applied to approximate a NOAEL for these
37 studies, which observed a high magnitude of response at the LOAEL. A database uncertainty factor,
38 UFo, of 3 was applied to account for database deficiencies, including the lack of a standard
This document is a draft for review purposes only and does not constitute Agency policy.
2-9 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 multigenerational study or extended 1-generation study that includes exposure from premating
2 through lactation, considering that benzo [a]pyrene has been shown to affect fertility in adult male
3 and female animals by multiple routes of exposure (see Section 1.1.2). Considering that decreased
4 fertility in adult male and female mice is observed following gestational exposure, it is assumed that
5 exposure occurring over this more comprehensive period of development could result in a lower
6 POD. Also, the lack of a study examining functional neurological endpoints following a more
7 comprehensive period of developmental exposure (i.e., gestation through lactation) is a data gap,
8 considering human and animal evidence indicating altered neurological development (see
9 Section 1.1.1).
10 Table 2-2 is a continuation of Table 2-1 and summarizes the application of UFs to each POD
11 to derive a candidate value for each data set. The candidate values presented in the table below are
12 preliminary to the derivation of the organ/system-specific reference values. These candidate
13 values are considered individually in the selection of a representative oral reference value for a
14 specific hazard and subsequent overall RfD for benzo[a]pyrene.
15 Table 2-2. Effects and corresponding derivation of candidate values
Endpoint and reference
PODHEDa
(mg/kg-d)
POD
type
UFA
UFH
UFL
UFS
UFD
Composite
UF
Candidate
value
(mg/kg-d)
Developmental
Neurobehavioral changes in
rats
Chen etal. (2012)
Cardiovascular effects in rats
Jules etal. (2012)
0.09
0.15
BMDL1SD
LOAEL
10
3
10
10
1
10
1
1
3
3
300
1,000
3 x ID'4
2 x 10"4
Reproductive
Decreased ovary weight in
rats
Xu etal. (2010)
Decreased intratesticular
testosterone in rats
Zheng etal. (2010)
Decreased sperm count and
motility in mice
Mohamed et al. (2010)
Cervical epithelial
hyperplasia in mice
Gao etal. (2011)
0.37
0.24
0.15
0.06
BMDL1SD
NOAEL
LOAEL
BMDL10
3
3
3
3
10
10
10
10
1
1
10
1
10
10
10
10
3
3
3
3
1,000
1,000
10,000
1,000
4 x 10"4
2 x ID'4
Not calculated
duetoUF
>3,000a
6 x 10"5
This document is a draft for review purposes only and does not constitute Agency policy,
2-10 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Endpoint and reference
PODHEDa
(mg/kg-d)
POD
type
UFA
UFH
UFL
UFS
UFD
Composite
UF
Candidate
value
(mg/kg-d)
Immunological
Decreased thymus weight in
rats
Kroese et al. (2001)
Decreased serum IgM in rats
De Jong etal. (1999)
Decreased serum IgA in rats
De Jong etal. (1999)
Decreased number of B cells
in rats
De Jong etal. (1999)
1.9
1.7
5.2
5.2
BMDL1SD
NOAEL
NOAEL
NOAEL
3
3
3
3
10
10
10
10
1
1
1
1
10
10
10
10
3
3
3
3
1,000
1,000
1,000
1,000
2 x 10"3
2 x 10"3
5 x 10"3
5 x 10"3
1
2
3
4
5
6
7
aAs recommended in EPA's A Review of the Reference Dose and Reference Concentration Processes (U.S. EPA,
2002), the derivation of a reference value that involves application of the full 10-fold UF in four or more areas of
extrapolation should be avoided.
Figure 2-1 presents graphically the candidate values, UFs, and PODs, with each bar
corresponding to one data set described in Tables 2-1 and 2-2.
This document is a draft for review purposes only and does not constitute Agency policy,
2-11 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Neurodevelopmental
alterations in rats
(Chen eta!.,2012)
Cardiovascular effects
in uts (Julesot al.,2012)
Composite UF
A Candidate value
• POD(HED)
Q
O
Q-
UJ
C£.
4' Ovary weight in rats
(Xuet al., 2010)
4-- Intratesticular
testosterone in rats
(Zheng eta I. ,2010)
4- Sperm count and
motility in mice
(Mohamedetal., 2010)
Cervical epithelial
hyperplasia in mice
(Gaoetal , 2011)
• 'is Thymus vvc-ight in rats
(Kroesectal.,2001)
1
2
3
4
5
6
ID
x b< run IgM in rats
(DeJonget al., 1999)
\_, Seruri IgA in rats
(DeJongetal., 1999)
4- Number of B cells in rats
(DeJongetal., 1999)
0.00001 0.0001 0.001 0.01 0.1
Doses (mg/kg-d)
10
Figure 2-1. Candidate values with corresponding PODs and composite UFs.
2.1.4. Derivation of Organ/System-Specific Reference Doses
Table 2-3 distills the candidate values from Table 2-2 into a single value for each organ or
system. These organ- or system-specific reference values may be useful for subsequent cumulative
risk assessments that consider the combined effect of multiple agents acting at a common site.
772/s document ;'s a draft for review purposes only and does not constitute Agency policy,
2-12 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2
Table 2-3. Organ/system-specific RfDs and proposed overall RfD for
benzo[a]pyrene
Effect
Developmental
Reproductive
Immunological
Proposed overall RfD
Basis
Neurobehavioral changes
Decreased ovary weight
Decreased thymus weight and serum IgM
Developmental toxicity
RfD (mg/kg-d)
3 x irj-4
4 x 10"4
2 x 10"3
3 x itr4
Study
exposure
description
Critical
window of
development
(postnatal)
Subchronic
Subchronic
Critical
window of
development
(postnatal)
Confidence
Medium
Medium
Low
Medium
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Developmental Toxicity
The candidate value based on neurobehavioral changes in rats [Chenetal.. 2012] was
selected as the organ/system-specific RfD representing developmental toxicity. This candidate
value was selected because it is associated with the application of the smaller composite UF and
because similar effects were replicated across other studies [Maciel etal.. 2014: Bouayedetal..
2009a: Bouayedetal.. 2009b: Grovaetal.. 20081
Reproductive Toxicity
Among the adverse reproductive effects associated with oral benzo[a]pyrene exposure,
decrements in sperm parameters, decreases in testosterone, and effects in the ovary were
supported by a large body of evidence. The data supporting cervical effects are limited to a single
study, and were therefore given less weight compared to the other reproductive effects. The
derivation of a candidate value based on decreased sperm count and motility [Mohamedetal..
2010] involved too much uncertainty (see Table 2-2] and the study used to derive a candidate value
based on decreased testosterone [Zheng etal.. 2010] did not observe a dose-response relationship
(a 15% decrease in testosterone was seen at the low and high doses, with statistical significance at
the high dose]. The study by Xu etal. [2010] observed a dose-response relationship for decreased
ovary weight (both doses were statistically significant]. Additionally, statistically significant
decreases in primordial follicles were observed at the high dose, supporting the ovaries as a target
of toxicity. Therefore, the candidate value based on decreased ovary weight in rats from the Xu et
al. (2010] study was selected as the organ/system-specific RfD representing reproductive toxicity.
The ovarian effects are supported by a large database of animal studies and human studies of
exposure to benzo[a]pyrene and PAH mixtures. While evidence in the benzo[a]pyrene database
This document is a draft for review purposes only and does not constitute Agency policy,
2-13 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 supports a male and female reproductive hazard, there is more confidence in the POD from Xu et al.
2 [2010] as the basis for an organ/system-specific RfD for reproductive effects.
3 Immunotoxicity
4 The candidate values based on decreased thymus weight [Kroese etal.. 2001] and serum
5 IgM levels in rats [De long et al.. 1999] were selected as the organ/system-specific RfD representing
6 immunotoxicity. The observed decreases in thymus weight, IgM and IgA levels, and number of
7 B cells associated with exposure to benzo[a]pyrene were determined to be representative of
8 immunotoxicity. In combination, these effects provide more robust evidence of immunotoxicity.
9 The candidate values for decreased thymus weight [Kroese etal., 2001] and serum IgM levels in
10 rats [De long etal.. 1999] were equal and provided the most sensitive POD; thus, these candidate
11 values were selected as the organ/system-specific RfD representing immunotoxicity.
12 2.1.5. Selection of the Proposed Overall Reference Dose
13 Multiple organ/system-specific reference doses were derived for effects identified as
14 human hazards or potential hazards from benzo[a]pyrene including developmental toxicity,
15 reproductive toxicity (representative of effects in both sexes], and immunotoxicity. To estimate an
16 exposure level below which effects from benzo[a]pyrene exposure are not expected to occur, the
17 lowest organ/system-specific RfD (3 x 1Q-4 mg/kg-day] is proposed as the overall RfD for
18 benzo[a]pyrene. This value, based on induction of neurobehavioral changes in rats exposed to
19 benzo[a]pyrene during a susceptible lifestage, is supported by several animal and human studies
20 (see Section 1.1.1].
21 The overall RfD is derived to be protective of all types of effects for a given duration of
22 exposure and is intended to protect the population as a whole including potentially susceptible
23 subgroups (U.S. EPA. 2002]. This value should be applied in general population risk assessments.
24 However, decisions concerning averaging exposures over time for comparison with the RfD should
25 consider the types of toxicological effects and specific lifestages of concern. For example,
26 fluctuations in exposure levels that result in elevated exposures during various lifestages could
27 potentially lead to an appreciable risk, even if average levels over the full exposure duration were
28 less than or equal to the RfD. Alternatively, developmental toxicity may not be a concern due to
29 exposure scenarios in which exposure is occurring outside of the critical window of development
30 2.1.6. Confidence Statement
31 A confidence level of high, medium, or low is assigned to the study used to derive the RfD,
32 the overall database, and the RfD itself, as described in Section 4.3.9.2 of EPA's Methods for
33 Derivation of Inhalation Reference Concentrations and Application of Inhalation Dosimetry (U.S. EPA.
34 1994].
35 Confidence in the principal study (Chen etal., 2012] is medium-to-high. The study design
36 included randomized experimental testing, blinded observations, culling of pups to account for
This document is a draft for review purposes only and does not constitute Agency policy.
2-14 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 nutritional availability, treatment-randomization, and controls for litter and nursing bias. Some
2 informative experimental details were, however, omitted including the sensitivity of some assays at
3 the indicated developmental ages, gender-specific data for all outcomes, and individual animal data.
4 Notably, these study limitations do not apply to the endpoint chosen to derive the RfD, and the
5 overall methods and reporting are considered sufficient. Confidence in the database is medium,
6 primarily due to the lack of a multigenerational reproductive toxicity study given the sensitivity to
7 benzo[a]pyrene during development. Reflecting medium-to-high confidence in the principal study
8 and medium confidence in the database, confidence in the RfD is medium.
9 2.1.7. Previous IRIS Assessment: Reference Dose
10 An RfD was not derived in the previous IRIS assessment
11 2.2. INHALATION REFERENCE CONCENTRATION FOR EFFECTS OTHER
12 THAN CANCER
13 The inhalation reference concentration (RfC) (expressed in units of mg/m3) is defined as an
14 estimate (with uncertainty spanning perhaps an order of magnitude) of a continuous inhalation
15 exposure to the human population (including sensitive subgroups) that is likely to be without an
16 appreciable risk of deleterious effects during a lifetime. It can be derived from a NOAEL, LOAEL, or
17 the 95 percent lower bound on the benchmark concentration (BMCL), with UFs generally applied to
18 reflect limitations of the data used.
19 2.2.1. Identification of Studies and Effects for Dose-Response Analysis
20 In Section 1.2.1, developmental and reproductive toxicities were identified as hazards of
21 benzo[a]pyrene exposure by the inhalation route. Studies within each effect category were
22 evaluated using general study quality characteristics (as discussed in Section 6 of the Preamble) to
23 help inform the selection of studies from which to derive toxicity values. Rationales for selecting
24 the studies and effects to represent each of these hazards are summarized below.
25 Human studies of environmental PAH mixtures across multiple cohorts have observed
26 developmental and reproductive effects following prenatal exposure. However, these studies are
27 limited by exposure to complex mixtures of PAHs; and, within individual studies, there may have
28 been more than one route of exposure. In addition, the available human studies that utilized
29 benzo[a]pyrene-DNA adducts as the exposure metric do not provide external exposure levels of
30 benzo[a]pyrene from which to derive an RfC. Although preferred for derivation of reference values,
31 human studies were not considered because of the contribution to the observed hazard of multiple
32 PAHs across multiple routes of exposure and uncertainty due to concurrent exposure to other PAHs
33 and other components of the mixtures (such as metals).
34 Animal studies were evaluated to determine which provided the most relevant routes and
35 durations of exposure, multiple exposure levels to provide information about the shape of the dose
36 response curve, and relative ability to detect effects at low exposure levels. The only chronic animal
This document is a draft for review purposes only and does not constitute Agency policy.
2-15 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 inhalation study available for benzo[a]pyrene, Thyssenetal. [1981], was designed as a cancer
2 bioassay and did not report other effects; however, the inhalation database for benzo[a]pyrene
3 includes several shorter duration studies that are sufficient for use in deriving reference values
4 [U.S. EPA. 2002). Specifically, several reproductive toxicity studies are available for the inhalation
5 route, including one subchronic study Archibong et al. [2008]. Furthermore, several developmental
6 studies are available that help identify hazards of exposure during sensitive developmental
7 windows [Wormley etal.. 2004: Archibong et al.. 2002]. In addition, a 4-week inhalation study in
8 rats is available that investigated, but did not detect, lung injury [Wolff etal.. 1989]. The inhalation
9 database for benzo[a]pyrene is less extensive than the database of studies by the oral route;
10 however, the types of noncancer effects observed are consistent between routes and are supported
11 by studies in human populations (see Sections 1.1.1,1.1.2, and 1.1.3].
12 Developmental Toxicity
13 Developmental toxicity, as represented by decreased fetal survival and developmental
14 neurotoxicity, was observed in several inhalation studies [Wormley et al.. 2004: Wuetal.. 2003a:
15 Archibong etal., 2002]. Wu etal. [2003a] was not considered for dose-response analysis due to
16 lack of study details related to number of dams and litters per group and lack of reporting of
17 numerical data. Wormley etal. [2004] was not considered for dose-response analysis, as this study
18 employed only a single exposure group at which overt toxicity was noted (a 66% reduction in fetal
19 survival].
20 Of the studies demonstrating developmental toxicity, the study conducted by Archibong et
21 al. [2002] was identified as the most informative study for dose-response analysis. Archibong et al.
22 [2002] observed decreased fetal survival at the lowest dose tested by the inhalation route on
23 CDs 11-20 (i.e., LOAEL of 25 [ig/m3]. This study indicates that the developing fetus is a sensitive
24 target following inhalation exposure to benzo[a]pyrene. The observed decrease in fetal survival is
25 supported by the oral database for benzo[a]pyrene (e.g., decreased survival of litters in mice
26 following in utero exposure to benzo[a]pyrene on CDs 7-16] [Mackenzie and Angevine. 1981].
27 Reproductive Toxicity
28 Reproductive toxicity, as represented by reductions in sperm quality, both count and
29 motility, and testis weights in adults, was observed by Archibong et al. [2008], Rameshetal. [2008]
30 and Archibong et al. f20021 Archibong et al. f20081 and Rameshetal. f20081 reported the results
31 of a single exposure, subchronic inhalation exposure study in male rats. This subchronic study was
32 of sufficient duration and possessed adequate power to detect effects, but utilized a single exposure
33 concentration, which is less informative for dose-response analysis than a design using multiple
34 exposure concentrations. However, this single-dose subchronic study is consistent with male
35 reproductive effects observed across multiple studies by the oral route and with human studies in
36 PAH exposed populations (see Section 1.1.2]. The endpoints of decreased testes weight and sperm
37 count and motility reported in Archibong et al. [2008] were selected for dose-response analysis as
This document is a draft for review purposes only and does not constitute Agency policy.
2-16 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 both represent sensitive endpoints of male reproductive toxicity and are indicators of potentially
2 decreased fertility. These effects are also consistent with human studies in PAH exposed
3 populations as effects on male fertility and semen quality have been demonstrated in
4 epidemiological studies of smokers [Spares and Melo. 2008).
5 2.2.2. Methods of Analysis
6 Data for decreased fetal survival from Archibong et al. [2002] were reported as litter means
7 and SDs. These data were not amenable to BMD modeling due to the pattern of variability in the
8 data set, and attempts to obtain the raw data from the study authors were unsuccessful. Therefore,
9 the LOAEL from this study was used as the POD for dose-response analysis. The study by
10 Archibong et al. [2008]. using only one exposure level, was judged not to support dose-response
11 modeling due to the lack of understanding of the underlying dose-response relationship. LOAELs
12 were also used as the PODs for dose-response analysis.
13 By definition, the RfC is intended to apply to continuous lifetime exposures for humans [U.S.
14 EPA. 1994]. EPA recommends that adjusted continuous exposures be used for inhalation
15 developmental toxicity studies as well as for studies of longer durations [U.S. EPA, 2002]. The
16 LOAELs identified from Archibong et al. [2002] and Archibong et al. [2008] were adjusted to
17 account for the discontinuous daily exposure as follows:
18
19 PODADj = POD x hours exposed per day/24 hours
20 = LOAEL x (duration of exposure/24 hours]
21 = PODAD,
22
23 Next, the human equivalent concentration [HEC] was calculated from the PODAoj by
24 multiplying by a DAF, which, in this case, was the regional deposited dose ratio [RDDRER] for
25 extrarespiratory (i.e., systemic] effects as described in Methods for Derivation of Inhalation
26 Reference Concentrations and Application of Inhalation Dosimetry [U.S. EPA. 1994]. The observed
27 developmental effects are considered systemic in nature (i.e., extrarespiratory] and the normalizing
28 factor for extrarespiratory effects of particles is body weight. The RDDRER was calculated as
29 follows:
30 BWA (VE)H (FTOT)H
31 where:
32 BW = body weight [kg];
33 VE = ventilation rate (L/min]; and
34 FTOT = total fractional deposition.
35
36 The total fractional deposition includes particle deposition in the nasal-pharyngeal,
37 tracheobronchial, and pulmonary regions. FTOT for both animals and humans was calculated using
This document is a draft for review purposes only and does not constitute Agency policy.
2-17 DRAFT— DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 the Multi-Path Particle Dosimetry (MPPD) model, a computational model used for estimating
2 human and rat airway particle deposition and clearance (MPPD; Version 2.0 © 2006, publicly
3 available through the Hamner Institute). FTOT was based on the average particle size of 1.7 ± 0.085
4 |im (mass median aerodynamic diameter [MMAD] ± geometric SD) as reported in Wu etal. (2003a)
5 for the exposure range 25-100 |im3. For the model runs, the Yeh-Schum 5-lobe model was used for
6 the human and the asymmetric multiple path model was used for the rat (see Appendix E for MPPD
7 model output). Both models were run under nasal breathing scenarios with the inhalability
8 adjustment selected. A geometric SD of 1 was used as the default by the model because the
9 reported geometric SD of 0.085 was <1.05.
10 The human parameters used in the model for calculating FTOT and in the subsequent
11 calculation of the PODnEc were as follows: human body weight, 70 kg; VE, 13.8 L/minute; breathing
12 frequency, 16 per minute; tidal volume, 860 mL; functional residual capacity, 3,300 mL; and upper
13 respiratory tract volume, 50 mL. Although the most sensitive population in Archibong et al. (2002)
14 is the developing fetus, the adult rat dams were directly exposed. Thus, adult rat parameters were
15 used in the calculation of the HEC. The parameters used for the rat were body weight, 0.25 kg (a
16 generic weight for male and female rats); VE, 0.18 L/minute; breathing frequency, 102 per minute;
17 tidal volume, 1.8 mL; functional residual capacity, 4 mL; and upper respiratory tract volume,
18 4.42 mL. All other parameters were set to default values (see Appendix E).
19 Under these conditions, the MPPD model calculated FTOT values of 0.621 for the human and
20 0.181 for the rat Using the above equation, the RDDRER was calculated to be 1.1.
21 From this, the PODHEc was calculated as follows:
22
23 PODHEc = PODAD, x RDDRER
24
25 Table 2-4 summarizes the sequence of calculations leading to the derivation of a human-
26 equivalent POD for each data set discussed above.
This document is a draft for review purposes only and does not constitute Agency policy.
2-18 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Table 2-4. Summary of derivation of PODs
Endpoint and
reference
Species/sex
Model
Developmental
Decreased fetal survival
Archibong et al. (2002)
Pregnant F344 rats
BMR
LOAEL(25u.g/m3)
BMC
(Hg/m3)
19% ^
BMCL
(Hg/m3)
PODADJa
(Hg/m3)
PODHECb
(Hg/m3)
4.2
4.6
Reproductive
Decreased testis weight
Archibong et al. (2008)
Decreased sperm count
and motility
Archibong et al. (2008)
Male F344 rats
Male F344 rats
LOAEL(75u.g/m3)
LOAEL(75u.g/m3
69% 4/sperm cou
73% 4, sperm mo
54% T* abnormal
34% ^
nt
tility
sperm
12.5
12.5
13.8
13.8
2
3 aPODs were adjusted for continuous daily exposure: PODADJ= POD x hours exposed per day/24 hours.
4 bPODHEc calculated by adjusting the PODADJ by the RDDR calculated using particle size reported in Hood et al. (2000)
5 using MPPD software as detailed in Section 2.2.2 and Appendix E in the Supplemental Information.
6 2.2.3. Derivation of Candidate Values
7 Under EPA's A Review of the Reference Dose and Reference Concentration Processes [U.S. EPA.
8 2002: Section 4.4.5], also described in the Preamble, five possible areas of uncertainty and
9 variability were considered. An explanation of the five possible areas of uncertainty and variability
10 follows:
11 An intraspecies uncertainty factor, UFH, of 10 was applied to account for variability and
12 uncertainty in toxicokinetic and toxicodynamic susceptibility within the subgroup of the human
13 population most sensitive to the health hazards of benzo[a]pyrene [U.S. EPA. 2002). In the case of
14 benzo[a]pyrene, the PODs were derived from studies in inbred animal strains and are not
15 considered sufficiently representative of the exposure and dose-response of the most susceptible
16 human subpopulations (in this case, the developing fetus). In certain cases, the toxicokinetic
17 component of this factor may be replaced when a PBPK model is available that incorporates the
18 best available information on variability in toxicokinetic disposition in the human population
19 (including sensitive subgroups). In the case of benzo[a]pyrene, insufficient information is available
20 to quantitatively estimate variability in human susceptibility; therefore, the full value for the
21 intraspecies UF was retained.
22 An interspecies uncertainty factor, UFA, of 3 (101/2 = 3.16, rounded to 3) was applied to
23 account for residual uncertainty in the extrapolation from laboratory animals to humans in the
24 absence of information to characterize toxicodynamic differences between rats and humans after
25 inhalation exposure to benzo[a]pyrene. This value is adopted by convention where an adjustment
26 from animal to a HEC has been performed as described in EPA's Methods for Derivation of Inhalation
27 Reference Concentrations and Application of Inhalation Dosimetry (U.S. EPA. 1994).
This document is a draft for review purposes only and does not constitute Agency policy.
2-19 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 A subchronic to chronic uncertainty factor, UFS, of 1 was applied when dosing occurred
2 during gestation [Archibongetal.. 2002] or the early postnatal period that is relevant to
3 developmental effects [U.S. EPA. 1991a). A value of 10 was applied when the POD is based on a
4 subchronic study to account for the possibility that longer exposure may induce effects at a lower
5 dose [Archibong et al.. 2008] was 60 days in duration). A UF for extrapolation from a LOAEL to a
6 NOAEL, UFL, of 10 was applied when a LOAEL was used as the POD (Archibongetal.. 2008:
7 Archibongetal.. 2002]. The data reported in these studies were not amenable to dose-response
8 modeling, which would have allowed for extrapolation to a minimally biologically significant
9 response level. At the LOAEL, these studies observed a high magnitude of response (see Table 2-4].
10 Therefore, a full UF of 10 was applied to approximate a NOAEL for studies that observed a high
11 magnitude of response at the LOAEL. For example, the LOAEL used as the POD for the
12 developmental effect observed in Archibongetal. (2002] was based on a 19% decrease in fetal
13 survival.
14 A database uncertainty factor, UFo, of 10 was applied to account for database deficiencies,
15 including the lack of a standard multigenerational study or extended 1-generation study that
16 includes exposure from premating through lactation, considering that benzo[a]pyrene has been
17 shown to affect fertility in adult male and female animals by multiple routes of exposure and that
18 decrements in fertility are greater following developmental exposure (see Section 1.1.2).
19 In addition, the lack of a study examining functional neurological endpoints following
20 inhalation exposure during development is also a data gap, considering human and animal evidence
21 indicating altered neurological development following exposure to benzo[a]pyrene alone or
22 through PAH mixtures (see Section 1.1.1).
23 The most sensitive POD for the RfC candidate values in Table 2-5 is based on the endpoint of
24 decreased fetal survival observed in Archibongetal. (2002). However, oral exposure studies have
25 demonstrated neurotoxicity at doses lower than those where decreased fetal survival was
26 observed. A statistically significant decrease in fetal survival was observed following treatment
27 with 160 mg/kg-day benzo[a]pyrene, but not at lower doses (Mackenzie and Angevine. 1981):
28 however, other oral studies observed statistically significant neurobehavioral effects at doses of
29 benzo[a]pyrene around 0.2-2 mg/kg-day (Chenetal.. 2012: Bouayed etal.. 2009a). Considering
30 the relative sensitivity of the systemic health effects observed in the oral database, it is likely that
31 neurodevelopmental toxicity would be expected to occur at exposure concentrations below the
32 POD for the RfC based on decreased fetal survival.
33 According to EPA's A Review of the Reference Dose and Reference Concentration Processes
34 (U.S. EPA. 2002: Section 4.4.5). the UFD is intended to account for the potential for deriving an
35 under-protective RfD/RfC as a result of an incomplete characterization of the chemical's toxicity,
36 but also including a review of existing data that may also suggest that a lower reference value might
37 result if additional data were available. Therefore, a database UF of 10 for the benzo[a]pyrene
This document is a draft for review purposes only and does not constitute Agency policy.
2-20 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2
3
4
5
6
7
inhalation database was applied to account for the lack of a multigenerational study and the lack of
a developmental neurotoxicity study.
Table 2-5 is a continuation of Table 2-4 and summarizes the application of UFs to each POD
to derive a candidate values for each data set The candidate values presented in the table below
are preliminary to the derivation of the organ/system-specific reference values. These candidate
values are considered individually in the selection of an RfC for a specific hazard and subsequent
overall RfC for benzo[a]pyrene.
Table 2-5. Effects and corresponding derivation of candidate values
Endpoint
PODHEC
(Hg/m3)
POD
type
UFA
UFH
UFL
UFS
UFD
Composite
UFb
Candidate
value3
(mg/m3)
Developmental
Decreased fetal survival
in rats
Archibong et al. (2002)
4.6
LOAEL
3
10
10
1
10
3,000
2 x 10"6
Reproductive
Decreased testis weight
in rats
Archibong et al. (2008)
Decreased sperm count
and motility in rats
Archibong et al. (2008)
13.8
13.8
LOAEL
LOAEL
3
3
10
10
10
10
10
10
10
10
30,000
30,000
Not calculated
duetoUF
>3,000
Not calculated
duetoUF
>3,000
9
10
11
12
13
14
15
16
aCandidate values were converted from u.g/m3 to mg/m3.
bAs recommended in EPA's A Review of the Reference Dose and Reference Concentration Processes (U.S. EPA,
2002), the derivation of a reference value that involves application of the full 10-fold UF in four or more areas of
extrapolation should be avoided.
Figure 2-2 presents graphically these candidate values UFs, and PODs, with each bar
corresponding to one data set described in Tables 2-4 and 2-5.
This document is a draft for review purposes only and does not constitute Agency policy,
2-21 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Q-
O
4- fetal survival in rats
(Archibong etal., 2002)
Composite UF
A Candidate value
• POD(HEC)
Q
O
cc
x]/ testis weight in rats
(Archibongetal., 2008)
4- sperm count and motility
in rats (Archibong et al., 2008)
0.001 0.01 0.1 1
Exposure concentration
10
100
1
2 Figure 2-2. Candidate values with corresponding PODs and composite UFs.
3 2.2.4. Derivation of Organ/System-Specific Reference Concentrations
4 Table 2-6 distills the candidate values from Table 2-5 into a single value for each organ or
5 system. These organ- or system-specific reference values may be useful for subsequent cumulative
6 risk assessments that consider the combined effect of multiple agents acting at a common site. The
7 candidate values for reproductive toxicity from Archibong et al. [2008] were not derived to
8 represent reproductive toxicity because as recommended in EPA's A Review of the Reference Dose
9 and Reference Concentration Processes [U.S. EPA. 2002]. the derivation of a reference value that
10 involves application of the full 10-fold UF in four or more areas of extrapolation should be avoided.
11 Table 2-6. Organ/system-specific RfCs and proposed overall RfC for
12 benzo[a]pyrene
Effect
Developmental
Reproductive
Proposed Overall RfC
Basis
Decreased fetal survival
Reductions in testes weight and
sperm parameters
Decreased fetal survival
RfC (mg/m3)
2 x 10"6
Not calculated
2 x 10"6
Study exposure
description
Critical window of
development
(prenatal)
Subchronic
Critical window of
development
(prenatal)
Confidence
Low-medium
NA
Low-medium
This document is a draft for review purposes only and does not constitute Agency policy,
2-22 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 2.2.5. Selection of the Proposed Reference Concentration
2 The derivation of multiple organ/system-specific reference concentrations were considered
3 for effects identified as human hazards of benzo[a]pyrene inhalation exposure, i.e., developmental
4 and reproductive toxicity. However, an organ/system-specific RfC to represent reproductive
5 toxicity could not be derived due to high uncertainty (i.e., a composite UF of >3,000).
6 An overall RfC of 2 x 10~6 mg/m3 was selected based on the hazard of developmental
7 toxicity. The study by Archibong et al. [2002] was selected as the study used for the derivation of
8 the proposed overall RfC, as it observed biologically significant effects at the lowest dose tested by
9 the inhalation route. This study indicates that the developing fetus is a sensitive target following
10 inhalation exposure to benzo[a]pyrene and the observed decreased fetal survival/litter is the most
11 sensitive noncancer effect observed following inhalation exposure to benzo[a]pyrene. Additional
12 support for this endpoint of decreased fetal survival is provided by a developmental/reproductive
13 study conducted via the oral route [Mackenzie and Angevine. 1981].
14 This overall RfC is derived to be protective of all types of effects for a given duration of
15 exposure and is intended to protect the population as a whole, including potentially susceptible
16 subgroups [U.S. EPA, 2002]. This value should be applied in general population risk assessments.
17 However, decisions concerning averaging exposures over time for comparison with the RfC should
18 consider the types of toxicological effects and specific lifestages of concern. For example,
19 fluctuations in exposure levels that result in elevated exposures during these lifestages could
20 potentially lead to an appreciable risk, even if average levels over the full exposure duration were
21 less than or equal to the RfC. Alternatively, developmental toxicity may not be a concern due to
22 exposure scenarios in which exposure is occurring outside of the critical window of development
23 2.2.6. Confidence Statement
24 A confidence level of high, medium, or low is assigned to the study used to derive the RfC,
25 the overall database, and the RfC itself, as described in Section 4.3.9.2 of EPA's Methods for
26 Derivation of Inhalation Reference Concentrations and Application of Inhalation Dosimetry [U.S. EPA.
27 1994].
28 The overall confidence in the RfC is low-to-medium. Confidence in the principal study
29 [Archibong et al., 2002] is medium. The conduct and reporting of this developmental study were
30 adequate; however, a NOAEL was not identified. Confidence in the database is low due to the lack
31 of a multigeneration toxicity study, lack of studies on developmental neurotoxicity and immune
32 endpoints, and lack of information regarding subchronic and chronic inhalation exposure.
33 However, confidence in the RfC is bolstered by consistent systemic effects observed by the oral
34 route (including reproductive and developmental effects] and similar effects observed in human
35 populations exposed to PAH mixtures. Reflecting medium confidence in the principal study and low
36 confidence in the database, confidence in the RfC is low-to-medium.
This document is a draft for review purposes only and does not constitute Agency policy.
2-23 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 2.2.7. Previous IRIS Assessment: Reference Concentration
2 An RfC was not derived in the previous IRIS assessment
3 2.2.8. Uncertainties in the Derivation of the RfD and RfC
4 The following discussion identifies uncertainties associated with the RfD and RfC for
5 benzo[a]pyrene. To derive the RfD, the UF approach (U.S. EPA. 2000a. 19941 was applied to a POD
6 based on neurobehavioral changes in rats treated developmentally. To derive the RfC, this same
7 approach was applied to a POD from a developmental study for the effect of decreased fetal
8 survival. UFs were applied to the POD to account for extrapolating from an animal bioassay to
9 human exposure, the likely existence of a diverse population of varying susceptibilities, and
10 database deficiencies. These extrapolations are carried out with default approaches given the lack
11 of data to inform individual steps.
12 The database for benzo[a]pyrene contains limited human data. The observation of effects
13 associated with benzo[a]pyrene exposure in humans is complicated by several factors including the
14 existence of benzo[a]pyrene in the environment as one component of complex mixtures of PAHs,
15 exposure to benzo[a]pyrene by multiple routes of exposure within individual studies, and the
16 difficulty in obtaining accurate exposure information. Data on the effects of benzo[a]pyrene alone
17 are derived from a large database of studies in animal models. The database for oral
18 benzo[a]pyrene exposure includes two lifetime bioassays in rats and mice, two developmental
19 studies in mice, and several subchronic studies in rats.
20 Although the database is adequate for RfD derivation, there is uncertainty associated with
21 the database including that the principal study for the RfD exposed animals during a relatively short
22 period of brain development potentially underestimating the magnitude of resulting neurological
23 effects. Also, the database lacks a comprehensive multi-generation reproductive/developmental
24 toxicity studies and immune system endpoints were not evaluated in the available chronic-duration
25 or developmental studies. Additionally, the only available chronic studies of oral or inhalational
26 exposure to benzo[a]pyrene focused primarily on neoplastic effects leaving non-neoplastic effects
27 mostly uncharacterized.
28 The only chronic inhalation study of benzo[a]pyrene was designed as a lifetime
29 carcinogenicity study and did not examine noncancer endpoints [Thyssenetal., 1981]. In addition,
30 subchronic and short-term inhalation studies are available, which examine developmental and
31 reproductive endpoints in rats. Developmental studies by the inhalation route identified
32 biologically significant reductions in the number of pups/litter and percent fetal survival and
33 possible neurodevelopmental effects (e.g., diminished electrophysiological responses to stimuli in
34 the hippocampus) following gestational exposures. Additionally, a 60-day oral study in male rats
35 reported male reproductive effects (e.g., decreased testes weight and sperm production and
36 motility), but provides limited information to characterize dose-response relationships with
37 chronic exposure scenarios.
This document is a draft for review purposes only and does not constitute Agency policy.
2-24 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 The study selected as the basis of the RfC provided limited information regarding the
2 inhalation exposures of the animals. Specifically, it is not clear whether the reported
3 concentrations were target values or analytical concentrations and the method used to quantify
4 benzo[a]pyrene in the generated aerosols was not provided. Requests to obtain additional study
5 details from the authors were unsuccessful; therefore, the assumption was made that the reported
6 concentrations were analytical concentrations.
7 One area of uncertainty in the database pertains to the lack of information regarding
8 fertility in animals exposed gestationally to benzo[a]pyrene, especially in light of developmental
9 studies by the oral route indicating reduced fertility in the Fl generation and decreased
10 reproductive organ weights. The database also lacks a multigenerational reproductive study via the
11 inhalation route. Areas of uncertainty include the lack of chronic inhalation studies focusing on
12 noncancer effects, limited data on dose-response relationships for impaired male or female fertility
13 with gestational exposure or across several generations, and limited data on immune system
14 endpoints with chronic exposure to benzo[a]pyrene.
15 The toxicokinetic and toxicodynamic differences for benzo[a]pyrene between the animal
16 species in which the POD was derived and humans are unknown. PBPK models can be useful for
17 the evaluation of interspecies toxicokinetics; however, the benzo[a]pyrene database lacks an
18 adequate model that would inform potential differences. There is some evidence from the oral
19 toxicity data that mice may be more susceptible than rats to some benzo[a]pyrene effects (such as
20 ovo toxicity) [Borman et al.. 2000). although the underlying mechanistic basis of this apparent
21 difference is not understood. Most importantly, it is unknown which animal species may be more
22 comparable to humans.
23 2.3. ORAL SLOPE FACTOR FOR CANCER
24 The carcinogenicity assessment provides information on the carcinogenic hazard potential
25 of the substance in question and quantitative estimates of risk from oral and inhalation exposure
26 may be derived. Quantitative risk estimates may be derived from the application of a low-dose
27 extrapolation procedure. If derived, the oral slope factor is a plausible upper bound on the estimate
28 of risk per mg/kg-day of oral exposure.
29 2.3.1. Analysis of Carcinogenicity Data
30 The database for benzo[a]pyrene contains numerous cancer bioassays that identify tumors,
31 primarily of the alimentary tract including the forestomach, following oral exposure in rodents.
32 Three 2-year oral bioassays are available that associate lifetime benzo[a]pyrene exposure with
33 carcinogenicity at multiple sites: forestomach, liver, oral cavity, jejunum, kidney, auditory canal
34 (Zymbal gland) tumors, and skin or mammary gland tumors in male and female Wistar rats [Kroese
35 etal., 2001): forestomach tumors in male and female Sprague-Dawley rats [Brune etal., 1981): and
This document is a draft for review purposes only and does not constitute Agency policy.
2-25 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 forestomach, esophageal, tongue, and larynx tumors in female B6C3Fi mice [Beland and Gulp, 1998:
2 Gulp etal.. 19981
3 In addition to these 2-year cancer bioassays, there are studies available that provide
4 supporting evidence of carcinogenicity but are less suitable for dose-response analysis due to one
5 or more limitations in study design: [1] no vehicle control group; (2) only one benzo[a]pyrene dose
6 group; or (3) a one-time exposure to benzo[a]pyrene [Benjamin et al.. 1988: Robinson et al.. 1987:
7 El-Bayoumy. 1985: Wattenberg. 1974: Roe etal.. 1970: Biancifiorietal.. 1967: Chouroulinkovetal..
8 1967: Berenblum and Haran. 1955]. Of the controlled, multiple dose-group, repeat-dosing studies
9 that remain, most treated animals for <1 year, which is less optimal for extrapolating to a lifetime
10 exposure [Weyandetal., 1995: Triolo etal., 1977: Fedorenko and Yansheva, 1967: Neal and Rigdon,
11 19671
12 Brune etal. [1981] dosed rats (32/sex/group] with benzo[a]pyrene in the diet or by gavage
13 in a 1.5% caffeine solution, sometimes as infrequently as once every 9th day, for approximately
14 2 years and observed increased forestomach tumors. This study was not selected for quantitation
15 due to the nonstandard treatment protocol in comparison to the Good Laboratory Practice (GLP]
16 studies conducted by Kroese etal. [2001] and Beland and Gulp [1998] and the limited reporting of
17 study methods.
18 The Kroese etal. f20011 and Beland and Gulp T19981 studies were selected as the best
19 available studies for dose-response analysis and extrapolation to lifetime cancer risk following oral
20 exposure to benzo[a]pyrene. The rat bioassay by Kroese etal. [2001] and the mouse bioassay by
21 Beland and Gulp [1998] were conducted in accordance with GLP as established by the Organisation
22 for Economic Co-operation and Development [OECD]. These studies included histological
23 examinations for tumors in many different tissues, contained three exposure levels and controls,
24 contained adequate numbers of animals per dose group [~50/sex/group], treated animals for up to
25 2 years, and included detailed reporting of methods and results (including individual animal data].
26 Details of the rat [Kroese et al.. 2001] and female mouse [Beland and Gulp. 1998] study
27 designs are provided in Appendix D of the Supplemental Information. Dose-related increasing
28 trends in tumors were noted at the following sites:
29 • Squamous cell carcinomas [SCCs] or papillomas of the forestomach or oral cavity in male
30 and female rats;
31 • SCCs or papillomas of the forestomach, tongue, larynx, or esophagus in female mice;
32 • Auditory canal carcinomas in male and female rats;
33 • Kidney urothelial carcinomas in male rats;
34 • Jejunum/duodenum adenocarcinomas in female and male rats;
35 • Hepatocellular adenomas or carcinomas in male and female rats; and
This document is a draft for review purposes only and does not constitute Agency policy.
2-26 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 • SCCs or basal cell tumors of the skin or mammary gland in male rats.
2 These tumors were generally observed earlier during the study with increasing exposure
3 levels, and showed statistically significantly increasing trends in incidence with increasing
4 exposure level (Cochran-Armitage trend test, p < 0.001). These data are summarized in Appendix D
5 of the Supplemental Information. As recommended by the National Toxicology Program (NTP)
6 [McConnell etal., 1986] and as outlined in EPA's Guidelines for Carcinogen Risk Assessment [U.S.
7 EPA. 2005a). etiologically similar tumor types (i.e., benign and malignant tumors of the same cell
8 type) were combined for these tabulations when it was judged that the benign tumors could
9 progress to the malignant form. In addition, when one tumor type occurred across several
10 functionally related tissues, as with squamous cell tumors in the tongue, esophagus, larynx, and
11 forestomach, or adenocarcinomas of the jejunum or duodenum, these incidences were also
12 aggregated as counts of tumor-bearing animals.
13 In the rat study [Kroese et al., 2001], the oral cavity and auditory canal were examined
14 histologically only if a lesion or tumor was observed grossly at necropsy. Consequently, dose-
15 response analysis for these sites was not straightforward. Use of the number of tissues examined
16 histologically as the number at risk would tend to overestimate the incidence, because the
17 unexamined animals were much less likely to have a tumor. On the other hand, use of all animals in
18 a group as the number at risk would tend to underestimate if any of the unexamined animals had
19 tumors that could only be detected microscopically. The oral cavity squamous cell tumors were
20 combined with those in the forestomach because both are part of the alimentary tract, recognizing
21 that there was some potential for underestimating this cancer risk.
22 The auditory canal tumors from the rat study were not considered for dose-response
23 analysis, for several reasons. First, the control and lower dose groups were not thoroughly
24 examined, similar to the situation described above for oral cavity tumors. Unlike the oral cavity
25 tumors, the auditory canal tumors were not clearly related to any other site or tumor type, as they
26 were described as a mixture of squamous and sebaceous cells derived from pilosebaceous units.
27 The tumors were observed mainly in the high-dose groups and were highly coincident with the oral
28 cavity and forestomach tumors. Because the only mid-dose male with an auditory canal tumor did
29 not also have a forestomach or oral cavity squamous cell tumor, and no auditory canal tumors were
30 observed in low-dose male or female rats, the data are insufficient to conclude that the auditory
31 canal tumors occur independently of other tumors. The investigators did not suggest that these
32 tumors were metastases from other sites (in which case, the auditory canal tumors would be
33 repetitions of other tumors, or statistically dependent). Therefore dose-response analysis was not
34 pursued for this site, either separately or in combination with another tumor type.
35 The incidence data that were modeled are provided in Tables E-9, E-10, andE-11 (Kroese et
36 al.. 2001: Beland and Gulp. 19981
This document is a draft for review purposes only and does not constitute Agency policy.
2-27 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 2.3.2. Dose-Response Analysis—Adjustments and Extrapolation Methods
2 EPA's Guidelines for Carcinogen Risk Assessment [U.S. EPA. 2005a] recommend that the
3 method used to characterize and quantify cancer risk from a chemical is determined by what is
4 known about the mode of action of the carcinogen and the shape of the cancer dose-response curve.
5 The dose response is assumed to be linear in the low-dose range, when evidence supports a
6 mutagenic mode of action because of DNA reactivity, or if another mode of action that is anticipated
7 to be linear is applicable. In this assessment, EPA concluded that benzo[a]pyrene carcinogenicity
8 involves a mutagenic mode of action (as discussed in Section 1.1.5). Thus, a linear approach to low-
9 dose extrapolation was used.
10 The high-dose groups of both the rat and mouse studies were dead or moribund by week 79
11 for female mice, week 72 for female rats, and week 76 for male rats. Due to the occurrence of
12 multiple tumor types, earlier occurrence with increasing exposure and early termination of the
13 high-dose group in each study, methods that can reflect the influence of competing risks and
14 intercurrent mortality on site-specific tumor incidence rates are preferred. In this case, EPA has
15 used the multistage-Weibull model, which incorporates the time at which death-with-tumor
16 occurred as well as the dose.
17 Adjustments for approximating human equivalent slope factors applicable for continuous
18 exposure were applied prior to dose-response modeling. First, continuous daily exposure for the
19 gavage study in rats [Kroese etal.. 2001] was estimated by multiplying each administered dose by
20 (5 days)/(7 days) = 0.71, under the assumption of equal cumulative exposure yielding equivalent
21 outcomes. Dosing was continuous in the mouse diet study [Beland and Gulp. 1998). so no
22 continuous adjustment was necessary. Next, consistent with the EPA's Guidelines for Carcinogen
23 Risk Assessment [U.S. EPA, 2005a], an adjustment for cross-species scaling was applied to address
24 toxicological equivalence across species. Following EPA's cross-species scaling methodology, the
25 time-weighted daily average doses were converted to HEDs on the basis of (body weight)3/4 (U.S.
26 EPA. 1992}. This was accomplished by multiplying administered doses by (animal body weight
27 (kg)/70 kg)1/4 (U.S. EPA. 1992}. where the animal body weights were TWAs from each group, and
28 the U.S. EPA (1988) reference body weight for humans is 70 kg. It was not necessary to adjust the
29 administered doses for lifetime equivalent exposure prior to modeling for the groups terminated
30 early, because the multistage-Weibull model characterizes the tumor incidence as a function of
31 time, from which it provides an extrapolation to lifetime exposure.
32 Details of the modeling and the model selection process can be found in Appendix E of the
33 Supplemental Information. PODs for estimating low-dose risk were identified at doses at the lower
34 end of the observed data, generally corresponding to 10% extra risk.
35 2.3.3. Derivation of the Oral Slope Factor
36 The PODs estimated for each tumor site are summarized in Table 2-7. The lifetime oral
37 cancer slope factor for humans is defined as the slope of the line from the lower 95% bound on the
This document is a draft for review purposes only and does not constitute Agency policy.
2-28 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 exposure at the POD to the control response (slope factor = 0.1/BMDLi0). This slope, a 95% upper
2 confidence limit represents a plausible upper bound on the true risk. Using linear extrapolation
3 from the BMDLio, human equivalent oral slope factors were derived for each gender/tumor site
4 combination and are listed in Table 2-7.
5 Table 2-7. Summary of the oral slope factor derivations
Tumor
Forestomach, oral cavity:
squamous cell tumors
Kroese et al. (2001)
Hepatocellular adenomas or
carcinomas
Kroese et al. (2001)
Jejunum/duodenum
adenocarcinomas
Kroese et al. (2001)
Kidney: urothelial carcinomas
Kroese et al. (2001)
Skin, mammary:
Basal cell tumors
Squamous cell tumors
Kroese et al. (2001)
Forestomach, oral cavity:
squamous cell tumors
Kroese et al. (2001)
Hepatocellular adenomas or
carcinomas
Kroese et al. (2001)
Jejunum/duodenum
adenocarcinomas
Kroese et al. (2001)
Forestomach, esophagus, tongue,
larynx (alimentary tract):
squamous cell tumors
Beland and Gulp (1998)
Species/
sex
MaleWistar
rats
MaleWistar
rats
MaleWistar
rats
MaleWistar
rats
MaleWistar
rats
Female
Wistar rats
Female
Wistar rats
Female
Wistar rats
Female
B6C3Fi
Mice
Selected
model
Multistage
Weibull
Multistage
Weibull
Multistage
Weibull
Multistage
Weibull
Multistage
Weibull
Multistage
Weibull
Multistage
Weibull
Multistage
Weibull
Multistage
Weibull
BMR
10%
10%
10%
10%
10%
10%
10%
10%
10%
BMD
(mg/kg-d)
0.453
0.651
3.03
4.65
2.86
2.64
0.539
0.575
3.43
0.127
POD =
BMDL
(mg/kg-d)
0.281
0.449
2.38
2.50
2.35
1.77
0.328
0.507
1.95
0.071
Slope factor3
(mg/kg-d)-1
0.4
0.2
0.04
0.04
0.04
0.06
0.3
0.2
0.05
1
0.5b
0.3b
1
6
7
8
9
10
aHuman equivalent slope factor = 0.1/BMDL10HED; see Appendix E of the Supplemental Information for details of
modeling results.
bSlope factor characterizing the risk of incurring at least one of the tumor types listed.
This document is a draft for review purposes only and does not constitute Agency policy,
2-29 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Oral slope factors derived from rat bioassay data varied by gender and tumor site
2 (Table 2-7). Values ranged from 0.04 per mg/kg-day, based on kidney tumors in males, to 0.4 per
3 mg/kg-day, based on alimentary tract tumors in males. Slope factors based on liver tumors in male
4 and female rats (0.2 per mg/kg-day) were only slightly lower than slope factors based on
5 alimentary tract tumors (0.2-0.3 per mg/kg-day). The oral slope factor for alimentary tract tumors
6 in female mice was highest at 1 per mg/kg-day (Table 2-7), which was approximately twofold
7 higher than the oral slope factor derived from the alimentary tract tumors in male rats.
8 Although the time-to-tumor modeling helps to account for competing risks associated with
9 decreased survival times and other causes of death including other tumors, considering the tumor
10 sites individually still does not convey the total amount of risk potentially arising from the
11 sensitivity of multiple sites—that is, the risk of developing any combination of the increased tumor
12 types. A method, for estimating overall risk, involving the assumption that the variability in the
13 slope factors could be characterized by a normal distribution, is detailed in Appendix E of the
14 Supplemental Information. The resulting composite slope factor for all tumor types for male rats
15 was 0.5 per mg/kg-day, about 25% higher than the slope factor based on the most sensitive tumor
16 site, oral cavity and forestomach, while for female rats, the composite slope factor was equivalent to
17 that for the most sensitive site (Table 2-7; see Appendix E of Supplemental Information for
18 composite slope factor estimates).
19 The overall risk estimates from rats and mice spanned about a threefold range. As there are
20 no data to support any one result as most relevant for extrapolating to humans, the most sensitive
21 result was used to derive the oral slope factor. The recommended slope factor for assessing human
22 cancer risk associated with lifetime oral exposure to benzo[a]pyrene is 1 per mg/kg-day, based on
23 the alimentary tract tumor response in female B6C3Fi mice. Note that the oral slope factor should
24 only be used with lifetime human exposures <0.1 mg/kg-day, because above this level, the dose-
25 response relationship is not expected to be proportional to benzo[a]pyrene exposure.
26 2.3.4. Uncertainties in the Derivation of the Oral Slope Factor
27 The oral slope factor for benzo[a]pyrene was based on the increased incidence of
28 alimentary tract tumors, including forestomach tumors, observed in a lifetime dietary study in mice
29 (Beland and Gulp, 1998). EPA has considered the uncertainty associated with the relevance of
30 forestomach tumors for estimating human risk from benzo[a]pyrene exposure. The rodent
31 forestomach serves to store foods and liquids for several hours before contents continue to the
32 stomach for further digestion (Clayson et al.. 1990: Grice et al.. 1986). Thus, tissue of the
33 forestomach in rodents may be exposed to benzo[a]pyrene for longer durations than analogous
34 human tissues in the oral cavity and esophagus. This suggests that the rodent forestomach may be
35 quantitatively more sensitive to the development of squamous epithelial tumors in the forestomach
36 compared to oral or esophageal tumors in humans.
37 Uncertainty in the magnitude of the recommended oral slope factor is reflected to some
38 extent in the range of slope factors among tumors sites and species; the oral slope factor based on
This document is a draft for review purposes only and does not constitute Agency policy.
2-30 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 the mouse alimentary tract data was about threefold higher than the overall oral slope factor based
2 on male rat data (Table 2-8). These comparisons show that the selection of target organ, animal
3 species, and interspecies extrapolation can impact the oral cancer risk estimate. However, all of the
4 activation pathways implicated in benzo[a]pyrene carcinogenicity have been observed in human
5 tissues, and associations have been made between the spectra of mutations in tumor tissues from
6 benzo[a]pyrene-exposed animals and humans exposed to complex PAH mixtures containing
7 benzo[a]pyrene (see Section 1.1.5).
8
9
Table 2-8. Summary of uncertainties in the derivation of cancer risk values
for benzo[a]pyrene oral slope factor
Consideration and
impact on cancer risk value
Decision
Justification and discussion
Selection of target organ
•^ oral slope factor, up to fivefold, if
alimentary tract tumors not selected
Alimentary tract tumors
(forestomach,
esophagus, tongue,
larynx)
Tumor site is concordant across rats and mice,
increasing support for its relevance to humans.
As there are no data to support any one result as
most relevant for extrapolating to humans, the
most sensitive result for alimentary tract tumors
was used to derive the oral slope factor.
Selection of data set
•^ oral slope factor ~threefold if rat
bioassay were selected for oral slope
factor derivation
Beland and Gulp (1998)
Beland and Gulp (1998) was a well-conducted
study and had the lowest HEDs of the available
cancer bioassays, reducing low-dose
extrapolation uncertainty.
Selection of dose metric
Alternatives could 4, or /T" slope
factor
Administered dose
Experimental evidence supports a role for
metabolism in toxicity, but actual responsible
metabolites have not been identified.
Interspecies extrapolation
Alternatives could 4, or 1" slope
factor (e.g., 3.5-fold 4, [scaling by
body weight] or 1" 2-fold [scaling by
BW2/3])
BW3/4 scaling (default
approach)
There are no data to support alternatives.
Because the dose metric was not an area under
the curve, BW scaling was used to calculate
equivalent cumulative exposures for estimating
equivalent human risks. While the true human
correspondence is unknown, this overall
approach is expected to neither over- nor
underestimate human equivalent risks.
Dose-response modeling
Alternatives could 4, or /T" slope
factor
Multistage-Weibull
model
No biologically based models for benzo[a]pyrene
were available. Because the multistage-Weibull
model could address additional available data
(time of death with tumor, and whether a tumor
caused the death of the animal), this model was
superior to other available models.
Low-dose extrapolation
•i, cancer risk estimate would be
expected with the application of
nonlinear low-dose extrapolation
Linear extrapolation
from POD (based on
mutagenic mode of
action)
Available mode-of-action data support linearity
(mutagenicity is a primary mode of action of
benzo[a]pyrene).
This document is a draft for review purposes only and does not constitute Agency policy,
2-31 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Consideration and
impact on cancer risk value
Statistical uncertainty at POD
4> oral slope factor 1.8-fold if BMD
used as the POD rather than BMDL
Sensitive subpopulations
1" oral slope factor to unknown
extent
Decision
BMDL (preferred
approach for calculating
plausible upper bound
slope factor)
ADAFs are
recommended for early
life exposures
Justification and discussion
Limited size of bioassay results in sampling
variability; lower bound is 95% Cl on administered
exposure at 10% extra risk of alimentary tract
tumors.
No chemical-specific data are available to
determine the range of human toxicodynamic
variability or sensitivity.
1 2.3.5. Previous IRIS Assessment: Oral Slope Factor
2 The previous cancer assessment for benzo[a]pyrene was posted on the IRIS database in
3 1992. At that time, benzo[a]pyrene was classified as a probable human carcinogen (Group B2)
4 based on inadequate data in humans and sufficient data in animals via several routes of exposure.
5 An oral slope factor was derived from the geometric mean of four slope factor estimates based on
6 studies of dietary benzo[a]pyrene administered in the diet for approximately 2 years in 10-week-
7 old Sprague-Dawley rats [Brune etal., 1981] and administered for up 7 months in 2-week-old to
8 5-month-old CFW-Swiss mice [Neal and Rigdon. 1967]. A single slope factor estimate of 11.7 per
9 mg/kg-day, using a linearized multistage procedure applied to the combined incidence of
10 forestomach, esophageal, and laryngeal tumors, was derived from the Brune etal. [1981] study (see
11 Section 1.1.5 for study details]. Three modeling procedures were used to derive risk estimates
12 from the Neal and Rigdon (1967] bioassay (see Section 1.1.5]. U.S. EPA (1991a] fit a two-stage
13 response model, based on exposure-dependent changes in both transition rates and growth rates of
14 preneoplastic cells, to derive a value of 5.9 per mg/kg-day. U.S. EPA(1991b] derived a value of
15 9.0 per mg/kg-day by linear extrapolation from the 10% response point to the background
16 response in a re-analysis of the 1990 model. Finally, using a Weibull-type model to reflect less-
17 than-lifetime exposure to benzo[a]pyrene, the same assessment (U.S. EPA. 1991b] derived an
18 upper-bound slope factor estimate of 4.5 per mg/kg-day. The four slope factor estimates, which
19 reflected extrapolation to humans assuming surface area equivalence (BW2/3 scaling] were within
20 threefold of each other and were judged to be of equal merit Consequently, the geometric mean of
21 these four estimates, 7.3 per mg/kg-day, was recommended, in 1992, as the oral slope factor.
22 2.4. INHALATION UNIT RISK FOR CANCER
23 The carcinogenicity assessment provides information on the carcinogenic hazard potential
24 of the substance in question and quantitative estimates of risk from oral and inhalation exposure
25 may be derived. Quantitative risk estimates may be derived from the application of a low-dose
26 extrapolation procedure. If derived, the inhalation unit risk is a plausible upper bound on the
27 estimate of risk per |J.g/m3 air breathed.
This document is a draft for review purposes only and does not constitute Agency policy,
2-32 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 2.4.1. Analysis of Carcinogenicity Data
2 The inhalation database demonstrating carcinogenicity of benzo[a]pyrene consists of a
3 lifetime inhalation bioassay in male hamsters [Thyssenetal.. 1981] and intratracheal instillation
4 studies, also in hamsters [Feron and Kruysse. 1978: Ketkar etal.. 1978: Feronetal.. 1973: Henry et
5 al.. 1973: Saffiotti etal.. 1972). The intratracheal instillation studies provide supporting evidence of
6 carcinogenicity of inhaled benzo[a]pyrene; however, the use of this exposure method alters the
7 deposition, clearance, and retention of substances, and therefore, studies utilizing this exposure
8 technique are not as useful for the quantitative extrapolation of cancer risk from the inhalation of
9 benzo[a]pyrene in the environment [Driscolletal., 2000].
10 The bioassay by Thyssenetal. [1981] represents the only lifetime inhalation cancer
11 bioassay available for describing exposure-response relationships for cancer from inhaled
12 benzo[a]pyrene. As summarized in Section 1.1.5, increased incidences of benign and malignant
13 tumors of the pharynx, larynx, trachea, esophagus, nasal cavity, or forestomach were seen with
14 increasing exposure concentration. In addition, survival was decreased relative to control in the
15 high-exposure group; mean survival times in the control, low-, and mid-concentration groups were
16 96.4, 95.2, and 96.4 weeks, respectively, compared to 59.5 weeks in the high-exposure group
17 animals [Thyssenetal.. 1981]. Overall, tumors occurred earlier in the highest benzo[a]pyrene
18 exposure group than in the mid-exposure group.
19 Strengths of the study included exposures until natural death, up to 2.5 years, multiple
20 exposure groups; histological examination of multiple organ systems, and availability of individual
21 animal pathology reports with time of death and tumor incidence data by site in the upper
22 respiratory and digestive tracts. In addition, the availability of weekly chamber air monitoring data
23 and individual times on study allowed the calculation of time-weighted average (TWA] lifetime
24 continuous exposures for each hamster. Group averages of these TWA concentrations were 0, 0.25,
25 1.01, and 4.29 mg/m3 fU.S. EPA. 19901
26 Several limitations concerning exposure conditions in the Thyssenetal. [1981] study were
27 evaluated for their impact on the derivation of an inhalation unit risk for benzo[a]pyrene. These
28 issues include minimal detail about the particle size distribution of the administered aerosols,
29 variability of chamber concentrations, and the use of a sodium chloride aerosol as a carrier.
30 First, particle distribution analysis of aerosols, in particular the MMAD and geometric SD,
31 was not reported, although the investigators did report that particles were within the respirable
32 range for hamsters, with >99% of the particles having diameters 0.2-0.5 um and >80% having
33 diameters 0.2-0.3 um.
34 Second, weekly averages of chamber concentration measurements varied two- to fivefold
35 from the overall average for each group, which exceeds the limit for exposure variability of <20%
36 for aerosols recommended by OECD [2009]. For risk assessment purposes, EPA generally assumes
37 that cancer risk is proportional to cumulative exposure, and therefore to lifetime average exposure
38 as estimated here, when there is no information to the contrary. Under this assumption, the
This document is a draft for review purposes only and does not constitute Agency policy.
2-33 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 variability of the chamber concentrations has little impact on the estimated exposure-response
2 relationship. The impacts of alternative assumptions are considered in Section 2.4.4.
3 Lastly, exposure occurred through the inhalation of benzo[a]pyrene adsorbed onto sodium
4 chloride aerosols, which might have irritant carrier effects, and may have a different deposition
5 than benzo[a]pyrene adsorbed onto carbonaceous particles (as is more typical in the environment).
6 The above study design and reporting issues concerning the particle size composition, exposure
7 variability, and deposition do not negate the robust tumor response following benzo[a]pyrene
8 inhalation exposure. Consequently, EPA concluded that the strengths of the study supported the
9 use of the data to derive an inhalation unit risk for benzo[a]pyrene. See Section 2.4.4 for a
10 discussion of uncertainties in the unit risk.
11 2.4.2. Dose-Response Analysis—Adjustments and Extrapolation Methods
12 Biologically based dose-response models for benzo[a]pyrene are not available. A simplified
13 version of the two-stage carcinogenesis model proposed by Moolgavkar and Venzon [1979] and
14 Moolgavkar and Knudson [1981] has been applied to the Thyssenetal. [1981] individual animal
15 data [U.S. EPA, 1990]. However, the simplifications necessary to fit the tumor incidence data
16 reduced that model to an empirical model (i.e., there were no biological data to inform estimates of
17 cell proliferation rates for background or initiated cells]. Sufficient data were available to apply the
18 multistage-Weibull model, as used for the oral slope factor (described in detail in Appendix E of the
19 Supplemental Information], specifically the individual times of death for each animal. Unlike in the
20 oral bioassays, Thyssenetal. (1981] did not determine cause of death for any of the animals. Since
21 the investigators for the oral bioassays considered some of the same tumor types to be fatal at least
22 some of the time, bounding estimates of the POD for these Thyssenetal. (1981] data were
23 developed by treating the tumors alternately as either all incidental to the death of an affected
24 animal or as causing the death of an affected animal.
25 The tumor incidence data used for dose-response modeling comprised the benign and
26 malignant tumors in the pharynx and respiratory tract (see Table E-17]. The tumors in these sites
27 were judged to be sufficiently similar to combine in overall incidences, based on the assumption
28 that the benign tumors could develop into malignancies, as outlined in EPA's Guidelines for
29 Carcinogen Risk Assessment (Section 2.2.2.1.2: U.S. EPA, 2005a]. Specifically, while the pharynx and
30 larynx are associated with the upper digestive tract and the upper respiratory tract, respectively,
31 these sites are close anatomically and in some cases where both tissues were affected, the site of
32 origin could not be distinguished (U.S. EPA. 1990]. In addition, the benign tumors (e.g., papillomas,
33 polyps, and papillary polyps] were considered early stages of the SCCs in these tissues (U.S. EPA.
34 1990]. Consequently, the overall incidence of SCCs or benign tumors judged to originate from the
35 same cell type (papillomas, polyps, or papillary polyps] were selected for dose-response modeling.
36 A toxicokinetic model to assist in cross-species scaling of benzo[a]pyrene inhalation
37 exposure was not available. EPA's RfC default dosimetry adjustments (U.S. EPA. 1994] were
38 utilized in the benzo[a]pyrene RfC calculation (see Section 2.2.2] but could not be applied to the
This document is a draft for review purposes only and does not constitute Agency policy.
2-34 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 aerosols generated for the inhalation bioassay by Thyssen et al. [1981] as the approaches
2 presented in the RfC methodology guidelines [U.S. EPA. 1994] were developed for insoluble and
3 nonhygroscopic particles, not the sodium chloride particle used in Thyssen etal. [1981].
4 Consequently, without data to inform a basis for extrapolation to humans, it was assumed that
5 equal risk for all species would be associated with equal concentrations in air, at least at anticipated
6 environmental concentrations. This is equivalent to assuming that any metabolism of
7 benzo[a]pyrene is directly proportional to breathing rate and that the deposition rate is equal
8 between species.
9 The multistage-Weibull model was fit to the TWA exposure concentrations and the
10 individual animal tumor and survival data for tumors in the larynx, pharynx, trachea, or nasal cavity
11 (tumors of the pharynx and upper respiratory tract], using the software program, MultiStage-
12 Weibull [U.S. EPA. 2010c]. Modeling results are provided in Appendix E of the Supplemental
13 Information. Because benzo[a]pyrene carcinogenicity involves a mutagenic mode of action, linear
14 low-exposure extrapolation from the BMCLio was used to derive the inhalation unit risk [U.S. EPA.
15 2005a].
16 2.4.3. Inhalation Unit Risk Derivation
17 The results from modeling the inhalation carcinogenicity data from Thyssen etal. [1981]
18 are summarized in Table 2-9. Taking the tumors to have been the cause of death of the
19 experimental animals with tumors, the BMCio and BMCLio values were 0.468 and 0.256 mg/m3,
20 respectively. Then, taking all of the tumors to have been incidental to the cause of death for each
21 animal with a tumor, the BMCio and BMCLio values were 0.254 and 0.163 mg/m3, respectively,
22 about twofold lower than the first case. Because the tumors were unlikely to have all been fatal, the
23 lower BMCLio from the incidental deaths analysis, 0.163 mg/m3, is recommended for the
24 calculation of the inhalation unit risk. Using linear extrapolation from the BMCLio (0.163 mg/m3],
25 an inhalation unit risk of 0.6 per mg/m3, or 6 x 1Q-* per ng/m3 (rounding to one significant digit],
26 was calculated. Note that the inhalation unit risk should only be used with lifetime human
27 exposures <0.3 mg/m3, the human equivalent POD, because above this level, the dose-response
28 relationship is not expected to be proportional to benzo[a]pyrene exposure.
This document is a draft for review purposes only and does not constitute Agency policy.
2-35 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Table 2-9. Summary of the inhalation unit risk derivation
Tumor site and context
Upper respiratory tract and pharynx;
all treated as cause of death
Thyssenetal. (1981)
Upper respiratory tract and pharynx;
all treated as incidental to death
Thyssenetal. (1981)
Species/
sex
Male
hamsters
Male
hamsters
Selected
model
Multistage
Weibull, 2°
Multistage
Weibull, 2°
BMR
10%
10%
BMC
(mg/m3)
0.468
0.254
POD =
BMCL
(mg/m3)
0.256
0.163
Unit risk3
(mg/m3)-1
0.4
0.6
2
3 aHuman equivalent unit risk = 0.10/BMCL10; see Appendix E for details of modeling results.
4 2.4.4. Uncertainties in the Derivation of the Inhalation Unit Risk
5 Table 2-10 summarizes uncertainties in the derivation of the inhalation unit risk for
6 benzo[a]pyrene; further detail is provided in the following discussion. Only one animal cancer
7 bioassay, in one sex, by the inhalation route is available that describes the exposure-response
8 relationship for respiratory tract tumors with lifetime inhalation exposure to benzo [a]pyrene
9 [Thyssenetal.. 1981). Although corroborative information on exposure-response relationships in
10 other animal species is lacking, the findings for upper respiratory tract tumors are consistent with
11 findings in other hamster studies with intratracheal administration of benzo[a]pyrene (upper and
12 lower respiratory tract tumors), and with some of the portal-of-entry effects in oral exposure
13 studies.
14 The hamster inhalation bioassay by Thyssen et al. [1981] observed upper respiratory tract
15 tumors, but not lung tumors. The lack of a lung tumor response in hamsters, given the strong
16 association of inhaled PAH mixtures with lung cancer in humans across many studies (see
17 Section 1.1.5) suggests that this study may not be ideal for extrapolating to humans. Hamsters have
18 an apparent lower sensitivity to lung carcinogenesis than rats and mice and a tendency to give false
19 negatives for particles classified as carcinogenic to humans by IARC (Mauderly. 1997). However,
20 hamster laryngeal tumors have been used as an indication of the carcinogenic hazard of cigarette
21 smoke for more than 50 years (IARC, 2002). For example, a large study investigating the inhalation
22 of cigarette smoke in hamsters (n = 4,400) indicated that the larynx was the most responsive tumor
23 site, which the authors indicated was due to a large difference in particle deposition between the
24 larynx and the lung (Dontenwill etal.. 1973). EPA's Guidelines for Carcinogen Assessment (U.S. EPA.
25 2005a] stress that site concordance between animals and humans need not always be assumed.
26 Therefore, the robust tumor response in the upper respiratory tract of Syrian golden hamsters was
27 considered to be supportive of the use of the Thyssenetal. (1981) study for the derivation of an
28 inhalation unit risk.
29 An additional uncertainty includes the inability to apply U.S. EPA (1994) dosimetry
30 approaches to extrapolate inhaled concentrations from animals to humans, due to the use of a
This document is a draft for review purposes only and does not constitute Agency policy.
2-36 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 soluble hygroscopic carrier particle (sodium chloride) for the delivery of benzo[a]pyrene. One
2 likely consequence of the use of hygroscopic carrier particles would be the growth of
3 benzo[a]pyrene-sodium chloride particles in the humid environment of the respiratory tract
4 resulting in increased particle diameter and resulting changes in particle deposition, specifically,
5 increased impaction in the upper respiratory tract and less deposition in the lung [Varghese and
6 Gangamma. 2009: Asgharian. 2004: Perron. 1994: XuandYu. 1985). In addition, sodium chloride
7 can be irritating to the respiratory tract, depending on concentration. The Thyssenetal. [1981]
8 study reported that vehicle controls were exposed to 240 |ig/m3, and it is unclear whether exposure
9 to this concentration of sodium chloride could have potentiated the tumor response seen in the
10 mid- and high-concentration benzo[a]pyrene groups. Exposure to benzo[a]pyrene in the
11 environment predominantly occurs via non-soluble, non-hygroscopic, carbonaceous particles (such
12 as soot and diesel exhaust particles). The potential impact of differences in carrier particle on the
13 magnitude of the inhalation unit risk is unknown.
14 Regarding uncertainty associated with exposure characterization, the individual exposure
15 chamber measurements varied from about an order of magnitude less than the target concentration
16 to about twofold higher than the target concentration. Weekly average analytical concentrations
17 were documented to vary by two- to fivefold in all exposed groups, with no particular trends over
18 time. Continuous time-weighted group average concentrations were used for dose-response
19 modeling under the assumption that equal cumulative exposures are expected to lead to similar
20 outcomes. This assumption is generally expected to lead to an unbiased estimate of risk when there
21 is incomplete information. However, it is possible that peak exposure above some concentration
22 may be more associated with the observed effects, or that deposition of particles may have reached
23 a maximum level or plateau, such as in the high-exposure group. Regarding the role of peak
24 exposures, the higher exposures for each group were distributed evenly throughout the study for
25 the most part, suggesting that any association of risk with peak exposures would also be
26 proportional to cumulative exposure. If particle deposition reached a plateau with the high-
27 exposure group, there is relatively less impact on the unit risk because the derivation relies on the
28 dose-response at lower exposure. But the actual dynamics of particle deposition at these or other
29 exposure levels are not well understood. There is not enough information available to estimate a
30 more quantitative impact on the estimated unit risk due to these uncertainties.
This document is a draft for review purposes only and does not constitute Agency policy.
2-37 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2
Table 2-10. Summary of uncertainties in the derivation of cancer risk values
for benzo[a]pyrene (inhalation unit risk)
Consideration and
impact on cancer risk value
Decision
Justification and discussion
Selection of data set and target organ
No inhalation unit risk if Thyssen et al.
(1981) not used
Respiratory tract tumors
from Thyssen et al.
(1981)
The Thyssen et al. (1981) bioassay is the only
lifetime inhalation cancer bioassay available for
describing exposure-response relationships for
cancer from inhaled benzo[a]pyrene.
Intratracheal installation studies support the
association of benzo[a]pyrene exposure with
respiratory tract tumors.
Selection of dose metric
Alternatives could 4, or
unit risk
Administered exposure
as TWA
Experimental evidence supports a role for
metabolism in toxicity, but actual responsible
metabolites are not identified. The
recommended unit risk is a reasonable estimate
if the proportion of the carcinogenic moiety
remains the same at lower exposures.
Interspecies extrapolation
Alternatives could 4, or ^ slope
factor
Cross-species scaling
was not applied. The
carrier particle used was
soluble and hygroscopic,
therefore the RfC
methodology (U.S. EPA,
1994) dosimetric
adjustments could not
be applied.
There are no data to support alternatives. Equal
risk per u.g/m3 is assumed.
Dose-response modeling
Alternatives could 4, or
factor
slope
Multistage-Weibull
model
No biologically based models for benzo[a]pyrene
were available. Because the multistage-Weibull
model could address additional available data
(time of death with tumor), this model was
superior to other available empirical models.
Low-dose extrapolation
•^ cancer risk estimate would be
expected with the application of
nonlinear low-dose extrapolation
Linear extrapolation
from the POD (based on
mutagenic mode of
action)
Available mode-of-action data support linearity
(mutagenicity is a primary mode of action of
benzo[a]pyrene).
Statistical uncertainty at POD
•^ inhalation unit risk 1.4-fold if BMC
used as the POD rather than BMCL
BMCL (preferred
approach for calculating
plausible upper bound
unit risk)
Limited size of bioassay results in sampling
variability; lower bound is 95% confidence
interval (Cl) on administered exposure at 10%
extra risk of respiratory tract tumors.
Sensitive subpopulations
"T* inhalation unit risk to unknown
extent
ADAFs are
recommended for early
life exposures
No chemical-specific data are available to
determine the range of human toxicodynamic
variability or sensitivity.
3
4
2.4.5. Previous IRIS Assessment: Inhalation Unit Risk
An inhalation unit risk for benzo[a]pyrene was not previously available on IRIS.
This document is a draft for review purposes only and does not constitute Agency policy,
2-38 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 2.5. DERMAL SLOPE FACTOR FOR CANCER
2 Human and animal studies of exposure to PAH mixtures or benzo[a]pyrene alone
3 demonstrate an increased incidence of skin tumors with increasing dermal exposure. This
4 assessment for benzo[a]pyrene derives a dermal slope factor, a quantitative risk estimate that is a
5 plausible upper bound on the estimate of risk per [J.g/day of lifetime dermal exposure. This
6 derivation provides the first dermal slope factor for the Integrated Risk Information System (IRIS)
7 database.
8 2.5.1. Analysis of Carcinogenicity Data
9 Skin cancer in humans has been documented to result from occupational exposure to
10 complex mixtures of PAHs including benzo[a]pyrene, such as coal tar pitches, non-refined mineral
11 oils, shale oils, and soot flARC. 2010: BaanetaL 2009: IPCS. 1998: Boffetta etal.. 1997: AT SDR.
12 1995]. Although studies of human exposure to benzo[a]pyrene alone are not available, repeated
13 application of benzo[a]pyrene to skin (in the absence of exogenous promoters) has been
14 demonstrated to induce skin tumors in guinea pigs, rabbits, rats, and mice. Given the availability of
15 lifetime bioassays of dermal benzo[a]pyrene exposure in mice, this analysis focuses on
16 carcinogenicity bioassays including repeated dermal exposure to benzo[a]pyrene for approximately
17 2 years. These studies involved 2- or 3-times/week exposure protocols, at least two exposure
18 levels plus controls, and histopathological examinations of the skin and other tissues (Sivaketal..
19 1997: Grimmer etal.. 1984: Habsetal.. 1984: Grimmer etal.. 1983: Habs etal.. 1980: Schmahl etal..
20 1977: Schmidt etal.. 1973: Roe etal.. 1970: Poel. 1963.1959]. These studies, in several strains of
21 mice, demonstrated primarily malignant skin tumors, as well as earlier occurrence of tumors with
22 increasing exposure levels (see Tables D-15 to D-23 in the Supplemental Information for study
23 details).
24 Other carcinogenicity studies in mice were judged to support the studies listed above, but
25 were not considered in the dose-response analysis because of the availability of lifetime studies
26 directly relevant to deriving a dermal slope factor. These other studies included: (1) early "skin
27 painting" studies of benzo[a]pyrene carcinogenicity in mouse skin that did not report sufficient
28 information to estimate the doses applied (e.g., Wynder and Hoffmann, 1959: Wynder etal., 1957]:
29 (2) bioassays with minimal dose-response information, either using just one benzo[a]pyrene dose
30 level or with only dose levels inducing 90-100% incidence of mice with tumors, which provide
31 relatively little information about the shape of the dose-response relationship especially for low
32 exposures (e.g.. Wilson and Holland. 1988): and (3) shorter studies (i.e., <1 year) (Higginbotham et
33 al.. 1993: Albert etal.. 1991: Nesnow etal.. 1983: Emmettetal.. 1981: Levin etal.. 1977). which
34 would tend to underestimate lifetime risk by overlooking the potential for the development of
35 tumors in later life.
36 The National Institute for Occupational Safety and Health (NIOSH) study fSivak etal.. 1997:
37 NIOSH. 1989] is a well-conducted and documented study and was determined to be the best
This document is a draft for review purposes only and does not constitute Agency policy.
2-39 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 available study for dose-response analysis. Specifically, mice were randomly assigned to treatment
2 groups while maintaining comparable distributions of body weights, housed singly (minimizing
3 grooming or other interference with application sites), and observed weekly for tumor status.
4 Histopathology was evaluated by two pathologists, both without knowledge of treatment group,
5 who reached a consensus diagnosis in each case. In addition, the availability of time-of-tumor
6 appearance and time of death for all animals provided a clearer characterization of the number at
7 risk of development of tumors and the extent of exposure associated with tumor development,
8 supporting more accurate estimation of lifetime cancer risk.
9 The other lifetime studies provide support for the NIOSH study. Overall, study designs
10 varied in terms of number of exposure levels used (two to nine, compared with three in typical NTP
11 bioassays) and in number of mice per group (from ~17 to 100 mice/dose group, compared with
12 50 used in most NTP bioassays). While the largest studies would be expected to have greater
13 ability to detect low responses at low doses (e.g.. Schmidt etal.. 1973). studies conducted at similar
14 doses, but smaller group sizes, showed significant dose-response trends (e.g.. Poel. 1959). For all of
15 the supporting studies, individual animal data were not available).
16 Some aspects of study design and conduct of the supporting studies were not consistently
17 reported. These study attributes included composition and purity of the test article, randomization
18 of animals to treatment groups, frequency of tumor evaluations, exact length of treatment period,
19 and blinding of treatment group from pathologists. As each study was affected by a similar number
20 of omissions, possibly reflecting reporting practices at the time they were conducted, all but one
21 study (Poel. 1963) were included in dose-response modeling for comparison (see Appendix E.2.4 of
22 the Supplemental Information).
23 Poel (1963) exposed three strains of mice until natural death. The study authors did not
24 report length of time that the control mice were on study, nor the duration of exposure for mice
25 that did not develop tumors following treatment Overall, while these three datasets support a
26 finding of dermal carcinogenicity, they did not provide sufficient information to estimate the extent
27 of exposure associated with the observed tumor incidence and were not used for dose-response
28 modeling.
29 2.5.2. Dose-Response Analysis—Adjustments and Extrapolation Methods
30 Biologically based dose-response models for dermal exposure to benzo[a]pyrene are not
31 available. As with the oral and inhalation benzo[a]pyrene carcinogenicity data, benzo[a]pyrene
32 dermal exposure carcinogenicity data were generally characterized by earlier occurrence of tumors
33 and increased mortality with increasing exposure level. Individual data were available to apply a
34 time-to-tumor model (the multistage-Weibull model, described in detail in Appendix E.2.1 of the
35 Supplemental Information) to the NIOSH study data, while each of the other lifetime dermal data
36 sets (Sivak etal.. 1997: Grimmer etal.. 1984: Habs etal.. 1984: Grimmer etal.. 1983: Habsetal..
37 1980: Schmahl etal.. 1977: Schmidt etal.. 1973: Roe etal.. 1970: Poel. 1963.19591 was modeled
38 using the multistage model. The following discussion describes the inputs for each model.
This document is a draft for review purposes only and does not constitute Agency policy.
2-40 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 For both dose-response models, following EPA's Guidelines for Carcinogen Risk Assessment
2 [Section 2.2.2.1.2: U.S. EPA. 2005a]. incidence of malignant or benign skin tumors was preferred for
3 dose-response modeling, based on evidence that skin papillomas can develop into malignant skin
4 tumors. For some studies, it was not clear whether related benign tumors were included in the
5 total tumors reported [e.g.. Habs et al.. 1980). In the NIOSH study, one tumor in the mid-dose group
6 that had been reported as a papilloma by Sivaketal. [1997] was listed as a keratoacanthoma in the
7 full report [NIOSH. 1989]. Since it is not clear whether keratoacanthomas can develop into
8 malignant skin tumors, two analyses were run for the NIOSH data set, one including the
9 keratoacanthoma and one omitting it altogether (see Table E-23 of the Supplemental Information].
10 For the other studies, the most inclusive tabulation available of mice with benign or malignant skin
11 tumors, without double-counting, was used.
12 Concerning the appropriate dose measure, we have used continuous daily exposure (see
13 Preamble, Section 7.2] unless there are data indicating that dose-rate effects are relevant. The
14 exposure protocols for the selected bioassays varied between two or three applications per week.
15 Although environmental dermal exposure may more likely occur intermittently than oral or
16 inhalation exposures, due to interruption of exposure through bathing or washing of affected areas,
17 the dermal slope factor was derived for use with estimates of constant daily lifetime exposure.
18 Therefore, all administered doses were converted to TWA daily doses using the equation:
19
20 Average daily dose/day = (ug/application] x (number of applications/week 4- 7 days/week]
21
22 A particular consideration for time-to-tumor modeling (affecting only the NIOSH study] is
23 the time of tumor observation. The NIOSH data set provided the time of first appearance of skin
24 tumors for each animal. Skin tumors were classified as incidental for the purposes of modeling
25 because death generally occurred weeks later.
26 For the lifetime studies without individual animal data, some refinements of the dose-
27 response information were made where possible before applying the multistage model, in order to
28 approximate what would be accomplished had a time-to-tumor model been viable. These
29 refinements involved adjustments for incidence (i.e., numbers of animals at risk were reduced if
30 mortality occurred before tumors first appeared] and dose (i.e., in the event of shortened exposure
31 duration due to 100% mortality in a dose group, an equivalent, lower lifetime exposure was
32 estimated]. These adjustments are described by study in Appendix E.2.3 of the Supplemental
33 Information.
34 The multistage-Weibull and multistage-cancer models were then fit to the respective data
35 sets and BMDi0s/BMDLi0s estimated (details in Appendix E.2.3 of the Supplemental Information].
36 Because benzo[a]pyrene is expected to cause cancer via a mutagenic mode of action, a linear
37 approach to low dose extrapolation from the PODs (i.e., BMDLio] was used (U.S. EPA, 2005a] to
38 derive the candidate dermal slope factors.
This document is a draft for review purposes only and does not constitute Agency policy.
2-41 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 A toxicokinetic model to assist in cross-species scaling of benzo[a]pyrene dermal exposure
2 was not available, nor was there any guidance to inform a basis for extrapolation of dermal dose-
3 response in mice to humans. Several alternative approaches to interspecies extrapolation were
4 developed that were linear, and that could be applied with equivalent results either before or after
5 dose-response modeling. Consequently, these adjustments were evaluated separately from the
6 dose-response modeling and applied after derivation of a dermal slope factor for mice (see
7 Section 2.5.4).
8 2.5.3. Derivation of the Dermal Slope Factor
9 The results from modeling the dermal carcinogenicity data separated by sex are
10 summarized in Table 2-11 (see Tables E-23 and E-24 in the Supplemental Information for more
11 details). Adequate fits to the NIOSH data were obtained using time-to-tumor modeling, whether or
12 not the keratoacanthoma in the mid-dose group was included. Excluding the keratoacanthoma, in
13 view of its seemingly random occurrence and the uncertainty whether such tumors would progress
14 to carcinomas, linear extrapolation from the BMDLio of 0.060 [J.g/day from the NIOSH study led to a
15 candidate dermal slope factor for mice of 1.7 per [ig/day.
16 Dermal slope factors calculated from the supporting studies (Sivaketal., 1997: Grimmer et
17 al.. 1984: Habsetal.. 1984: Grimmer etal.. 1983: Habs etal.. 1980: Schmahl etal.. 1977: Schmidt et
18 al.. 1973: Roe etal.. 1970: Poel. 1963.1959) using the multistage model and linear extrapolation
19 from the BMDLio values ranged from 0.25 to 1.8 per [ig/day, a roughly sevenfold range
20 (Table 2-11). Values ranged from 0.9 to 1.7 per [ig/day for male mice, and from 0.25 to 0.67 per
21 [ig/day for female mice. These results suggest that some female mouse strains may be as sensitive
22 as some male mouse strains, but the associated uncertainties—e.g., increased extent of low-dose
23 extrapolation and incomplete exposure information—provide less support for many of these
24 values. Four female mice data sets were considered to be the most uncertain because of dose
25 ranges covered and incomplete information regarding length of exposure (Grimmer et al.. 1984:
26 Habs etal.. 1984: Grimmer etal.. 1983: Habs etal.. 19801 In particular, the data set reported by
27 Habs etal. (1984) yielded the highest but most uncertain result, with only two dose-response
28 points; the slope estimate is particularly affected by the characterization of the high exposure level.
29 There was insufficient information to conclude that males were more sensitive because both sexes
30 were not tested for any mouse strain.
This document is a draft for review purposes only and does not constitute Agency policy.
2-42 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2
Table 2-11. Summary of dermal slope factor derivations, unadjusted for
interspecies differences
Reference
Mouse
strain
Selected
model3
BMR
BMD
(Hg/d)
POD =
BMDL
(Hg/d)
Candidate
dermal
slope
factors'5
(Hg/dr1
Comments
Male mice
Sivak et al.
(1997); NIOSH
(1989)
Poel (1959)°
C3H/HeJ
C57L
Multistage-
Weibull2°
Multistage 3°
10%
10%
0.11
0.13
0.060
0.078
1.7
1.3
Well-conducted and
reported study, including
individual times on study
and individual tumor
diagnoses
Grouped survival data
reported
Female mice
Roe etal. (1970)
Schmidt et al.
(1973)
Schmidt et al.
(1973)
Schmahl et al.
(1977)
Habs etal. (1980)
(Habsetal.,
1984)
Grimmer et al.
(1983)
Grimmer et al.
(1984)
Swiss
Swiss
NMRI
NMRI
NMRI
NMRI
CFLP
CFLP
Multistage 2°
Multistage 3°
Multistage 2°
Multistage 2°
Multistage 4°
Multistage 1°
Multistage 1°
Log-logistic
10%
10%
10%
10%
10%
30%
10%
50%
10%
40%
70%
0.69
0.28
0.33
0.23
0.36
0.49
0.078
0.51
0.24
1.2
1.07
0.39
0.22
0.29
0.15
0.24
0.44
0.056
0.37
0.21
1.0
0.48
0.25
0.45
0.34
0.67
0.42
0.69
1.8
1.4
0.48
0.40
1.5
Grouped survival data
reported
No characterization of
exposure duration
No characterization of
exposure duration
No characterization of
exposure duration
Higher overall exposure
range; unclear overall
duration of exposure
No characterization of
exposure duration for high
exposure; high response at
lowest exposure limits
usefulness of low-dose
extrapolation
No characterization of
exposure duration
No characterization of
exposure duration; high
response at lowest exposure
limits usefulness of low-dose
extrapolation
3
4
5
6
aSee Appendix E.2.4 (Supplemental Information) for modeling details.
bUnadjusted for interspecies differences. Slope factor = R/BMDLR, where R is the BMR expressed as a fraction.
cHigh exposure groups with 100% mortality omitted prior to dose-response modeling.
This document is a draft for review purposes only and does not constitute Agency policy,
2-43 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
2 2.5.4. Dermal Slope Factor Cross-Species Scaling
3 Different methodologies have been established for interspecies scaling of PODs used to
4 derive oral slope factors and inhalation unit risks. Cross-species adjustment of oral doses is based
5 on allometric scaling using the % power of body weight. This adjustment accounts for more rapid
6 distribution, metabolism, and clearance in small animals [U.S. EPA. 2005a). Cross-species
7 extrapolation of inhalation exposures is based on standard dosimetry models that consider factors
8 such as solubility, reactivity, and persistence [U.S. EPA. 1994] in addition to species differences in
9 physiology. Although no established methodology exists to adjust for interspecies differences in
10 dermal toxicity at the point of contact, allometric scaling using body weight to the % power was
11 selected based on known species differences in dermal metabolism and penetration of
12 benzo[a]pyrene. In vitro skin permeation was highest in the mouse, compared to rat, rabbit, and
13 human, and was enhanced by induction of CYP enzymes [Kao etal.. 1985). Using this approach,
14 rodents and humans exposed to the same daily dose of a carcinogen, adjusted for BW3/4, would be
15 expected to have equal lifetime risks of cancer.
16 Alternative approaches were also evaluated, including: (1) assuming that a given mass of
17 benzo[a]pyrene, applied daily, would pose the same risk in an animal or in humans, regardless of
18 whether it is applied to a small surface area or to a larger surface area at a proportionately lower
19 concentration (e.g., no interspecies adjustment); (2) assuming that equal mass per day, if applied to
20 equal fractions of total skin surface will have similar cancer risks; and (3) assuming that risk is
21 directly proportional to dose expressed as mass per kg body weight per day. A comparison of these
22 alternatives is provided in Appendix E of the Supplemental Information.
23 The PODM derived from the NIOSH study [Sivak etal.. 1997: NIOSH. 1989] is adjusted to a
24 HED as follows:
25 POD HED (Jig/day) = PODM (^g/day) x (BWH / BWM)3/4
26 = 0.060 ^g/day x (70 kg / 0.035 kg)3/4
27 = 17.9 ^g/day
28
29 The resulting PODnED is used to calculate the dermal slope factor for benzo[a]pyrene:
30
31 Dermal slope factor = BMR/PODHED = 0.1/(17.9 ng/day) = 0.006 per Hg/day.
32
33 Note that the dermal slope factor should only be used with lifetime human exposures
34 <18 |ig/day, the human equivalent of the PODM, because above this level the dose-response
35 relationship is not expected to be proportional to the mass of benzo[a]pyrene applied.
36 Several assumptions are made in the use of this scaling method. First, it is assumed that the
37 toxicokinetic processes in the skin will scale similarly to interspecies differences in whole-body
38 toxicokinetics. Secondly, it is assumed that the risk at low doses ofbenzo[a]pyrene is linear.
This document is a draft for review purposes only and does not constitute Agency policy.
2-44 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Although one study indicates that at high doses of benzo[a]pyrene carcinogenic potency is related
2 to mass applied per unit skin and not to total mass [Davies. 1969}. this may be due to promotional
3 effects, such as inflammation, that are observed at high doses of benzo[a]pyrene.
4 The dermal slope factor has been developed for a local effect and it is not intended to
5 estimate systemic risk of cancer following dermal absorption of benzo[a]pyrene into the systemic
6 circulation. Although some information suggests thatbenzo[a]pyrene metabolites can enter
7 systemic circulation following dermal exposure in humans [Godschalk et al.. 1998a]. lifetime skin
8 cancer bioassays that have included pathological examination of other organs have not found
9 elevated incidences of tumors at distal sites [Habs etal., 1980: Schmahl etal., 1977: Schmidt etal.,
10 1973: Roe etal., 1970: Poel, 1959]. This may be because benzo[a]pyrene tends to bind to targets
11 within the skin rather than enter the plasma receptor fluid (a surrogate measure of systemic
12 absorption) in in vitro human skin experiments. These data are consistent with metabolism of
13 benzo[a]pyrene to reactive metabolites within the viable layers of the skin [Wester et al.. 1990).
14 Some studies indicate that the fraction of benzo[a]pyrene left within the viable layers of the skin is
15 a large portion of the applied dose [Moody et al.. 2007: Moody and Chu. 1995]. Taken together,
16 these data support the conclusion that the risk of skin cancer following dermal exposure likely
17 outweighs cancer risks at distal organs.
18 2.5.5. Uncertainties in the Derivation of the Dermal Slope Factor
19 Table 2-12 summarizes uncertainties in the derivation of the dermal slope factor for
20 benzo[a]pyrene; further detail is provided in the following discussion. Uncertainty in the
21 recommended dermal slope factor is partly reflected in the range of POD values derived from the
22 modeled mouse skin tumor data sets: the lowest and highest BMDLio values listed in Table 2-11
23 show a sevenfold difference (0.056-0.39 [ig/day] in magnitude. However, many of the studies
24 considered had incomplete information concerning time on study, which would tend to lead to
25 underestimates of risk. Low-dose extrapolation uncertainty for several of these studies also
26 decreases confidence in their results. Reliance on the NIOSH study, a relatively well-conducted
27 study with the low exposure levels having low early mortality, exposures continuing for full
28 lifetimes, and individual times on study minimizes this source of uncertainty.
29 Human dermal exposure to benzo[a]pyrene in the environment likely occurs predominantly
30 through soil contact The available mouse dermal bioassays of benzo[a]pyrene relied on delivery of
31 benzo[a]pyrene to the skin in a solvent solution (typically acetone or toluene]. The use of a volatile
32 solvent likely results in a larger dose of benzo[a]pyrene available for uptake into the skin
33 (compared to soil]. Consequently, reliance on these studies may overestimate the risk of skin
34 tumors from benzo[a]pyrene contact through soil; however, cancer bioassays delivering
35 benzo[a]pyrene through a soil matrix are not available.
36 There is uncertainty in extrapolating from the intermittent exposures in the mouse assays
37 to daily exposure scenarios. All of the dermal bioassays that were considered treated animals
38 2-3 times/week. This assessment makes the assumption that risk is proportional to total
This document is a draft for review purposes only and does not constitute Agency policy.
2-45 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 cumulative exposure. However, this may overestimate risk if duration-adjusted doses are below
2 doses that saturate or diminish detoxifying metabolic steps.
3 The relative impact of a particular vehicle on benzo[a]pyrene carcinogenicity and the
4 relative sensitivity of male and female mice to benzo [a]pyrene exposure was difficult to ascertain in
5 the supporting lifetime studies. Overall, the study designs for these studies included different
6 mouse strains, sexes, and vehicles, but for any given mouse strain, even across multiple studies,
7 only one sex and only one vehicle was tested. Thus, it was not possible to evaluate the relative
8 impact of particular vehicles onbenzo[a]pyrene carcinogenicity, or the relative sensitivity of male
9 and female mice.
10 The available data were not useful to determine which animal species may be the best
11 surrogate for human dermal response to benzo[a]pyrene. In extrapolation of the animal dermal
12 information to humans, the assumption is made that equal lifetime risks of cancer would follow
13 from exposure to the same daily dose adjusted for BW3/4. Qualitatively, the toxicokinetics and
14 toxicodynamics in mouse and human skin appear to be similar [Knafla etal.. 2011: Bickers etal..
15 1984]. Specifically, all of the activation pathways implicated in benzo[a]pyrene carcinogenicity
16 have been observed in mouse and human skin, and associations have been made between the
17 spectra of mutations in tumor tissues from benzo[a]pyrene-exposed animals and humans exposed
18 to complex PAH mixtures containing benzo[a]pyrene (see Section 1.1.5).
19 The dermal slope factor for benzo[a]pyrene is based on skin cancer and does not represent
20 systemic cancer risk from dermal exposure. It is unclear whether dermal exposure to
21 benzo[a]pyrene would result in elevated risk of systemic tumors. Some studies in humans suggest
22 that although the skin may be responsible for a "first pass" metabolic effect, benzo[a]pyrene-
23 specific adducts have been detected in white blood cells (WBCs) following dermal exposure to
24 benzo[a]pyrene, indicating that dermally applied benzo[a]pyrene enters systemic circulation
25 [Godschalketal.. 1998a). Although none of the lifetime dermal bioassays in mice, which included
26 macroscopic examination of internal organs, reported an elevation of systemic tumors in
27 benzo[a]pyrene-treated mice compared to controls [Higginbothametal.. 1993: Habs etal.. 1980:
28 Schmahl etal.. 1977: Schmidt etal.. 1973: Roe etal.. 1970: Poel. 1959]. most of these studies
29 attempted to remove animals with grossly observed skin tumors from the study before the death of
30 the animal, possibly minimizing the development of more distant tumors with longer latency. The
31 risk of benzo [a]pyrene-induced point-of-contact tumors in the skin possibly competes with
32 systemic risk of tumors. Currently, the potential contribution of dermally absorbed benzo[a]pyrene
33 to systemic cancer risk is unclear.
This document is a draft for review purposes only and does not constitute Agency policy.
2-46 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2
Table 2-12. Summary of uncertainties in the derivation of cancer risk values
for benzo[a]pyrene dermal slope factor
Consideration and
impact on cancer risk value
Decision
Justification and discussion
Selection of data set
4, dermal slope factor if alternative
data set were selected
NIOSH (1989)
Study included lowest doses among available
studies (where intercurrent mortality was less
likely to impact the number at risk).
Selection of target organ
No dermal slope factor if skin tumor
studies not used
Selection of skin tumors
Skin tumors were replicated in numerous studies
of male or female mice. No studies were
available indicating that other tumors occur
following dermal exposure.
Selection of dose metric
Alternatives could 4, or
factor
slope
Administered dose, as
TWA in u.g/d
Experimental evidence supports a role for
metabolism in toxicity, but actual responsible
metabolites are not identified.
Interspecies extrapolation
Alternatives could 4, or ^ slope
factor
Total daily dose scaled
by BW3/4
Alternatives discussed in Appendix E. An
established methodology does not exist to adjust
for interspecies differences in dermal toxicity at
the point of contact. Benzo[a]pyrene
metabolism is known to occur in the dermal
layer. Viewing the skin as an organ, and without
evidence to the contrary, metabolic processes
were assumed to scale allometrically.
Dose-response modeling
Alternatives could 4, or 1
factor
slope
Multistage-Weibull
model
No biologically based models for benzo[a]pyrene
were available. The multistage-Weibull model is
consistent with biological processes,
incorporates timecourse information, and is
preferred for IRIS cancer assessments when
individual data are available (Gehlhaus et al.,
2011).
Low-dose extrapolation
•^ cancer risk estimate would be
expected with the application of
nonlinear low-dose extrapolation
Linear extrapolation
from POD (based on
mutagenic mode of
action)
Available mode of action data support linearity
(mutagenicity is a primary mode of action of
benzo[a]pyrene).
Sensitive subpopulations
1" dermal slope factor to unknown
extent
ADAFs are
recommended for early
life exposures
No chemical-specific data are available to
determine the range of human toxicodynamic
variability or sensitivity.
4 2.5.6. Previous IRIS Assessment: Dermal Slope Factor
5 A dermal slope factor for benzo[a]pyrene was not previously available on IRIS.
6 2.6. APPLICATION OF AGE-DEPENDENT ADJUSTMENT FACTORS (ADAFS)
7 Based on sufficient support in laboratory animals and relevance to humans, benzo[a]pyrene
8 is determined to be carcinogenic by a mutagenic mode of action. According to the Supplemental
This document is a draft for review purposes only and does not constitute Agency policy,
2-47 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Guidance for Assessing Susceptibility from Early Life Exposure to Carcinogens ("Supplemental
Guidance"} [U.S. EPA. 2005b). individuals exposed during early life to carcinogens with a mutagenic
mode of action are assumed to have increased risk for cancer. The oral slope factor of 1 per
mg/kg-day, inhalation unit risk of 0.6 per mg/m3, and dermal slope factor of 0.006 per [J.g/day for
benzo[a]pyrene, calculated from data applicable to adult exposures, do not reflect presumed early
life susceptibility to this chemical. Although chemical-specific data exist for benzo[a]pyrene that
quantitatively demonstrate increased early life susceptibility to cancer [Vesselinovitchetal.. 1975].
these data were not considered sufficient to develop separate risk estimates for childhood
exposure, as they used acute i.p. exposures [U.S. EPA, 2005b]. In the absence of adequate chemical-
specific data to evaluate differences in age-specific susceptibility, the Supplemental Guidance [U.S.
EPA. 2005b) recommends that ADAFs be applied in estimating cancer risk.
The Supplemental Guidance [U.S. EPA. 2005b) establishes ADAFs for three specific age
groups. These ADAFs and their corresponding age groupings are: 10 for individuals exposed at
<2 years of age, 3 for exposed individuals at 2-<16 years of age, and 1 for exposed individuals
>16 years of age. The 10- and 3-fold adjustments are combined with age-specific exposure
estimates when estimating cancer risks from early life (<16 years of age) exposures to
benzo[a]pyrene. To illustrate the use of the ADAFs established in the Supplemental Guidance [U.S.
EPA. 2005b). sample calculations are presented for three exposure duration scenarios, including
full lifetime, assuming a constant benzo[a]pyrene exposure of 0.001 mg/kg-day (Table 2-13).
Table 2-13. Sample application of ADAFs for the estimation of benzo[a]pyrene
cancer risk following lifetime (70-year) oral exposure
Age group
0-<2 yrs
2-<16 yrs
>16 yrs
ADAF
10
3
1
Unit risk (per
mg/kg-d)
1
1
1
Sample exposure
concentration (mg/kg-d)
0.001
0.001
0.001
Duration
adjustment
2 yrs/70 yrs
14 yrs/70 yrs
54 yrs/70 yrs
Total risk
Cancer risk for age-
specific exposure
period
0.0003
0.0006
0.0008
0.002
22
23
24
25
26
27
28
29
30
The example exposure duration scenarios include full lifetime exposure (assuming a
70-year lifespan). Table 2-13 lists the four factors (ADAFs, cancer risk estimate, assumed exposure,
and duration adjustment) that are needed to calculate the age-specific cancer risk based on the
early age-specific group. The cancer risk for each age group is the product of the four factors in
columns 2-5. Therefore, the cancer risk following daily benzo[a]pyrene oral exposure in the age
group 0-<2 years is the product of the values in columns 2-5 or 10 x 1 x 0.001 x 2/70 = 3 x 1Q-4.
The cancer risk for specific exposure duration scenarios that are listed in the last column are added
together to get the total risk. Thus, a 70-year (lifetime) risk estimate for continuous exposure to
This document is a draft for review purposes only and does not constitute Agency policy,
2-48 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Toxicological Review ofBenzo[a]pyrene
0.001 mg/kg-day benzo[a]pyrene is 2 x 1Q-3, which is adjusted for early-life susceptibility and
assumes a 70-year lifetime and constant exposure across age groups.
In calculating the cancer risk for a 30-year constant exposure to benzo[a]pyrene at an
exposure level of 0.001 mg/kg-day for ages 0-30 years, the duration adjustments would be 2/70,
14/70, and 14/70, and the age-specific risks for the three age groups would be 3 x 1Q-4, 6 x 1Q-4,
and 2 x 10~4, which would result in a total risk estimate of 1 x 1Q-3.
In calculating the cancer risk for a 30-year constant exposure to benzo[a]pyrene at an
exposure level of 0.001 mg/kg-day for ages 20-50 years, the duration adjustments would be 0/70,
0/70, and 30/70. The age-specific risks for the three groups are 0, 0, and 4 x 10~4, which would
result in a total risk estimate of 4 x 10~4.
Consistent with the approaches for the oral route of exposure (Table 2-13), the ADAFs
should also be applied when assessing cancer risks for subpopulations with early life exposures to
benzo[a]pyrene via the inhalation and dermal routes (presented in Tables 2-14 and 2-15).
Table 2-14. Sample application of ADAFs for the estimation of benzo[a]pyrene
cancer risk following lifetime (70-year) inhalation exposure
Age group
0-<2 yrs
2-<16 yrs
>16 yrs
ADAF
10
3
1
Unit risk
(per ug/m3)
6 x 10"4
6 x 10"4
6 x 10"4
Sample exposure
concentration (ug/m3)
0.1
0.1
0.1
Duration
adjustment
2 yrs/70 yrs
14 yrs/70 yrs
54 yrs/70 yrs
Total risk
Cancer risk for
age-specific
exposure period
0.00002
0.00004
0.00005
0.00010
16
17
Table 2-15. Sample application of ADAFs for the estimation of benzo[a]pyrene
cancer risk following lifetime (70-year) dermal exposure
Age group
0-<2 yrs
2-<16 yrs
>16 yrs
ADAF
10
3
1
Unit risk
(per ng/d)
0.006
0.006
0.006
Sample exposure
concentration (ng/d)
0.001
0.001
0.001
Duration
adjustment
2 yrs/70 yrs
14 yrs/70 yrs
54 yrs/70 yrs
Total risk
Cancer risk for
age-specific
exposure period
2 x 10"6
4 x 10"6
5 x 10"6
1 x 10"5
18
This document is a draft for review purposes only and does not constitute Agency policy,
2-49 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2 REFERENCES
3 Abedi-Ardekani, B; Kamangar, F; Hewitt, SM; Hainaut, P; Sotoudeh, M; Abnet, CC; Taylor, PR; Boffetta, P;
4 Malekzadeh, R; Dawsey, SM. (2010). Polycyclic aromatic hydrocarbon exposure in oesophageal
5 tissue and risk of oesophageal squamous cell carcinoma in north-eastern Iran. Gut 59: 1178-
6 1183. http://dx.doi.org/10.1136/gut.2010.210609
7 Aklillu, E; 0vreb0, S; Botnen, IV; Otter, C; Ingelman-Sundberg, M. (2005). Characterization of common
8 CYP1B1 variants with different capacity for benzo[a]pyrene-7,8-dihydrodiol epoxide formation
9 from benzo[a]pyrene. Cancer Res 65: 5105-5111. http://dx.doi.org/10.1158/0008-5472.CAN-05-
10 0113
11 Albert. RE; Miller. ML; Cody. T; Andringa. A; Shukla. R; Baxter. CS. (1991). Benzo[a]pyrene-induced skin
12 damage and tumor promotion in the mouse. Carcinogenesis 12: 1273-1280.
13 http://dx.doi.0rg/10.1093/carcin/12.7.1273
14 Albert, RE; Miller, ML; Cody, TE; Talaska, G; Underwood, P; Andringa, A. (1996). Epidermal cytokinetics,
15 DNA adducts, and dermal inflammation in the mouse skin in response to repeated
16 benzo[a]pyrene exposures. Toxicol Appl Pharmacol 136: 67-74.
17 Alexandrov. K; Cascorbi. I; Rojas. M; Bouvier. G; Kriek. E; Bartsch. H. (2002). CYP1A1 and GSTM1
18 genotypes affect benzo[a]pyrene DNA adducts in smokers' lung: comparison with
19 aromatic/hydrophobic adduct formation [Review]. Carcinogenesis 23: 1969-1977.
20 Alexandrov, K; Rojas, M; Geneste, O; Castegnaro, M; Camus, AM; Petruzzelli, S; Giuntini, C; Bartsch, H.
21 (1992). An improved fluorometric assay for dosimetry of benzo(a)pyrene diol-epoxide-DNA
22 adducts in smokers' lung: Comparisons with total bulky adducts and aryl hydrocarbon
23 hydroxylase activity. Cancer Res 52: 6248-6253.
24 Allan, LL; Schlezinger, JJ; Shansab, M; Sherr, PH. (2006). CYP1A1 in polycyclic aromatic hydrocarbon-
25 induced B lymphocyte growth suppression. Biochem Biophys Res Commun 342: 227-235.
26 http://dx.doi.0rg/10.1016/i.bbrc.2006.01.131
27 Amos. Cl; Xu. W; Spitz. MR. (1999). Is there a genetic basis for lung cancer susceptibility? [Review].
28 Recent Results Cancer Res 151: 3-12.
29 Andreassen. A; Kure. EH; Nielsen. PS; Autrup. H; Haugen. A. (1996). Comparative synchronous
30 fluorescence spectrophotometry and 32P-postlabeling analysis of PAH-DNA adducts in human
31 lung and the relationship to TP53 mutations. Mutat Res 368: 275-282.
32 Arafa, HM; Aly, HA; Abd-Ellah, MF; EI-Refaey, HM. (2009). Hesperidin attenuates benzo[alpha] pyrene-
33 induced testicular toxicity in rats via regulation of oxidant/antioxidant balance. Toxicol Ind
34 Health 25: 417-427. http://dx.doi.org/10.1177/0748233709106624
35 Archibong, AE; Inyang, F; Ramesh, A; Greenwood, M; Nayyar, T; Kopsombut, P; Hood, DB; Nyanda, AM.
36 (2002). Alteration of pregnancy related hormones and fetal survival in F-344 rats exposed by
37 inhalation to benzo(a)pyrene. Reprod Toxicol 16: 801-808.
38 Archibong, AE; Ramesh, A; Niaz, MS; Brooks, CM; Roberson, SI; Lunstra, DP. (2008). Effects of
39 benzo(a)pyrene on intra-testicular function in F-344 rats. Int J Environ Res Public Health 5: 32-
40 40. http://dx.doi.org/10.3390/iierph5010032
41 Armstrong, B; Tremblay, C; Baris, D; Theriault, G. (1994). Lung cancer mortality and polynuclear aromatic
42 hydrocarbons: A case-cohort study of aluminum production workers in Arvida, Quebec, Canada.
43 Am J Epidemiol 139: 250-262.
This document is a draft for review purposes only and does not constitute Agency policy.
R-l DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Armstrong, BG; Gibbs, G. (2009). Exposure-response relationship between lung cancer and polycyclic
2 aromatic hydrocarbons (PAHs). Occup Environ Med 66: 740-746.
3 http://dx.doi.org/10.1136/oem.2008.043711
4 Arthur D Little (Arthur D Little Inc.). (1989). Assessment of the cocarcinogenic/promoting activity of
5 asphalt fumes. National Institute for Occupational Safety and Health.
6 http://www.ntis. gov/search/product.aspx?ABBR=PB91110213
7 Asgharian, B. (2004). A model of deposition of hygroscopic particles in the human lung. Aerosol Sci
8 Technol 38: 938-947. http://dx.doi.org/10.1080/027868290511236
9 ASRM (American Society of Reproductive Medicine). (2008). The clinical utility of sperm DNA integrity
10 testing [Review]. Fertil Steril 90: S178-S180. http://dx.doi.Org/10.1016/i.fertnstert.2008.08.054
11 ATS PR (Agency for Toxic Substances and Disease Registry). (1995). Toxicological profile for polycyclic
12 aromatic hydrocarbons [ATSDR Tox Profile]. Washington, DC: U.S. Department of Health and
13 Human Services, http://www.atsdr.cdc.gov/toxprofiles/tp69.pdf
14 ATSDR (Agency for Toxic Substances and Disease Registry). (2011). The priority list of hazardous
15 substances that will be the subject of toxicological profiles.
16 http://www.atsdr.cdc.gov/SPL/index.html
17 Baan, R; Grosse, Y; Straif, K; Secretan, B; El Ghissassi, F; Bouvard, V; Benbrahim-Tallaa, L; Guha, N;
18 Freeman, C; Galichet, L; Cogliano, V. (2009). A review of human carcinogens-part F: Chemical
19 agents and related occupations. Lancet Oncol 10: 1143-1144. http://dx.doi.org/10.1016/S1470-
20 2045(09)70358-4
21 Balu. N; Padgett. WT; Lambert. GR; Swank. AE; Richard. AM; Nesnow. S. (2004). Identification and
22 characterization of novel stable deoxyguanosine and deoxyadenosine adducts of
23 benzo[a]pyrene-7,8-quinone from reactions at physiological pH. Chem Res Toxicol 17: 827-838.
24 http://dx.doi.org/10.1021/tx034207s
25 Barry, G; Cook, JW; Haslewood, GAD; Hewett, CL; Hieger, I; Kennaway, EL. (1935). The Production of
26 Cancer by Pure Hydrocarbons.-Part III. Proc Biol Sci 117: 318-351.
27 http://dx.doi.org/10.1098/rspb.1935.0032
28 Bartsch, H; Nair, U; Risch, A; Rojas, M; Wikman, H; Alexandrov, K. (2000). Genetic polymorphism of CYP
29 genes, alone or in combination, as a risk modifier of tobacco-related cancers [Review]. Cancer
30 Epidemiol Biomarkers Prev 9: 3-28.
31 Bartsch, H; Rojas, M; Alexandrov, K; Risch, A. (1998). Impact of adduct determination on the assessment
32 of cancer susceptibility. In M Schwab; HJ Senn; P Schlag; K Munk; H Rabes (Eds.), Recent results
33 in cancer research, v!54: Genes and environment in cancer (pp. 86-96). New York, NY: Springer.
34 Bartsch, H; Rojas, M; Nair, U; Nair, J; Alexandrov, K. (1999). Genetic cancer susceptibility and DNA
35 adducts: studies in smokers, tobacco chewers, and coke oven workers. Cancer Detect Prev 23:
36 445-453.
37 Beland, F; Gulp, S. (1998). Chronic bioassay of two composite samples from selected manufactured gas
38 plant waste sites [unpublished report]. (Technical Report 6722.02). Jefferson, AK: National
39 Center for Toxicological Research.
40 Benbrahim-Tallaa. L; Baan. RA; Grosse. Y; Lauby-Secretan. B; El ghissassi. F; Bouvard. V; Guha. N; Loomis.
41 D; Straif. K. (2012). Carcinogenicity of diesel-engine and gasoline-engine exhausts and some
42 nitroarenes. Lancet Oncol 13: 663-664. http://dx.doi.org/10.1016/S1470-2045(12)70280-2
43 Benjamin, H; Storkson, J; Tallas, PG; Pariza, MW. (1988). Reduction of benzo[a]pyrene-induced
44 forestomach neoplasms in mice given nitrite and dietary soy sauce. Food Chem Toxicol 26: 671-
45 678. http://dx.doi.org/10.1016/0278-6915(88)90066-X
46 Bennett, WP; Alavanja, MC; Blomeke, B; Vahakangas, KH; Castren, K; Welsh, JA; Bowman, ED; Khan, MA;
47 Flieder, DB; Harris, CC. (1999). Environmental tobacco smoke, genetic susceptibility, and risk of
48 lung cancer in never-smoking women. J Natl Cancer Inst 91: 2009-2014.
This document is a draft for review purposes only and does not constitute Agency policy.
R-2 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Berenblum, I; Haran, N. (1955). The influence of croton oil and of polyethylene glycol-400 on
2 carcinogenesis in theforestomach of the mouse. Cancer Res 15: 510-516.
3 Berger, J; Manz, A. (1992). Cancer of the stomach and the colon-rectum among workers in a coke gas
4 plant. Am J Ind Med 22: 825-834. http://dx.doi.org/10.1002/aiim.4700220605
5 Biancifiori, C; Caschera, F; Giornelli-Santulli, F; Bucciarelli, E. (1967). The action of oestrone and four
6 chemical carcinogens in intact and ovariectomized BALB/c/Cb/Se mice. Br J Cancer 21: 452-459.
7 Bickers, PR; Mukhtar, H; Dutta-Choudhury, T; Marcelo, CL; Voorhees, JJ. (1984). Aryl hydrocarbon
8 hydroxylase, epoxide hydrolase, and benzo[a]-pyrene metabolism in human epidermis:
9 Comparative studies in normal subjects and patients with psoriasis. J Invest Dermatol 83: 51-56.
10 http://dx.doi.org/10.llll/1523-1747.epl2261680
11 Bingham, E; Falk, HL (1969). Environmental carcinogens. The modifying effect of cocarcinogens on the
12 threshold response. Arch Environ Health 19: 779-783.
13 Binkova, B; Chvatalova, I; Lnenickova, Z; Milcova, A; Tulupova, E; Farmer, PB; Sram, RJ. (2007). PAH-DNA
14 adducts in environmentally exposed population in relation to metabolic and DNA repair gene
15 polymorphisms. Mutat Res-Fundam Mol Mech Mutagen 620: 49-61.
16 http://dx.doi.0rg/10.1016/i.mrfmmm.2007.02.022
17 Blaha, L; Kapplova, P; Vondracek, J; Upham, B; Machala, M. (2002). Inhibition of gap-junctional
18 intercellular communication by environmentally occurring polycyclic aromatic hydrocarbons.
19 Toxicol Sci 65: 43-51. http://dx.doi.Org/10.1093/toxsci/65.l.43
20 Blanton. RH; Lyte. M; Myers. MJ; Bick. PH. (1986). Immunomodulation by polyaromatic hydrocarbons in
21 mice and murine cells. Cancer Res 46: 2735-2739.
22 Boffetta. P; Burstyn. I; Partanen. T; Kromhout. H; Svane. O; Langard. S; Jarvholm. B; Frentzel-Beyme. R;
23 Kauppinen. T; Stucker. I; Shaham. J; Heederik. D; Ahrens. W; Bergdahl. IA; Cenee. S; Ferro. G;
24 Heikkila, P; Hooiveld, M; Johansen, C; Randem, BG; Schill, W. (2003). Cancer mortality among
25 European asphalt workers: an international epidemiological study II Exposure to bitumen fume
26 and other agents. Am J Ind Med 43: 28-39. http://dx.doi.org/10.1002/aiim.10182
27 Boffetta, P; Jourenkova, N; Gustavsson, P. (1997). Cancer risk from occupational and environmental
28 exposure to polycyclic aromatic hydrocarbons [Review]. Cancer Causes Control 8: 444-472.
29 http://dx.doi.Org/10.1023/A:1018465507029
30 Bolton, JL; Trush, MA; Penning, TM; Dryhurst, G; Monks, TJ. (2000). Role of quinones in toxicology
31 [Review]. Chem Res Toxicol 13: 135-160. http://dx.doi.org/10.1021/tx9902082
32 Borini, A; Tarozzi, N; Bizzaro, D; Bonu, MA; Fava, L; Flamigni, C; Coticchio, G. (2006). Sperm DNA
33 fragmentation: paternal effect on early post-implantation embryo development in ART. Hum
34 Reprod 21: 2876-2881. http://dx.doi.org/10.1093/humrep/del251
35 Borman, SM; Christian, PJ; Sipes, IG; Hoyer, PB. (2000). Ovotoxicity in female Fischer rats and B6 mice
36 induced by low-dose exposure to three polycyclic aromatic hydrocarbons: comparison through
37 calculation of an ovotoxic index. Toxicol Appl Pharmacol 167: 191-198.
38 http://dx.doi.org/10.1006/taap.2000.9006
39 Bosetti. C; Boffetta. P; La Vecchia. C. (2007). Occupational exposures to polycyclic aromatic
40 hydrocarbons, and respiratory and urinary tract cancers: a quantitative review to 2005 [Review].
41 Ann Oncol 18: 431-446. http://dx.doi.org/10.1093/annonc/mdll72
42 Bostrom. CE; Gerde. P; Hanberg. A; Jernstrom. B; Johansson. C; Kyrklund. T; Rannug. A; Tornqvist. M;
43 Victorin, K; Westerholm, R. (2002). Cancer risk assessment, indicators, and guidelines for
44 polycyclic aromatic hydrocarbons in the ambient air [Review]. Environ Health Perspect 110
45 Suppl 3: 451-488.
46 Bouayed, J; Bohn, T; Tybl, E; Kiemer, AK; Soulimani, R. (2012). Benzo[a]pyrene-lnduced Anti-Depressive-
47 like Behaviour in Adult Female Mice: Role of Monoaminergic Systems. Basic & Clinical
This document is a draft for review purposes only and does not constitute Agency policy.
R-3 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Pharmacology & Toxicology Online Pharmacology Online 110: 544-550.
2 http://dx.doi.0rg/10.llll/i.1742-7843.2011.00853.x
3 Bouayed, J; Desor, F; Rammal, H; Kiemer, AK; Tybl, E; Schroeder, H; Rychen, G; Soulimani, R. (2009a).
4 Effects of lactational exposure to benzo[alpha]pyrene (B[alpha]P) on postnatal
5 neurodevelopment, neuronal receptor gene expression and behaviour in mice. Toxicology 259:
6 97-106. http://dx.doi.0rg/10.1016/i.tox.2009.02.010
7 Bouayed, J; Desor, F; Soulimani, R. (2009b). Subacute oral exposure to benzo[alpha]pyrene (B[alpha]P)
8 increases aggressiveness and affects consummatory aspects of sexual behaviour in male mice. J
9 Hazard Mater 169: 581-585. http://dx.doi.Org/10.1016/i.ihazmat.2009.03.131
10 Broekman, BF; Chan, YH; Chong, YS; Quek, SC; Fung, D; Low, YL; Ooi, YP; Gluckman, PD; Meaney, MJ;
11 Wong, TY; Saw, SM. (2009). The influence of birth size on intelligence in healthy children.
12 Pediatrics 123: e!011-el016. http://dx.doi.org/10.1542/peds.2008-3344
13 Brown, J. R.; Thornton, JL (1957). Percivall Pott (1714-1788) and chimney sweepers' cancer of the
14 scrotum. Br J Ind Med 14: 68-70.
15 Brown, LA; Khousbouei, H; Goodwin, JS; Irvin-Wilson, CV; Ramesh, A; Sheng, L; Mcca I lister, MM; Jiang,
16 GC; Aschner, M; Hood, DB. (2007). Down-regulation of early ionotrophic glutamate receptor
17 subunit developmental expression as a mechanism for observed plasticity deficits following
18 gestational exposure to benzo(a)pyrene. Neurotoxicology 28: 965-978.
19 http://dx.doi.0rg/10.1016/i.neuro.2007.05.005
20 Brune. H; Deutsch-Wenzel. RP; Habs. M; Ivankovic. S; Schmahl. D. (1981). Investigation of the
21 tumorigenic response to benzo(a)pyrene in aqueous caffeine solution applied orally to Sprague-
22 Dawley rats. J Cancer Res Clin Oncol 102: 153-157. http://dx.doi.org/10.1007/BF00410666
23 Bryan. WR; Shimkin. MB. (1943). Quantitative analysis of dose-response data obtained with three
24 carcinogenic hydrocarbons in strain C3H male mice. J Natl Cancer Inst 3: 503-531.
25 Buening, MK; Wislocki, PG; Levin, W; Yagi, H; Thakker, PR; Akagi, H; Koreeda, M; Jerina, DM; Conney,
26 AH. (1978). Tumorigenicity of the optical enantiomers of the diastereomeric benzo[a]pyrene
27 7,8-diol-9,10-epoxides in newborn mice: exceptional activity of (+)-7"beta", 8"alpha"-dihydroxy-
28 9"alpha",10"alpha"-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene. PNAS 75: 5358-5361.
29 Burdick. AD; Davis. JW; Liu. KJ; Hudson. LG; Shi. H; Monske. ML; Burchiel. SW. (2003). Benzo(a)pvrene
30 quinones increase cell proliferation, generate reactive oxygen species, and transactivate the
31 epidermal growth factor receptor in breast epithelial cells. Cancer Res 63: 7825-7833.
32 Burns-Naas, LA; Hastings, KL; Ladies, GS; Makris, SL; Parker, GA; Holsapple, MP. (2008). What's so special
33 about the developing immune system? [Review]. Int J Toxicol 27: 223-254.
34 http://dx.doi.org/10.1080/10915810801978110
35 Burstyn, I; Kromhout, H; Johansen, C; Langard, S; Kauppinen, T; Shaham, J; Ferro, G; Boffetta, P. (2007).
36 Bladder cancer incidence and exposure to polycyclic aromatic hydrocarbons among asphalt
37 pavers. Occup Environ Med 64: 520-526. http://dx.doi.org/10.1136/oem.2006.029801
38 Burstyn. I; Kromhout. H; Kauppinen. T; Heikkila. P; Boffetta. P. (2000). Statistical modelling of the
39 determinants of historical exposure to bitumen and polycyclic aromatic hydrocarbons among
40 paving workers. Ann Occup Hyg 44: 43-56.
41 Burstyn. I; Kromhout. H; Partanen. T; Svane. O; Langard. S; Ahrens. W; Kauppinen. T; Stucker. I; Shaham.
42 J; Heederik. D; Ferro. G; Heikkila. P; Hooiveld. M; Johansen. C; Randem. BG; Boffetta. P. (2005).
43 Polycyclic aromatic hydrocarbons and fatal ischemic heart disease. Epidemiology 16: 744-750.
44 http://dx.doi.org/10.1097/01.ede.0000181310.65043.2f
45 Busby, WF, Jr; Goldman, ME; Newberne, PM; Wogan, GN. (1984). Tumorigenicity of fluoranthene in a
46 newborn mouse lung adenoma bioassay. Carcinogenesis 5: 1311-1316.
47 http://dx.doi.0rg/10.1093/carcin/5.10.1311
This document is a draft for review purposes only and does not constitute Agency policy.
R-4 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Busby, WF, Jr; Stevens, EK; Martin, CM; Chow, FL; Garner, RC. (1989). Comparative lung tumorigenicity of
2 parent and mononitro-polynuclear aromatic hydrocarbons in the BLU:Ha newborn mouse assay.
3 Toxicol Appl Pharmacol 99: 555-563. http://dx.doi.org/10.1016/0041-008X(89)90162-2
4 Cal/EPA (California Environmental Protection Agency). (2010). Public health goals for chemicals in
5 drinking water. Benzo(a)pyrene. http://oehha.ca.gov/water/phg/pdf/091610Benzopyrene.pdf
6 Car0e, TK; Ebbeh0j, NE; Wulf, HC; Agner, T. (2013). Occupational skin cancer may be underreported. 60:
7 A4624.
8 Cavalieri, E; Rogan, E; Cremonesi, P; Higginbotham, S; Salmasi, S. (1988a). Tumorigenicity of 6-
9 halogenated derivatives of benzo[a]pyrene in mouse skin and rat mammary gland. J Cancer Res
10 Clin Oncol 114: 10-15. http://dx.doi.org/10.1007/BF00390479
11 Cavalieri, E; Rogan, E; Higginbotham, S; Cremonesi, P; Salmasi, S. (1988b). Tumor-initiating activity in
12 mouse skin and carcinogenicity in rat mammary gland of fluorinated derivatives of
13 benzo[a]pyrene and 3-methylcholanthrene. J Cancer Res Clin Oncol 114: 16-22.
14 Cavalieri, E; Rogan, E; Sinha, D. (1988c). Carcinogenicity of aromatic hydrocarbons directly applied to rat
15 mammary gland. J Cancer Res Clin Oncol 114: 3-9. http://dx.doi.org/10.1007/BF00390478
16 Cavalieri, E; Rogan, E; Toth, B; Munhall, A. (1981). Carcinogenicity of the environmental pollutants
17 cyclopenteno-[cd]pyrene and cyclopentano[cd]pyrene in mouse skin. Carcinogenesis 2: 277-281.
18 Cavalieri, EL; Devanesan, PD; Rogan, EG. (1988d). Radical cations in the horseradish peroxidase and
19 prostaglandin H synthase mediated metabolism and binding of benzo[a]pyrene to
20 deoxyribonucleic acid. Biochem Pharmacol 37: 2183-2187. http://dx.doi.org/10.1016/0006-
21 2952(88)90579-5
22 Cavalieri. EL; Higginbotham. S; Ramakrishna. NV; Devanesan. PD; Todorovic. R; Rogan. EG; Salmasi. S.
23 (1991). Comparative dose-response tumorigenicity studies of dibenzo[alpha,l]pyrene versus
24 7,12-dimethylbenz[alpha]anthracene, benzo[alpha]pyrene and two dibenzo[alpha,l]pyrene
25 dihydrodiols in mouse skin and rat mammary gland. Carcinogenesis 12: 1939-1944.
26 http://dx.doi.org/10.1093/carcin/12.10.1939
27 Cavalieri, EL; Rogan, EG. (1992). The approach to understanding aromatic hydrocarbon Carcinogenesis.
28 The central role of radical cations in metabolic activation [Review]. Pharmacol Ther 55: 183-199.
29 Cavalieri, EL; Rogan, EG. (1995). Central role of radical cations in metabolic activation of polycyclic
30 aromatic hydrocarbons [Review]. Xenobiotica 25: 677-688.
31 http://dx.doi.org/10.3109/00498259509061885
32 Cavalieri, EL; Rogan, EG; Cremonesi, P; Devanesan, PD. (1988e). Radical cations as precursors in the
33 metabolic formation of quinones from benzo[a]pyrene and 6-fluorobenzo[a]pyrene. Fluoro
34 substitution as a probe for one-electron oxidation in aromatic substrates. Biochem Pharmacol
35 37: 2173-2182.
36 CDC (Centers for Disease Control and Prevention). (2004). The health consequences of smoking: A report
37 of the Surgeon General. Washington, DC: U.S. Department of Health and Human Services.
38 http://www.surgeongeneral.gov/library/smokingconsequences/
39 Chakravarti. D; Pelling. JC; Cavalieri. EL; Rogan. EG. (1995). Relating aromatic hydrocarbon-induced DNA
40 adducts and c-H-ras mutations in mouse skin papillomas: The role of apurinic sites. PNAS 92:
41 10422-10426.
42 Charles. GD; Bartels. MJ; Zacharewski. TR; Gollapudi. BB; Freshour. ML; Carney. EW. (2000). Activity of
43 benzo[a]pyrene and its hydroxylated metabolites in an estrogen receptor-alpha reporter gene
44 assay. Toxicol Sci 55: 320-326. http://dx.doi.Org/10.1093/toxsci/55.2.320
45 Chau, N; Bertrand, JP; Mur, JM; Figueredo, A; Patris, A; Moulin, JJ; Pham, QT. (1993). Mortality in retired
46 coke oven plant workers. Br J Ind Med 50: 127-135.
47 Chen, C; Tang, Y; Jiang, X; Qi, Y; Cheng, S; Qiu, C; Peng, B; Tu, B. (2012). Early postnatal benzo(a)pyrene
48 exposure in Sprague-Dawley rats causes persistent neurobehavioral impairments that emerge
This document is a draft for review purposes only and does not constitute Agency policy.
R-5 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 postnatally and continue into adolescence and adulthood. Toxicol Sci 125: 248-261.
2 http://dx.doi.org/10.1093/toxsci/kfr265
3 Chen, G; Gingerich, J; Soper, L; Douglas, GR; White, PA. (2010). Induction of lacZ mutations in
4 MutaMouse primary hepatocytes. Environ Mol Mutagen 51: 330-337.
5 http://dx.doi.org/10.1002/em.20540
6 Chen, X; An, H; Ao, L; Sun, L; Liu, W; Zhou, Z; Wang, Y; Cao, J. (2011). The combined toxicity of dibutyl
7 phthalate and benzo(a)pyrene on the reproductive system of male Sprague Dawley rats in vivo. J
8 Hazard Mater 186: 835-841. http://dx.doi.Org/10.1016/i.ihazmat.2010.ll.078
9 Cheng, L; Spitz, MR; Hong, WK; Wei, Q. (2000). Reduced expression levels of nucleotide excision repair
10 genes in lung cancer: A case-control analysis. Carcinogenesis 21: 1527-1530.
11 Chengzhi, C; Yan, T; Shuqun, C; Xuejun, J; Youbin, Q; Yinyin, X; Qian, T; Baijie, T. (2011). New candidate
12 proteins for benzo(a)pyrene-induced spatial learning and memory deficits. J Toxicol Sci 36: 163-
13 171.
14 Chouroulinkov, I; Gentil, A; Guerin, M. (1967). [Study of the carcinogenic activity of 9,10-dimethyl-
15 benzanthracene and of 3,4-benzopyrene given orally]. Bull Cancer 54: 67-78.
16 Chung. JY; Kim. YJ; Kim. JY; Lee. SG; Park. JE; Kim. WR; Yoon. YD; Yoo. KS; Yoo. YH; Kim. JM. (2011).
17 Benzo[a]pyrene reduces testosterone production in rat Leydig cells via a direct disturbance of
18 testicular steroidogenic machinery. Environ Health Perspect 119: 1569-1574.
19 http://dx.doi.org/10.1289/ehp.1003391
20 Clavson. DB; Iverson. F; Nera. EA; Lok. E. (1990). The significance of induced forestomach tumors. Annu
21 Rev Pharmacol Toxicol 30: 441-463.
22 Clement Associates. Inc. . (1990). Comparative potency approach for estimation of the total cancer risk
23 associated with exposures to mixtures of polycyclic aromatic hydrocarbons in the environment:
24 Appendices.
25 Cooper, AR; Moley, KH. (2008). Maternal tobacco use and its preimplantation effects on fertility: More
26 reasons to stop smoking [Review]. Seminars in Reproductive Medicine 26: 204-212.
27 http://dx.doi.org/10.1055/s-2008-1042959
28 Costantino, JP; Redmond, CK; Bearden, A. (1995). Occupationally related cancer risk among coke oven
29 workers: 30 years of follow-up. J Occup Environ Med 37: 597-604.
30 Cresteil, T. (1998). Onset of xenobiotic metabolism in children: Toxicological implications. Food Addit
31 Contam 15: 45-51. http://dx.doi.org/10.1080/02652039809374614
32 Gulp, SJ; Gaylor, DW; Sheldon, WG; Goldstein, LS; Beland, FA. (1996). DNA Adduct Measurements in
33 Relation to Small Intestine and Forestomach Tumor Incidence during the Chronic Feeding of
34 Coal TAR or Benzo[ A ]Pyrene to Mice. Polycycl Aromat Compd 11: 161-168.
35 http://dx.doi.org/10.1080/10406639608544662
36 Gulp, SJ; Gaylor, DW; Sheldon, WG; Goldstein, LS; Beland, FA. (1998). A comparison of the tumors
37 induced by coal tar and benzo[a]pyrene in a 2-year bioassay. Carcinogenesis 19: 117-124.
38 Gulp. SJ; Warbritton. AR; Smith. BA; Li. EE; Beland. FA. (2000). DNA adduct measurements, cell
39 proliferation and tumor mutation induction in relation to tumor formation in B6C3F1 mice fed
40 coal tar or benzo[a]pyrene. Carcinogenesis 21: 1433-1440.
41 http://dx.doi.0rg/10.1093/carcin/21.7.1433
42 Das. M; Asokan. P; Don. PS; Krueger. GG; Bickers. PR; Mukhtar. H. (1986). Carcinogen metabolism in
43 human skin grafted onto athymic nude mice: a model system for the study of human skin
44 Carcinogenesis. Biochem Biophys Res Commun 138: 33-39.
45 Davies, RF. (1969). Tumorigenicity of very fresh tobacco smoke condensate to mouse skin. Br J Cancer
46 23:858-860.
This document is a draft for review purposes only and does not constitute Agency policy.
R-6 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 De Jong, WH; Kroese, ED; Vos, JG; Van Loveren, H. (1999). Detection of immunotoxicity of
2 benzo[a]pyrene in a subacute toxicity study after oral exposure in rats. Toxicol Sci 50: 214-220.
3 http://dx.doi.0rg/10.1093/toxsci/50.2.21
4 De Waal. EJ; Schuurman. HJ; Van Loveren. H; Vos. JG. (1997). Differential effects of 2,3,7,8-
5 tetrachlorodibenzo-p-dioxin, bis(tri-n-butyltin) oxide and cyclosporine on thymus
6 histophysiology [Review]. Crit Rev Toxicol 27: 381-430.
7 http://dx.doi.org/10.3109/10408449709089900
8 Dean, JH; Luster, Ml; Boorman, GA; Lauer, LD; Leubke, RW; Lawson, L (1983). Selective
9 immunosuppression resulting from exposure to the carcinogenic congener of benzopyrene in
10 B6C3F1 mice. Clin Exp Immunol 52: 199-206.
11 DeMarini, DM; Landi, S; Tian, D; Hanley, NM; Li, X; Hu, F; Roop, BC; Mass, MJ; Keohavong, P; Gao, W;
12 Olivier, M; Hainaut, P; Mumford, JL (2001). Lung tumor KRAS and TP53 mutations in
13 nonsmokers reflect exposure to PAH-rich coal combustion emissions. Cancer Res 61: 6679-6681.
14 Denissenko, MF; Pao, A; Tang, M; Pfeifer, GP. (1996). Preferential formation of benzo[a]pyrene adducts
15 at lung cancer mutational hotspots in P53. Science 274: 430-432.
16 http://dx.doi.org/10.1126/science.274.5286.430
17 Dietert, RR. (2008). Developmental immunotoxicology (DIT): windows of vulnerability, immune
18 dysfunction and safety assessment. J Immunotoxicol 5: 401-412.
19 http://dx.doi.org/10.1080/15476910802483324
20 Dietert. RR; Etzel. RA; Chen. D; Halonen. M; Holladay. SD; Jarabek. AM; Landreth. K; Peden. DB;
21 Pinkerton. K; Smialowicz. RJ; Zoetis. T. (2000). Workshop to identify critical windows of exposure
22 for children's health: immune and respiratory systems work group summary [Review]. Environ
23 Health Perspect 108 Suppl 3: 483-490.
24 Dietert, RR; Piepenbrink, MS. (2006). Perinatal immunotoxicity: Why adult exposure assessment fails to
25 predict risk [Review]. Environ Health Perspect 114: 477-483.
26 http://dx.doi.org/10.1289/ehp.8566
27 Dobbing, J; Sands, J. (1973). Quantitative growth and development of human brain. Arch Dis Child 48:
28 757-767.
29 Dobbing, J; Sands, J. (1979). Comparative aspects of the brain growth spurt. Early Hum Dev 3: 79-83.
30 Dong, MH; Redmond, CK; Mazumdar, S; Costantino, JP. (1988). A multistage approach to the cohort
31 analysis of lifetime lung cancer risk among steelworkers exposed to coke oven emissions. Am J
32 Epidemiol 128: 860-873.
33 Dontenwill, W; Chevalier, HJ; Harke, HP; Lafrenz, U; Reckzeh, G; Schneider, B. (1973). Investigations on
34 the effects of chronic cigarette-smoke inhalation in Syrian golden hamsters. J Natl Cancer Inst
35 51: 1781-1832. http://dx.doi.Org/10.1093/inci/51.6.1781
36 Driscoll, KE; Costa, PL; Hatch, G; Henderson, R; Oberdorster, G; Salem, H; Schlesinger, RB. (2000).
37 Intratracheal instillation as an exposure technique for the evaluation of respiratory tract toxicity:
38 Uses and limitations. Toxicol Sci 55: 24-35.
39 EI-Bayoumy. K. (1985). Effects of organoselenium compounds on induction of mouse forestomach
40 tumors by benzo(a)pyrene. Cancer Res 45: 3631-3635.
41 EI-Bayoumy. K; Hecht. SS; Hoffmann. D. (1982). Comparative tumor initiating activity on mouse skin of 6-
42 nitrobenzo[a]pyrene, 6-nitrochrysene, 3-nitroperylene, 1-nitropyrene and their parent
43 hydrocarbons. Cancer Lett 16: 333-337. http://dx.doi.org/10.1016/0304-3835(82)90015-5
44 EI-Nemr, A; Al-Shawaf, T; Sabatini, L; Wilson, C; Lower, AM; Grudzinskas, JG. (1998). Effect of smoking on
45 ovarian reserve and ovarian stimulation in in-vitro fertilization and embryo transfer. Hum
46 Reprod 13: 2192-2198.
47 Emmett, EA; Bingham, EM; Barkley, W. (1981). A carcinogenic bioassay of certain roofing materials. Am J
48 Ind Med 2: 59-64. http://dx.doi.org/10.1002/aiim.4700020110
This document is a draft for review purposes only and does not constitute Agency policy.
R-7 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Esser, C. (2009). The immune phenotype of AhR null mouse mutants: not a simple mirror of xenobiotic
2 receptor over-activation [Review]. Biochem Pharmacol 77: 597-607.
3 http://dx.doi.0rg/10.1016/i.bcp.2008.10.002
4 Evanoff, BA; Gustavsson, P; Hogstedt, C. (1993). Mortality and incidence of cancer in a cohort of Swedish
5 chimney sweeps: an extended follow up study. Occup Environ Med 50: 450-459.
6 Everall.JD; Dowd, PM. (1978). Influence of environmental factors excluding ultra violet radiation on the
7 incidence of skin cancer. Bull Cancer 65: 241-247.
8 Fedorenko, Z; Yansheva, N. (1967). [Experimental reproduction of tumors of the antral part of the
9 stomach in mice by administration of various dose of 3,4-benzpyrene]. Gig Sanit 32: 168-173.
10 Fernandez-Salguero, P; Pineau, T; Hilbert, DM; McPhail, T; Lee, SS; Kimura, S; Nebert, DW; Rudikoff, S;
11 Ward, JM; Gonzalez, FJ. (1995). Immune system impairment and hepatic fibrosis in mice lacking
12 the dioxin-binding Ah receptor. Science 268: 722-726.
13 Feron, VJ; de Jong, D; Emmelot, P. (1973). Letter: Dose-response correlation for the induction of
14 respiratory-tract tumours in Syrian golden hamsters by intratracheal instillations of
15 benzo(a)pyrene. Eur J Cancer 9: 387-390.
16 Feron, VJ; Kruysse, A. (1978). Effects of exposure to furfural vapour in hamsters simultaneously treated
17 with benzo[alpha] pyrene or diethylnitrosamine. Toxicology 11: 127-144.
18 Perron, GA. (1994). Aerosol properties and lung deposition [Editorial]. Eur Respir J 7: 1392-1394.
19 Fertuck. KC; Matthews. JB; Zacharewski. TR. (2001). Hydroxylated benzo[a]pyrene metabolites are
20 responsible for in vitro estrogen receptor-mediated gene expression induced by
21 benzo[a]pyrene, but do not elicit uterotrophic effects in vivo. Toxicol Sci 59: 231-240.
22 http://dx.doi.0rg/10.1093/toxsci/59.2.231
23 Fine. JS; Silverstone. AE; Gasiewicz. TA. (1990). Impairment of prothymocyte activity by 2,3,7,8-
24 tetrachlorodibenzo-p-dioxin. J Immunol 144: 1169-1176.
25 Flowers, L; Bleczinski, WF; Burczynski, ME; Harvey, RG; Penning, TM. (1996). Disposition and biological
26 activity of benzo[a]pyrene-7,8-dione. A genotoxic metabolite generated by dihydrodiol
27 dehydrogenase. Biochemistry 35: 13664-13672. http://dx.doi.org/10.1021/bi961077w
28 Flowers, L; Ohnishi, ST; Penning, TM. (1997). DNA strand scission by polycyclic aromatic hydrocarbon o-
29 quinones: role of reactive oxygen species, Cu(ll)/Cu(l) redox cycling, and o-semiquinone anion
30 radicals. Biochemistry 36: 8640-8648. http://dx.doi.org/10.1021/bi970367p
31 Francis, EZ; Kimmel, CA; Rees, DC. (1990). Workshop on the qualitative and quantitative comparability of
32 human and animal developmental neurotoxicity: Summary and implications. Neurotoxicol
33 Teratol 12: 285-292. http://dx.doi.org/10.1016/0892-0362(90)90101-H
34 Friesen, MC; Benke, G; Del Monaco, A; Dennekamp, M; Fritschi, L; de Klerk, N; Moving, JL; Macfarlane, E;
35 Sim, MR. (2009). Relationship between cardiopulmonary mortality and cancer risk and
36 quantitative exposure to polycyclic aromatic hydrocarbons, fluorides, and dust in two prebake
37 aluminum smelters. Cancer Causes Control 20: 905-916. http://dx.doi.org/10.1007/sl0552-009-
38 9329-8
39 Friesen. MC; Demers. PA; Spinelli. JJ; Eisen. EA; Lorenzi. MF; Le. ND. (2010). Chronic and acute effects of
40 coal tar pitch exposure and cardiopulmonary mortality among aluminum smelter workers. Am J
41 Epidemiol 172: 790-799. http://dx.doi.org/10.1093/aie/kwq208
42 Friesen. MC; Macnab. YC; Marion. SA; Demers. PA; Davies. HW; Teschke. K. (2006). Mixed models and
43 empirical bayes estimation for retrospective exposure assessment of dust exposures in Canadian
44 sawmills. Ann Occup Hyg 50: 281-288. http://dx.doi.org/10.1093/annhyg/mei076
45 Gale, CR; O'Callaghan, FJ; Bredow, M; Martyn, CM; Team, ALSoPaCS. (2006). The influence of head
46 growth in fetal life, infancy, and childhood on intelligence at the ages of 4 and 8 years. Pediatrics
47 118: 1486-1492. http://dx.doi.org/10.1542/peds.2005-2629
This document is a draft for review purposes only and does not constitute Agency policy.
R-8 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Galvan, N; Page, TJ; Czuprynski, G; Jefcoate, CR. (2006). Benzo(a)pyrene and 7,12-
2 dimethylbenz(a)anthrecene differentially affect bone marrow cells of the lymphoid and myeloid
3 lineages. Toxicol Appl Pharmacol 213: 105-116. http://dx.doi.Org/10.1016/i.taap.2005.09.018
4 Gao, M; Li, Y; Sun, Y; Shah, W; Yang, S; Wang, Y; Long, J. (2011). Benzo[a]pyrene exposure increases toxic
5 biomarkers and morphological disorders in mouse cervix. Basic & Clinical Pharmacology &
6 Toxicology Online Pharmacology Online 109: 398-406. http://dx.doi.org/lO.llll/j. 1742-
7 7843.2011.00755.x
8 Garcon, G; Gosset, P; Garry, S; Marez, T; Hannothiaux, MH; Shirali, P. (2001a). Pulmonary induction of
9 proinflammatory mediators following the rat exposure to benzo(a)pyrene-coated onto Fe2O3
10 particles. Toxicol Lett 121: 107-117.
11 Garcon, G; Zerimech, F; Hannothiaux, M; Gosset, P; Martin, A; Marez, T; Shirali, P. (2001b). Antioxidant
12 defense disruption by polycyclic aromatic hydrocarbons-coated onto Fe(2)O(3) particles in
13 human lung cells (A549). Toxicology 166: 129-137.
14 Gaspari, L; Chang, SS; Santella, RM; Garte, S; Pedotti, P; Taioli, E. (2003). Polycyclic aromatic
15 hydrocarbon-DNA adducts in human sperm as a marker of DNA damage and infertility. Mutat
16 Res 535: 155-160.
17 Geacintov, NE; Cosman, M; Hingerty, BE; Amin, S; Broyde, S; Patel, DJ. (1997). NMR solution structures
18 of stereoisometric covalent polycyclic aromatic carcinogen-DNA adduct: principles, patterns,
19 and diversity [Review]. Chem Res Toxicol 10: 111-146. http://dx.doi.org/10.1021/tx9601418
20 Gehlhaus. MW; Gift. JS; Hogan. KA; Kopylev. L; Schlosser. PM; Kadry. AR. (2011). Approaches to cancer
21 assessment in EPA's Integrated Risk Information System [Review]. Toxicol Appl Pharmacol 254:
22 170-180. http://dx.doi.0rg/10.1016/i.taap.2010.10.019
23 Centner. NJ; Weber. LP. (2011). Intranasal benzo[a]pyrene alters circadian blood pressure patterns and
24 causes lung inflammation in rats. Arch Toxicol 85: 337-346. http://dx.doi.org/10.1007/s00204-
25 010-0589-6
26 Gibbs, GW. (1985). Mortality of aluminum reduction plant workers, 1950 through 1977. J Occup Environ
27 Med 27: 761-770.
28 Gibbs, GW; Armstrong, B; Sevigny, M. (2007). Mortality and cancer experience of Quebec aluminum
29 reduction plant workers. Part 2: mortality of three cohorts hired on or before January 1, 1951. J
30 Occup Environ Med 49: 1105-1123. http://dx.doi.org/10.1097/JOM.Ob013e318157d34a
31 Gibbs, GW; Horowitz, I. (1979). Lung cancer mortality in aluminum reduction plant workers. J Occup
32 Med 21: 347-353.
33 Gibbs, GW; Sevigny, M. (2007a). Mortality and cancer experience of Quebec aluminum reduction plant
34 workers, part 4: cancer incidence. J Occup Environ Med 49: 1351-1366.
35 http://dx.doi.org/10.1097/JOM.Ob013e318156ecbc
36 Gibbs, GW; Sevigny, M. (2007b). Mortality and cancer experience of Quebec aluminum reduction plant
37 workers. Part 3: monitoring the mortality of workers first employed after January 1, 1950. J
38 Occup Environ Med 49: 1269-1287. http://dx.doi.org/10.1097/JOM.Ob013e3181593da8
39 Ginsberg. G; Hattis. D; Sonawane. B. (2004). Incorporating pharmacokinetic differences between
40 children and adults in assessing children's risks to environmental toxicants [Review]. Toxicol
41 Appl Pharmacol 198: 164-183. http://dx.doi.Org/10.1016/i.taap.2003.10.010
42 GLC (Great Lakes Commission). (2007). Assessment of benzo(a)pyrene air emissions in the Great Lakes
43 region: A report of the Great Lakes Regional Toxic Air Emissions Inventory Steering Committee.
44 Ann Arbor, Ml.
45 Godschalk, R; Curfs, D; Bartsch, H; Van Schooten, FJ; Nair, J. (2003). Benzo[a]pyrene enhances lipid
46 peroxidation induced DNA damage in aorta of apolipoprotein E knockout mice. Free Radic Res
47 37: 1299-1305.
This document is a draft for review purposes only and does not constitute Agency policy.
R-9 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Godschalk, R; Nair, J; van Schooten, FJ; Risch, A; Drings, P; Kayser, K; Dienemann, H; Bartsch, H. (2002).
2 Comparison of multiple DNA adduct types in tumor adjacent human lung tissue: effect of
3 cigarette smoking. Carcinogenesis 23: 2081-2086.
4 Godschalk, RW; Maas, LM; Kleinjans, JC; Van Schooten, FJ. (1998a). Influences of DNA isolation and RNA
5 contamination on carcinogen-DNA adduct analysis by 32P-postlabeling. Environ Mol Mutagen
6 32:344-350. http://dx.doi.org/10.1002/(SICI)1098-2280(1998)32:4<344::AID-EM8>3.0.CO;2-P
7 Godschalk, RW; Ostertag, JU; Moonen, EJ; Neumann, HA; Kleinjans, JC; van Schooten, FJ. (1998b).
8 Aromatic DNA adducts in human white blood cells and skin after dermal application of coal tar.
9 Cancer Epidemiol Biomarkers Prev 7: 767-773.
10 Godschalk, RW; Ostertag, JU; Zandsteeg, AM; Van Agen, B; Neuman, HA; Van Straaten, H; Van Schooten,
11 Fl (2001). Impact of GSTM1 on aromatic-DNA adducts and p53 accumulation in human skin and
12 lymphocytes. Pharmacogenetics 11: 537-543.
13 Graem, N. (1986). Epidermal changes following application of 7,12-dimethylbenz(a)anthracene and 12-
14 O-tetradecanoylphorbol-13-acetate to human skin transplanted to nude mice studied with
15 histological species markers. Cancer Res 46: 278-284.
16 Grant, G; Roe, FJ. (1963). The effect of phenanthrene on tumour induction by 3,4-benzopyrene
17 administered to newly born mice. Br J Cancer 17: 261-265.
18 Grice, HC; Clayson, DB; Flamm, WG; Ito, N; Kroes, R; Newberne, PM; Scheuplein, R. (1986). Panel
19 discussion: Possible mechanisms of BHA carcinogenicity from a consideration of its chemical and
20 biological properties. Food Chem Toxicol 24: 1235-1242.
21 Grimmer. G; Brune. H; Deutsch-Wenzel. R; Dettbarn. G; Misfeld. J. (1984). Contribution of polycyclic
22 aromatic hydrocarbons to the carcinogenic impact of gasoline engine exhaust condensate
23 evaluated by implantation into the lungs of rats. J Natl Cancer Inst 72: 733-739.
24 Grimmer, G; Brune, H; Deutsch-Wenzel, R; Naujack, KW; Misfeld, J; Timm, J. (1983). On the contribution
25 of polycyclic aromatic hydrocarbons to the carcinogenic impact of automobile exhaust
26 condensate evaluated by local application onto mouse skin. Cancer Lett 21: 105-113.
27 Grova, N; Schroeder, H; Farinelle, S; Prodhomme, E; Valley, A; Muller, CP. (2008). Sub-acute
28 administration of benzo[a]pyrene (B[a]P) reduces anxiety-related behaviour in adult mice and
29 modulates regional expression of N-methyl-D-aspartate (NMDA) receptors genes in relevant
30 brain regions. Chemosphere 73: S295-S302.
31 http://dx.doi.0rg/10.1016/i.chemosphere.2007.12.037
32 Grova, N; Valley, A; Turner, JD; Morel, A; Muller, CP; Schroeder, H. (2007). Modulation of behavior and
33 NMDA-R1 gene mRNA expression in adult female mice after sub-acute administration of
34 benzo(a)pyrene. Neurotoxicology 28: 630-636. http://dx.doi.Org/10.1016/i.neuro.2007.01.010
35 Gustavsson, P; Jakobsson, R; Johansson, H; Lewin, F; Norell, S; Rutkvist, LE. (1998). Occupational
36 exposures and squamous cell carcinoma of the oral cavity, pharynx, larynx, and oesophagus: A
37 case-control study in Sweden. Occup Environ Med 55: 393-400.
38 Gustavsson. P; Plato. N; Lidstrom. EB; Hogstedt. C. (1990). Lung cancer and exposure to diesel exhaust
39 among bus garage workers. Scand J Work Environ Health 16: 348-354.
40 Guvatt. GH; Oxman. AD; Kunz. R; Vist. GE; Falck-Ytter. Y; Schunemann. HJ. (2008a). GRADE: What is
41 "quality of evidence" and why is it important to clinicians? [Review]. BMJ 336: 995-998.
42 http://dx.doi.org/10.1136/bmi.39490.551019.BE
43 Guvatt, GH; Oxman, AD; Vist, GE; Kunz, R; Falck-Ytter, Y; Alonso-Coello, P; Schunemann, HJ. (2008b).
44 GRADE: An emerging consensus on rating quality of evidence and strength of recommendations.
45 BMJ 336: 924-926. http://dx.doi.org/10.1136/bmi.39489.470347.AD
46 Habs, M; Jahn, SA; Schmahl, D. (1984). Carcinogenic activity of condensate from coloquint seeds
47 (Citrullus colocynthis) after chronic epicutaneous administration to mice. J Cancer Res Clin Oncol
48 108: 154-156. http://dx.doi.org/10.1007/BF00390988
This document is a draft for review purposes only and does not constitute Agency policy.
R-10 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Habs, M; Schmahl, D; Misfeld, J. (1980). Local carcinogenicity of some environmentally relevant
2 polycyclic aromatic hydrocarbons after lifelong topical application to mouse skin. Arch
3 Geschwulstforsch 50: 266-274.
4 Hackman, P; Hou, SM; Nyberg, F; Pershagen, G; Lambert, B. (2000). Mutational spectra at the
5 hypoxanthine-guanine phosphoribosyltransferase (HPRT) locus in T-lymphocytes of nonsmoking
6 and smoking lung cancer patients. Mutat Res 468: 45-61.
7 Hainaut, P; Pfeifer, GP. (2001). Patterns of p53 G->T transversions in lung cancers reflect the primary
8 mutagenic signature of DNA-damage by tobacco smoke. Carcinogenesis 22: 367-374.
9 Hakim, FT; Memon, SA; Cepeda, R; Jones, EC; Chow, CK; Kasten-Sportes, C; Odom, J; Vance, BA;
10 Christensen, BL; Mackall, CL; Gress, RE. (2005). Age-dependent incidence, time course, and
11 consequences of thymic renewal in adults. J Clin Invest 115: 930-939.
12 http://dx.doi.org/10.1172/JCI22492
13 Hales, DB. (2002). Testicular macrophage modulation of Leydig cell steroidogenesis [Review]. J Reprod
14 Immunol57:3-18.
15 Hammond, EC; Selikoff, IJ; Lawther, PL; Seidman, H. (1976). Inhalation of benzpyrene and cancer in man.
16 Ann N YAcad Sci 271: 116-124. http://dx.doi.Org/10.llll/i.1749-6632.1976.tb23100.x
17 Hanahan, D; Weinberg, RA. (2011). Hallmarks of cancer: The next generation. Cell 144: 646-674.
18 http://dx.doi.0rg/10.1016/i.cell.2011.02.013
19 Hansen. ES. (1989). Cancer incidence in an occupational cohort exposed to bitumen fumes. Scand J Work
20 Environ Health 15: 101-105.
21 Hansen. ES. (1991). Mortality of mastic asphalt workers. Scand J Work Environ Health 17: 20-24.
22 http://dx.doi.org/10.5271/siweh.1739
23 Hardin. JA; Hinoshita. F; Sherr. PH. (1992). Mechanisms by which benzo[a]pyrene, an environmental
24 carcinogen, suppresses B cell lymphopoiesis. Toxicol Appl Pharmacol 117: 155-164.
25 Harvey, RG. (1996). Mechanisms of carcinogenesis of polycyclic aromatic hydrocarbons. Polycycl Aromat
26 Compd 9: 1-23. http://dx.doi.org/10.1080/10406639608031196
27 Hattis, D; Goble, R; Russ, A; Chu, M; Ericson, J. (2004). Age-related differences in susceptibility to
28 carcinogenesis: A quantitative analysis of empirical animal bioassay data. Environ Health
29 Perspect 112: 1152-1158. http://dx.doi.org/10.1289/ehp.6871
30 Henry, MC; Port, CD; Bates, RR; Kaufman, DG. (1973). Respiratory tract tumors in hamsters induced by
31 benzo(a)pyrene. Cancer Res 33: 1585-1592.
32 HEW (U.S. Department of Health, Education and Welfare). (1964). Smoking and health: Report of the
33 advisory committee to the surgeon general of the public health service. Washington, DC: U.S.
34 Department of Health, Education, and Welfare.
35 http://profiles.nlm.nih.gov/ps/retrieve/ResourceMetadata/NNBBMQ
36 Higginbotham, S; Ramakrishna, NVS; Johansson, SL; Rogan, EG; Cavalieri, EL. (1993). Tumor-initiating
37 activity and carcinogenicity of dibenzo[a,l]pyrene versus 7,12-dimethylbenz[a]anthracene and
38 benzo[a]pyrene at low doses in mouse skin. Carcinogenesis 14: 875-878.
39 http://dx.doi.0rg/10.1093/carcin/14.5.875
40 Hill. AB. (1965). The environment and disease: Association or causation? Proc R Soc Med 58: 295-300.
41 Hoffmann. D; Rathkamp. G; Nesnow. S; Wynder. EL. (1972). Fluoranthenes: Quantitative determination
42 in cigarette smoke, formation by pyrolysis, and tumor-initiating activity. J Natl Cancer Inst 49:
43 1165-1175.
44 Hogstedt, C; Jansson, C; Hugosson, M; Tinnerberg, H; Gustavsson, P. (2013). Cancer incidence in a cohort
45 of Swedish chimney sweeps, 1958-2006. Am J Public Health 103: 1708-1714.
46 http://dx.doi.org/10.2105/AJPH.2012.300860
This document is a draft for review purposes only and does not constitute Agency policy.
R-ll DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Holladay, SD; Smialowicz, RJ. (2000). Development of the murine and human immune system:
2 Differential effects of immunotoxicants depend on time of exposure [Review]. Environ Health
3 Perspect 108: 463-473.
4 Holladav, SD; Smith, BJ. (1994). Fetal hematopoietic alterations after maternal exposure to
5 benzo[a]pyrene: A cytometric evaluation. J Toxicol Environ Health 42: 259-273.
6 http://dx.doi.org/10.1080/15287399409531878
7 Holladav, SD; Smith, BJ. (1995). Benzo[a]pyrene-induced alterations in total immune cell number and
8 cell-surface antigen expression in the thymus, spleen and bone marrow of B6C3F1 mice. Vet
9 Hum Toxicol 37: 99-104.
10 Homburger, F; Hsueh, SS; Kerr, CS; Russfield, AB. (1972). Inherited susceptibility of inbred strains of
11 Syrian hamsters to induction of subcutaneous sarcomas and mammary and gastrointestinal
12 carcinomas by subcutaneous and gastric administration of polynuclear hydrocarbons. Cancer
13 Res 32: 360-366.
14 Hood, DB; Nayyar, T; Ramesh, A; Greenwood, M; Inyang, F. (2000). Modulation in the developmental
15 expression profile of Spl subsequent to transplacental exposure of fetal rats to desorbed
16 benzo[a]pyrene following maternal inhalation. Inhal Toxicol 12: 511-535.
17 http://dx.doi.org/10.1080/089583700402897
18 HSDB (Hazardous Substances Data Bank). (2012). Benzo(a)pyrene [Database]. Bethesda, MD: National
19 Library of Medicine. Retrieved from http://toxnet.nlm.nih.gov/cgi-bin/sis/htmlgen7HSDB
20 Hsu. PC: Chen. IY: Pan. CH: Wu. KY: Pan. MH: Chen. JR: Chen. CJ: Chang-Chien. GP: Hsu. CH: Liu. CS: Wu.
21 MT. (2006). Sperm DNA damage correlates with polycyclic aromatic hydrocarbons biomarker in
22 coke-oven workers. Int Arch Occup Environ Health 79: 349-356.
23 http://dx.doi.org/10.1007/s00420-005-0066-3
24 IARC (International Agency for Research on Cancer). (1973). Certain polycyclic aromatic hydrocarbons
25 and heterocyclic compounds. (HEEP/74/11596). Lyon, France.
26 IARC (International Agency for Research on Cancer). (1983). Polynuclear aromatic compounds, part 1:
27 Chemical, environmental and experimental data. Lyon, France: World Health Organization.
28 http://monographs.iarc.fr/ENG/Monographs/vol32/volume32.pdf
29 IARC (International Agency for Research on Cancer). (2002). Section 3. Carcinogenicity studies in
30 experimental animals [IARC Monograph] (pp. 973-1003). Lyon, France.
31 http://monographs.iarc.fr/ENG/Monographs/vol83/mono83-6C.pdf
32 IARC (International Agency for Research on Cancer). (2003). Predictive value of rodent forestomach and
33 gastric neuroendocrine tumours in evaluating carcinogenic risks to humans: Views and expert
34 opinions of an IARC working group, Lyon, 29 November -1 December 1999. (IARC Technical
35 Publication No. 39). Lyon, France.
36 http://monographs.iarc.fr/ENG/Publications/techrep39/IARCrep39.pdf
37 IARC (International Agency for Research on Cancer). (2004). Tobacco smoke and involuntary smoking
38 [IARC Monograph] (pp. 1-1438). (RISKLINE/2004080004). Lyon, France.
39 http://monographs.iarc.fr/ENG/Monographs/vol83/mono83.pdf
40 IARC (International Agency for Research on Cancer). (2006). Preamble to the IARC monographs. Lyon,
41 France. http://monographs.iarc.fr/ENG/Preamble/
42 IARC (International Agency for Research on Cancer). (2010). Some non-heterocyclic polycyclic aromatic
43 hydrocarbons and some related exposures [IARC Monograph] (pp. 1-853). Lyon, France.
44 http://monographs.iarc.fr/ENG/Monographs/vol92/mono92.pdf
45 IARC (International Agency for Research on Cancer). (2012). Soot, as found in occupational exposure of
46 chimney sweeps [IARC Monograph]. In Chemical agents and related occupations A review of
47 human carcinogens (pp. 209-214). Lyon, France.
48 http://monographs.iarc.fr/ENG/Monographs/vollOOF/monolOOF-21.pdf
This document is a draft for review purposes only and does not constitute Agency policy.
R-12 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Ickstadt. K; Schafer. M; Fritsch. A; Schwender. H; Abel. J; Bolt. HM; Bruning. T; Ko. YD; Vetter. H; Harth.
2 \A (2008). Statistical methods for detecting genetic interactions: a head and neck squamous-cell
3 cancer study. J Toxicol Environ Health A 71: 803-815.
4 http://dx.doi.org/10.1080/15287390801985745
5 Inyang, F; Ramesh, A; Kopsombut, P; Niaz, MS; Hood, DB; Nvanda, AM; Archibong, AE. (2003). Disruption
6 of testicular steroidogenesis and epididymal function by inhaled benzo(a)pyrene. Reprod Toxicol
7 17:527-537.
8 IOM (Institute of Medicine). (2008). Improving the presumptive disability decision-making process for
9 veterans. In JM Samet; CC Bodurow (Eds.). Washington, DC: National Academies Press.
10 http://www.nap.edu/openbook.php7record id=11908
11 I PCS (International Programme on Chemical Safety). (1998). Selected non-heterocyclic polycyclic
12 aromatic hydrocarbons [WHO EHC]. In Environmental Health Criteria (pp. CHF 174).
13 (CIS/00/00657). Geneva, Switzerland: World Health Organization.
14 http://www.inchem.org/documents/ehc/ehc/ehc202.htm
15 Jayasekara, S; Sharma, RP; Drown, DB. (1992). Effects of benzo[a]pyrene on steady-state levels of
16 biogenic amines and metabolizing enzymes in mouse brain regions. Ecotoxicol Environ Saf 24: 1-
17 12. http://dx.doi.org/10.1016/0147-6513(92)90029-3
18 Jeffrey, AM; Jennette, KW; Blobstein, SH; Weinstein, IB; Beland, FA; Harvey, RG; Kasal, H; Miura, I;
19 Nakanishi. K. (1976). Letter: Benzo[a]pyrene-nucleic acid derivative found in vivo: structure of a
20 benzo[a]pyrenetetrahydrodiol epoxide-guanosine adduct. J Am Chem Soc 98: 5714-5715.
21 Jeng. HA; Yordt. D; Davis. S; Swanson. JR. (2013). Assessment of alteration of reproductive system in vivo
22 induced by subchronic exposure to benzo(a)pyrene via oral administration. Environ Toxicol.
23 http://dx.doi.org/10.1002/tox.21889
24 Jensen, TK; Henriksen, TB; Hjollund, NH; Scheike, T; Kolstad, H; Giwercman, A; Ernst, E; Bonde, JP;
25 Skakkebaek, NE; Olsen, J. (1998). Adult and prenatal exposures to tobacco smoke as risk
26 indicators of fertility among 430 Danish couples. Am J Epidemiol 148: 992-997.
27 Jerina, DM; Chadha, A; Cheh, AM; Schurdak, ME; Wood, AW; Sayer, JM. (1991). Covalent bonding of
28 bay-region diol epoxides to nucleic acids [Review]. Adv Exp Med Biol 283: 533-553.
29 Jiang, H; Gelhaus, SL; Mangal, D; Harvey, RG; Blair, IA; Penning, TM. (2007). Metabolism of
30 benzo[a]pyrene in human bronchoalveolar H358 cells using liquid chromatography-mass
31 spectrometry. Chem Res Toxicol 20: 1331-1341. http://dx.doi.org/10.1021/tx700107z
32 Jiang, H; Shen, YM; Quinn, AM; Penning, TM. (2005). Competing roles of cytochrome P450 1A1/1B1 and
33 aldo-keto reductase 1A1 in the metabolic activation of (+/-)-7,8-Dihydroxy-7,8-dihydro-
34 benzo[a]pyrene in human bronchoalveolar cell extracts. Chem Res Toxicol 18: 365-374.
35 http://dx.doi.org/10.1021/tx0497245
36 Jules, GE; Pratap, S; Ramesh, A; Hood, DB. (2012). In utero exposure to benzo(a)pyrene predisposes
37 offspring to cardiovascular dysfunction in later-life. Toxicology 295: 56-67.
38 http://dx.doi.0rg/10.1016/i.tox.2012.01.017
39 Kao. J; Patterson. FK; Hall. J. (1985). Skin penetration and metabolism of topically applied chemicals in
40 six mammalian species, including man: an in vitro study with benzo[a]pyrene and testosterone.
41 Toxicol Appl Pharmacol 81: 502-516.
42 Kapitulnik. J; Wislocki. PG; Levin. W; Yagi. H; Jerina. DM; Conney. AH. (1978). Tumorigenicity studies with
43 diol-epoxides of benzo(a)pyrene which indicate that "(+/-)"-trans-7"beta",8"alpha"-dihydroxy-
44 9"alpha" 10"alpha"-epoxy-7,8,9,10-tetrahydrobenzo(a)pyrene is an ultimate carcinogen in
45 newborn mice. Cancer Res 38: 354-358.
46 Kappes, U; Schliemann-Willers, S; Bankova, L; Heinemann, C; Fischer, TW; Ziemer, M; Schubert, H;
47 Norgauer, J; Fluhr, JW; Eisner, P. (2004). The quality of human skin xenografts on SCID mice: A
This document is a draft for review purposes only and does not constitute Agency policy.
R-13 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 noninvasive bioengineering approach. Br J Dermatol 151: 971-976.
2 http://dx.doi.0rg/10.llll/i.1365-2133.2004.06191.x
3 Karakaya, A; Ates, I; Yucesoy, B. (2004). Effects of occupational polycyclic aromatic hydrocarbon
4 exposure on T-lymphocyte functions and natural killer cell activity in asphalt and coke oven
5 workers. Hum Exp Toxicol 23: 317-322.
6 Karakaya, A; Yucesoy, B; Turhan, A; Erdem, O; Burgaz, S; Karakaya, AE. (1999). Investigation of some
7 immunological functions in a group of asphalt workers exposed to polycyclic aromatic
8 hydrocarbons. Toxicology 135: 43-47.
9 Karlehagen, S; Andersen, A; Ohlson, CG. (1992). Cancer incidence among creosote-exposed workers.
10 Scand J Work Environ Health 18: 26-29. http://dx.doi.org/10.5271/siweh.1612
11 Kee, K; Flores, M; Cedars, Ml; Reijo Pera, RA. (2010). Human primordial germ cell formation is
12 diminished by exposure to environmental toxicants acting through the AHR signaling pathway.
13 Toxicol Sci 117: 218-224. http://dx.doi.org/10.1093/toxsci/kfql79
14 Keohavong, P; Lan, Q; Gao, WM; DeMarini, DM; Mass, MJ; Li, XM; Roop, BC; Weissfeld, J; Tian, D;
15 Mumford, JL. (2003). K-ras mutations in lung carcinomas from nonsmoking women exposed to
16 unvented coal smoke in China. Lung Cancer 41: 21-27. http://dx.doi.org/10.1016/S0169-
17 5002(03)00125-9
18 Ketkar, M; Reznik, G; Schneider, P; Mohr, U. (1978). Investigations on the carcinogenic burden by air
19 pollution in man. Intratracheal instillation studies with benzo(a)pyrene in bovine serum albumin
20 in Syrian hamsters. Cancer Lett 4: 235-239. http://dx.doi.org/10.1016/S0304-3835(78)94787-0
21 Klemme. JC; Mukhtar. H; Elmets. CA. (1987). Induction of contact hypersensitivity to
22 dimethylbenz(a)anthracene and benzo(a)pyrene in C3H/HeN mice. Cancer Res 47: 6074-6078.
23 Knafla. A; Petrovic. S; Richardson. M; Campbell. J; Rowat. C. (2011). Development and application of a
24 skin cancer slope factor for exposures to benzo[a]pyrene in soil. Regul Toxicol Pharmacol 59:
25 101-110. http://dx.doi.0rg/10.1016/i.vrtph.2010.09.011
26 Knuckles, ME; Inyang, F; Ramesh, A. (2001). Acute and subchronic oral toxicities of benzo[a]pyrene in F-
27 344 rats. Toxicol Sci 61: 382-388.
28 Kolb, B; Whishaw, IQ. (1989). Plasticity in the neocortex: mechanisms underlying recovery from early
29 brain damage. Prog Neurobiol 32: 235-276.
30 Kong, LY; Luster, Ml; Dixon, D; O'Grady, J; Rosenthal, GJ. (1994). Inhibition of lung immunity after
31 intratracheal instillation of benzo(a)pyrene. Am J Respir Crit Care Med 150: 1123-1129.
32 Koreeda, M; Moore, PD; Wislocki, PG; Levin, W; Yagi, H; Jerina, DM. (1978). Binding of benzo[a]pyrene
33 7,8-diol-9,10-epoxides to DNA, RNA, and protein of mouse skin occurs with high
34 stereoselectivity. Science 199: 778-781.
35 Kristensen, P; Eilertsen, E; Einarsdottir, E; Haugen, A; Skaug, V; Ovrebo, S. (1995). Fertility in mice after
36 prenatal exposure to benzo[a]pyrene and inorganic lead. Environ Health Perspect 103: 588-590.
37 http://dx.doi.org/10.1289/ehp.95103588
38 Kroese. ED; Muller. JJA; Mohn. GR; Dortant. PM; Wester. PW. (2001). Tumorigenic effects in Wistar rats
39 orally administered benzo[a]pyrene for two years (gavage studies): Implications for human
40 cancer risks associated with oral exposure to polycyclic aromatic hydrocarbons. (658603 010).
41 Bilthoven, The Netherlands: National Institute for Public Health and the Environment (RIVM).
42 http://www.rivm.nl/bibliotheek/rapporten/658603010.pdf
43 Kummer, V; Maskova, J; Zraly, Z; Matiasovic, J; Faldyna, M. (2007). Effect of postnatal exposure to
44 benzo[a]pyrene on the uterus of immature rats. Exp Toxicol Pathol 59: 69-76.
45 http://dx.doi.0rg/10.1016/i.etp.2007.02.007
46 Kummer, V; Maskova, J; Zraly, Z; Neca, J; Simeckova, P; Vondracek, J; Machala, M. (2008). Estrogenic
47 activity of environmental polycyclic aromatic hydrocarbons in uterus of immature Wistar rats.
48 Toxicol Lett 180: 212-221. http://dx.doi.Org/10.1016/i.toxlet.2008.06.862
This document is a draft for review purposes only and does not constitute Agency policy.
R-14 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Kuper, CF; Beems, RB; Bruijntjes, JP; Schuurman, HJ; Vos, JG. (1992). Normal development, growth, and
2 aging of the thymus. In U Mohr; DL Dungworth; CC Capen (Eds.), Pathobiology of the aging rat,
3 volume 1 (pp. 25-49). Washington, DC: ILSI press.
4 Kuper, CF; De Heer, E; Van Loveren, H; Vos, JG. (2002). Immune system. In MA Wallig (Ed.), Handbook of
5 toxicologic pathology (2nd ed., pp. 585-644). San Diego, CA: Academic Press.
6 Landreth, KS. (2002). Critical windows in development of the rodent immune system [Review]. Hum Exp
7 Toxicol 21: 493-498.
8 Landreth, KS; Dodson, SVM. (2005). Development of the rodent immune system. In SD Holladay (Ed.),
9 Developmental immunotoxicology (pp. 3-19). Boca Raton, FL: CRC Press.
10 http://dx.doi.org/10.1201/9781420024036.ptl
11 Lavoie, EJ; Amin, S; Hecht, SS; Furuya, K; Hoffmann, D. (1982). Tumour initiating activity of dihydrodiols
12 of benzo[b]fluoranthene, benzo[j]fluoranthene, and benzo[k]fluoranthene. Carcinogenesis 3:
13 49-52. http://dx.doi.0rg/10.1093/carcin/3.l.49
14 Lavoie, EJ; Braley, J; Rice, JE; Rivenson, A. (1987). Tumorigenic activity of non-alternant polynuclear
15 aromatic hydrocarbons in newborn mice. Cancer Lett 34: 15-20.
16 Lavoie, EJ; Cai, ZW; Meschter, CL; Weyand, EH. (1994). Tumorigenic activity of fluoranthene, 2-
17 methylfluoranthene and 3-methylfluoranthene in newborn CD-I mice. Carcinogenesis 15: 2131-
18 2135. http://dx.doi.org/10.1093/carcin/15.10.2131
19 Lavoue. J; Gerin. M; Cote. J; Lapointe. R. (2007). Mortality and cancer experience of Quebec aluminum
20 reduction plant workers. Part I: The reduction plants and coal tar pitch volatile (CTPV) exposure
21 assessment. J Occup Environ Med 49: 997-1008.
22 http://dx.doi.org/10.1097/JOM.Ob013e3181484cf3
23 Letzel. S; Drexler. H. (1998). Occupationally related tumors in tar refinery workers. J Am Acad Dermatol
24 39: 712-720. http://dx.doi.org/10.1016/S0190-9622(98)70043-X
25 Levin, W; Wood, AW; Wislocki, PG; Kapitulnik, J; Yagi, H; Jerina, DM; Conney, AH. (1977). Carcinogenicity
26 of benzo-ring derivatives of benzo(a)pyrene on mouse skin. Cancer Res 37: 3356-3361.
27 Li, D; Firozi, PF; Wang, LE; Bosken, CH; Spitz, MR; Hong, WK; Wei, Q. (2001). Sensitivity to DNA damage
28 induced by benzo(a)pyrene diol epoxide and risk of lung cancer: a case-control analysis. Cancer
29 Res 61: 1445-1450.
30 Li, Z; Chadalapaka, G; Ramesh, A; Khoshbouei, H; Maguire, M; Safe, S; Rhoades, RE; Clark, R; Jules, G;
31 Mccallister, M; Aschner, M; Hood, DB. (2012). PAH particles perturb prenatal processes and
32 phenotypes: protection from deficits in object discrimination afforded by dampening of brain
33 oxidoreductase following in utero exposure to inhaled benzo(a)pyrene. Toxicol Sci 125: 233-247.
34 http://dx.doi.org/10.1093/toxsci/kfr261
35 Liang, J; Zhu, H; Li, C; Ding, Y; Zhou, Z; Wu, Q. (2012). Neonatal exposure to benzo[a]pyrene decreases
36 the levels of serum testosterone and histone H3K14 acetylation of the StAR promoter in the
37 testes of SD rats. Toxicology 302: 285-291. http://dx.doi.Org/10.1016/i.tox.2012.08.010
38 Lindelof. B; Sigurgeirsson. B. (1993). PUVA and cancer: a case-control study. Br J Dermatol 129: 39-41.
39 http://dx.doi.0rg/10.llll/i.1365-2133.1993.tb03308.x
40 Liu. M; Grant. SG; Macina. OT; Klopman. G; Rosenkranz. HS. (1997). Structural and mechanistic bases for
41 the induction of mitotic chromosomal loss and duplication (malsegregation1) in the yeast
42 Saccharomyces cerevisiae: Relevance to human Carcinogenesis and developmental toxicology.
43 Mutat Res-Fundam Mol Mech Mutagen 374: 209-231. http://dx.doi.org/10.1016/S0027-
44 5107(96)00236-9
45 Liu, S; Abdelrahim, M; Khan, S; Ariazi, E; Jordan, VC; Safe, S. (2006). Aryl hydrocarbon receptor agonists
46 directly activate estrogen receptor alpha in MCF-7 breast cancer cells. Biol Chem 387: 1209-
47 1213. http://dx.doi.org/10.1515/BC.2006.149
This document is a draft for review purposes only and does not constitute Agency policy.
R-15 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Liu, SH; Wang, JH; Chuu, JJ; Lin-Shiau, SY. (2002). Alterations of motor nerve functions in animals
2 exposed to motorcycle exhaust. J Toxicol Environ Health A 65: 803-812.
3 http://dx.doi.org/10.1080/00984100290071144
4 Liu, Z; Muehlbauer, KR; Schmeiser, HH; Hergenhahn, M; Belharazem, D; Hollstein, MC. (2005). p53
5 mutations in benzo(a)pyrene-exposed human p53 knock-in murine fibroblasts correlate with
6 p53 mutations in human lung tumors. Cancer Res 65: 2583-2587.
7 http://dx.doi.org/10.1158/0008-5472.CAN-04-3675
8 Luebke, RW; Chen, DH; Dietert, R; Yang, Y; King, M; Luster, Ml. (2006). The comparative immunotoxicity
9 of five selected compounds following developmental or adult exposure [Review]. J Toxicol
10 Environ Health B Crit Rev 9: 1-26. http://dx.doi.org/10.1080/15287390500194326
11 Lummus, ZL; Henningsen, G. (1995). Modulation of T-cell ontogeny by transplacental benzo(a)pyrene.
12 International Journal of Immunopharmacology 17: 339-350.
13 Luster, Ml; Portier, C; Pait, DG; White, KL, Jr; Gennings, C; Munson, AE; Rosenthal, GJ. (1992). Risk
14 assessment in immunotoxicology: I: Sensitivity and predictability of immune tests. Fundam Appl
15 Toxicol 18: 200-210. http://dx.doi.org/10.1016/0272-0590(92)90047-L
16 Lyte, M; Bick, PH. (1985). Differential immunotoxic effects of the environmental chemical
17 benzo[a]pyrene in young and aged mice. Mech Ageing Dev 30: 333-341.
18 http://dx.doi.org/10.1016/0047-6374(85)90122-8
19 Ma. Q; Lu. AY. (2007). CYP1A induction and human risk assessment: an evolving tale of in vitro and in
20 vivo studies [Review]. Drug Metab Dispos 35: 1009-1016.
21 http://dx.doi.org/10.1124/dmd.107.015826
22 Maciel. ES; Biasibetti. R; Costa. AP; Lunardi. P; Schunck. RV; Becker. GC; Arbo. MD; Dallegrave. E;
23 Goncalves. CA; Saldiva. PH; Garcia. SC; Leal. RB; Leal. MB. (2014). Subchronic oral administration
24 of Benzo[a]pyrene impairs motor and cognitive behavior and modulates S100B levels and
25 MAPKs in rats. Neurochem Res 39: 731-740. http://dx.doi.org/10.1007/sll064-014-1261-y
26 Mackall, CL; Fleisher, TA; Brown, MR; Andrich, MP; Chen, CC; Feuerstein, IM; Horowitz, ME; Magrath, IT;
27 Shad, AT; Steinberg, SM. (1995). Age, thymopoiesis, and CD4+ T-lymphocyte regeneration after
28 intensive chemotherapy. N Engl J Med 332: 143-149.
29 http://dx.doi.org/10.1056/NEJM199501193320303
30 Mackenzie, KM; Angevine, DM. (1981). Infertility in mice exposed in utero to benzo(a)pyrene. Biol
31 Reprod 24: 183-192.
32 Mahler, BJ; Van Metre, PC; Bashara, TJ; Wilson, JT; Johns, DA. (2005). Parking lot sealcoat: an
33 unrecognized source of urban polycyclic aromatic hydrocarbons. Environ Sci Technol 39: 5560-
34 5566. http://dx.doi.org/10.1021/es0501565
35 Malaveille, C; Kuroki, T; Sims, P; Grover, PL; Bartsch, H. (1977). Mutagenicity of isomeric diol-epoxides of
36 benzo[a]pyrene and benz[a]anthracene in S. typhimurium TA98 and TA100 and in V79 Chinese
37 hamster cells. Mutat Res 44: 313-326.
38 Mancini. R; Romano. G; Sgambato. A; Flamini. G; Giovagnoli. MR; Boninsegna. A; Carraro. C; Vecchione.
39 A; Cittadini. A. (1999). Polycyclic aromatic hydrocarbon-DNA adducts in cervical smears of
40 smokers and nonsmokers. Gynecol Oncol 75: 68-71. http://dx.doi.org/10.1006/gyno.1999.5525
41 Marshall. G; Vousden. KH; Phillips. DH. (1984). Activation of c-Ha-ras-1 proto-oncogene by in vitro
42 modification with a chemical carcinogen, benzo(a)pyrene diol-epoxide. Nature 310: 586-589.
43 http://dx.doi.org/10.1038/310586aO
44 Mass, MJ; Jeffers, AJ; Ross, JA; Nelson, G; Galati, AJ; Stoner, GD; Nesnow, S. (1993). Ki-ras oncogene
45 mutations in tumors and DNA adducts formed by benz[j]aceanthrylene and benzo[a]pyrene in
46 the lungs of strain A/J mice. Mol Carcinog 8: 186-192.
47 http://dx.doi.org/10.1002/mc.2940080309
This document is a draft for review purposes only and does not constitute Agency policy.
R-16 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Matiasovic, J; Leva, L; Maskova, J; Kummer, V; Faldyna, M. (2008). Effects of postnatal exposure to
2 benzo[a]pyrene on the immunity of immature rats. Veterinarni Medicina 53: 93-100.
3 Matikainen, T; Perez, Gl; Jurisicova, A; Pru, JK; Schlezinger, JJ; Ryu, HY; Laine, J; Sakai, T; Korsmeyer, SJ;
4 Casper, RF; Sherr, DH; Tilly, JL. (2001). Aromatic hydrocarbon receptor-driven Bax gene
5 expression is required for premature ovarian failure caused by biohazardous environmental
6 chemicals. Nat Genet 28: 355-360. http://dx.doi.org/10.1038/ng575
7 Matikainen, TM; Moriyama, T; Morita, Y; Perez, Gl; Korsmeyer, SJ; Sherr, DH; Tilly, JL. (2002). Ligand
8 activation of the aromatic hydrocarbon receptor transcription factor drives Bax-dependent
9 apoptosis in developing fetal ovarian germ cells. Endocrinology 143: 615-620.
10 Mattison, PR; Thorgeirsson, SS. (1977). Genetic differences in mouse ovarian metabolism of benzo
11 [a]pyrene and oocyte toxicity. Biochem Pharmacol 26: 909-912.
12 Mattison, PR; Thorgeirsson, SS. (1979). Ovarian aryl hydrocarbon hydroxylase activity and primordial
13 oocyte toxicity of polycyclic aromatic hydrocarbons in mice. Cancer Res 39: 3471-3475.
14 Mattison, PR; White, 1MB; Nightingale, MR. (1980). The effect of benzo(a)pyrene on fertility, primordial
15 oocyte number, and ovarian response to pregnant mare's serum gonadotropin. Pediatr
16 Pharmacol 1: 143-151.
17 Matullo, G; Peluso, M; Polidoro, S; Guarrera, S; Munnia, A; Krogh, V; Masala, G; Berrino, F; Panico, S;
18 Tumino, R; Vineis, P; Palli, D. (2003). Combination of DNA repair gene single nucleotide
19 polymorphisms and increased levels of DNA adducts in a population-based study. Cancer
20 Epidemiol Biomarkers Prev 12: 674-677.
21 Mauderly. JL. (1997). Relevance of particle-induced rat lung tumors for assessing lung carcinogenic
22 hazard and human lung cancer risk [Review]. Environ Health Perspect 105 (Suppl 5): 1337-1346.
23 Maughan. WZ; Muller. SA; Perry. HO; Pittelkow. MR; O'Brien. PC. (1980). Incidence of skin cancers in
24 patients with atopic dermatitis treated with ocal tar: A 25-year follow-up study. J Am Acad
25 Dermatol 3: 612-615.
26 McCallister, MM; Maguire, M; Ramesh, A; Qiao, AM; Sheng, L; Khoshbouei, H; Aschner, M; Ebner, FF;
27 Hood, DB. (2008). Prenatal exposure to benzo(a)pyrene impairs later-life cortical neuronal
28 function. Neurotoxicology 29: 846-854. http://dx.doi.Org/10.1016/i.neuro.2008.07.008
29 McConnell, EE; Solleveld, HA; Swenberg, JA; Boorman, GA. (1986). Guidelines for combining neoplasms
30 for evaluation of rodent carcinogenesis studies. J Natl Cancer Inst 76: 283-289.
31 Mccoull, KD; Rindgen, D; Blair, IA; Penning, TM. (1999). Synthesis and characterization of polycyclic
32 aromatic hydrocarbon o-quinone depurinating N7-guanine adducts. Chem Res Toxicol 12: 237-
33 246. http://dx.doi.org/10.1021/tx980182z
34 Melikian, AA; Sun, P; Prokopczyk, B; EI-Bayoumy, K; Hoffmann, D; Wang, X; Waggoner, S. (1999).
35 Identification of benzo[a]pyrene metabolites in cervical mucus and DNA adducts in cervical
36 tissues in humans by gas chromatography-mass spectrometry. Cancer Lett 146: 127-134.
37 http://dx.doi.org/10.1016/S0304-3835(99)00203-7
38 Midgette. AS; Baron. JA. (1990). Cigarette smoking and the risk of natural menopause [Review].
39 Epidemiology 1: 474-480.
40 Mielzynska. D; Siwinska. E; Kapka. L; Szyfter. K; Knudsen. LE; Merlo. DF. (2006). The influence of
41 environmental exposure to complex mixtures including PAHs and lead on genotoxic effects in
42 children living in Upper Silesia, Poland. Mutagenesis 21: 295-304.
43 http://dx.doi.org/10.1093/mutage/gel037
44 Miller, KP; Ramos, KS. (2001). Impact of cellular metabolism on the biological effects of benzo[a]pyrene
45 and related hydrocarbons [Review]. Drug Metab Rev 33: 1-35. http://dx.doi.org/10.1081/DMR-
46 100000138
This document is a draft for review purposes only and does not constitute Agency policy.
R-17 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Miller, MM; Plowchalk, PR; Weitzman, GA; London, SIM; Mattison, DR. (1992). The effect of
2 benzo(a)pyrene on murine ovarian and corpora lutea volumes. Am J Obstet Gynecol 166: 1535-
3 1541.
4 Mitropoulos, P; Norman, R. (2005). Occupational nonsolar risk factors of squamous cell carcinoma of the
5 skin: a population-based case-controlled study. Dermatol Online J 11: 5.
6 Mohamed. E; Song. WH; Oh. SA; Park. YJ; You. YA; Lee. S; Choi. JY; Kim. YJ; Jo. I; Pang. MG. (2010). The
7 transgenerational impact of benzo(a)pyrene on murine male fertility. Hum Reprod 25: 2427-
8 2433. http://dx.doi.org/10.1093/humrep/deq205
9 Moody, RP; Chu, I. (1995). Dermal exposure to environmental contaminants in the Great Lakes [Review].
10 Environ Health Perspect 103 Suppl 9: 103-114.
11 Moody, RP; Joncas, J; Richardson, M; Chu, I. (2007). Contaminated soils (I): In vitro dermal absorption of
12 benzo[a]pyrene in human skin. J Toxicol Environ Health A 70: 1858-1865.
13 http://dx.doi.org/10.1080/15287390701459296
14 Moolgavkar, SH; Knudson, AG, Jr. (1981). Mutation and cancer: A model for human carcinogenesis. J
15 Natl Cancer Inst 66: 1037-1052.
16 Moolgavkar, SH; Venzon, DJ. (1979). Two-event models for carcinogenesis: Incidence curves for
17 childhood and adult tumors. Math Biosci 47: 55-77. http://dx.doi.org/10.1016/0025-
18 5564(79)90005-1
19 Moulin. JJ; Wild. P; Mur. JM; Lafontaine. M; Lefer. M; Mercier-Gallav. M; Villemot. P; Whebi. V; Coulon.
20 JP. (1989). Risk of lung, larynx, pharynx and buccal cavity cancers among carbon electrode
21 manufacturing workers. Scand J Work Environ Health 15: 30-37.
22 http://dx.doi.org/10.5271/siweh.1885
23 Munson. AE; White. KL; Lvsy. HH. (1985). Effects of Subchronic Fourteen Day Exposure to Benzopyrene
24 in B6C3F1 Female Mice on Host Resistance. Submitted under TSCA Section FYI.
25 Munson, AE; White, KU. (1983). Effects of subchronic 14-day exposure to benzopyrene in b6c3fl male
26 and female mice on specific immunological parameters. American Petroleum Institute.
27 N'Diaye, M; Le Ferrec, E; Lagadic-Gossmann, D; Corre, S; Gilot, D; Lecureur, V; Monteiro, P; Rauch, C;
28 Galibert, MD; Fardel, O. (2006). Aryl hydrocarbon receptor- and calcium-dependent induction of
29 the chemokine CCL1 by the environmental contaminant benzo[a]pyrene. J Biol Chem 281:
30 19906-19915. http://dx.doi.org/10.1074/ibc.M601192200
31 Nanez, A; Ramos, IN; Ramos, KS. (2011). A mutant Ahr allele protects the embryonic kidney from
32 hydrocarbon-induced deficits in fetal programming. Environ Health Perspect 119: 1745-1753.
33 http://dx.doi.org/10.1289/ehp.1103692
34 Naslund, I; Rubio, CA; Auer, GU. (1987). Nuclear DNA changes during pathogenesis of squamous
35 carcinoma of the cervix in 3,4-benzopyrene-treated mice. Anal Quant Cytol Histol 9: 411-418.
36 Naufal. Z; Zhiwen. L; Zhu. L; Zhou. GD; Mcdonald. T; He. LY; Mitchell. L; Ren. A; Zhu. H; Finnell. R;
37 Donnelly, KC. (2010). Biomarkers of exposure to combustion by-products in a human population
38 in Shanxi, China. J Expo Sci Environ Epidemiol 20: 310-319.
39 http://dx.doi.org/10.1038/ies.2009.19
40 Neal. J; Rigdon. RH. (1967). Gastric tumors in mice fed benzo(a)pyrene: A quantitative study. Tex Rep
41 Biol Med 25: 553-557.
42 Neal. MS; Mulligan Tuttle. AM; Casper. RF; Lagunov. A; Foster. WG. (2010). Aryl hydrocarbon receptor
43 antagonists attenuate the deleterious effects of benzo[a]pyrene on isolated rat follicle
44 development. Reprod Biomed Online 21: 100-108.
45 http://dx.doi.0rg/10.1016/i.rbmo.2010.03.025
46 Neal, MS; Zhu, J; Foster, WG. (2008). Quantification of benzo[a]pyrene and other PAHs in the serum and
47 follicular fluid of smokers versus non-smokers. Reprod Toxicol 25: 100-106.
48 http://dx.doi.0rg/10.1016/i.reprotox.2007.10.012
This document is a draft for review purposes only and does not constitute Agency policy.
R-18 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Neal, MS; Zhu, J; Holloway, AC; Foster, WG. (2007). Follicle growth is inhibited by benzo-[a]-pyrene, at
2 concentrations representative of human exposure, in an isolated rat follicle culture assay. Hum
3 Reprod 22: 961-967. http://dx.doi.org/10.1093/humrep/del487
4 Nebert, DW; Puga, A; Vasiliou, V. (1993). Role of the Ah receptor and the dioxin-inducible [Ah] gene
5 battery in toxicity, cancer, and signal transduction [Review]. Ann N Y Acad Sci 685: 624-640.
6 http://dx.doi.0rg/10.llll/i.1749-6632.1993.tb35928.x
7 Nesnow, S; Mass, MJ; Ross, JA; Galati, AJ; Lambert, GR; Gennings, C; Carter, WH; Stoner, GD. (1998a).
8 Lung tumorigenic interactions in strain A/J mice of five environmental polycyclic aromatic
9 hydrocarbons. Environ Health Perspect 106 Suppl 6: 1337-1346.
10 Nesnow, S; Ross, JA; Mass, MJ; Stoner, GD. (1998b). Mechanistic relationships between DNA adducts,
11 oncogene mutations, and lung tumorigenesis in strain A mice. Exp Lung Res 24: 395-405.
12 Nesnow, S; Ross, JA; Stoner, GD; Mass, MJ. (1995). Mechanistic linkage between DNA adducts,
13 mutations in oncogenes and tumorigenesis of carcinogenic environmental polycyclic aromatic
14 hydrocarbons in strain A/J mice. Toxicology 105: 403-413. http://dx.doi.org/10.1016/0300-
15 483X(95)03238-B
16 Nesnow, S; Ross, JA; Stoner, GD; Mass, MJ. (1996). Tumorigenesis of Carcinogenic Environmental
17 Polycyclic Aromatic Hydrocarbons in Strain A/J Mice: Linkage to DNA Adducts and Mutations in
18 Oncogenes. Polycycl Aromat Compd 10: 259-266.
19 http://dx.doi.org/10.1080/10406639608034705
20 Nesnow. S; Triplett. LL; Slaga. TJ. (1983). Mouse skin tumor initiation-promotion and complete
21 carcinogenesis bioassays: mechanisms and biological activities of emission samples. Environ
22 Health Perspect 47: 255-268.
23 Nettesheim. P; Griesemer. RA; Martin. DH; Caton. JE. Jr. (1977). Induction of preneoplastic and
24 neoplastic lesions in grafted rat tracheas continuously exposed to benzo(a)pyrene. Cancer Res
25 37: 1272-1278.
26 Newbold, RF; Brookes, P. (1976). Exceptional mutagenicity of a benzo(a)pyrene diol epoxide in cultured
27 mammalian cells. Nature 261: 52-54. http://dx.doi.org/10.1038/261052aO
28 Nikonova, TV. (1977). [Transplacental effect of benz(a)pyrene and pyrene]. Biull Eksp Biol Med 84: 88-
29 91.
30 NIOSH (National Institute for Occupational Safety and Health). (1989). Assessment of the
31 cocarcinogenic/promoting activity of asphalt fumes.
32 http://www.ntis.gov/search/product.aspx?ABBR=PB91110213
33 Niu, Oj Zhang, H; Li, X; Li, M. (2010). Benzo[a]pyrene-induced neurobehavioral function and
34 neurotransmitter alterations in coke oven workers. Occup Environ Med 67: 444-448.
35 http://dx.doi.org/10.1136/oem.2009.047969
36 NRC (National Research Council). (1983). Risk assessment in the federal government: Managing the
37 process. Washington, DC: National Academies Press.
38 http://www.nap.edu/openbook.php7record id=366&page=Rl
39 NRC (National Research Council). (2009). Science and decisions: Advancing risk assessment. Washington,
40 DC: National Academies Press, http://www.nap.edu/catalog/12209.html
41 NRC (National Research Council). (2011). Review of the Environmental Protection Agency's draft IRIS
42 assessment of formaldehyde. Washington, DC: National Academies Press.
43 http://www.nap.edu/catalog/13142.html
44 OECD (Organisation for Economic Co-operation and Development). (2009). Series on testing and
45 assesment: Number 39. Guidance document on acute inhalation toxicity testing [OECD SIDS].
46 (ENV/JM/MONO(2009)28; JT03268144). Paris, Fr.
47 http://search.oecd. org/officialdocuments/displavdocumentpdf/?cote=env/im/mono%282009%
48 2928&doclanguage=en
This document is a draft for review purposes only and does not constitute Agency policy.
R-19 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Okey, AB; Riddick, PS; Harper, PA. (1994). Molecular biology of the aromatic hydrocarbon (dioxin)
2 receptor [Review]. Trends Pharmacol Sci 15: 226-232. http://dx.doi.org/10.1016/0165-
3 6147(94)90316-6
4 Olsen, AK; Andreassen, A; Singh, R; Wiger, R; Duale, N; Farmer, PB; Brunborg, G. (2010). Environmental
5 exposure of the mouse germ line: DNA adducts in spermatozoa and formation of de novo
6 mutations during spermatogenesis. PLoS ONE 5: el!349.
7 http://dx.doi.org/10.1371/iournal.pone.0011349
8 Olsson, A; Kromhout, H; Agostini, M; Hansen, J; Lassen, CF; Johansen, C; Kjaerheim, K; Langard, S;
9 Stucker, I; Ahrens, W; Behrens, T; Lindbohm, ML; Heikkila, P; Heederik, D; Portengen, L; Shaham,
10 J; Ferro, G; de Vocht, F; Burstyn, I; Boffetta, P. (2010). A case-control study of lung cancer nested
11 in a cohort of European asphalt workers. Environ Health Perspect 118: 1418-1424.
12 http://dx.doi.org/10.1289/ehp.0901800
13 ONS (Office for National Statistics). (2003). Cancer statistics registrations: Registrations of cancer
14 diagnosed in 2003, England. London, UK: Palgrave Macmillan.
15 http://www.ons.gov.uk/ons/rel/vsobl/cancer-statistics-registrations--england--series-mbl-/no-
16 -34-2003/index.html
17 Park, JH; Gelhaus, S; Vedantam, S; Oliva, AL; Batra, A; Blair, IA; Troxel, AB; Field, J; Penning, TM. (2008).
18 The pattern of p53 mutations caused by PAH o-quinones is driven by 8-oxo-dGuo formation
19 while the spectrum of mutations is determined by biological selection for dominance. Chem Res
20 Toxicol 21: 1039-1049. http://dx.doi.org/10.1021/tx700404a
21 Park. JH; Troxel. AB; Harvey. RG; Penning. TM. (2006). Polycyclic aromatic hydrocarbon (PAH) o-
22 quinones produced by the aldo-keto-reductases (AKRs) generate abasic sites, oxidized
23 pyrimidines, and 8-oxo-dGuo via reactive oxygen species. Chem Res Toxicol 19: 719-728.
24 http://dx.doi.org/10.1021/tx0600245
25 Pate Capps, N; Stewart, A; Burns, C. (2009). The interplay between secondhand cigarette smoke,
26 genetics, and cervical cancer: a review of the literature [Review]. Biol Res Nurs 10: 392-399.
27 http://dx.doi.org/10.1177/1099800408330849
28 Pavanello, S; Favretto, D; Brugnone, F; Mastrangelo, G; Dal Pra, G; Clonfero, E. (1999).
29 HPLC/fluorescence determination of anti-BPDE-DNA adducts in mononuclear white blood cells
30 from PAH-exposed humans. Carcinogenesis 20: 431-435.
31 Pavanello, S; Pulliero, A; Siwinska, E; Mielzynska, D; Clonfero, E. (2005). Reduced nucleotide excision
32 repair and GSTMl-null genotypes influence anti-B[a]PDE-DNA adduct levels in mononuclear
33 white blood cells of highly PAH-exposed coke oven workers. Carcinogenesis 26: 169-175.
34 http://dx.doi.org/10.1093/carcin/bgh303
35 Pavanello, S; Siwinska, E; Mielzynska, D; Clonfero, E. (2004). GSTM1 null genotype as a risk factor for
36 anti-BPDE-DNA adduct formation in mononuclear white blood cells of coke-oven workers. Mutat
37 Res 558: 53-62. http://dx.doi.Org/10.1016/i.mrgentox.2003.10.019
38 Penit. C; Vasseur. F. (1989). Cell proliferation and differentiation in the fetal and early postnatal mouse
39 thymus. J Immunol 142: 3369-3377.
40 Penning. TM; Burczynski. ME; Hung. CF; Mccoull. KD; Palackal. NT;Tsuruda. LS. (1999). Dihydrodiol
41 dehydrogenases and polycyclic aromatic hydrocarbon activation: generation of reactive and
42 redox active o-quinones [Review]. Chem Res Toxicol 12: 1-18.
43 http://dx.doi.org/10.1021/tx980143n
44 Penning, TM; Ohnishi, ST; Ohnishi, T; Harvey, RG. (1996). Generation of reactive oxygen species during
45 the enzymatic oxidation of polycyclic aromatic hydrocarbon trans-dihydrodiols catalyzed by
46 dihydrodiol dehydrogenase. Chem Res Toxicol 9: 84-92. http://dx.doi.org/10.1021/tx950055s
This document is a draft for review purposes only and does not constitute Agency policy.
R-20 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Perera, F; Li, TY; Lin, C; Tang, D. (2012a). Effects of prenatal polycyclic aromatic hydrocarbon exposure
2 and environmental tobacco smoke on child IQ in a Chinese cohort. Environ Res 114: 40-46.
3 http://dx.doi.0rg/10.1016/i.envres.2011.12.011
4 Perera, FP; Li, Z; Whyatt, R; Hoepner, L; Wang, S; Camann, D; Rauh, V. (2009). Prenatal airborne
5 polycyclic aromatic hydrocarbon exposure and child IQ at age 5 years. Pediatrics 124: e!95-
6 e202. http://dx.doi.org/10.1542/peds.2008-3506
7 Perera, FP; Rauh, V; Whyatt, RM; Tang, D; Tsai, WY; Bernert, JT; Tu, YH; Andrews, H; Barr, DB; Camann,
8 DE; Diaz, D; Dietrich, J; Reyes, A; Kinney, PL. (2005a). A summary of recent findings on birth
9 outcomes and developmental effects of prenatal ETS, PAH, and pesticide exposures [Review].
10 Neurotoxicology 26: 573-587. http://dx.doi.Org/10.1016/i.neuro.2004.07.007
11 Perera, FP; Rauh, V; Whyatt, RM; Tsai, WY; Bernert, JT; Tu, YH; Andrews, H; Ramirez, J; Qu, L; Tang, D.
12 (2004). Molecular evidence of an interaction between prenatal environmental exposure and
13 birth outcomes in a multiethnic population. Environ Health Perspect 112: 626-630.
14 Perera. FP; Tang. D; Rauh. V; Lester. K; Tsai. WY; Tu. YH; Weiss. L; Hoepner. L; King. J; Del Priore. G;
15 Lederman, SA. (2005b). Relationships among polycyclic aromatic hydrocarbon-DNA adducts,
16 proximity to the World Trade Center, and effects on fetal growth. Environ Health Perspect 113:
17 1062-1067.
18 Perera, FP; Tang, D; Wang, S; Vishnevetskv, J; Zhang, B; Diaz, D; Camann, D; Rauh, V. (2012b). Prenatal
19 Polycyclic Aromatic Hydrocarbon (PAH) Exposure and Child Behavior at Age 6-7 Years. Environ
20 Health Perspect 120: 921-926. http://dx.doi.org/10.1289/ehp.1104315
21 Perera. FP; Weinstein. IB. (2000). Molecular epidemiology: Recent advances and future directions
22 [Review]. Carcinogenesis 21: 517-524. http://dx.doi.Org/10.1093/carcin/21.3.517
23 Perrin. J; Tassistro. V; Mandon. M; Grillo. JM; Botta. A; Sari-Minodier. I. (2011a). Tobacco consumption
24 and benzo(a)pyrene-diol-epoxide-DNA adducts in spermatozoa: in smokers, swim-up procedure
25 selects spermatozoa with decreased DNA damage. Fertil Steril 95: 2013-2017.
26 http://dx.doi.0rg/10.1016/i.fertnstert.2011.02.021
27 Perrin, J; Tassistro, V; Paulmyer-Lacroix, O; Courbiere, B; Botta, A; Sari-Minodier, I. (2011b). In smokers,
28 swim-up and discontinuous gradient centrifugation recover spermatozoa with equally lower
29 amounts of DNA damage than spermatozoa obtained from neat semen. Fertil Steril 95: 2680-
30 2682. http://dx.doi.0rg/10.1016/i.fertnstert.2011.04.092
31 Petrie, HT. (2002). Role of thymic organ structure and stromal composition in steady-state postnatal T-
32 cell production [Review]. Immunol Rev 189: 8-19.
33 Pfeifer, GP; Denissenko, MF; Olivier, M; Tretvakova, N; Hecht, SS; Hainaut, P. (2002). Tobacco smoke
34 carcinogens, DNA damage and p53 mutations in smoking-associated cancers [Review].
35 Oncogene 21: 7435-7451. http://dx.doi.org/10.1038/si.onc.1205803
36 Pfeifer, GP; Hainaut, P. (2003). On the origin of G —> T transversions in lung cancer [Review]. Mutat Res
37 526: 39-43. http://dx.doi.org/10.1016/S0027-5107(03)00013-7
38 Pfeiffer. CA; Allen. E. (1948). Attempts to produce cancer in rhesus monkeys with carcinogenic
39 hydrocarbons and estrogens. Cancer Res 8: 97-127.
40 Pfeiffer. EH. (1977). Oncogenic interaction of carcinogenic and non-carcinogenic polycyclic aromatic
41 hydrocarbons in mice. In Air pollution and cancer in man. Lyon, France: IARC.
42 Phillips. PH. (2002). Smoking-related DNA and protein adducts in human tissues [Review].
43 Carcinogenesis 23: 1979-2004.
44 Pittelkow, MR; Perry, HO; Muller, SA; Maughan, WZ; O'Brien, PC. (1981). Skin cancer in patients with
45 psoriasis treated with coal tar: A 25-year follow-up study. Arch Dermatol 117: 465-468.
46 Poel, WE. (1959). Effect of carcinogenic dosage and duration of exposure on skin-tumor induction in
47 mice. J Natl Cancer Inst 22: 19-43.
This document is a draft for review purposes only and does not constitute Agency policy.
R-21 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Poel, WE. (1963). Skin as a test site for the bioassay of carcinogens and carcinogen precursors [IARC
2 Monograph]. In Biology of cutaneous cancers (pp. 611-631). Poel, WE.
3 Postlind, H; Vu, TP; Tukey, RH; Quattrochi, LC. (1993). Response of human CYPl-luciferase plasmids to
4 2,3,7,8-tetrachlorodibenzo-p-dioxin and polycyclic aromatic hydrocarbons. Toxicol Appl
5 Pharmacol 118: 255-262. http://dx.doi.org/10.1006/taap.1993.1031
6 Pru, JK; Kaneko-Tarui, T; Jurisicova, A; Kashiwagi, A; Selesniemi, K; Tilly, JL. (2009). Induction of
7 proapoptotic gene expression and recruitment of p53 herald ovarian follicle loss caused by
8 polycyclic aromatic hydrocarbons. Reproductive Sciences 16: 347-356.
9 http://dx.doi.org/10.1177/1933719108327596
10 Puisieux, A; Lim, S; Groopman, J; Ozturk, M. (1991). Selective targeting of p53 gene mutational hotspots
11 in human cancers by etiologically defined carcinogens. Cancer Res 51: 6185-6189.
12 Pukkala, E. (1995). Cancer risk by social class and occupation : a survey of 109,000 cancer cases among
13 Finns of working age. In J Wahrendorf (Ed.). Basel, Switzerland: Karger.
14 Pukkala, E; Martinsen, J; Lynge, E; Gunnarsdottir, H; Sparen, P; Trvggvadottir, L; Weiderpass, E;
15 Kjaerheim, K. (2009). Occupation and cancer - follow-up of 15 million people in five Nordic
16 countries. Acta Oncol 48: 646-790. http://dx.doi.org/10.1080/02841860902913546
17 OJu, C; Cheng, S; Xia, Y; Peng, B; Tang, Q; Tu, B. (2011). Effects of subchronic benzo(a)pyrene exposure
18 on neurotransmitter receptor gene expression in the rat hippocampus related with spatial
19 learning and memory change. Toxicology 289: 83-90.
20 http://dx.doi.0rg/10.1016/i.tox.2011.07.012
21 Oju. C; Peng. B; Cheng. S; Xia. Y; Tu. B. (2013). The effect of occupational exposure to benzo[a]pyrene on
22 neurobehavioral function in coke oven workers. Am J Ind Med 56: 347-355.
23 http://dx.doi.org/10.1002/aiim.22119
24 Ramesh, A; Inyang, F; Lunstra, DP; Niaz, MS; Kopsombut, P; Jones, KM; Hood, DB; Hills, ER; Archibong,
25 AE^pOOS). Alteration of fertility endpoints in adult male F-344 rats by subchronic exposure to
26 inhaled benzo(a)pyrene. Exp Toxicol Pathol 60: 269-280.
27 http://dx.doi.0rg/10.1016/i.etp.2008.02.010
28 Ramesh, A; Walker, SA; Hood, DB; Guillen, MD; Schneider, K; Weyand, EH. (2004). Bioavailability and risk
29 assessment of orally ingested polycyclic aromatic hydrocarbons [Review]. Int J Toxicol 23: 301-
30 333. http://dx.doi.org/10.1080/10915810490517063
31 Ramos, KS; Moorthy, B. (2005). Bioactivation of polycyclic aromatic hydrocarbon carcinogens within the
32 vascular wall: implications for human atherogenesis [Review]. Drug Metab Rev 37: 595-610.
33 http://dx.doi.org/10.1080/03602530500251253
34 Rask-Nielsen, R. (1950). On the susceptibility of the thymus, lung, subcutaneous and mammary tissues in
35 strain street mice to direct application of small doses of four different carcinogenic
36 hydrocarbons. Br J Cancer 4: 108-116.
37 Raveh, D; Slaga, TJ; Huberman, E. (1982). Cell-mediated mutagenesis and tumor-initiating activity of the
38 ubiquitous polycyclic hydrocarbon, cyclopenta[c,d]pyrene. Carcinogenesis 3: 763-766.
39 http://dx.doi.Org/10.1093/carcin/3.7.763
40 Raychoudhury. SS; Kubinski. D. (2003). Polycyclic aromatic hydrocarbon-induced cytotoxicity in cultured
41 rat Sertoli cells involves differential apoptotic response. Environ Health Perspect 111: 33-38.
42 Revel. A; Raanani. H; Younglai. E; Xu. J; Han. R; Savouret. JF; Casper. RF. (2001). Resveratrol, a natural
43 aryl hydrocarbon receptor antagonist, protects sperm from DNA damage and apoptosis caused
44 by benzo(a)pyrene. Reprod Toxicol 15: 479-486.
45 Rice, JE; Makowski, GS; Hosted, TJ; Lavoie, EJ. (1985). Methylene-bridged bay region chrysene and
46 phenanthrene derivatives and their keto-analogs: mutagenicity in Salmonella typhimurium and
47 tumor-initiating activity on mouse skin. Cancer Lett 27: 199-206.
48 http://dx.doi.org/10.1016/0304-3835(84)90119-8
This document is a draft for review purposes only and does not constitute Agency policy.
R-22 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Robinson, M; Laurie, RD; Bull, RJ; Stober, JA. (1987). Carcinogenic effects in A/J mice of particulate of a
2 coal tar paint used in potable water systems. Cancer Lett 34: 49-54.
3 http://dx.doi.org/10.1016/0304-3835(87)90072-3
4 Robles. R; Morita. Y; Mann. KK; Perez. Gl; Yang. S; Matikainen. T; Sherr. DH; Tilly. JL (2000). The aryl
5 hydrocarbon receptor, a basic helix-loop-helix transcription factor of the PAS gene family, is
6 required for normal ovarian germ cell dynamics in the mouse. Endocrinology 141: 450-453.
7 Rodriguez, JW; Kirlin, WG; Wirsiy, YG; Matheravidathu, S; Hodge, TW; Urso, P. (1999). Maternal
8 exposure to benzo[a]pyrene alters development of T lymphocytes in offspring.
9 Immunopharmacol Immunotoxicol 21: 379-396. http://dx.doi.org/10.3109/08923979909052769
10 Roe, FJ; Peto, R; Kearns, F; Bishop, D. (1970). The mechanism of carcinogenesis by the neutral fraction of
11 cigarette smoke condensate. Br J Cancer 24: 788-806.
12 Roe, FJ; Waters, MA. (1967). Induction of hepatoma in mice by carcinogens of the polycyclic
13 hydrocarbon type. Nature 214: 299-300.
14 Roelofzen, JH; Aben, KK; Oldenhof, UT; Coenraads, PJ; Alkemade, HA; van de Kerkhof, PC; van der Valk,
15 PG; Kiemeney, LA. (2010). No increased risk of cancer after coal tar treatment in patients with
16 psoriasis or eczema. J Invest Dermatol 130: 953-961. http://dx.doi.org/10.1038/iid.2009.389
17 Rogan, EG; Devanesan, PD; Ramakrishna, NV; Higginbotham, S; Padmavathi, NS; Chapman, K; Cavalieri,
18 EL; Jeong, H; Jankowiak, R; Small, GJ. (1993). Identification and quantitation of benzo[a]pyrene-
19 DNA adducts formed in mouse skin. Chem Res Toxicol 6: 356-363.
20 http://dx.doi.org/10.1021/tx00033a017
21 Rojas. M; Alexandrov. K; Cascorbi. I; Brockmoller. J; Likhachev. A; Pozharisski. K; Bouvier. G; Auburtin. G;
22 Mayer. L; Kopp-Schneider. A; Roots. I; Bartsch. H. (1998). High benzo[a]pyrene diol-epoxide DNA
23 adduct levels in lung and blood cells from individuals with combined CYP1A1 Mspl/Msp-
24 GSTM1*0/*0 genotypes. Pharmacogenetics 8: 109-118.
25 Rojas, M; Cascorbi, I; Alexandrov, K; Kriek, E; Auburtin, G; Mayer, L; Kopp-Schneider, A; Roots, I; Bartsch,
26 hL (2000). Modulation of benzo[a]pyrene diolepoxide-DNA adduct levels in human white blood
27 cells by CYP1A1, GSTM1 and GSTT1 polymorphism. Carcinogenesis 21: 35-41.
28 http://dx.doi.0rg/10.1093/carcin/21.l.35
29 Rojas, M; Godschalk, R; Alexandrov, K; Cascorbi, I; Kriek, E; Ostertag, J; Van Schooten, FJ; Bartsch, H.
30 (2001). Myeloperoxidase-463A variant reduces benzo[a]pyrene diol epoxide DNA adducts in
31 skin of coal tar treated patients. Carcinogenesis 22: 1015-1018.
32 http://dx.doi.0rg/10.1093/carcin/22.7.1015
33 Rojas, M; Marie, B; Vignaud, JM; Martinet, N; Siat, J; Grosdidier, G; Cascorbi, I; Alexandrov, K. (2004).
34 High DNA damage by benzo[a]pyrene 7,8-diol-9,10-epoxide in bronchial epithelial cells from
35 patients with lung cancer: comparison with lung parenchyma. Cancer Lett 207: 157-163.
36 http://dx.doi.0rg/10.1016/i.canlet.2003.ll.016
37 Rothman, KJ; Greenland, S. (1998). Modern epidemiology (2nd ed.). Philadelphia, PA: Lippincott,
38 Williams, &Wilkins.
39 Rowe. AA; O'Connor. TP. (2011). Assessment of water quality of runoff from sealed asphalt surfaces.
40 (EPA/600/R-10/178). Cincinnati, OH: U.S. Environmental Protection Agency.
41 Rubes. J; Rybar. R; Prinosilova. P; Veznik. Z; Chvatalova. I; Solansky. I; Sram. RJ. (2010). Genetic
42 polymorphisms influence the susceptibility of men to sperm DNA damage associated with
43 exposure to air pollution. Mutat Res 683: 9-15.
44 http://dx.doi.0rg/10.1016/i.mrfmmm.2009.09.010
45 Ruchirawat, M; Navasumrit, P; Settachan, D. (2010). Exposure to benzene in various susceptible
46 populations: co-exposures to 1,3-butadiene and PAHs and implications for carcinogenic risk.
47 Chem Biol Interact 184: 67-76. http://dx.doi.Org/10.1016/i.cbi.2009.12.026
This document is a draft for review purposes only and does not constitute Agency policy.
R-23 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Ruggeri, B; Dirado, M; Zhang, SY; Bauer, B; Goodrow, T; Kleinszanto, AJP. (1993). Benzofalpyrene-
2 induced murine skin tumors exhibit frequent and characteristic G to T mutations in the p53
3 gene. PNAS90: 1013-1017. http://dx.doi.Org/10.1073/pnas.90.3.1013
4 Sadeu, JC; Foster, WG. (2011). Cigarette smoke condensate exposure delays follicular development and
5 function in a stage-dependent manner. Fertil Steril 95: 2410-2417.
6 http://dx.doi.0rg/10.1016/i.fertnstert.2011.03.072
7 Saffiotti, U; Montesano, R; Sellakumar, AR; Kaufman, DG. (1972). Respiratory tract carcinogenesis
8 induced in hamsters by different dose levels of benzo[a]pyrene and ferric oxide. J Natl Cancer
9 Inst 49: 1199-1204.
10 Sakkas, D; Alvarez, JG. (2010). Sperm DNA fragmentation: mechanisms of origin, impact on reproductive
11 outcome, and analysis [Review]. Fertil Steril 93: 1027-1036.
12 http://dx.doi.0rg/10.1016/i.fertnstert.2009.10.046
13 Saunders, CR; Das, SK; Ramesh, A; Shockley, DC; Mukherjee, S. (2006). Benzo(a)pyrene-induced acute
14 neurotoxicity in the F-344 rat: role of oxidative stress. J Appl Toxicol 26: 427-438.
15 http://dx.doi.org/10.1002/iat.1157
16 Saunders, CR; Ramesh, A; Shockley, DC. (2002). Modulation of neurotoxic behavior in F-344 rats by
17 temporal disposition of benzo(a)pyrene. Toxicol Lett 129: 33-45.
18 Saunders, CR; Shockley, DC; Knuckles, ME. (2001). Behavioral effects induced by acute exposure to
19 benzo(a)pyrene in F-344 rats. Neurotox Res 3: 557-579. http://dx.doi.org/10.1007/BF03033211
20 Schmahl. D; Schmidt. KG; Habs. M. (1977). Syncarcinogenic action of polycyclic hydrocarbons in
21 automobile exhaust gas condensates. In U Mohr; D Schmahl; L Tomatis; W Davis (Eds.), Air
22 pollution and cancer in man (pp. 53-59). Lyon, France: IARC.
23 Schmidt. JV; Su. GH; Reddy. JK; Simon. MC; Bradfield. CA. (1996). Characterization of a murine Ahr null
24 allele: involvement of the Ah receptor in hepatic growth and development. PNAS 93: 6731-6736.
25 Schmidt, KG; Schmahl, D; Misfeld, J. (1973). Investigations on the carcinogenic burden by air pollution in
26 man. VI. Experimental investigations to determine a dose-response relationship and to estimate
27 a threshold dose of benzo(a)pyrene in the skin of two different mouse strains. Zentralbl
28 Bakteriol, Parasitenkd, Infektionskrankh Hyg, Abt 1: Orig, Reihe B 158: 62-68.
29 Schnizlein, CT; Munson, AE; Rhoades, RA. (1987). Immunomodulation of local and systemic immunity
30 after subchronic pulmonary exposure of mice to benzo(a)pyrene. International Journal of
31 Immunopharmacology 9: 99-106.
32 Schonland, SO; Zimmer, JK; Lopez-Benitez, CM; Widmann, T; Ramin, KD; Goronzy, JJ; Weyand, CM.
33 (2003). Homeostatic control of T-cell generation in neonates. Blood 102: 1428-1434.
34 http://dx.doi.org/10.1182/blood-2002-ll-3591
35 Secretan, B; Straif, K; Baan, R; Grosse, Y; El Ghissassi, F; Bouvard, V; Benbrahim-Tallaa, L; Guha, N;
36 Freeman, C; Galichet, L; Cogliano, V. (2009). A review of human carcinogens-Part E: tobacco,
37 areca nut, alcohol, coal smoke, and salted fish. Lancet Oncol 10: 1033-1034.
38 Seli. E; Gardner. DK; Schoolcraft. WB; Moffatt. O; Sakkas. D. (2004). Extent of nuclear DNA damage in
39 ejaculated spermatozoa impacts on blastocyst development after in vitro fertilization. Fertil
40 Steril 82: 378-383. http://dx.doi.Org/10.1016/i.fertnstert.2003.12.039
41 Shamsuddin. AK; Gan. R. (1988). Immunocytochemical localization of benzo(a)pyrene-DNA adducts in
42 human tissue. Hum Pathol 19: 309-315. http://dx.doi.org/10.1016/S0046-8177(88)80524-0
43 Sharovskava, J; Kobliakova, I; Solomatina, N; Kobliakov, V. (2006). Effect of some carcinogenic and non-
44 carcinogenic polycyclic aromatic hydrocarbons on gap junction intercellular communication in
45 hepatoma cell cultures. Eur J Cell Biol 85: 387-397. http://dx.doi.Org/10.1016/i.eicb.2005.ll.006
46 Shen. H; Spitz. MR; Qiao. Y; Guo. Z; Wang. LE; Bosken. CH; Amos. Cl; Wei. Q. (2003). Smoking, DNA
47 repair capacity and risk of nonsmall cell lung cancer. Int J Cancer 107: 84-88.
48 http://dx.doi.org/10.1002/iic.11346
This document is a draft for review purposes only and does not constitute Agency policy.
R-24 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Shen, YM; Troxel, AB; Vedantam, S; Penning, TM; Field, J. (2006). Comparison of p53 mutations induced
2 by PAH o-quinones with those caused by anti-benzo[a]pyrene diol epoxide in vitro: role of
3 reactive oxygen and biological selection. Chem Res Toxicol 19: 1441-1450.
4 http://dx.doi.org/10.1021/tx0601206
5 Sheng, L, iu; Ding, X; Ferguson, M; Mccallister, M; Rhoades, R; Maguire, M; Ramesh, A; Aschner, M;
6 Campbell, D; Levitt, P, at; Hood, DB. (2010). Prenatal Polycyclic Aromatic Hydrocarbon Exposure
7 Leads to Behavioral Deficits and Downregulation of Receptor Tyrosine Kinase, MET. Toxicol Sci
8 118: 625-634. http://dx.doi.org/10.1093/toxsci/kfq304
9 Shimizu, Y; Nakatsuru, Y; Ichinose, M; Takahashi, Y; Kume, H; Mimura, J; Fujii-Kuriyama, Y; Ishikawa, T.
10 (2000). Benzo[a]pyrene carcinogenicity is lost in mice lacking the aryl hydrocarbon receptor.
11 PNAS97: 779-782. http://dx.doi.Org/10.1073/pnas.97.2.779
12 Silverstone, AE; Cantor, H; Goldstein, G; Baltimore, D. (1976). Terminal deoxynucleotidyl transferase is
13 found in prothymocytes. J Exp Med 144: 543-548.
14 Sivak, A; Niemeier, R; Lynch, D; Beltis, K; Simon, S; Salomon, R; Latta, R; Belinky, B; Menzies, K; Lunsford,
15 A; Cooper, C; Ross, A; Bruner, R. (1997). Skin carcinogenicity of condensed asphalt roofing fumes
16 and their fractions following dermal application to mice. Cancer Lett 117: 113-123.
17 http://dx.doi.org/10.1016/S0304-3835(97)00214-0
18 Slaga, TJ; Gleason, GL; Mills, G; Ewald, L; Fu, PP; Lee, HM; Harvey, RG. (1980). Comparison of the skin
19 tumor-initiating activities of dihydrodiols and diol-epoxides of various polycyclic aromatic
20 hydrocarbons. Cancer Res 40: 1981-1984.
21 Slaga. TJ; Huberman. E; Selkirk. JK; Harvey. RG; Bracken. WM. (1978). Carcinogenicity and mutagenicity
22 of benz(a)anthracene diols and diol-epoxides. Cancer Res 38: 1699-1704.
23 Scares. SR; Melo. MA. (2008). Cigarette smoking and reproductive function [Review]. Curr Opin Obstet
24 Gynecol 20: 281-291. http://dx.doi.org/10.1097/GCO.Ob013e3282fc9cle
25 Soballe, PW; Montone, KT; Satyamoorthy, K; Nesbit, M; Herlyn, M. (1996). Carcinogenesis in human skin
26 grafted to SCID mice. Cancer Res 56: 757-764.
27 Spinelli, JJ; Band, PR; Svirchev, LM; Gallagher, RP. (1991). Mortality and cancer incidence in aluminum
28 reduction plant workers. J Occup Med 33: 1150-1155.
29 Spinelli, JJ; Demers, PA; Le, IMP; Friesen, MD; Lorenzi, MF; Fang, R; Gallagher, RP. (2006). Cancer risk in
30 aluminum reduction plant workers (Canada). Cancer Causes Control 17: 939-948.
31 http://dx.doi.org/10.1007/sl0552-006-0031-9
32 Steiner, PE. (1955). Carcinogenicity of multiple chemicals simultaneously administered. Cancer Res 15:
33 632-635.
34 Stephanou, P; Konstandi, M; Pappas, P; Marselos, M. (1998). Alterations in central monoaminergic
35 neurotransmission induced by polycyclic aromatic hydrocarbons in rats. European Journal of
36 Drug Metabolism and Pharmacokinetics 23: 475-481. http://dx.doi.org/10.1007/BF03189998
37 Stern, RS; Laird, N. (1994). The carcinogenic risk of treatments for severe psoriasis. Photochemotherapy
38 Follow-up Study. Cancer 73: 2759-2764.
39 Stern. RS; Liebman. EJ; Vakeva. L. (1998). Oral psoralen and ultraviolet-A light (PUVA) treatment of
40 psoriasis and persistent risk of nonmelanoma skin cancer. PUVA Follow-up Study. J Natl Cancer
41 Inst 90: 1278-1284.
42 Stern. RS; Zierler. S; Parrish. JA. (1980). Skin carcinoma in patients with psoriasis treated with topical tar
43 and artificial ultraviolet radiation. Lancet 1: 732-735.
44 Straif, K; Baan, R; Grosse, Y; Secretan, B; El Ghissassi, F; Cogliano, V. (2005). Carcinogenicity of polycyclic
45 aromatic hydrocarbons. Lancet Oncol 6: 931-932. http://dx.doi.org/10.1016/S1470-
46 2045(05)70458-7
47 Sugiyama, T. (1973). Chromosomal aberrations and carcinogenesis by various benz[a]anthracene
48 derivatives. Cancer Sci 64: 637-639.
This document is a draft for review purposes only and does not constitute Agency policy.
R-25 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Suter, M; Abramovici, A; Showalter, L; Hu, M; Shope, CD; Varner, M; Aagaard-Tillery, K. (2010). In utero
2 tobacco exposure epigenetically modifies placental CYP1A1 expression. Metabolism 59: 1481-
3 1490. http://dx.doi.0rg/10.1016/i.metabol.2010.01.013
4 Swartz, WJ; Mattison, DR. (1985). Benzo(a)pyrene inhibits ovulation in C57BL/6N mice. Anat Rec 212:
5 268-276. http://dx.doi.org/10.1002/ar.1092120308
6 Szczeklik, A; Szczeklik, J; Galuszka, Z; Musial, J; Kolarzyk, E; Targosz, D. (1994). Humoral
7 immunosuppression in men exposed to polycyclic aromatic hydrocarbons and related
8 carcinogens in polluted environments. Environ Health Perspect 102: 302-304.
9 Szymariska, K; Matos, E; Hung, RJ; Wunsch-Filho, V; Eluf-Neto, J; Menezes, A; Daudt, AW; Brennan, P;
10 Boffetta, P. (2010). Drinking of mate and the risk of cancers of the upper aerodigestive tract in
11 Latin America: a case-control study. Cancer Causes Control 21: 1799-1806.
12 http://dx.doi.org/10.1007/sl0552-010-9606-6
13 Takizawa, K; Yagi, H; Jerina, DM; Mattison, DR. (1984). Murine strain differences in ovotoxicity following
14 intraovarian injection with benzo(a)pyrene, (+)-(7R,8S)-oxide, (-)-(7R,8R)-dihydrodiol, or (+)-
15 (7R,8S)-diol-(9S,10R)-epoxide-2. Cancer Res 44: 2571-2576.
16 Talaska, G; Ginsburg, D; Ladow, K; Puga, A; Dalton, T; Warshawsky, D. (2006). Impact of Cypla2 or Ahr
17 gene knockout in mice: implications for biomonitoring studies. Toxicol Lett 162: 246-249.
18 http://dx.doi.0rg/10.1016/i.toxlet.2005.09.020
19 Talaska. G; Jaeger. M; Reilman. R; Collins. T; Warshawsky. D. (1996). Chronic, topical exposure to
20 benzo[a]pyrene induces relatively high steady-state levels of DNA adducts in target tissues and
21 alters kinetics of adduct loss. PNAS 93: 7789-7793.
22 Tamaki. A; Havashi. H; Nakajima. H; Takii. T; Katagiri. D; Miyazawa. K; Hirose. K; Onozaki. K. (2004).
23 Polycyclic aromatic hydrocarbon increases mRNA level for interleukin 1 beta in human
24 fibroblast-like synoviocyte line via aryl hydrocarbon receptor. Biol Pharm Bull 27: 407-410.
25 Tang, D; Li, TY; Liu, JJ; Chen, YH; Qu, L; Perera, F. (2006). PAH-DNA adducts in cord blood and fetal and
26 child development in a Chinese cohort. Environ Health Perspect 114: 1297-1300.
27 http://dx.doi.org/10.1289/ehp.8939
28 Tang, D; Warburton, D; Tannenbaum, SR; Skipper, P; Santella, RM; Cereijido, GS; Crawford, FG; Perera,
29 FP. (1999). Molecular and genetic damage from environmental tobacco smoke in young
30 children. Cancer Epidemiol Biomarkers Prev 8: 427-431.
31 Tang. PL; Li. TY; Liu. JJ; Zhou. ZJ; Yuan. T; Chen. YH; Rauh. VA; Xie. J; Perera. F. (2008). Effects of prenatal
32 exposure to coal-burning pollutants on children's development in China. Environ Health
33 Perspect 116: 674-679. http://dx.doi.org/10.1289/ehp.10471
34 Tang, Cf Xia, Y; Cheng, S; Tu, B. (2011). Modulation of behavior and glutamate receptor mRNA
35 expression in rats after sub-chronic administration of benzo(a)pyrene. Biomed Environ Sci 24:
36 408-414. http://dx.doi.org/10.3967/0895-3988.2011.04.012
37 Temple, L; Kawabata, TT; Munson, AE; White, KL. (1993). Comparison of ELISA and Plaque-Forming Cell
38 Assays for Measuring the Humoral Immune Response to SRBC in Rats and Mice Treated with
39 Benzo[a]pyrene or Cyclophosphamide. Toxicol Sci 21: 412-419.
40 http://dx.doi.0rg/10.1093/toxsci/21.4.412
41 Thirman. MJ; Albrecht. JH; Krueger. MA; Erickson. RR; Cherwitz. PL; Park. SS; Gelboin. HV; Holtzman. JL.
42 (1994). Induction of cytochrome CYPIA1 and formation of toxic metabolites of benzo[a]pyrene
43 by rat aorta: a possible role in atherogenesis. PNAS 91: 5397-5401.
44 Thyssen, J; Althoff, J; Kimmerle, G; Mohr, U. (1981). Inhalation studies with benzo[a]pyrene in Syrian
45 golden hamsters. J Natl Cancer Inst 66: 575-577.
46 Topping, DC; Martin, DH; Nettesheim, P. (1981). Determination of cocarcinogenic activity of
47 benzo[e]pyrene for respiratory tract mucosa. Cancer Lett 11: 315-321.
48 http://dx.doi.org/10.1016/0304-3835(81)90097-5
This document is a draft for review purposes only and does not constitute Agency policy.
R-26 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Torinuki, W; Tagami, H. (1988). Incidence of skin cancer in Japanese psoriatic patients treated with
2 either methoxsalen phototherapy, Goeckerman regimen, or both therapies: A 10-year follow-up
3 study. J Am Acad Dermatol 18: 1278-1281.
4 Tornqvist, S; Norell, S; Ahlbom, A; Knave, B. (1986). Cancer in the electric power industry. Br J Ind Med
5 43: 212-213.
6 Triolo, AJ; Aponte, GE; Herr, PL (1977). Induction of aryl hydrocarbon hydroxylase and forestomach
7 tumors ben benzo(a)pyrene. Cancer Res 37: 3018-3021.
8 U.S. Congress. (2011). Consolidated Appropriations Act, 2012. (Pub. L No. 112-74; 125 STAT. 786). 112th
9 U.S. Congress. http://www.gpo.gov/fdsys/pkg/PLAW-112publ74/pdf/PLAW-112publ74.pdf
10 U.S. EPA (U.S. Environmental Protection Agency). (1986a). Guidelines for mutagenicity risk assessment.
11 (EPA/630/R-98/003). Washington, DC: U.S. Environmental Protection Agency, Risk Assessment
12 Forum, http://www.epa.gov/iris/backgrd.html
13 U.S. EPA (U.S. Environmental Protection Agency). (1986b). Guidelines for the health risk assessment of
14 chemical mixtures. Fed Reg 51: 34014-34025.
15 U.S. EPA (U.S. Environmental Protection Agency). (1986c). Guidelines for the health risk assessment of
16 chemical mixtures. (EPA/630/R-98/002). Washington, DC: U.S. Environmental Protection
17 Agency, Risk Assessment Forum. http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=22567
18 U.S. EPA (U.S. Environmental Protection Agency). (1988). Recommendations for and documentation of
19 biological values for use in risk assessment. (EPA/600/6-87/008). Cincinnati, OH: U.S.
20 Environmental Protection Agency, National Center for Environmental Assessment.
21 http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=34855
22 U.S. EPA (U.S. Environmental Protection Agency). (1990). Development of a dose-response model for
23 inhaled B(a)P [EPA Report]. (EPA-600/R-92-135).
24 http://www.ntis.gov/search/product.aspx?ABBR=PB93161016
25 U.S. EPA (U.S. Environmental Protection Agency). (1991a). Dose-response analysis of ingested
26 Benzo(a)pyrene (CAS No. 50-32-8) [EPA Report]. (EPA600R92045).
27 http://www.ntis.gov/search/product.aspx?ABBR=PB93167484
28 U.S. EPA (U.S. Environmental Protection Agency). (1991b). Drinking water criteria document for
29 polycyclic aromatic hydrocarbons (PAHs). (ECAOCIND010).
30 http://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=9100CSXZ.txt
31 U.S. EPA (U.S. Environmental Protection Agency). (1991c). Guidelines for developmental toxicity risk
32 assessment. (EPA/600/FR-91/001). Washington, DC: U.S. Environmental Protection Agency, Risk
33 Assessment Forum, http://www.epa.gov/raf/publications/guidelines-dev-toxicity-risk-
34 assessment.htm
35 U.S. EPA (U.S. Environmental Protection Agency). (1992). Draft report: A cross-species scaling factor for
36 carcinogen risk assessment based on equivalence of mg/kg(3/4)/day. Fed Reg 57: 24151-24173.
37 U.S. EPA (U.S. Environmental Protection Agency). (1994). Methods for derivation of inhalation reference
38 concentrations and application of inhalation dosimetry. (EPA/600/8-90/066F). Research Triangle
39 Park, NC: U.S. Environmental Protection Agency, Environmental Criteria and Assessment Office.
40 http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=71993
41 U.S. EPA (U.S. Environmental Protection Agency). (1996). Guidelines for reproductive toxicity risk
42 assessment. (EPA/630/R-96/009). Washington, DC: U.S. Environmental Protection Agency, Risk
43 Assessment Forum. http://www.epa.gov/raf/publications/pdfs/REPRO51.PDF
44 U.S. EPA (U.S. Environmental Protection Agency). (1998). Guidelines for neurotoxicity risk assessment.
45 (EPA/630/R-95/001F). Washington, DC: U.S. Environmental Protection Agency, Risk Assessment
46 Forum. http://www.epa.gov/raf/publications/pdfs/NEUROTOX.PDF
This document is a draft for review purposes only and does not constitute Agency policy.
R-27 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 U.S. EPA (U.S. Environmental Protection Agency). (2000a). Science policy council handbook: risk
2 characterization. (EPA/100/B-00/002). Washington, D.C.: U.S. Environmental Protection Agency,
3 Office of Science Policy, http://www.epa.gov/osa/spc/pdfs/rchandbk.pdf
4 U.S. EPA (U.S. Environmental Protection Agency). (2000b). Supplementary guidance for conducting
5 health risk assessment of chemical mixtures. (EPA/630/R-00/002). Washington, DC: U.S.
6 Environmental Protection Agency, Risk Assessment Forum.
7 http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=20533
8 U.S. EPA (U.S. Environmental Protection Agency). (2002). A review of the reference dose and reference
9 concentration processes. (EPA/630/P-02/002F). Washington, DC: U.S. Environmental Protection
10 Agency, Risk Assessment Forum. http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=51717
11 U.S. EPA (U.S. Environmental Protection Agency). (2004). Risk Assessment Guidance for Superfund
12 (RAGS), Volume I: Human health evaluation manual, (part E: Supplemental guidance for dermal
13 risk assessment): Final. (EPA/540/R/99/005). Washington, DC.
14 http://www.epa.gov/oswer/riskassessment/ragse/index.htm
15 U.S. EPA (U.S. Environmental Protection Agency). (2005a). Guidelines for carcinogen risk assessment.
16 (EPA/630/P-03/001F). Washington, DC: U.S. Environmental Protection Agency, Risk Assessment
17 Forum, http://www.epa.gov/cancerguidelines/
18 U.S. EPA (U.S. Environmental Protection Agency). (2005b). Supplemental guidance for assessing
19 susceptibility from early-life exposure to carcinogens. In US Environmental Protection Agency,
20 Risk Assessment Forum (pp. 1125-1133). (EPA/630/R-03/003F). Washington, DC: U.S.
21 Environmental Protection Agency, Risk Assessment Forum.
22 http://www.epa.gov/cancerguidelines/guidelines-carcinogen-supplement.htm
23 U.S. EPA (U.S. Environmental Protection Agency). (2006a). Approaches for the application of
24 physiologically based pharmacokinetic (PBPK) models and supporting data in risk assessment
25 (Final Report) [EPA Report]. (EPA/600/R-05/043F). Washington, DC: U.S. Environmental
26 Protection Agency, National Center for Environmental assessment.
27 http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=157668
28 U.S. EPA (U.S. Environmental Protection Agency). (2006b). A framework for assessing health risk of
29 environmental exposures to children. (EPA/600/R-05/093F). Washington, DC: U.S.
30 Environmental Protection Agency, National Center for Environmental Assessment.
31 http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=158363
32 U.S. EPA (U.S. Environmental Protection Agency). (2010a). Consumer factsheet on benzo(a)pyrene:
33 Ground water and drinking water [Fact Sheet].
34 http://water.epa.gov/drink/contaminants/basicinformation/historical/upload/Archived-
35 Consumer-Factsheet-on-Benzo.pdf
36 U.S. EPA (U.S. Environmental Protection Agency). (2010b). Integrated science assessment for carbon
37 monoxide [EPA Report]. (EPA/600/R-09/019F). Research Triangle Park, NC.
38 http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=218686
39 U.S. EPA (U.S. Environmental Protection Agency). (2010c). Revised draft: Multistage Weibull time-to-
40 tumor model in EPA's benchmark dose software (BMDS): Methodology description. Washington,
41 DC.
42 U.S. EPA (U.S. Environmental Protection Agency). (2011). Recommended use of body weight 3/4 as the
43 default method in derivation of the oral reference dose. (EPA/100/R11/0001). Washington, DC:
44 U.S. Environmental Protection Agency, Risk Assessment Forum.
45 http://www.epa.gov/raf/publications/interspecies-extrapolation.htm
46 U.S. EPA (U.S. Environmental Protection Agency). (2012a). Advances in inhalation gas dosimetry for
47 derivation of a reference concentration (RFC) and use in risk assessment. (EPA/600/R-12/044).
48 Washington, DC. http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=244650
This document is a draft for review purposes only and does not constitute Agency policy.
R-28 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 U.S. EPA (U.S. Environmental Protection Agency). (2012b). Benchmark Dose Software (BMDS). Version
2 2.2. Retrieved from http://www.epa.gov/NCEA/bmds/index.html
3 U.S. EPA (U.S. Environmental Protection Agency). (2012c). Benchmark dose technical guidance.
4 (EPA/100/R-12/001). Washington, DC: Risk Assessment Forum.
5 http://www.epa.gov/raf/publications/pdfs/benchmark dose guidance.pdf
6 U.S. EPA (U.S. Environmental Protection Agency). (2012d). EPA announces NAS' review of IRIS
7 Assessment development process. Available online at
8 http://vosemite.epa.gOV/opa/admpress.nsf/0/lce2a7875daf093485257aOOOQ54df547OpenDoc
9 ument
10 U.S. EPA (U.S. Environmental Protection Agency). (2014). EPAs Integrated Risk Information System:
11 Assessment development process [EPA Report]. Washington, DC.
12 http://epa.gov/iris/process.htm
13 Urano, K; Katakai, Y; Tokuda, Y; Ueyama, Y; Nomura, T; Yamamoto, S. (1995). Failure of genotoxic
14 carcinogens to produce tumors in human skin xenografts transplanted to SCID mice.
15 Carcinogenesis 16: 2223-2226. http://dx.doi.Org/10.1093/carcin/16.9.2223
16 Urso, P; Gengozian, N. (1980). Depressed humoral immunity and increased tumor incidence in mice
17 following in utero exposure to benzo[alpha]pyrene. J Toxicol Environ Health 6: 569-576.
18 http://dx.doi.org/10.1080/15287398009529874
19 Urso. P; Gengozian. N. (1982). Alterations in the humoral immune response and tumor frequencies in
20 mice exposed to benzo[a]pyrene and X-rays before or after birth. J Toxicol Environ Health 10:
21 817-835. http://dx.doi.org/10.1080/15287398209530297
22 Urso. P; Gengozian. N. (1984). Subnormal expression of cell-mediated and humoral immune responses
23 in progeny disposed toward a high incidence of tumors after in utero exposure to
24 benzo[a]pyrene. J Toxicol Environ Health 14: 569-584.
25 http://dx.doi.org/10.1080/15287398409530606
26 Urso, P; Johnson, RA. (1987). Early changes in T lymphocytes and subsets of mouse progeny defective as
27 adults in controlling growth of a syngeneic tumor after in utero insult with benzo(a)pyrene.
28 Immunopharmacology 14: 1-10.
29 Urso, P; Johnson, RA. (1988). Quantitative and functional change in T cells of primiparous mice following
30 injection of benzo(a)pyrene at the second trimester of pregnancy. Immunopharmacol
31 Immunotoxicol 10: 195-217. http://dx.doi.org/10.3109/08923978809014333
32 Urso, P; Kramer, MK. (2008). The role of adherent cells in the immunosuppressed state of mouse
33 progeny transplacentally exposed to benzo(alpha)pyrene. In Vitro Cell Dev Biol Anim 44: 273-
34 279. http://dx.doi.org/10.1007/sll626-008-9112-2
35 Urso, P; Ryan, MC; Bennett, JS. (1988). Changes in peripheral blood cells in mice after injection with
36 benzo-a-pyrene during pregnancy. Immunopharmacol Immunotoxicol 10: 179-194.
37 http://dx.doi.org/10.3109/08923978809014332
38 Urso. P; Zhang. W; Cobb. JR. (1992). Immunological consequences from exposure to benzo(a)pyrene
39 during pregnancy. Scand J Immunol Suppl 11: 203-206.
40 van den Berg. FA; Baken. KA; Vermeulen. JP; Gremmer. ER; van Steeg. H; van Loveren. H. (2005). Use of
41 the local lymph node assay in assessment of immune function. Toxicology 211: 107-114.
42 http://dx.doi.0rg/10.1016/i.tox.2005.03.003
43 van Lipzig, MM; Vermeulen, IMP; Gusinu, R; Legler, J; Frank, H; Seidel, A; Meerman, JH. (2005). Formation
44 of estrogenic metabolites of benzo[a]pyrene and chrysene by cytochrome P450 activity and
45 their combined and supra-maximal estrogenic activity. Environ Toxicol Pharmacol 19: 41-55.
46 http://dx.doi.0rg/10.1016/i.etap.2004.03.010
This document is a draft for review purposes only and does not constitute Agency policy.
R-29 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Van Metre, PC; Mahler, BJ. (2010). Contribution of PAHs from coal-tar pavement sealcoat and other
2 sources to 40 U.S. lakes. Sci Total Environ 409: 334-344.
3 http://dx.doi.0rg/10.1016/i.scitotenv.2010.08.014
4 Varghese, SK; Gangamma, S. (2009). Particle deposition in human respiratory system: deposition of
5 concentrated hygroscopic aerosols. Inhal Toxicol 21: 619-630.
6 http://dx.doi.org/10.1080/08958370802380792
7 Vesselinovitch, SD; Kyriazis, AP; Mihailovich, N; Rao, KV. (1975). Conditions modifying development of
8 tumors in mice at various sites by benzo(a)pyrene. Cancer Res 35: 2948-2953.
9 Vesselinovitch, SD; Rao, KV; Mihailovich, N. (1979). Neoplastic response of mouse tissues during
10 perinatal age periods and its significance in chemical carcinogenesis [Review]. Natl Cane Inst
11 Monogr239-250.
12 Vineis, P; Anttila, S; Benhamou, S; Spinola, M; Hirvonen, A; Kiyohara, C; Garte, SJ; Puntoni, R; Rannug, A;
13 Strange, RC; Taioli, E. (2007). Evidence of gene gene interactions in lung carcinogenesis in a large
14 pooled analysis [Review]. Carcinogenesis 28: 1902-1905.
15 http://dx.doi.org/10.1093/carcin/bgm039
16 Vineis, P; Perera, F. (2007). Molecular epidemiology and biomarkers in etiologic cancer research: the
17 new in light of the old [Review]. Cancer Epidemiol Biomarkers Prev 16: 1954-1965.
18 http://dx.doi.org/10.1158/1055-9965.EPI-07-0457
19 Voelter-Mahlknecht. S; Scheriau. R; Zwahr. G; Koch. B; Escobar Pinzon. LC; Drexler. H; Letzel. S. (2007).
20 Skin tumors among employees of a tar refinery: the current data and their implications. Int Arch
21 Occup Environ Health 80: 485-495. http://dx.doi.org/10.1007/s00420-006-0155-v
22 Vondracek. J; Kozubfk. A; Machala. M. (2002). Modulation of estrogen receptor-dependent reporter
23 construct activation and GO/Gl-S-phase transition by polycyclic aromatic hydrocarbons in
24 human breast carcinoma MCF-7 cells. Toxicol Sci 70: 193-201.
25 Wang, F; He, Y; Guo, H; Li, J; Yang, Y; Wu, Z; Zheng, H; Wu, T. (2010). Genetic variants of nucleotide
26 excision repair genes are associated with DNA damage in coke oven workers. Cancer Epidemiol
27 Biomarkers Prev 19: 211-218. http://dx.doi.org/10.1158/1055-9965.EPI-09-0270
28 Wattenberg, LW. (1974). Inhibition of carcinogenic and toxic effects of polycyclic hydrocarbons by
29 several sulfur-containing compounds. J Natl Cancer Inst 52: 1583-1587.
30 Waylen, AL; Metwally, M; Jones, GL; Wilkinson, AJ; Ledger, WL. (2009). Effects of cigarette smoking upon
31 clinical outcomes of assisted reproduction: a meta-analysis [Review]. Hum Reprod Update 15:
32 31-44. http://dx.doi.org/10.1093/humupd/dmn046
33 Wei, Q; Cheng, L; Amos, Cl; Wang, LE; Guo, Z; Hong, WK; Spitz, MR. (2000). Repair of tobacco
34 carcinogen-induced DNA adducts and lung cancer risk: a molecular epidemiologic study. J Natl
35 Cancer Inst 92: 1764-1772.
36 Wei. SJ; Chang. RL; Merkler. KA; Gwynne. M; Cui. XX; Murthy. B; Huang. MT; Xie. JG; Lu. YP; Lou. YR;
37 Jerina, DM; Conney, AH. (1999). Dose-dependent mutation profile in the c-Ha-ras proto-
38 oncogene of skin tumors in mice initiated with benzo[a]pyrene. Carcinogenesis 20: 1689-1696.
39 Weinberg. CR; Wilcox. AJ; Baird. DP. (1989). Reduced fecundability in women with prenatal exposure to
40 cigarette smoking. Am J Epidemiol 129: 1072-1078.
41 Wester. PW; Kroes. R. (1988). Forestomach carcinogens: pathology and relevance to man [Review].
42 Toxicol Pathol 16: 165-171.
43 Wester, RC; Maibach, HI; Bucks, DA; Sedik, L; Melendres, J; Liao, C; Dizio, S. (1990). Percutaneous
44 absorption of [14C]DDT and [14C]benzo[a]pyrene from soil. Fundam Appl Toxicol 15: 510-516.
45 Weyand, EH; Chen, YC; Wu, Y; Koganti, A; Dunsford, HA; Rodriguez, LV. (1995). Differences in the
46 tumorigenic activity of a pure hydrocarbon and a complex mixture following ingestion:
47 Benzo[a]pyrene vs manufactured gas plant residue. Chem Res Toxicol 8: 949-954.
48 http://dx.doi.org/10.1021/tx00049a008
This document is a draft for review purposes only and does not constitute Agency policy.
R-30 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Weyand, EH; He, ZM; Ghodrati, F; Wu, Y; Marshall, MV; Lavoie, EJ. (1992). Effect of fluorine substitution
2 on benzo[j]fluoranthene genotoxicity. Chem Biol Interact 84: 37-53.
3 Weyand, EH; Lavoie, EJ. (1988). Comparison of PAH DNA adduct formation and tumor initiating activity
4 in newborn mice. Proc Am Assoc Cancer Res 29: 98.
5 White, KL; Lysy, HH; Holsapple, MP. (1985). Immunosuppression by polycyclic aromatic hydrocarbons: a
6 structure-activity relationship in B6C3F1 and DBA/2 mice. Immunopharmacology 9: 155-164.
7 WHO (World Health Organization). (2012). Guidance for immunotoxicity risk assessment for chemicals.
8 (Harmonization Project Document No. 10). Geneva, Switzerland.
9 http://www.inchem.org/documents/harmproi/harmproi/harmproilO.pdf
10 Whyatt, RM; Santella, RM; Jedrychowski, W; Garte, SJ; Bell, DA; Ottman, R; Gladek-Yarborough, A;
11 Cosma, G; Young, TL; Cooper, TB; Randall, MC; Manchester, DK; Perera, FP. (1998). Relationship
12 between ambient air pollution and DNA damage in Polish mothers and newborns. Environ
13 Health Perspect 106: 821-826.
14 Wilson, JS; Holland, LM. (1988). Periodic response difference in mouse epidermis chronically exposed to
15 crude-oils or BaP: Males vs. females. Toxicology 50: 83-94. http://dx.doi.org/10.1016/0300-
16 483X(88)90123-0
17 Winker, N; Tuschl, H; Kovac, R; Weber, E. (1997). Immunological investigations in a group of workers
18 exposed to various levels of polycyclic aromatic hydrocarbons. J Appl Toxicol 17: 23-29.
19 Wislocki. PG: Bagan. ES: Lu. AYH: Doolev. KL: Fu. PP: Han-Hsu. H: Beland. FA: Kadlubar. FF. (1986).
20 Tumorigenicity of nitrated derivatives of pyrene, benz[a]anthracene, chrysene, and
21 benzo[a]pyrene in the newborn mouse assay. Carcinogenesis 7: 1317-1322.
22 http://dx.doi.0rg/10.1093/carcin/7.8.1317
23 Wolff. RK: Griffith. WC: Henderson. RF: Hahn. FF: Harkema. J. R.: Rebar. AH: Eidson. AF: Mcclellan. RO.
24 (1989). Effects of repeated inhalation exposures to 1-nitropyrene, benzo(a)pyrene, Ga2O3
25 particles, and SO2 alone and in combinations on particle clearance, bronchoalveolar lavage fluid
26 composition, and histopathology. J Toxicol Environ Health 27: 123-138.
27 http://dx.doi.org/10.1080/15287398909531283
28 Wood, AW; Levin, W; Chang, RL; Huang, MT; Ryan, DE; Thomas, PE; Lehr, RE; Kumar, S; Koreeda, M;
29 Akagi, H; Ittah, Y; Dansette, P; Yagi, H; Jerina, DM; Conney, AH. (1980). Mutagenicity and tumor-
30 initiating activity of cyclopenta(c,d)pyrene and structurally related compounds. Cancer Res 40:
31 642-649.
32 Wormley, DP; Chirwa, S; Nayyar, T; Wu, J; Johnson, S; Brown, LA; Harris, E; Hood, DB. (2004). Inhaled
33 benzo(a)pyrene impairs long-term potentiation in the Fl generation rat dentate gyrus. Cell Mol
34 Biol (Noisy-le-grand) 50: 715-721.
35 Wu, J; Hou, H; Ritz, B; Chen, Y. (2010). Exposure to polycyclic aromatic hydrocarbons and missed
36 abortion in early pregnancy in a Chinese population. Sci Total Environ 408: 2312-2318.
37 http://dx.doi.0rg/10.1016/i.scitotenv.2010.02.028
38 Wu. J; Ramesh. A; Navyar. T; Hood. DB. (2003a). Assessment of metabolites and AhR and CYP1A1 mRNA
39 expression subsequent to prenatal exposure to inhaled benzo(a)pyrene. Int J Dev Neurosci 21:
40 333-346.
41 Wu. MT: Pan. CH: Huang. YL: Tsai. PJ: Chen. CJ: Wu. TN. (2003b). Urinary excretion of 8-hydroxy-2-
42 deoxyguanosine and 1-hydroxypyrene in coke-oven workers. Environ Mol Mutagen 42: 98-105.
43 http://dx.doi.org/10.1002/em.10176
44 Wynder, EL; Fritz, L; Furth, N. (1957). Effect of concentration of benzopyrene in skin carcinogenesis. J
45 Natl Cancer Inst 19: 361-370.
46 Wynder, EL; Hoffmann, D. (1959). A study of tobacco carcinogenesis. VII. The role of higher polycyclic
47 hydrocarbons. Cancer 12: 1079-1086. http://dx.doi.org/10.1002/1097-
48 0142(195911/12)12:6<1079::AID-CNCR2820120604>3.0.CO;2-I
This document is a draft for review purposes only and does not constitute Agency policy.
R-31 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Xia. Y; Cheng. S; He. J; Liu. X; Tang. Y; Yuan. H; He. L; Lu. T; Tu. B; Wang. Y. (2011). Effects of subchronic
2 exposure to benzo[a]pyrene (B[a]P) on learning and memory, and neurotransmitters in male
3 Sprague-Dawley rat. Neurotoxicology 32: 188-198.
4 http://dx.doi.0rg/10.1016/i.neuro.2010.12.015
5 Xia. Y; Zhu. P; Han. Y; Lu. C; Wang. S; Gu. A; Fu. G; Zhao. R; Song. L; Wang. X. (2009). Urinary metabolites
6 of polycyclic aromatic hydrocarbons in relation to idiopathic male infertility. Hum Reprod 24:
7 1067-1074. http://dx.doi.org/10.1093/humrep/dep006
8 Xu. C; Chen. J; Qiu. Z; Zhao. Q; Luo. J; Yang. L; Zeng. H; Huang. Y; Zhang. L; Cao. J; Shu. W. (2010).
9 Ovotoxicity and PPAR-mediated aromatase downregulation in female Sprague-Dawley rats
10 following combined oral exposure to benzo[a]pyrene and di-(2-ethylhexyl) phthalate. Toxicol
11 Lett 199: 323-332. http://dx.doi.Org/10.1016/i.toxlet.2010.09.015
12 Xu, D; Penning, TM; Blair, IA; Harvey, RG. (2009). Synthesis of phenol and quinone metabolites of
13 benzo[a]pyrene, a carcinogenic component of tobacco smoke implicated in lung cancer. J Org
14 Chem 74: 597-604. http://dx.doi.org/10.1021/io801864m
15 Xu, GB; Yu, CP. (1985). Theoretical lung deposition of hygroscopic NaCI aerosols. Aerosol Sci Technol 4:
16 455-461. http://dx.doi.org/10.1080/02786828508959070
17 Xu, Z; Brown, LM; Pan, GW; Liu, TF; Gao, GS; Stone, BJ; Cao, RM; Guan, DX; Sheng, JH; Yan, ZS; Dosemeci,
18 M; Fraumeni, JF; Blot, WJ. (1996). Cancer risks among iron and steel workers in Anshan, China,
19 Part II: Case-control studies of lung and stomach cancer. Am J Ind Med 30: 7-15.
20 http://dx.doi.org/10.1002/(SICI)1097-0274(199607)30:l<:7::AID-AJIM2>:3.0.CO:2-#
21 Xue. W; Warshawsky. D. (2005). Metabolic activation of polycyclic and heterocyclic aromatic
22 hydrocarbons and DNA damage: a review [Review]. Toxicol Appl Pharmacol 206: 73-93.
23 http://dx.doi.0rg/10.1016/i.taap.2004.ll.006
24 Ye, X; Skjaerven, R; Basso, O; Baird, DP; Eggesbo, M; Cupul Uicab, LA; Haug, K; Longnecker, MP. (2010).
25 In utero exposure to tobacco smoke and subsequent reduced fertility in females. Hum Reprod
26 25: 2901-2906. http://dx.doi.org/10.1093/humrep/deq235
27 Young, C; Cherrie, J; Van Tongeren, M; Fortunato, L; Hutchings, S; Rushton, L. (2012). The burden of
28 occupational cancer in Great Britain: Non-melanoma skin cancer. Sudbury, Suffolk: Health and
29 Safety Executive (HSE). http://www.hse.gov.uk/research/rrhtm/rr928.htm
30 Yu, D; Berlin, JA; Penning, TM; Field, J. (2002). Reactive oxygen species generated by PAH o-quinones
31 cause change-in-function mutations in p53. Chem Res Toxicol 15: 832-842.
32 Zaccaria, KJ; McClure, PR. (2013). Using immunotoxicity information to improve cancer risk assessment
33 for polycyclic aromatic hydrocarbon mixtures [Review]. Int J Toxicol 32: 236-250.
34 http://dx.doi.org/10.1177/1091581813492829
35 Zandi-Nejad, K; Luyckx, VA; Brenner, BM. (2006). Adult hypertension and kidney disease - The role of
36 fetal programming [Review]. Hypertension 47: 502-508.
37 http://dx.doi.org/10.1161/01.HYP.0000198544.09909.la
38 Zenzes. MT; Puy. LA; Bielecki. R. (1998). Immunodetection of benzo[a]pyrene adducts in ovarian cells of
39 women exposed to cigarette smoke. Mol Hum Reprod 4: 159-165.
40 http://dx.doi.0rg/10.1093/molehr/4.2.159
41 Zenzes. MT; Puv. LA; Bielecki. R; Reed. TE. (1999). Detection of benzo[a]pyrene diol epoxide-DNA
42 adducts in embryos from smoking couples: evidence for transmission by spermatozoa. Mol Hum
43 Reprod 5: 125-131.
44 Zhang, HM; Nie, JS; Li, X; Niu, Q. (2012). Characteristic analysis of peripheral blood mononuclear cell
45 apoptosis in coke oven workers. J Occup Health 54: 44-50. http://dx.doi.org/10.1539/ioh.ll-
46 0155-OA
This document is a draft for review purposes only and does not constitute Agency policy.
R-32 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Zhang, Y; Ramos, KS. (1997). The induction of proliferative vascular smooth muscle cell phenotypes by
2 benzo[a]pyrene does not involve mutational activation of ras genes. Mutat Res 373: 285-292.
3 http://dx.doi.org/10.1016/S0027-5107(96)00213-8
4 Zhang, YJ; Li, Y; Deleo, VA; Santella, RM. (1990). Detection of DNA adducts in skin biopsies of coal tar-
5 treated psoriasis patients: immunofluorescence and 32P postlabeling. Skin Pharmacol Physiol 3:
6 171-179. http://dx.doi.org/10.1159/000210867
7 Zheng, SJ; Tian, HJ; Cao, J; Gao, YQ. (2010). Exposure to di(n-butyl)phthalate and benzo(a)pyrene alters
8 IL-1(3 secretion and subset expression of testicular macrophages, resulting in decreased
9 testosterone production in rats. Toxicol Appl Pharmacol 248: 28-37.
10 http://dx.doi.0rg/10.1016/i.taap.2010.07.008
11
This document is a draft for review purposes only and does not constitute Agency policy.
R-33 DRAFT—DO NOT CITE OR QUOTE
------- |